Instrukciya F5-A E H Ver.4.4 (En)
Instrukciya F5-A E H Ver.4.4 (En)
Instrukciya F5-A E H Ver.4.4 (En)
Mat.No. Rev.
00F5AEA-K440 1A
Table of contents
2. Summary Features, operating conditions and intended use of the KEB COMBIVERT.
The basic operation of the KEB COMBIVERT like password input, para-
4. Operation meter and set selection.
Selection of Operating
5. Operating modes of the KEB COMBIVERT
Mode
Gives support with regard to the initial start-up and shows possibilities
6. Initial Start-up and techniques for the optimization of the drive.
To make the programming easier all inverter functions and the parame-
7. Functions ters belonging to it are comprised in this chapter.
6.2.1.2 Vector controlled operation with encoder feedback without motor model............. 6.2 - 6
6.2.1.3 Vector controlled operation with encoder feedback with motor model.................. 6.2 - 8
6.2.1.4 Start-up F5H-M (ASCL/ vector-controlled without encoder feedback with motor
model)................................................................................................................. 6.2 - 11
6.2.2 Start-up of a synchronous motor.................................................................... 6.2 - 15
6.2.2.1 Start-up F5A-S.................................................................................................... 6.2 - 16
6.2.2.2 Start-up F5E-S (SCL).......................................................................................... 6.2 - 18
7. Functions................................................................................................7.1 - 3
7.1 Operating and appliance data....................................................................... 7.1 - 3
7.1.1 Overview of the ru-Parameters........................................................................ 7.1 - 3
7.1.2 Overview of the In-Parameters........................................................................ 7.1 - 4
7.1.3 Overview of the Sy-Parameters....................................................................... 7.1 - 4
7.1.4 Explanation to parameter description.............................................................. 7.1 - 5
7.1.5 Description of the ru-Parameters..................................................................... 7.1 - 5
7.1.6 Description of the In-Parameters................................................................... 7.1 - 16
7.1.7 Description of the SY (System) - Parameters................................................ 7.1 - 20
7.2 Analog in- and outputs.................................................................................. 7.2 - 3
7.2.1 Brief description analog inputs......................................................................... 7.2 - 3
7.2.2 Interface selection ........................................................................................... 7.2 - 4
7.2.2.1 AN1 / AN2 (An.00, An.10)...................................................................................... 7.2 - 4
7.2.2.2 AN3 (An.20)........................................................................................................... 7.2 - 5
7.2.3 Noise filter (An.01, An.11, An.21)..................................................................... 7.2 - 5
7.2.4 Save mode (An.02, An.12, An.22).................................................................... 7.2 - 5
7.2.4.1 Input selection (An.03, An.13, An.23).................................................................... 7.2 - 6
7.2.5 Zero clamp (An.04, An.14, An.24).................................................................... 7.2 - 7
7.2.6 Amplifier of the input characteristic (An.05...07, An.15...17, An.25...27).......... 7.2 - 8
7.2.7 Lower- and upper limit (An.08, An.09, An.18, An.19, An.28, An.29)............... 7.2 - 10
7.2.8 Selection REFinput / AUX function (An.30).................................................... 7.2 - 10
7.2.9 Brief description analog outputs..................................................................... 7.2 - 11
7.2.10 Output signals ............................................................................................... 7.2 - 12
7.2.11 Analog Output / Display (ru.33...34 / ru.35...36)............................................ 7.2 - 13
7.2.12 ANOUT 1 / -2 / -3 / -4 / function (An.31 / An.36 / An.41, An.47)..................... 7.2 - 14
7.2.13 Amplifier of the output characteristic (An.33...35 / An.38...40 / An.43...45 /
An.49...51)..................................................................................................... 7.2 - 15
7.2.14 ANOUT 1...4 Digital setting (An.32 / 37 / 42 / 48).......................................... 7.2 - 16
7.3 Digital in- and outputs................................................................................. 7.3 - 19
7.3.1 Summary description digital inputs................................................................ 7.3 - 19
7.3.2 Input signals PNP / NPN selection (di.00)..................................................... 7.3 - 20
7.3.3 Setting of digital inputs by software (di.01, di.02).......................................... 7.3 - 20
7.3.4 Input terminal state (ru.21), internal input state (ru.22).................................. 7.3 - 21
7.3.5 Digital noise filter (di.03), fast digital noise filter (di.23).................................. 7.3 - 22
7.3.6 Inversion of Inputs (di.04).............................................................................. 7.3 - 22
7.3.7 Delay activation / deactivation digital inputs.................................................. 7.3 - 22
7.3.8 Input trigger (di.05)......................................................................................... 7.3 - 23
7.3.9 Strobe-dependent Inputs (di.06, di.07, di.08)................................................. 7.3 - 24
7.3.10 Reset / input selection (di.09) and reset / input slope selction (di.10)............ 7.3 - 25
7.5 Motor data and controller adjustment of the asynchronous motor.......... 7.5 - 3
7.5.1 Open loop operation (V/f characteristic).......................................................... 7.5 - 4
7.5.1.1 Rated frequency (uF.00), boost (uF.01) and delta boost (uF.04 / uF.05)............... 7.5 - 4
7.5.1.2 Maximal voltage mode (uF.10).............................................................................. 7.5 - 4
7.5.1.3 Additional rated point (uF.02 / uF.03)..................................................................... 7.5 - 5
7.5.1.4 Voltage stabilisation (uF.09).................................................................................. 7.5 - 5
7.5.1.5 Switching frequency (uF.11).................................................................................. 7.5 - 7
7.5.1.6 Energy saving function (uF.06...08)....................................................................... 7.5 - 7
7.5.1.7 SMM (sensorless motor management)................................................................. 7.5 - 7
7.5.2 Speed-controlled operation............................................................................ 7.5 - 11
7.5.2.1 Initial settings....................................................................................................... 7.5 - 11
7.5.2.2 Vector controlled operation without motor model................................................ 7.5 - 14
7.5.2.3 Vector controlled operation with motor model (with encoder feedback).............. 7.5 - 15
7.5.2.4 Vector control without speed feedback (ASCL)................................................... 7.5 - 27
7.5.2.5 Special function: Rotor adaption......................................................................... 7.5 - 32
7.5.3 Block diagram................................................................................................ 7.5 - 34
7.6 Adjustments of the synchronous motor...................................................... 7.6 - 3
7.6.1 Initial settings................................................................................................... 7.6 - 3
7.12.2.9 Approach to reference point with subsequent drive to zero signal.................... 7.12 - 13
7.12.2.10 Approach to reference point with overdriving of the reference limit switch....... 7.12 - 13
7.12.2.11 Position reset in the encoder............................................................................. 7.12 - 14
7.12.3 Synchronous mode...................................................................................... 7.12 - 14
7.12.3.1 Synchronous mode / principle........................................................................... 7.12 - 14
7.12.3.2 Synchronous mode / premise............................................................................ 7.12 - 15
7.12.3.3 Synchronous mode / position normalisation...................................................... 7.12 - 17
7.12.3.4 Synchronous mode / selection of operating mode............................................ 7.12 - 18
7.12.3.5 Synchronous mode / activation and synchronization........................................ 7.12 - 19
7.12.3.6 Gear factor........................................................................................................ 7.12 - 24
7.12.3.7 Angular correction............................................................................................. 7.12 - 25
7.12.3.8 Angular reset..................................................................................................... 7.12 - 26
7.12.4 Posi mode.................................................................................................... 7.12 - 27
7.12.4.1 Selection of operating mode.............................................................................. 7.12 - 27
7.12.4.2 Posi mode / principle......................................................................................... 7.12 - 27
7.12.4.3 Posi mode / premise.......................................................................................... 7.12 - 29
7.12.4.4 Position normalisation....................................................................................... 7.12 - 30
7.12.4.5 Posi mode / actual position............................................................................... 7.12 - 34
7.12.4.6 Posi mode / set and target position................................................................... 7.12 - 35
7.12.4.7 Posi mode / single positioning........................................................................... 7.12 - 35
7.12.4.8 Posi mode / sequential positioning.................................................................... 7.12 - 38
7.12.4.9 Posi mode / Positioning with set changeover.................................................... 7.12 - 52
7.12.4.10 Posi mode / rotary table.................................................................................... 7.12 - 52
7.12.4.11 Posi mode / defined stop................................................................................... 7.12 - 57
7.12.4.12 Posi mode / remaining distance positioning...................................................... 7.12 - 58
7.12.4.13 Posi mode / flying referencing with correction................................................... 7.12 - 59
7.12.4.14 Posi mode / start positioning............................................................................. 7.12 - 64
7.12.4.15 Posi mode / inaccessible positions.................................................................... 7.12 - 68
7.12.4.16 Posi mode / stop positioning............................................................................. 7.12 - 70
7.12.4.17 Analog position setting...................................................................................... 7.12 - 70
7.12.4.18 Analog position output....................................................................................... 7.12 - 71
7.12.4.19 Target window................................................................................................... 7.12 - 71
7.12.4.20 Backlash............................................................................................................ 7.12 - 72
7.12.4.21 Position scan..................................................................................................... 7.12 - 72
7.12.4.22 Teach function................................................................................................... 7.12 - 72
7.12.4.23 Functions and displays for the positioning mode.............................................. 7.12 - 74
7.12.5 Contouring control mode.............................................................................. 7.12 - 77
7.12.5.1 Contouring control mode / premises................................................................. 7.12 - 77
7.12.5.2 Contouring control mode / settings.................................................................... 7.12 - 78
7.12.5.3 Contouring control mode / write / read data...................................................... 7.12 - 78
7.12.5.4 Contouring control mode / speed precontrol..................................................... 7.12 - 79
7.12.5.5 Contouring control mode / watchdog................................................................. 7.12 - 79
7.12.5.6 Contouring control mode / example.................................................................. 7.12 - 80
7.12.6 Position controller........................................................................................ 7.12 - 82
7.13 Protective functions.................................................................................... 7.13 - 4
7.13.1 Error and warning messages......................................................................... 7.13 - 4
7.13.1.1 underpotential...................................................................................................... 7.13 - 5
7.13.1.2 Overvoltage ........................................................................................................ 7.13 - 6
7.13.1.3 overcurrent.......................................................................................................... 7.13 - 6
7.13.1.4 overload............................................................................................................... 7.13 - 6
7.13.1.5 Inverter over temperature.................................................................................... 7.13 - 7
7.13.1.6 External fault....................................................................................................... 7.13 - 7
7.13.1.7 Bus error.............................................................................................................. 7.13 - 7
7.13.1.8 Limit switch error................................................................................................. 7.13 - 8
1. Introduction
2. Summary 2
3. Hardware
4. Operation
Selection of Operating
5.
Mode
7. Functions
8. Error Assistance
9. Project Design
10. Networks
12. Annex
Autoboost
2. Summary
Prog. Operator menu
2.1 Product description PID controller
2.1.1 Features of KEB COM- Encoder interface 14 Parameter groups
BIVERT DC braking
Software in-/outputs
Speed search
2
Jogging function (prog.) KEB 2 prog. relay outputs
electr. motor protection COMBIVERT 8 prog. digital inputs
Slip compensation 2 prog. digital inputs
HSP5 interface 2 prog. digital outputs
Energy saving function
Power-off function 2 prog. analog outputs
Protective equipment Hardware current limit
Prog. Prog. filter for analog and digital inputs
Adjustable balancing of the ramps
Hour meter
2.1.2 Function principle The power circuit of a frequency inverter consists basically of a mains rectifier, the
DC-link and an inverter at the output. The mains rectifier consists of an uncontrolled
single or three-phase bridge connection, the single-phase design is restricted to small
powers. It converts the AC-voltage of the mains into a DC-voltage, which is smoothed
by the DC-link capacitor, thus in the ideal case (inverter unloaded) the DC-link is
charged with a voltage of UZK = √2 . UN.
Since during the charging of the DC-link capacitor very high currents flow for a short
time which would lead to the tripping of the input fuses or even to the destruction of
the mains rectifier, the charging current must be limited to a permissible level. This is
achieved by using an inrush current limiting resistor in series to the capacitor. After
the charging of the capacitor is completed the limiting resistor is bridged, for example,
by a relay and is therefore only active at the switch-on of the inverter.
As the smoothing of the DC-link voltage requires a large capacity, the capacitor still
has a high voltage for some time after the disconnection of the inverter from the mains.
The actual task of the frequency inverter, to produce an output voltage variable in
frequency and amplitude for the control of the three-phase AC motor, is taken over
by the converter at the output. It makes available a 3-phase output voltage according
to the principle of the pulse-width modulation, which generates a sinusoidal current
at the three-phase asynchronous motor.
U
L1 UN R UZK
V M
L2 C
(L3) W 3~
15.F5.A1E–35DA
at FI: Cooling A-Z: as digits, varnished at Servos: motor cooling
0: Standard A: 0 varnished (standard) 0: Self-cooling
1: Flat rear B: 1 varnished (Flat Rear) 1: Separate cooling 2
2: Water-cooled C: 2 varnished (water cooling)
3: Convection D: 3 varnished (convection)
4: Special fan E: 4 varnished (special fan)
Input identification
0: 1ph 230V AC/DC 5: 400V DC A: 6ph 400V AC
1: 3ph 230V AC/DC 6: 1ph 230V AC z: 230V AC or AC/DC
2: 1/3ph 230V AC/DC 7: 3ph 230V AC Y: 400V AC or AC/DC
3: 3ph 400V AC/DC 8: 1/3ph 230V AC W: 230V DC
4: 230V DC 9: 3ph 400V AC V: 400V DC
Housing type A, B, D, E, G, H, P, R, U, W
Accessory
0: without 4: PFC 2) integrated
1: GTR 7 1) 5: GTR 7 1), PFC 2) integrated
2: integrated filter 6: integrated filter , integrated PFC 2)
3: GTR 7 1), integrated filter 7: GTR 7 1), integrated filter, integrated PFC 2)
Control type
A: Standard software for operation of asynchronous motors with vector control and V/f characteristic
control, synchronous motors with vector control
E: Software for the speed-controlled operation of synchronous motors without speed feedback
H: Software for the speed-controlled operation of asynchronous motors without speed feedback
Series F5
1. Introduction
2. Summary
3. Hardware 3
4. Operation
Selection of Operating
5.
Mode
7. Functions
8. Error Assistance
9. Project Design
10. Networks
12. Annex
3. Hardware
3.1 Control circuit F5-A, -E, -H
3.1.1 Terminal strip X2A
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
In order to prevent a malfunction caused by interference voltage supply on the control inputs, the following
directions should be observed:
*
+
0...±10 VDC R = 3...10 k
Inputs X2A.3 and X2A.4 can be programmed and assigned as setpoint input (see chapter 7.2).
The supply of the control circuit through an external voltage source keeps the control in operational condition
even if the power stage is switched off. To prevent undefined conditions at external power supply the basic
procedure is to first switch on the power supply and after that the inverter.
3.2 Control circuit A-servo
3.2.1 Terminal strip X2A
Setpoint signal
0…±10 V DC
Setpoint potentiometer
3…10 kΩ
Analog output 3
±10 V DC
max. 5 mA
external voltage supply
20…30 V DC ±0 %
1 A
max. 30 V DC
0.01…1 A
1) Connect potential equalizing line only if a potential difference of > 30 V exists between the
controls. The internal resistance is reduced to 30 kΩ.
In case of inductive load on the relay outputs a protective wiring must be provided (e.g. free-wheeling diode) !
The control card must always be supplied with an external power source. Thus the control remains
in operation even when the power unit is switched off. To prevent undefined conditions at external power
supply the basic procedure is to first switch on the power supply and after that the inverter.
The connections of the control terminal blocks and encoder inputs have safe isolation according to EN 50178.
1. Introduction
2. Summary
3. Hardware
Selection of Operating
5. 4.2 Password input
Mode
6. Initial Start-up
7. Functions
8. Error Assistance
9. Project Design
10. Networks
12. Annex
4. Operation
The following chapter describes the fundamentals of the software structure as well as the operating of the unit.
4.1 Fundamentals
Parameters are values changeable by the operator in a program, which have an influence on the program flow.
A parameter consists of:
Example:
A conveyor belt shall be used with 3 different speeds. A parameter set is programmed for each "speed" ...acceleration,
deceleration etc. can be adjusted individually.
The blinking point indicates the changeable area. By pressing the ENTER-key the blinking point is shifted.
ENTER ENTER
F/R F/R
FUNC
Changes between parameter value SPEED
and parameter designation.
Changing of
Standard-Parameters ENTER-Parameters
Parameter values can be changed only, when the parameter set is not adjusted to "Active parameter
set" (A)! (see 4.1.6)
4.1.4 ENTER-Parameter
For some parameters it is not sensible that the selected values become active immediately. For that reason
they are called ENTER-Parameters, they do not become active until the ENTER key is pressed.
Example: At digital setting of rotation direction the rotation reverse (r) shall be selected from standstill (LS).
As seen above it must be switched at this in forward direction of rotation (F). The drive shall start only if the
direction of rotation reverse is selected and confirmed with ENTER.
Certain parameters are not programmable, as their value must be the same in all sets (e.g. bus address or
baud rate). For an easy identification of these parameters the parameter set number is missing in the para-
meter identification. For all non-programmable parameters the same value is valid independent of the
selected parameter set!
4
4.1.6 Resetting of error messages
If a malfunction occurs during operation, the actual display is overwritten by a by a blinking error message. The
error message can be deleted by pressing the ENTER key, so the initial value is displayed again.
ATTENTION! The resetting of the error message with ENTER is no error reset, i.e. the error status in the in-
verter is not reset. Thus is possible to correct adjustments before the error reset. An error reset is only possible
by the reset terminal or control release.
To permit conclusions on the operational performance of the drive, parameters are provided that indicate the
peak values. Peak value means that the highest measured value is stored for the ON-time of the inverter
(slave pointer principle). The peak value is cancelled by ▲ or ▼ and the actual measured value is displayed .
To monitor the correct execution of an action some parameters send a status signal. For example, "PASS"
is displayed after copying a set to indicate that the function was completed error-free. These status signals
must be acknowledged with ENTER.
1. Introduction
2. Summary
3. Hardware
Selection of Operating
5. 4.2 Password structure
Mode
6. Initial Start-up
7. Functions
8. Error Assistance
9. Project Design
10. Networks
12. Annex
Depending on the actual operating mode the password can be entered in the following parameters:
The parameter value of the above parameters shows the actual password level. Following indications are
possible:
CP - read only Only the customer parameter group is visible, ex-
cept for CP. 0 all parameters are in the read-only
status (see chapter 4.3).
CP - on Only the customer parameter group is visible. All
parameters can be changed.
4.2.2 Passwords
By selecting one of the following passwords you can switch to the respective password level:
Drive-Mode
To finish the drive mode press ENTER + FUNCT key for approx. 3 sec.
Example 1:
Switching from CP mode
to the application mode
4
With the exception of
the service password all
entered password levels
are generally stored non-
volatile!
Example 1:
Switching from applica-
tion mode to the CP-
read-only mode
1. Introduction
2. Summary
3. Hardware
4. Operation
Selection of Operating
5. 5.1 Selection of Operating Mode 5
Mode
6. Initial Start-up
7. Functions
8. Error Assistance
9. Project Design
10. Networks
12. Annex
4: F5-M / 4000rpm
Software for the speed-controlled operation of asynchronous motors without
5: F5-M / 8000rpm speed feedback with a maximum speed upto 128000rpm
6: F5-M / 16000rpm - control type 7 or is for a maximum speed of 32000rpm instead of 500rpm
F5H V2.31
7: F5-M / 32000rpm - additional control types 12 and 13
12: F5-M / 64000rpm
(Attention: limitations see high frequency software F5A V4.31)
13: F5-M / 128000rpm
8: F5-S / 4000rpm Software for the speed-controlled operation of synchronous motors without
9: F5-S / 8000rpm speed feedback
F5E V2.30 - The measured speed is replaced by an estimated speed actual value, which is formed
10: F5-S / 16000rpm by means of a mathematical model of the synchronous motor
11: F5-S / 500rpm - Operation with speed feedback also possible
8: F5-S / 4000rpm
Software for the speed-controlled operation of synchronous motors without
9: F5-S / 8000rpm speed feedback with a maximum speed upto 128000rpm
10: F5-S / 16000rpm - control type 11 or is for a maximum speed of 32000rpm instead of 500rpm
F5E V2.31
11: F5-S / 32000rpm - additional control types 14 and 15
14: F5-S / 64000rpm
(Attention: limitations see high frequency software F5A V4.31)
15: F5-S / 128000rpm
This operating instruction is not valid for the G mode (V/f - characteristic open-loop mode, because the dis-
play is scaled in Hz, not in rpm). I.e. this instruction is not valid for controller types, which selects the G-mode
(F5-G / xxxHz).
Attention:
If a download list is downloaded to an inverter with another controller type, or if the COMBIVIS Config-File is
used for another controller type some parameters (e.g. set speed, speed limits, etc..) are not correctly displa-
yed.
COMBIVIS recognizes the use of inappropriate lists and select the correct config file automatically. Unintended
settings and wrong displays can occur if the warning messages are ignored.
The standardisation of some parameters is depending on the speed range of the control type.
Some parameters (ramp adjustments) have reference values, which are depending on the selected speed
range (500, 4000, 8000, 16000, 32000, 64000, 128000 rpm.
Attention:
The description of single parameter adjustments provides (unless otherwise mentioned)
a speed range of 4000 rpm (ud.02 = 4 or 8).
Chapters 7.5 to 7.10 are not valid for all operating modes. 5
Which chapters are valid is depending on the software type, control type and the selection in cS.00 ad cS.01.
1. Introduction
2. Summary
4. Operation
Selection of Operating
5.
Mode
6. Initial Start-up 6
7. Functions
8. Error Assistance
9. Project Design
6.2 Start-up
10. Networks
12. Annex
6. Start-up
But because of the complex application possibilities we must restrict ourselves to explaining the start-up of
standard applications.
After unpacking the goods and checking them for complete delivery following measures are to be carried out:
The EMC-conform installation of the inverter is described in the Instruction Manual Part 1. Installation and con-
nection instructions are found in the instruction manual part 2. 6
The mounting surface of the inverter must be bright.
If necessary, use contact lacquer as protection against corrosion.
Connect the earthing strip to central point in the control cabinet
ENTE
R
F/R
STAR
T
STO
P
FUN
C.
SPEE
D
C
O
M
B
IV
E
R
T
L1
L2
L3
+
-
PB
U
V
W
0402
9508 C-1220
5.S0
07.F
1. Introduction
2. Summary
3. Hardware
4. Operation
Selection of Operating
5.
Mode
6. Initial start-up 6
7. Functions
8. Error Assistance
10. Networks
12. Annex
The following descriptions suppose that the frequency inverter is on the password level "application mode"
(ud.01 = application mode). The selection of the password level is described in the manual chapter 4. The start-
up should be executed with COMBIVIS in order to have a short start-up time.
Operating lists are available on the KEB homepage (www.keb.de). This lists contain the necessary parameters
for start-up.
Attention: The start-up instruction manual can only give a short overview of the parameter adjustments which
are mandatory necessary to start-up the motor.
Thus it represents a check list and not a complete parameter description.
The appropriate chapters of the application manual must be read carefully for exact information about the pa-
rameters, additionally points to consider and application-specific adjustments!
- in-phase connection (inverter terminals U, V, W must be connected at the motor terminal board with the
appropriate contacts)
If the wiring is correct the following direction of rotation occurs at setting "clockwise rotation":
6
- large surface of shield connection, well grounding (see „EMC conform wiring“ in the manual „...before
starting“)
The following chapters describe the start-up of an asynchronous motor in the 4 available modes:
⇒ uF.01 Boost
6.2.1.2 Vector controlled operation with encoder feedback without motor model
With this start-up description it is provided that an incremental encoder to encoder interface 1 (15pole Sub D
socket X3A) is used for speed feedback.
Read chapter 7.11 "Speed measurement" for necessary adjustments of your speed encoder when using ano-
ther encoder type.
Also it must be secured that the motor phases and the incremental encoder are correct wired (in phase, large
surface of shield connection, well earth connection). The motor temperature sensor must be connected.
A controlled start-up (U/f - characteristic) of the drive can be executed for examination in case of insecurity
about the phase allocation (motor and incremental encoder).
Values dr.00 to dr.05 must be taken from the motor name Note:
plate. The equivalent circuit data dr.06...dr.10
are without meaning.
⇒ dr.00
⇒ dr.01
⇒ dr.02
⇒ dr.03
⇒ dr.04
⇒ dr.05
⇒ dS.04 = 24
6.2.1.3 Vector controlled operation with encoder feedback with motor model
Attention: With this start-up description it is provided that an incremental encoder to encoder interface 1 (15pole
Sub D socket X3A) is used for speed feedback.
Also it must be secured that the motor phases and the incremental encoder are correct wired (in phase, large
surface of shield connection, well grounding).
The motor temperature sensor must be connected.
A controlled start-up (U/f - characteristic) of the drive can be executed for examination in case of insecurity
about the phase allocation (motor and incremental encoder).
If error messages occur during the start-up phase, read chapter 8 „Fault diagnosis“.
6.2.1.4 Start-up F5H-M (ASCL/ vector-controlled without encoder feedback with motor model)
If error messages occur during the start-up phase, read chapter 8 „Fault diagnosis“.
The following chapter describes the initial start-up of a synchronous motor splitted in speed controlled systems
with encoder feedback (F5A-S) and speed controlled systems without encoder feedback (F5E-S).
Steps 1 to 10 can be skiped when using a system consisting of KEB COMBIVERT F5 and KEB servo motor.
The following steps describe a standard start-up when using a "customer-specific motor“. Enocer interface 1 is
used as encoder feedback.
The following steps describe a standard start-up when using a KEB motor or "customer-specific motor“. An
encoder feedback is not used.
Special adjustments must be done for the operation of special or high-frequency motors.
Please contact KEB for this case.
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp setting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
7. Functions
7.1 Operating and appliance data
The parameter groups „ru“, „In“ and „SY“ are described in this chapter. They serve for the operational monito-
ring, error analysis and evaluation as well as for the unit identification.
The In- (Information) parameter group contains data and information on the identification of the hardware and
software as well as on the type and number of the errors that occurred. Following parameters are available:
In.00 inverter type
In.01 rated Inverter current
In.03 max. carrier frequency
In.04 rated carrier frequency
In.06 Software version
In.07 Software date
In.10 serial no. (date)
In. 11 serial no. (count)
In.12 serial no. (AB-no. high)
In.13 serial no. (AB-no. low)
In.14 customer no. (high)
In.15 customer no. (low)
In.16 QS no.
In. 17 Temperature mode
In.22 user parameter 1
In.23 user parameter 2
In. 24 Last error
In. 25 Error Assistance
In.26 E.OC error counter
In.27 E.OL error counter
In.28 E.OP error counter
In.29 E.OH error counter
In.30 E.OHI error counter
In.31 KEB Hiperface
In.32 Interface software date
The Sy- (system) parameter group contains system specific parameters. Following parameters are available:
SY.01 Watchdog cycle SY.44 Status word long
Sy.02 inverter identifier SY.50 Control word low
Sy.03 Power unit code Sy. 51 Status word low
Sy.06 inverter address SY.52 Set speed value
Sy.07 baud rate ext. bus SY.53 actual speed display
SY.08 bus synchron time Sy.56 Start display address
Sy.09 HSP5 watchdog time
Sy.11 baud rate int. bus
SY.32 Scope timer
SY.41 Control word high
SY.42 Status word high
SY.43 Control word long
A counter-clockwise rotating field (reverse) is represented by a negative sign. Precondition is the phase-correct
connection of the motor.
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp setting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
By selecting an input interface (An.00/ 10) input AN1 or AN2 can be adjusted to the applied input signal. By
An.20 the third analog input can be switched additionally to AN1. Subsequently the analog inputs are smoothed
in an electronic filter (An.01/ 11/ 21) by averaging. A save mode can be adjusted with An.02/12/22 and activated
with a programmable input (An.03/13/23). To avoid voltage fluctuations and ripple voltages around the zero
point the analog signal can be faded out around the zero point up to ±10 % (An.04/14/24). In the characteristic
amplifier the input signals can be influenced in X and Y direction as well as in the rise (An.05...07/ 15...17/
25...27). At the output of the characteristic amplifier the signal can be limited to minimum and maximum value
(An.08, 09/18, 19/28, 29). At the output of the block it can be defined with An.30 which analog signal serves
as reference value and which one serves as auxiliary value. The ru-Parameters are used for the display of the
analog signal pre and post amplifier. The internal values are limited to ±400%.
Depending on the adjusted interface (An.00 / An.10) the analog inputs AN1 and AN2 can process following
input signals:
Ri = 55 kOhm
X2A. 1 2 3 4 5 6 7 8 9 PE
1)
+ - + -
SPS SPS
0...±10 VDC
1) Connect equipotential bonding conductor only, if a potential difference of > 30 V exists between the con-
trols. > 30 V besteht. The internal resistance is reduced to 30 kOhm.
X2A. 1 2 3 4 5 6 7 8 9 PE
R = 3...10 k
0...10V DC Ri=30kΩ (An.00 / An.10 = 0) The output CRF Terminal X2A.7 may be loaded with max. 6mA!
X2A. 1 2 3 4 5 6 7 8 9 PE
+ +
0...±20 mADC
4...20 mADC
Ri = 250 Ω
With An.20 it is determined from where the 3. analog input value is received. Following values can be defined:
An.20: AN3
Va- Function
lue
0 Analog option (AN3)
1 Analog input 1 (AN1)
The noise filters shall suppress disturbances and ripples of the input signals. If the noise filter is switched off the
analog inputs are queried every 1 ms and the recorded value is transferred then. The noise filter adjustments
preset the number of sampled values for the averaging.
Coming from the input filter the save mode can be switched on with An.02 / An.12 / An.22 . If now the pro-
grammable digital input (value 1) is set the analog signal is processed directly and written parallel into the non-
volatile memory. As soon as the digital input is disconnected (value 0), the inverter continuos to run with value
stored in the memory. Moreover, with An.02/ An.12/ An.22 it can be determined whether the memory contents
are saved or deleted upon switch off.
An.02/12/22 Bit 0
0: Direktmode
1: Speichermode
vom zum
Eingangsfilter Kennlinien-
nichtflüchtiger verstärker
Speicher
7.2.4.1 Input selection (An.03, An.13, An.23)
The digital inputs for storing are selected with An.03 / An.13 / An.23 in accordance with the table „Input selec-
tion“ (see also chapter 7.3.11 „Assignment of the inputs“). In order to store an analog value, the save mode
must be switchted on (An.02/12/22 = 1) and the selected input must be activated.
Through capacitive as well as inductive coupling on the input lines or voltage fluctuations of the signal source,
the motor connected to the inverter can still drift (tremble) during standstill in spite of the analog input filter. It is
the task of the zero point hysteresis to suppress this.
With the parameters An.04 / An.14 / An.24 the respective analog signals can be faded out within a range of
0...±10%. The adjusted value is valid for positive and negative input signals.
If a negative percent value is adjusted the hysteresis acts in addition to the zero point around the current set-
point. Setpoint changes are accepted only if they are larger than the adjusted hysteresis.
10%
Value range
With these parameters the input signals can be adapted in X and Y direction as well as in the rise to the
requirements. In the case of factory setting no zero point offset is adjusted, the rise (gain) is 1, i.e. the input
value corresponds to the output value of this step (see Fig. 7.2.6.a). The output value is calculated according
to following formula:
-100%
By means of some examples, we want to show the possibilities of the function. According to Fig. 7.2.6.b
1. adjustment of the X-Offset for input AN1 to 50 (%)
2. adjustment of the amplification to 2
1. 100% 2. 100%
An.5
An.6
-100% 50% 100% -100% 100%
-100% -100%
With these settings the entire speed range can be driven with 0...10 V via input AN1. (rotation direction = ±ana-
log)
7
0% In corresponds to -100% Out
50% In corresponds to 0% Out
100% In corresponds to 100% Out
According to Fig. 7.2.6.c
Fig. 7.2.6.c X-Offset (An.06) =75%; Y-Offset (An.07) = 100%; Ampl. (An.5) = -1.00
An.5
An.6=75%
-100% 100% -100% 100% -100% 100%
7.2.7 Lower- and upper limit (An.08, An.09, An.18, An.19, An.28, An.29)
These parameters serve for limiting of the analog signals after the amplifier stage. All parameters are adju-
stable in the range of -400...400 %. Since no mutual locking exists, it is to be ensured, that the lower limit is
adjusted smaller than the upper limit.
400%
An.9 / An.19 / An.29
-400%
400%
-400%
The reference value for the calculation of the AUX signal from the encoder values of channel 1 or 2 is depen-
ding on ud.02:
The KEB COMBIVERT has four programmable analog outputs (ANOUT1, 2 and ANOUT3, 4). Parameters
An.31/ 36 allow the selection of one size which is given out at the outputs X2A.5/ 6 . ANOUT 3 and ANOUT 4
(An.41 / 47) can be output as switching condition 42 or 43 with the digital outputs as PWMsignal. The analog
signal can be adapted to the requirements with the characteristic amplifier (An.33...35 / An.38...40 / 43...45/
49...51). The ru-parameters show the current size before and after the amplification. The period time for the
PWM-signal can be adjusted with An.46/52.
The reference values for mode 0-3 and 18-20 changes depending on ud.02.
A voltage of 0...±11.5 VDC represents the selected size in the range of 0...±115 % with a resolution of 10 bit
at the output. In order to be able to balance load-dependent voltage drops, the limitation at the output of the
characteristic amplifiers is ±115 %.
Uout = 0...±11,5 V
+ -
Imax = 5mA
Ri < 100Ω
ANOUT1 (X2A.5) / AGND
ANOUT2 (X2A.6) (X2A.8 / X2A.9)
RB
ANOUT 3/4
Following parameters are used for the display of the analog outputs, before and after the characteristic ampli-
fication:
These parameters define the function which controls the respective output . Following adjustments are possi-
ble:
An.31/ An.36/ An.41/ An.47
Va- Function Output of 100 % corresponds to
lue
0 Absolute actual value ru.07 Amount of the speed actual value
Amount of speed set value before ramp
1 Absolute set value ru.01
generator 3000 rpm 2)
2 Actual value ru.07 Speed actual value
3 Set value ru.01 Speed set value
4 Output voltage ru.20 Output voltage 0...500 V
5 DC voltage ru.18 actual DC voltage 0...1000 V
6 Apparent current ru.15 apparent current
0...2 x inverter current (In.01)
7 Active current ru.17 active current
8 Digital setting with An.32/ 37/ 42/ 48 with An.32/ 37/ 42/ 48 preset value
9 External PID output ru.52 Output value of the PID controller
0...100 %
Amount of the PID controller output va-
10 Absolute ext. PIDoutput ru.52
lue
11 Absolute active current ru.17 Amount of the active current 0...2 x inverter current (In.01)
12 Heat sink temperature ru.38 Power module temperature
0...100 °C
13 Motor temperature ru.46 Motor temperature
14 Actual torque (F5-M/S) Actual torque
0...3 x rated torque
15 Absolute actual torque (F5-M/S) Amount actual torque
only in closed- DASM: dr.14
16 Set torque (F5-M/S) Set torque
loop operation DSM: dr.27
17 Absolute set torque (F5-M/S) Amount set torque
18 system deviation of the speed controller system deviation of the speed controller
19 Speed reference variable ru.02 Speed set value after ramp generator
0...3000 rpm
Absolute speed reference variable
20 Angle difference
ru.02
0... number of increments for
21 Angle difference (ru.58) Angle difference
a revolution
Analog input 1 pre amplifier display
22 Value of AN.01 at terminal
(ru.27)
Analog input 1 post amplifier display Value of AN.01 after analog value pro-
23
(ru.28) cessing
0...100 %
Analog input 2 pre amplifier display
24 Value of AN.02 at terminal
(ru.29)
Analog input 2 post amplifier display Value of AN.02 after analog value pro-
25
(ru.30) cessing
0...2 x rated motor power
26 Active power (ru.81) Active power DASM: dr.03
DSM: dr.32
27 Actual position (ru.54) Actual position Ref-position 0 % (PS.41)
28 Set position (ru.56) Set position Ref-position 100 % (PS.42)
act. torque, referring to the max. permis-
29 Max. torque in % (ru.90) 0...100 %
sible torque of the drive chain
1) 2) 3)
dependent on the inverter rated current (In.1), dependent on ud.2, dependent on the motor
The characteristic amplifier are following after selecting the signal to be given out (see Fig. 7.2.9). With these
parameters the input signals can be adapted in X and Y direction as well as in the rise to the requirements. No
zero offset is adjusted at factory setting, the gain is 1, i.e. 100% of the output variable correspond to 10V at the
analog output (see fig. 7.2.14.a).
Function ANOUT1 -2 -3 -4 Value range Resolution Default
A m p l i f i c a - An.33 An.38 An.43 An.49 ±20,00 0,01 1,00
tion
X-Offset An.34 An.39 An.44 An.50 ±100,0% 0,1% 0,0%
Y-Offset An.35 An.40 An.45 An.51 ±100,0% 0,1% 0,0%
Output voltage
100% 10V displayable
range
An.33
variable to be indica-
-100%
An.34 ted
100%
-100%
An.35
7
Inverting the analog output
An.35
100% 10V
An.33
An.34
-100% 100%
-100%
An example for using the analog output as 0/10V switch is shown in picture 7.2.13.c
1. adjustment of the gain (An.33) to 20.00
2. adjustment of the X-Offset (An.34) to the desired switching level
100% 10V
An.34 An.33
-100% 100%
-100%
Because of the high amplification the analog output switches in a relative small switching window.
Since the analog output always works firmly onto the values defined in 7.2.12, one can adjust the characteristic
with the aid of the amplification that the complete range 0...±10V is used.
defined value
––––––––––––––– = Amplification (An.33/ 38/ 43/ 49)
desired value
100Hz
––––– = 1,47
68Hz
Analog values can be preset in percent for the respective input with parameters An.32/ An.37/ An.42/ An.48.
For that purpose value 8: „digital setting“ must be adjusted as process variable. The setting is done within the
range ±100 %.
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp setting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
The KEB COMBIVERT has 8 external digital inputs and 4 internal inputs (IA...ID). All inputs can be assigned
to one or several functions.
Coming from the terminal strip it can be defined with parameter di.00 whether external inputs shall be con-
trolled in PNP or NPN wiring (not at safety relais). Parameter ru.21 shows the currently controlled input. Each
input can optionally (di.01) be set via terminal strip or by means of software with di.02. A digital filter (di.03,
di.23) reduces the noise sensitivity of the inputs. The inputs can be inverted with di.04 and with di.05 one
can switch to edge-triggering. A strobe-mode can be activated with parameters di.06...di.08. The input status
(ru.22) shows the inputs that are actually set for processing. The function(s) carried out by a programmed input
is defined by means of the input selection of the corresponding function or di.11...22.
For safety reasons the control release (ST) must generally be switched by means of hardware. Edge-trigge-
ring, inversion and strobe signal can be adjusted but have no influence.
0
1
a)
D Q 7
b) Clk /Q D Q
>=1
Latch /Q
di.02
1
Bit X
>=1
. D Q
sy.41 .
Bit Y .
Clk /Q
Digital inputs without external wiring can be set with parameters di.01 „select signal source“ and di.02 „digital
input setting“.
The control release must generally be switched by means of hardware even if one switches by software (see
Fig. 7.3.3 AND-operation wtih di.02 and sy.50)!
ST 1 ST
Terminal strip & &
I4 128 I4
IA 256 IA
Înternal inputs
IB 512 IB
IC 1024 IC
ID 2048 ID
As shown in Fig. 7.3.3, it can be selected with di.01 , whether the inputs shall be switched from the terminal
strip (default) or by way of parameter di.02 . Both parameters are bit-coded, i.e. according to following table, the
appropriate value for the input is to be entered. In the case of several inputs the sum is to be formed.
(Exception: Control release must always be bridged at the terminal strip).
The terminal state (ru.21) displays the logical level at the input terminals. It is unimportant, whether the inputs
are internally active or not. If a terminal is controlled, the appropriate decimal value according to table "terminal
state" is output. If several terminals are active, then the sum of the decimal values is output.
The internal input state (ru.22) displays the logical state of the digital inputs internally set for further processing.
If an input is set, the appropriate decimal value according to table 7.3.1 is output. If several inputs are set, then
the sum of the decimal values is output.
7.3.5 Digital noise filter (di.03), fast digital noise filter (di.23)
The digital noise filter reduces the susceptibility to interferences on the digital inputs. Only hardware inputs can
be filtered. Each input has a separate filter counter that counts up at active input and down at inactive input.
The output of the filter is set on reaching the filter time and reset when zero is reached.
Priority of the filter times: The larger of the two times is used.
With parameter di.04 it can be adjusted whether a signal is 1- or 0-active (inverted). The parameter is bit-coded,
i.e. the value corresponding to the input mut be entered. If several inputs shall be inverted, then the sum is to
be formed. (Exception: An inversion of the control release remains without function).
Value range: 0 (off) ... 32.00 s; Resolution 0.01s; Default value : 0 (off), not set-programmable
Function:
The function can be switched off with the value di.40 – 55 = 0.
The ON delay starts with the positive edge of the input, the OFF delay with the negative edge.
In the following example I1 and I2 are first switched individually and then simultaneously.
The four required times are different adjusted.
7
7.3.8 Input trigger (di.05)
As a standard the inverter is controlled with static signals, i.e. an input is set for as long as a signal is applied.
However, practice has shown that a signal may be available for a limited time only, but the input shall still
remain set. In that case the input or several inputs can be adjusted to edge-triggered flip-flop. Then a rising
edge with a pulse duration that is longer than the response time of the digital filter is sufficient for switch-on.
Switch-off is effected with the next rising edge.
Control release (ST) can be set to edge-triggered flip-flop, but this remains without affect on the function, since
it is a pure static signal.
Input trigger
t
A strobe signal is used mainly for triggering the input signals. For example, two inputs shall be used for the
parameter set selection. But the signals for the control do not arrive exactly even, so for a short time it would
be switched into an unintended set. With active strobe (scanning signal) the current input signals of the strobe-
dependent inputs are accepted and kept until the next scanning.
Which inputs are switched by strobe?
With di.08 any input can be selected as strobe-dependent input. With the control release di.08 has no function
since this is a static input.
The strobe input is adjusted with parameter di.06. If several inputs are adjusted as strobe they are linked in
OR-operation.
As a standard the strobe is edge-active, i.e. the input conditions on the strobe input are accepted with rising
edge and maintained until the next rising edge. For some applications it is sensible to use the strobe in a man-
ner of a gate function. In that case the strobe signal is static, i.e. the input signals are accepted for as long as
the strobe signal is set (or for as long as the gate is open).
di.07 Strobe-mode
Strobe input
t
Resulting
Strobe signal t
Signal at
terminals t
Input status t
input signal
t
Strobe signal
t
Input status
t
7.3.10 Reset / input selection (di.09) and reset / input slope selction (di.10)
The reset input is defined with di.09 according table 7.3.1 . If the reset input shall react to a negative edge, one
or several of the reset inputs defined with di.09 can be switched to negative edge evaluation with di.1.
a.) One or several inputs can be assigned to each function. This means an input can be selected at the
single functions (positioning, fixed value selection, etc.) which activates this function.
b.) One or several functions can be assigned to each digital input. This means, in parameters di.11...di.22
„Function“ and parameters di.24...di.35 „+ Function“ one or several functions can be assigned to each
single digital input. In parameters di.11...di.22 several functions can be assigned to the respective in-
puts, only one can be selected in parameters di.24...di.35.
Both variants are locked against each other; if an input is assigned to a function, also parameters di.11...di.22
and di.24...di.35 are accordingly adjusted.
- with the functional programming of the inputs, the function's parametrization also permits selecting
which inputs will activate the function,
- the input-related display gives an overview of the complete function of an input and finally it can be
checked if there are undesired function overlappings.
The following table shows a list of the parameters with which the various functions can be assigned digital
inputs:
An. 03 AN1 save trigger / input selection oP.20 Fixed value / input selection 2
An.13 AN2 save trigger / input selection oP.56 Motorpoti increase / input selection
An.23 AN3 save trigger / input selection oP.57 Motorpoti decrease / input selection
cn.11 PID reset / input selection oP.58 Motorpoti Reset / input selection
cn.12 I reset / input selection oP.60 Input selection /clockwise rotation
cn.13 Fade in reset / input selection oP.61 input selection / direction reverse
di.09 Reset / input selection Pn.04 Input selection external error
di.36 Software ST input selection Pn.23 LAD stop / input selection
di.37 Locking ST input selection Pn.29 DC brake / input selection
di.39 disable ST input selection Pn.64 GTR7 / input selection
dr.61 Rs correction auto temp input selection PS.02 Pos/syn / input selection
Ec.48 scan channel 2 input selection PS.03 Shift. slave / input selection
Ec.49 scan channel 1+2 input selection PS.10 Shift. slave inv. / input selection
Ec.50 scan position Ec.60 PS.18 Reference switch / input selection
Ec.51 scan position Ec.61 PS.19 Start reference / input selection
Fr.07 Parameter set / input selection PS.29 Start posi / input selection
Fr.11 Reset set / input selection PS.36 Teach index input selection
LE.17 Timer 1 Start / input selection PS.37 Position scan input selection
LE.19 Timer 1 Reset / input selection PS.38 relative positioning f/r input selection
LE.22 Timer 2 Start / input selection PS.43 Correction reference point
LE.24 Timer 2 Reset / input selection uF.08 Energy saving function / input selection
oP.19 Fixed value / input selection 1 uF.21 Dead time compensation off input selec-
tion
The following table gives an overview of all functions which can be assigned to a digital input with parameters
1)
the column „Fct. Para“ indicates the function-related parameter, corresponding to the value in di.11...di.22.
* is scanned with 250µs
The next table indicates an overview of the functions, which can be assigned additionally to a digital input with
parameters di.24.. di.35 (only one auxilliary function is possible per input / bit 31 „I+ function“ must be activated
for the appropriate input):
1)
the column „Fct. Para“ indicates the function-related parameter, corresponding to the value in di.11...di.22
* is scanned with 250µs
di.36 software ST, di.37 locking ST, di.38 turn off ST delay time
The function is switched off, if no input is selected in di.36. ST can not be selected as software ST or input for
locking.
With the locking function the control release can be controlled in case of voltage failure (even if the controlled
PLC is failure) as long as e.g. Power off needs for stopping the drive.
Condition: terminal ST must be bridged !
Switching off an input (selection in di.36) is decelerated for the time adjusted in di.38. Within this time the lo-
cking input (selection in di.37) must be active in order to secure the function.
A software input e.g. (IA-ID) can be assigned with the function Power Off (do.00...do.07 = 17, switching condi-
tion for OA-OD) as locking input.
Eingang
Software-
Reglerfreigabe
(di.36)
di.38
verzögerte
Reglerfreigabe
(di.36,38)
Eingang
Selbsthaltung
Reglerfreigabe
(di.37)
Funktion
Software-
Reglerfreigabe
7.3.13 Deactivation of the digital control release
With the digital input selection (di.01 / di.02) or the control word (SY.43/ SY.50) the control release can be set
digital (e.g. via bus system). Additionally terminal ST must always be activated.
In parameter di.39, "turn off ST input selection", an input can be selected for deactivating the digital setting of
the control release . Thus only terminal ST is effective.
Thus it is possible to
8 do.11 do.28 4
do.19 do.36 23=8
Bit 6-7 Relay
1 23=8 1
9 1 8 1 8 8
0 R2=O1
31 do.3 8 1 2 & 8 1 2 & 64 R2=O2 output
10 1
1 of 31 ... 8 ... 8 >1 128 R2=R1 (R2)
>1
11 1 128 1 128 192 R2=R2 X2A.27...29
8 7
12
do.12 do.29
31 do.4 do.20 do.37 24=16
13 1 24=16 1
1 of 31 1 8 1 8 16
1 & 1 & Internal
14 8 2 8 2 1
... 8 ... 8 >1
output OA
>1
15 1 128 1 128
16 31 do.5 Internal
1 of 31 do.13 do.30
17
do.21 5
2 =32
do.38 25=32 output OB
1 1 8 1 1 8 32
8 1 2 & 8 1 2 &
18 1
31 do.6 ... 8 >1 ... 8 >1
19 1
1 128 128 Internal
1 of 31
20 output OC
do.14 do.31
21 do.22 do.39 26=64
1 26=64 1
1 8 1 8 64
31 do.7 & &
22 8 1 2 8 1 2 1
1 of 31
... 8 ... 8 Internal
23 >1 >1
1 128 1 128 output OD
91 do.15 do.32
do.23 27=128
do.40 27=128
1 1 8 1 1 8 128
1 2 & 1 2 &
8 8 1
... 8 >1 ... 8 >1
1 128 1 128
8 8
Description
For the switching of the digital outputs one can choose up to 8 conditions from the 91 different conditions.
These are entered in do.00...do.07. Switching condition 0 and 1 can be filtered by do.43 and do.44. Parameter
ru.23 displays, if one or several of these conditions are met . For each flag it can now be selected which of the 8
conditions shall apply to it (do.16...do.23). Each condition can still be inverted before selection (do.08...do.15).
As a standard all conditions (if several are selected) are OR operated. With do.24 this can be changed to AND-
operation . i.e. all conditions selected for this flag must be fulfilled before it is set. Parameter ru.24 displays the
flags which are set in this stage. do.33...40 form a second logic step where a selection of the flags from logic
step 1 can be made. Each individual flag can be inverted with do.25...32. do.41 adjusts the type of the linkage
(AND/OR). Parameter do.42 is used for inverting one or several outputs. With do.51 the output signals are
assigned to the terminals. ru.80 serves for the display of the digital output state, thereafter ru.25. The internal
outputs OA...OD are directly connected with the internal inputs IA...ID.
The total current of X2A.18,19 is limited to 50mA. In case of inductive load at the relay outputs or at the transi-
stor output a protective wiring is to be provided (free-wheeling diode)!
A filter can be set for switching condition 0 with do.43. For switching condition 1 with do.44. The change of a
switching condition must be applied for the filter time, then it becomes active at the output of the filter. If the
change of a switching condition is cancelled during the filter time, the filter time is reset and restarted with the
next change. The filter time can be adjusted in a range of 0 (off)...1000 ms.
From the following switching conditions one can select up to 8 for further processing. The values are then
entered in the parameters do.00...do.07.
The selection of inputs to be linked occurs via the comparison level parame-
ters LE.00...LE.07.
62 Inputs NOR linked (ru.22)
Input ST RST F R I1 I2 I3 I4 IA IB IC ID
Value 1 2 4 8 16 32 64 128 256 512 1024 2048
With parameters do.08...do.15 each of the 8 switching conditions (do.00...do.07) can be inverted separately
for each flag. With this function it is possible to set any chosen switching condition as non-condition. The pa-
rameter is bit-coded. According to Fig. 7.3.15 theweighting for the switching conditions to be inverted must be
entered in do.08...do.15. If several conditions shall be inverted, the sum is to be formed.
Example:
Output X2A.19 shall be set when the inverter is not accelerating! In this case we assign the switching condition
21 (inverter accelerates) for example to do.01 (enter value 21). We invert the switching condition (do.01) with
do.09, so enter value „2“.
Parameters do.16...do.23 serve for the selection of the 8 preassigned switching conditions. The selection is
done foreach flag separately, where one can choose between no one and up to all 8 switching conditions. The
value of the selected switching condition must be entered in do.16...do.23 in accordance with fig. 7.3.15 . If
several conditions shall be inverted, the sum is to be formed.
After the switching conditons are selected for each output, it can now be determined, how these are connected.
As a default all conditions are OR connected, i.e. the flag is set if one of the selected conditions is fulfilled.
Another possibility is the AND connection which can be adjusted with do.24. AND connection means that all
selected conditions must be fulfilled before the flag is set.
Parameter do.24 is bit-coded. The table under 7.3.17 shows the assignment.
8
do.23 & 1 Bit 7
Flag 7
>1 0
With parameters do.25...do.32 each of the 8 flags (bit 0...7) from logic step 1 can be inverted separately.
With this function it is possible to set any chosen flag as non-flag. The parameter is bit-coded. According to
Fig. 7.3.18 the weighting of the flag to be inverted must be entered in do.25...do.32. If several flags shall be
inverted, the sum is to be formed.
In the second logic step a selection of the flags of the first logic step can be made. The selection is done for
each output separately, where one can choose between none and up to all 8 flags.The value of the selected
switching conditions must be entered in do.33...do.40 in accordance with fig. 7.3.18. The sum is to be formed
if several flags shall be selected.
After the flags are selected for each output, now it can be determined how these are linked. As a default all
flags are OR operated, i.e. if one of the selected flags is set, the output switches.Another possibility is the AND
operation which can be adjusted with do.41. AND operation means, all selected flags must be set before the
output switches.
do.41
8
do.33 & 1 Bit 0
O1
>1 0
8
do.34 & 1 Bit 1
O2
>1 0
8
do.40 & 1 Bit 7
OD
>1 0
As shown in Fig. 7.3.23b, the outputs can be inverted again after linking with parameter do.42. The parameter is
bit-coded, i.e. according to following table the value belonging to this output must be entered. If several outputs
shall be inverted, the sum is to be formed.
7.3.24 Output terminal state (ru.25) and digital output state (ru.80)
Parameter ru.25 displays the logic condition of the digital outputs after the allocation by do.51. Parameter ru.80
displays the logic condition before the allocation. If an output is setthe appropriate decimal value according to
the table below, is output. If several outputs are set, then the sum of the decimal values is output.
The output signals are allocated to output terminals O1, O2, R1 and R2 with do.51. The assignment is done
according to following table:
Solution proposal:
The adjustment of do.24 is independent for this example, as only one condition each is adjusted at do.16...18.
Adjust flags
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp setting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
The setpoints of the KEB COMBIVERT F5 can be preadjusted analog as well as digital. The AUX-function adds
or multiplies an analog setpoint to/with other setpoint settings.
The setpoint and rotation selection links the different setpoint sources with the possible sources of rotation
direction. The signal thus obtained is used for further setpoint calculation.
Only after interrogation of the absolute setpoint limits, all the data that is required for the ramp calculation is
available.
Setpoint limits
Setpoint calculation
ru.01: Set value display
Set value display
before ramp
Ramp generator 7
ru.02:ramp output display
Set value display
after ramp
control
The cycle time of the software is 1 ms. During this time the analog input/output status is updated once. Addi-
tionally the inverter requires a processing time of 1... 3 ms before the new setpoint value is calculated. If the
inverter is used as secondary final control element of a superior control, this time can impair the dynamics of
the entire closed-loop control system.
In such cases the analog setpoint value can be processed directly to the control processor (direct setpoint
adjustment). Thus a scan time of 250 µs is possible. To enable this fast response to an analog setpoint value,
some restrictions must be accepted:
• The calculation formula of the analog setpoint value changes. Parameters oP.06 / oP.07 are without influence
on the setpoint calculation.
This value is limited with oP.14 for for both directions of rotation.
7
• The reference limitations oP.06 / oP.07 / oP.11 do not have any function; the absolute max. reference is only
limited by oP.14 (for both directions).
• The acceleration / deceleration and S-curve time have no effect; it is operated internally without ramps.
• The parameters An.01...04 and An.07...09 are without any function
• The stop position controller cannot be activated.
The setpoint setting with the settings oP.00 = 0...9 is 16 Bit wide. This results in a maximum resolution of 0.125
rpm in 4000-rpm mode (ud.02 = 4 and 8).
For applications needing a higher resolution, the high resolution setpoint setting was introduced. Here, the
setpoint is set as a 32-Bit-value.
Since only a 16-Bit-value can be output, the low-order 16 bits of the ramp output value are upsampled. On
overrun, the base value is increased for one cycle (1 ms) (by 0.125 rpm in 4000-rpm mode). These setpoint
fluctuations are smoothed mechanically, leading to the higher average resolution.
To achieve the highest possible resolution for the application, two parameters can be used:
Parameter oP.64 adjusts the reference value of the calculation and is dependent on ud.02.
oP.63 x oP.64
Setpoint = ———————
230
That means: If value 230 is preset for oP.63, the setpoint is equal to oP.64 "ref. value high-resolution"
oP.64
High resolution = ——————
230
If oP.64 is set to 2000rpm, i.e., the half maximum value (4000 rpm mode), the high resolution is:
2000 rpm
High resolution = —————— = 1,86 x 10-6 rpm
230
desired setpoint
oP.63 = ——————————— x 230
oP.64
Example 1
0.140624 rpm
oP.63 = —————— x 230 = 75497
2000 rpm
Example 2
Reference value (oP.64): 2000 rpm
requested setpoint: 32.37843 rpm
32.37843 rpm
oP.63 = ——————— x 230 = 17383037
2000 rpm
Resolution and scaling factor are the same as for oP.63. Due to internal rounding, the value of ru.82 may be 1
less than the set value oP.63.
The selection of rotation direction determines the manner in which the rotation direction is adjusted. The fol-
lowing possibilities are available:
0 limited:
negative setpoints are set to zero, i.e. only positive setpoints are driven in accordance with the selected rotation
direction
absolute:
no sign of the set value is evaluated and it is always driven with the amount in accordance with the selected
rotation direction.
0-limited absolute
Setpoint Setpoint
Setpoint Setpoint
adjust- adjust-
ment ment
The rotation selection via terminal strip allows the adjustment of the direction of rotation via switch or from a
primary control.
Direction forward input selection (Run / Stop) oP.60, direction reverse (forward / reverse) oP.61
With parameter oP.60 one input is determined for rotation direction forward (or run/stop) and with oP.61 one
input for rotation direction reverse (or forward/reverse). (see chapter 7.3)
In forward/reverse rotation setting (oP.01= "2" or "3") the inputs determined with oP.60 and oP.61 work as fol-
lows:
In run/stop and forward/reverse rotation setting (oP.01= "4" or "5") the inputs determined with oP.60 and oP.61
work as follows:
The direction of rotation can be defined with the preadjusted set value signal. In the case of analog signals
through adjustment of positive or negative voltages. In the case of digital signals through adjustment of positive
values (without sign) or negative values (negative sign in the display).
In this case „F“ or „R“ must be set via a digital input, digital via oP.02 or "start" via control word SY.50 in order for
the inverter to modulate. It is unimportant which rotation setting is used, as the direction of rotation is dependent
on the setpoint.
oP.01 = 10: The rotation direction release is done exclusively via the control word Run/Stop.
oP01 = 11: Function is like at value 10 only without deceleration ramps. The modulation of the frequen- 7
cy inverter is disabled and the motor will coast down if the direction of rotation is disabled.
In this case the inverter modulates always. No direction of rotation needs to be adjusted.
The control word is used for the state control of the inverter via bus. In order for the inverter to react to the con-
trol word, the respective control process must be enabled (oP.01=8 or 9). When setting the direction of rotation
via the control word, the setpoint can be evaluated 0-limited (oP.01 = 8) or absolute (oP.01 = 9).
If run/stop is to be adjusted via the control word, oP.02 must be set to "0". The terminals F/R may
not be wired (OR-connection of terminal, oP.02 and SY.50).
7.4.4 Acceleration of speed reversal
The KEB COMBIVERT supports up to 3 fixed frequencies for each parameter set, which can be selected via
two digital inputs. The required inputs for the selection are defined with oP.19 and oP.20 (also see "Digital in-
puts" chapt. 7.3.11). The rotation direction source for fixed value mode is defined with oP.18. The adjustment
is independent of oP.01 and is valid exclusively for the fixed frequencies. The adjustment of a fixed frequency
has priority over the "normal" setpoint adjustment.
4 5
3 6
oP.19 Input selection oP.20 Input selection 2 7
Fixed value 1 Fixed value 2
1 8
0...4095 0...4095
(according to table) (according to table) 0 9
oP. 18
Step value
rotation source
Fixed frequency 1 Step value 2
oP.21: oP.22:
-4000...4000 rpm -4000...4000 rpm
Step value 3 7
oP.23: -4000...4000 rpm
Setpoint limits
Setpoint
f3
Fixed value f1...f3 f2
f1
t
The rotation source is determined with oP.18 atactive fixed value . The function and the value range correspond
to oP.1.
The three step values oP.21...23 are set-programmable and can be adjusted in the range of -4000...4000 rpm.
-100 oP.6
min. setpoint forward
Setpoint [%] oP.7*1 (oP.6) +100 Setpoint [%]
min. setpoint reverse
-n [rpm]
*1
If the value "=For" is adjusted in these parameters (limit values rotation direction reverse), then the adjusted 7
values for rotation direction forward (oP.06, oP.10 and oP.14) are valid.
In case of analog and percentaged setpoint adjustment in percent the minimal and maximal frequencies form
the characteristic for the frequency calculation (0% = min. reference; 100% = max. reference). In case of digi-
tal setpoint adjustment or fixed value the minimal and maximal frequencies limit the setpoint. Separate limits
can be adjusted for both rotation directions. If the value "For" is adjusted for rotation direction "Reverse", then
the values for "Forward" are valid.
Max. output val. forward (oP.40) / max. output val. reverse (oP.41)
All other limitations (oP.10 / oP.11 "max. reference" and oP.14 / oP.15 "abs. max. reference") limit exclusively
the speed setpoint .
This function is only active if in parameter Ec.42 "encoder alarm mode" the alarm for the used
encoder channel is activated (alarm = on). In vector-controlled operation without speed feedback,
speed limiting is always active.
The state „58: ERROR! Overspeed“ (E.OS) is triggered, if ru.07 "actual value display" exceeds either the value
of oP.40 / oP.41 "max. output val." or the value of ru.79 „abs. speed EMC“ (only for synchronous motors).
The user defines limits with oP.40/oP.41 that may not be exceeded by the application under any circumstances.
ru.79 shows the maximum speed for a synchronous motor which, if exceeded, leads to an EMC of the motor
high enough to damage the DC-intermediate circuit of the inverter.
Reason for the occurence of excessive speed can be too small a distance between the maximum setpoint and
the speed limit, so that overshoots can trigger the error. Other causes can be (e.g., caused by EMC) malfunc-
tions in the speed measurement or a noisy, insufficiently smoothed speed estimate in the encoderless control
(SCL or ASCL).
-100% +100%
-n [rpm]
With the adjusted setpoint limits the speed range 0%...100% is defined. In this case the adjustment of 0%
corresponds to the minimal speed and 100% to the maximal speed.
The speed after the setpoint limiting is calculated according to following formula:
oP.10-oP.06
positive setpoint = oP.06 + (setpoint setting [%] x −− )
100%
oP.11-oP.07
negative setpoint = oP.07 + (setpoint setting [%] x −− )
100%
The absolute setpoint adjustment, i.e. the setpoint is directly adjusted as speed and limited through the
corresponding minimal and maximal values as well as through the absolute maximal values.
Setting ranges are faded out with this function, in order to avoid resonances. The target is pass through with
the ramp. The setpoint value is always adjusted to the upper or lower limit of the target.
Parameter:
The ramp generator assigns an adjustable time to a speed change, during this time the change stall take
lace. The acceleration time (for pos. speed changes) and deceleration time (for neg. speed changes) can be
preset separately for each direction of rotation.
The different ramp functions can be adjusted separately for every frequency change (acceleration forward,
deceleration forward,...etc.). The selection is made with oP.27 and is adjustable separately in each set.
Mode „constant ascent “ concerns to the standard ramp generator with defined acceleration, deceleration and
jerk values (see chapter 7.4.7.2).
Mode „constant time “is needed only in exceptional cases, if acceleration / deceleration shall be executed al-
ways indepent of the setpoint in the same time (see chapter 7.4.7.3).
Mode „ogive run “is a special form of the mode „constant ascent “, which is particulary suitable for lift and tra-
versing drives (see chapter 7.4.7.4).
The more exact explanation of each operation mode is done in the respective sub-chapters.
oP.27, Bit 8
If the function is activated, oP.10 is considered as reference value for the modes "constant ascent" and "ogive
run" for forward direction and oP.11 for reverse direction. The specified reference value applies furthermore in
the mode for constant time.
Limitations:
The scaling factor of the ramp time is very comprehensive. Therefore all involved parameters are accepted
neither for analog setting nor as process data.
Since this is not valid for oP.10 and oP.11, there are the following restrictions for compatibility reasons:
This mode is the KEB factory setting. The acceleration / deceleration values are defined with parameters oP.28
to oP.31.
The jerk (i.e. the permissible acceleration / deceleration change) is defined with parameters oP.32...oP.35 and
oP.70... oP.73.
The linear ramps are parameterized with parameters oP.28 „acc.time for.“, oP.29 „acc. time rev.“, oP.30 „dec.
time for.“and oP.31 „dec. time rev.“.
1000
Rotation direction Rotation direction
forward
500
∆n ∆n oP.29 *1 oP.31 *1
∆t ∆t ∆t ∆t
0
oP.28 oP.30
reverse
∆n ∆n
800
1000
-n [rpm]
oP.28 Acceleration time forward *1 If the value "=For" is adjusted in these parameters (accelerati-
oP.29 *1
Acceleration time reverse on and deceleration times for rotation direction reverse), then
the values of rotation direction forward (oP.28 and oP.30) are
oP.30 *2 Deceleration time forward valid.
1
oP.31 * Deceleration time reverse
∆n Speed change *2 If the value "=Acc" is adjusted, then the value of acceleration
∆ t Acceleration time for ∆n forward (oP.28) is valid.
5 s * 1000 rpm
oP.28 = = 5,56 s
900 rpm
For some applications it is of advantage when the drive starts and stops jerk-free. This function
is achieved through a straightening of the acceleration and deceleration ramps.
Parameters oP.32 „s-curve time acc. for.“ to oP.35 „s-curve time dec. rev.“, and oP.70 „s-curve up time acc.for.“
to oP.73 „s-curve up time dec. rev.“ define the time for acceleration from 0 to the maximum value and decele-
ration from the maximum value to 0.
The maximum value for acceleration / deceleration is defined by the linear ramp times oP.28... oP.31.
1000
Rotation direction Rotation direction
forward
oP.29 oP.31
oP.28 oP.30
-1000
7
-n [rpm]
At the ramp with constant time oP.28… oP31 adjusts the time the inverter accelerates from speed 0 to the ac-
tual setpoint (ramp mode = 1) and/or decelerates from the last setpoint to speed 0 (ramp mode = 2). Then the
acceleration / deceleration time at start/stop operation is independent from the setpoint. In this operating mode
s-curves are not possible.
oP.28 oP.30
Setpoint
time time
Attention Ramp mode „constant time / actual setpoint“ should always be selected for acceleration and "con-
stant time / last setpoint" for deceleration.
The other adjustments are programmable and can be used if it shall be operated between diffe-
rent setpoint speeds (except 0).
When starting from 0 and/or deceleration to 0, they have the following effects:
If the mode „constant time / actual setpoint“ is selected for deceleration, deceleration is calculated
to:
That means: the drive don´t decelerate, it keeps running with the last setpoint before stop com-
mand.
Minimum acceleration / deceleration is limited programatically to:
That means: the drive would not continue to run constantly, but it decelerates very slowly.
In the mode "constant ascent", a change in setpoint while the inverter is still in the acceleration / deceleration
phase will lead to the fastest possible response.
If the new setpoint requires e.g., a change from acceleration to deceleration, the acceleration ramp is inter-
rupted and the deceleration ramp is started immediately.
This can lead to an undefined jerk. 7
If ogive run is selected, the programmed s-curve times are always used, the acceleration / deceleration change
continuously and no undefined jerk occurs.
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp setting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
The controlled operation can be carried out with or without motor model:
7
- Vector controlled operation with motor model without encoder feedback (ASCL)
During vector controlled operation of an asynchronous motor without encoder feedback (Asynchronous Sen-
sorless Closed Loop => ASCL), the speed is estimated with a mathematical model of the asynchronous
machine.
Standard version F5A does not contain operating mode ASCL. It needs the special software F5H.
7.5.1.1 Rated frequency (uF.00), boost (uF.01) and delta boost (uF.04 / uF.05)
The voltage/frequency characteristic (U/f) is adjusted with the rated frequency (uF.00) and the boost (uF.01).
The rated frequency adjusts the frequency at which 100 % modulation depth (~input voltage) are achieved. The
boost adjusts the output voltage to 0 Hz. Depending on uF.10 the modulation limit can be further increased in
this stage up to 200 % (see Fig.7.5.1.1).
The delta-boost is a time-limited boost used to overcome large breakaway torques. The delta-boost acts adding
to the boost; but the sum is limited to 25.5 %.
uF.04
uF.01
t
uF.05
By changing the maximal voltage mode more torque can be released free above the rated frequency through
overmodulation (110% voltage). Raising the v/f-characteristic has an influence at activated energy saving func-
tion or at voltage stabilisation.
To adapt the V/f-characteristic to special conditions an additional rated point can be specified with uF.02 and
uF.03. uF.02 defines the frequency and uF.03 the voltage. At uF.02 = 0 Hz the adjustment is ignored. The
parabolic characteristic is activated with uF.02 = „-1: parabolic characteristic". Then parameter uF.03 has no
function.
100%
uF.03 7
uF.02 = -1: Parabolic characteristic
0,0...400 Hz; Default = 0,0 Hz
uF.03 = 0,0...100,0 %; Default = 0,0 %
uF.01
fout
uF.02 uF.00
The DC link voltage and thereby the directly dependent output voltage can be changed by fluctuations of the
mains voltage or the load. In the case of enabled voltage stabilization the fluctuations of the output voltage are
compensated. I.e., 100% output voltage correspond to the value set in uF.09 , but maximally 110% · (UZK / √
2), depending on the setting of uf.10. This function further allows operation of motors with a low nominal voltage
at the inverter.
f
CP.16=50 Hz
uF.00 = 50 Hz
Motor voltage
Utilisation
Utilisation
Motor speed
Zwischenkreisspannung Zwischenkreisspannung
Motorspannung Motorspannung
Auslastung Auslastung
Information on the carrier frequencies can be found in chapter 7.10.3 "Carrier frequencies and Derating".
The energy saving function allows the lowering or raising of the current output voltage. Corresponding to the
activation conditions defined in uF.06, the voltage corresponding to the V/f characteristic is scaled by the energy
saving factor (uF.07).
If torque compensation is active (s. chapter 7.5.1.7), the energy saving function is used for control optimization.
The V/f characteristic will then not be affected.
The maximum output voltage cannot be higher than the input voltage, even for a factor > 100 %.
The function is used for example in cyclic executed load/no-load applications. During the no-load phase the
speed is maintained, but energy is saved as a result of the voltage reduction.
The SMM-function (sensorless motor management) includes the torque and slip compensation. These two
functions can be activated separately. For an optimal control characteristic, the combination of both functions
is required.
Setting the correct motor data is required, since they are used in calculations needed by the inverter to achieve
the best possible results in the control of boost and slip.
Torque compensation
Torque compensation adapts the voltage at variable load torques in such a way that the magnetizing current is
kept constant. With it a higher maximum torque is achieved at small output frequencies compared to uncom-
pensated operation (block diagram see chapter 7.5.3.).
Following parameters can be taken directly from the name plate and entered:
!! Parameter dr.00 and dr.02 must be adjusted according to the used connection (star/delta).
The following parameters can be taken from the corresponding data sheet or can be determined from measure-
ments:
The stator resistance can either be measured with an ohmmeter or determined automatically.
In this way the ohmic line resistance is registered simultaneously (important in the case of long incoming lines).
For the measurement with an ohmmeter, the connection between motor and inverter has to be broken. The
measurement is carried out on a warm motor, between 2 phases of the motor feed cable, independent of the
motor wiring (Δ / Y). For a more accurate result, all 3 values (U/V, U/W and V/W) should be measured and the
values then be averaged.
The automatic determination can be carried out for each parameter set separately. Thus a parameter set can
be programmed for example as "Warm-up set" for particularly critical applications.
During the determination the status display (ru.00) indicates "Cdd". Upon successful determination the motor
stator resistance is entered in dr.06. If an error occurs during determination then error signal "E.Cdd" is output.
After input of the rating plate data of a new motor or after automatic measurement of the stator resistance, an
automatic optimisation of the torque and slip compensation can be carried out with Fr.10 (see chapter 7.5.1.7).
The optimisation is started by writing value "3" on Fr.10. Thereby the inverter must be in status „noP“ (no control
release). Provided that only one motor is used, the measurement can occur with direct set programming for all
parameters at once.
The adaption covers approx. 90 % of the applications. For an application-specific adjustment a manual fine
adjustment can now still be carried out for an individual case.
7.5.1.7.4 Adjustment of the slip compensation (cS.00, cS.01, cS.04, cS.06, cS.09)
The integrated speed controller is used at cS.00 = 2 for slip compensation. The rotor speed calculated from the
motor model is selected as the actual controller value by cS.01 = 2.
The slip compensation can be configured with bits 3-6 in cS.00.
During the standard slip compensation, the slip is proportional calculated from the effective current. This calcu-
lation becomes imprecise above the nominal setpoint and in generatoric operation.
For the improved slip compensation, the slip calculation during motor operation above the nominal setpoint is
approximated to the real M/n-characteristic with a parabolic function. Greater inaccuracies will then occur only
above twice the rated torque.
During generatoric operation, the linear dependency is preserved. The steepness of the characteristic can be
adjusted with cS.03.
With uF.16 and uF.17 the torque compensation is activated and configured.
Magnetising current setpoint and actual value are calculated in the motor model.
ATTENTION! Through overcompensation increased motor currents can occur particularly with small frequen-
cies.
Vector controlled operation is activated by inputting the values 4, 5 or 6 into the category "control mode" of the
parameter "controller configuration" (cS.00).
The following adjustments are required in speed-controlled operation for all modes (with / without encoder re-
spectively with / without motor model):
Input of the motor rating plate data is at the beginning of each start-up:
After input of this data, the operator must switch to closed-loop operation (cS.00 = 4) and enter Fr.10 = 1 or 2
(explanation see below) once.
The inverter must be in status "noP", i.e., the input "control release" (ST) may not be set. Thus the following
parameter are pre-charged dependent on the motor and inverter data:
Definition of magnetisation:
- dr.19 Flux adaption factor
- dr.20 Field weakening curve
Current controller
- dS.00 KP current
- dS.01 KI current
Torque limits:
- cS.19 abs. torque ref
- cS.20...cS.23 torque limit (clockwise rotation motor operation, counter clockwise rotation motor operati-
on, clockwise rotation generator operation, counter clockwise rotation generator operation)
- Pn.61 quick stop torque limit
Flux controller:
- dS.11 KP flux
- dS.12 KI flux
- dS.13 magnetising current limit
Inertia:
- cS.25 inertia (kg x cm2)
Speed controller (preloaded only if automatic speed controller setting is activated by cS.26 ≠ 0):
- cS.06 KP speed
- cS.09 KI speed
Some of these parameters (e.g., the limiting characteristic) depend upon the available voltage.
During vector controlled operation, the voltage stabilization generally should be "off". The software-integrated
current controllers control the voltages and a simultaneous intervention of the voltage stabilization increases
the system´s vibrational tendencies.
With Fr.10 = 1, precharging occurs dependent on the voltage class of the inverter (400V or 230V)
The actual DC link voltage of the frequency inverter, which is proportional to the supply input voltage is consi-
dered for the calculations at Fr.10 = 2.
If the parameter „voltage stabilisation“ (uF.09) is not set to the default value „1120: off“, then the value in uF.09
is taken as reference voltage for the calculations for settings Fr.10 = 1 or 2.
If the drive is to be operated at a different voltage then during initial start-up, proceed as follows:
In parameter uF.09, enter the nominal voltage to be used later, activate Fr.10 = 1 and reset parameter uF.09 to
"off".
Attention:
After completion of a possible "fine tuning", i.e., the manual adjustment of controller parameters, torque limits,
etc., parameter Fr.10 may not be activated anymore. Otherwise, the manually adjusted parameters will be
overwritten by the calculated values!
The actual value source for the speed must be selected in parameter cS.01.
Possible values for drives with speed encoder are 0 (speed measurement via encoder interface channel 1) or
1 (speed measurement via encoder interface channel 2).
Description of the correct parameter setting of the encoder interfaces is made in chapter 7.11 "Speed measure-
ment".
If operation without tachometer generator is desired, cS.01 = 2 (calculated actual value) must be selected.
This setting is only possible for open-loop V/f-characteristic operation (for software type F5-A) or for control via
motor model (for software type F5-H and F5-E, respectively).
With activation of the system inversion it is reached that the motor with selected rotation direction "forward"
(e.g. by setpoint- or rotation setting) has the physically direction "reverse" respectively at setting "reverse" the
physical rotation "forward". Precondition is a correct wiring of motor and speed feedback (if available).
One possible application of this function is, e.g., the deployment of 2 drive units, where facing motors drive the
same shaft. If system inversion is activated for a drive, the same setpoint can be set for both via a control, even
if one motor rotates clockwise and the other counter clockwise.
For applications with encoder feedback, the same function can be activated by switching on system inversion
in parameter Ec.06 (see chapter 7.11).
For motors that don´t allow identification of the motor data (e.g. no-load operation of the motor not realisable),
vector controlled operation without motor model must be selected.
Parameters dr.06...dr.10 have no function in speed-controlled operation without motor model. If the drive shall
be operated with motor model, chapter 7.5.2.2 can be skipped.
The slip is affected significantly by the rated speed in speed-controlled operation without motor model. If the
drive requires too much current for a certain load, or if is can be seen that the output voltage at high load gets
too small, an incorrect (too low) rated speed may be the cause.
In this case, the rated speed must be adjusted in small increments until the optimum is found.
Since the motor voltage is proportional to frequency * flux, the flux must be lowered according to a 1/x function
above the rated point (maximum voltages reached) to keep the voltage constant.
In the base speed range of the motor, the maximum torque is limited by the current the inverter is able to supply.
In the field weakening range, the achievable torque is additionally limited by the voltage.
Since the motor parameters, like main inductance, change in the field weakening range, the flux does not follow
the desired 1/x-characteristic during control without motor model in the field weakening range.
This change in the main inductance can partially be compensated for with the default setting of the amplification
factor field weakening (dr.20) of 1.2 instead of 1.
Motorfluss
1
Faktor Feldschwächung (dr.20) = 1
=> 1 / x Flussabsenkung
0,5
Faktor Feldschwächung = 1,2
dr.18 2 * dr.18
Istfrequenz [U/min]
For large motors, the automatic calculation of the magnetising current occasionally returns values that are too 7
large. This value can be reduced by adjusting the parameter "factor flux adaption" (dr.19).
Whether the automatically calculated magnetising current is too large, can be tested by accelerating the drive
to the field weakening speed (dr.18) in speed-controlled operation with no load. At this speed, the average va-
lue of the modulation factor should not exceed 90%. If this value is exceeded, the factor "factor flux adaption"
(dr.19) should be reduced.
7.5.2.3 Vector controlled operation with motor model (with encoder feedback)
The vector controlled operation with motor model is possible only if the electrical characteristic data of a motor
are known. For this operating mode, the motor model calculation must be activated in parameter dS.04.
The electrical characteristic data of the motor must be known for vector controlled operation with motor model.
The parameters DASM stator resistance (dr.06), DASM leakage inductance (dr.07) and DASM rotor resistance
(dr.08) can be taken from a motor data sheet or they can be automatically determined by the KEB COMBIVERT
using the motor identification. For motors with high power, the resistances are very small (a few mΩ). This can
lead to error in the automatically identification. For these motor, it may be sensible to use the value from the
The required equivalent circuit data for the motor model can be determined by the KEB COMBIVERT itself.
First the motor data must be entered and the motor adaption must be executed according to chapters 7.6.1.
- Writing of parameter dr.48 in inverter state "stop (mod. off)", measurement is starting auto-
matically.
- Writing of parameter dr.48 in inverter state "nop" with subsequent control release.
The direction of rotation during identification of the main inductance is always "clockwise rotation"!
Value „82: calculate drive data / Cdd“ is output in inverter state ru.00 during the measurement. After success-
ful measurement the display is ru.00 = „127: drive data calculated / Cddr“.
If the measurement is interrupted with an error, ru.00 = 60 ERROR! drive data/ E.Cdd" is displayed.
If the internal brake handling is used in the application, then it must be deactivated for the identification. For
safety reasons the output signal "brake release" is not set during measurement, since the motor cannot gene-
rate a defined torque in this time. Stator resistance, rotor resistance and leakage inductance can be measured
also at engaged brake.
For the identification of the main inductance, the drive must be decoupled from the load and the output swit-
ching condition which is assigned to the break control must be set to value "1" (= always active), setting the
brake permanently open.
Use generally the automatic mode for the identification of the parameters.
Automatic mode is the most simple method of parameter identification.
Measurement of the dead time compensation characteristics, as well as the stator- and rotor resistance and
the leakage inductance is done in standstill. A small rotation of the motor caused by the test signals is possi-
ble.
It is necessary for the identification of the main inductance, that the motor accelerates to the speed for maxi-
mum torque (dr.17) and then it operates in no-load operation.
There is a special ramp "motor identification ramp time" (dr.49) for identification.
This ramp applies during calibration of the main inductance for the acceleration to dr.17 and the deceleration
at the end of the identification.
7
The speed controller must be parameterised reasonable (select small Ki), the drive may not vibrate during
the identification.
The following chapter, "single identification", contains more detailed information with respect to the separate
steps of the identification and can be skipped if automatic mode is chosen. In the chapter after the next, "ad-
ditional trimmings", two further identifications are described which are not part of the automatic mode and that
are unnecessary in many cases.
The explanations of the parameters required to be set continues in chapter 7.5.2.3.3 "generally required set-
tings for operation with motor model".
Single identifications should not be used for the first measurement of the motor parameters, since invalid
measuring results can occur in case of a wrong identification sequence or omitting of individual points.
Single identification can always be used if a complete automatic measurement was executed and only indivi-
dual parameters shall be identified. For example this can be a resistance measurement in warm condition or a
new measurement of main inductance after changing parameter dr.19 "factor flux adaption".
192: 15,625Hz
224: 7,8125Hz
* at dr.48 = 8 auto-identification
With dr.48 = 1 (calculation of the main inductance (ASM) / EMC(SM)), a starting value for the main inductance
is calculated from the rating plate data.
Measurement of the leakage inductance (dr.07) occurs at standstill with a test signal. The frequency of the
measuring signal is adjustable via bit 5...7 in parameter dr.48.
Since the inverter determines automatically the ideal measuring frequency, value 0 should be always selected
for bits 5... 7.
7
Stator resistance measurement (dr.48 = 3)
Measurement of the rotor resistance (dr.08) occurs at standstill with a test signal. The frequency of the measu-
ring signal is adjustable via bit 5...7 in parameter dr.48.
Since the inverter determines automatically the ideal measuring frequency, value 0 should be always selected
for bits 5... 7.
Since the measurement frequency occasionally has to be reduced to 7.8125 Hz for better measurement accu-
racy, the motor may rotate.
With dr.48 = 5, the internal model parameters as well as current-, flux- and speed calculation controller parame-
ters are calculated from the equivalent circuit data. If a mode other than automatic is used for the identification,
this action should be executed after the measurement of the leakage inductance, rotor and stator resistance,
but before the identification of the main inductance, in order that the controllers are correctly parametrised for
the speed ramp-up.
It is necessary for the identification of the main inductance that the motor accelerates to the speed for maximum
torque (dr.17). The speed controller must be parameterised reasonable (select small Ki), the drive may not
vibrate during the identification.
The motor must be able to rotate in no-load operation. After the main inductance has been identified, the drive
stops automatically.
There is a special ramp "motor identification ramp time" (dr.49) for identification. This ramp applies for accele-
ration at the beginning and deceleration at the end of the identification.
The dead time detection only works as single identification if the stator resistance is correct preset.
The measured dead time-values can be read out via In.39 and In.40.
The calibrated dead time compensation characteristics are in force if uF.18 = 3 is set.
In error case parameter "motor identification error code" (dr.66) displays the reason for this error:
For applications with particularly high demands on the accuracy of the torque display, this can be calibrated.
As a standard, the torque display does not display a value of 0 in encoderless operation during no-load opera-
tion. The reason for this is switching frequency-dependent losses in the inverter and friction losses due to the
application.
If the torque display has to be corrected for this offset, the torque offset of the whole drive can be calibrated with
dr.48 = 14...18 for the various switching frequencies. 7
Thereby the drive accelerates in stepwise with the adjusted ramp in dr.49 to maximum 1,3-fold synchronous
speed. The speed limits set in the oP-parameters remain operative during this phase.
The calibrated no-load torque is stored as correction characteristic. During operation, the display of the actual
torque in ru.12 is corrected using this characteristic.
The torque offset-characteristic can be read with parameter dr.58/ dr.59.
The characteristic is not part of the data backup created by read out of a complete list.
This should be executed only if the application really requires increased torque accuracy. Since the trimming
values are not contained in the complete list, porting the data to a different inverter is labour-intensive.
As a standard, the current offset from the inverter is permanently ascertained and balanced, as long as the
modulation is switched off. Therefore, the current offset-detection via dr.48 is usually not required.
In some cases, one achieves more accurate current offset values if one carries out the trimming with current
in the motor.
If dr.48 = 19 is selected, the inverter provides a test signal to the motor and so carries out the trimming once.
A disadvantage of this current offset detection is that it is carried out only once and therefore temperature and
ageing effects are not taken into account.
To preserve the identified offset, automatic measurement is deactivated with dr.48 = 19.
ATTENTION! Since the automatic measurement can only be reactivated by the KEB service personnel, the
current offset detection should preferably be carried out only in consultation with KEB.
The drive is only ready for operation after switching the modulation if the flux is build up. If one starts earlier,
the drive can display undefined behaviours (erroneous torque display, too high currents, poorer controller be-
haviour).
Bit 7 in dS.04 ("Wait for magnetisation (ASM)") must therefore always be set (value 128). Thus the setpoint
setting is only released if the flux is build up to 95%.
100%
Fluss / Nennfluss
80%
Sollwertanzeige (ru.01)
60%
Reglerfreigabe (ST)
20%
0%
0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1
Zeit [s]
The flux controller must be activated for the operation with motor model.
The parametrisation of the controller (KP flux / dS.11, KI flux / dS.12, limit magnetising current / dS.13) is carried
out automatically by Fr.10 and after the motor identification (dr.48).
During operation with speed feedback, the flux controller must be activated over the whole speed range, i.e.,
the value 32 must be chosen in dS.04 in the item "flux control".
Value 64 or 96 should be selected during operation without speed feedback.
With Fr.10, the parameter dS.13 "Limit magnetising current" is set to half the rated motor current. If the flux
build-up time is to be shortened or if particularly high demands are made on the dynamics in the field weakening
range, this value can be changed to the rated motor current (dr.00).
The inverter can only provide the standstill current at speed 0. Error OL2 is released shortly if the current is
higher. Thereby this can lead to problems during magnetizing at some motor/inverter combinations. In these
cases, the setting dS.04 Bit 5, 6 = 64 "flux controller not active during boot" must be chosen.
The drive has also measured the dead time compensation characteristic during automatic identification. The
calibrated characteristic must be activated for the control with motor model by the setting "dead time compen-
sation mode" (uF.18) = 3: „automatic“. „automatisch“ aktiviert werden.
Further available kinds of the dead time compensation are only required for special applications (applications
with high frequencies, some special motors) or in other operating modes (e.g. V/f characteristics controlled).
The dead time compensation can be switched off via a digital input. The digital input is selected with parameter
uF.21. This disconnection is only required for special applications with high frequency.
For large motors, the automatic calculation of the magnetising current occasionally returns values that are too
large. This way, the dynamic operation in the field weakening range may worsen.
Whether the automatically calculated magnetising current is too large, can be tested by accelerating the drive to
the field weakening speed (dr.18) with no load. At this speed, the voltage limit (modulation factor 100%) should 7
not be reached yet. Othwise the „factor flux adaption“ (dr.19) should be reduced until the modulation factor is
approx. 90 - 95%.
Subsequently, a new identification of the main inductance must be carried out (dr.48 = 6) and the controller
must be adapted to the new main inductance.with dr.48 = 5.
The new "factor flux adaption" must then be checked with a new ramp-up.
Attention: If the factor is reduced too much, the available voltage will not be fully exploited anymore (modulation
factor ru.42 even for high speed and a load always smaller than 95%), and the motor current increases!
A special start-up must be executed for motors with output frequencies > 200 Hz.
The rated speed is not be indicated on the type plate of the spindles. When driving with motor model this rated
speed ist only important for calculation of the pole-pair number and the model tripping level in parameter ds.19
(default value, 2*slip speed).
If no value is indicated here, then 98.5% of the synchronous speed can be accepted.
nn = fn * 60 * 0,985 / ppn
fn = rated frequency
nn = rated speed
The output frequency should not exceed 1/10 of the switching frequency.
Thus the following applies:
switching fre- max. output frequency output speed (for pole-pair number = 2)
quency
2 kHz 200Hz 6000 rpm
4 kHz 400Hz 12000 rpm
8 kHz 800Hz 24000 rpm
12 kHz ---- ---
16 kHz 1600Hz 48000 rpm
Note: That a spindle e.g. with leakage inductance dr.07 = 1.4 mH and output frequency of 800 Hz (24000 rpm)
can be driven in practice also with 4kHz should not be considered for the dimensioning of the inverter.
An additional voltage vector can be output for 8 and16 kHz. The current controller is calculated only all 12 µs,
but the transformation angle is changed by 62.5 µs
(dS.18 bit 6 = on ).
If the motor model is activated, HCL generally should be deactivated via uf.15 = 0 =off.
• Main inductance:
There can be problems at measurement of the main inductance in lower speed range and when reaching
the target speed.
Measurement of the main inductance is started with a value for the inductance which is calculated from
the motor data. The lower speed range must be passed through speedy because the mode of calculation
can only be estimated and additionally the motor data of the manufacturers are problematically . For this
the additional ramp in dS.21 and dS.22 make sense.
• Leakage inductance:
Measurement of the leakage inductance can lead to wrong values, if the inductance of the motor has a
pronounced current dependence. The current has a „bell-shaped“ curve at setting of sine-wave voltage.
This can often be observed at spindle motors. The magnetising current can be watched via Combivis in
ru.87 during the rotor resistance measurement, to find out whether the motor to be start up is concerned.
Depending on the deviation of the current from the sine form the inductance is reduced to 85..70% of the
identified value.
A correction of the current controller adjustment is eventually necessary if the identified leakage in-
ductance (ds.07) is < 1,4mH:
The motor model (dS.04 bit 0) must be active for this. Control to the model current (dS.18 bit 3) has the advan-
tage that disturbances of the measured currents enter filtered to the motor model and thus the calculated model
current and the current control are smoother. The disadvantage is in the danger that the drive could change
to OC when there is a difference between measured and calculated current. Thus pay special attention to the
identification of the equivalent circuit data.
Observer:
The observer (dS.18 bit5) adjusts the model currents dependent on the measured currents by the adjusted
factor in parameter dS.23 „observer factor".
If the drive shall be accelerated at torque limit, it is mandatory necessary that the max. breakdown torque is not
exceeded. For this parameter (dS.03 bit 1) must be activated and the breakdown torque must be entered at
corner speed (dr.18) in parameter dr.16.
The corner speed (dr.18) should be set to rated speed (dr.01). If the breakdown torque should not be arise from
the data sheet, the breakdown torque must be calculated from the equivalent circuit data.
The dead time compensation is mandatory required up to 200 Hz, above 200 Hz it should be switched off. For
this a digital software output e.g. (do.04 = „27: actual speed > level“) level (LE.04 = 12000 rpm (ppz=1) with
hysterese LE.12 = 500 rpm and the input function uf.21 „dead time compensation off“ = 256 must be assigned
to the corresponding software input. With uf.25 in version 4.1 or 2.1 you can define a time where the dead
time is soft switched off.
In order to unload the current controller at high output frequencies there are two possibilities:
After the PI speed controller, a PT1 element can be activated via parameter cS.29.
A PT1 time of 2...8ms is recommended. Starting from version 4.1/2.1 parameter cS.29 is considered by the
mass-moment of inertia in the calculation for the controller parameter of the speed controller.
A maximum current can be preset in dr. 37 in order to protect the inverter against overcurrent error. dS.3 bit0
must be set additionally.
• Current ripple dependent on the switching frequency (ft) and leakage inductance (Ls). A calculation of this
part is possible, but very extensive.
Rough formula : I Ripple = 46,4 / ft / Ls * kHz * mH *A (Ls in mH / ft in kHz)
• overshoot of the current controller, approx. 10% of the selected maximum current.
• the „300Hz“ reload voltage ripple (at 50Hz mains frequency) in the DC link. Serves for a superimposed
current oscillation in the output frequency (=output freq. - 300Hz). This part is depending on many factors
(size of DC capacitors (C), line supply impedance, leakage inductance of the motor (LS), active power (Pw).
Rough formula : IRipple_dc = Pw2 / C / Ls * 105 * µF * mH / kW2 (Pw in kW / Ls in mH / C in µF
This part can be reduced by an input choke: 15% reserve to the OC level.
Example:
Calculation of the OC level of the inverter and maximum current (dr.37) if the motor shall be accelerated with
preset torque.
Parameterizing of the maximum current: Imax (dr.37) = Mmax/Mn • In_mot = 1.65 • 30 A = 49.5 A
OC_Level = 83 A
A 18.F5.H with 8 kHz rated frequency and 90 A OC levels must be selected for this spindle with 30 A rated cur-
rent and 14.5 kW rated power.
This chapter must be read only if an asynchronous motor without speed feedback is to be operated. Since the
speed can be calculated only with the aid of a mathematical model, this operating mode may only be used with
the following limitations:
The operation at small speed is a critical range which should be passed very quickly.
The size of this range cannot be indicated universally valid. It is strongly dependent on the used motors.
The usable speed range for standard-asynchronous motors is approx.:
In order to leave the critical range of small speed at starting and stopping there is an additional ramp for this
range.
The ramps is defined by parameter dS.21 "auxilliary ramp/ speed limit" and dS.22 "auxilliary ramp/time".
Parameter dS.21 indicates the speed range for which the start ramp applies. dS.22 indicates the acceleration-/
deceleration time.
Anzeige Rampenausgang
( ru.02 )
Example:
0,2 s Zeit
If the drive is to be stopped, the critical range of low frequencies must be passed again.
The additional problem of the drive not stopping completely, but instead running permanently at a low frequency
with a very high current occurs here, leading to a miscalculation of the speed.
Under the following conditions, therefore, the mode is switched from vector controlled to current regulated,
frequency controlled operation:
- Drive decelerates
- the estimated output frequency is smaller than dS.19 ("speed limit model switch-off DEC")
The parameter dS.19 is loaded with a default value by the identification or by Fr.10 "reset field-oriented control-
ler parameters". Should problems still occur during deceleration, the value for dS.19 can be increased.
If the drive is stopped by switching off the rotation direction release, the modulation is switched off after rea-
ching of output frequency 0.
If the drive is stopped by setting the setpoint to 0, the current is reduced to the magnetising current after rea-
ching the output frequency = 0.
At this point, the real speed of the motor is not yet 0 in some cases.
Therefore, the time for which the higher constant current is set can be increased with parameter dS.20 "ASCL
model switch-off current follow-up time".
dS.19: Limit uf-control decele-
ration
=> Switching to the frequency
controlled range
dS.20: ASCL
Model switch-off
Current follow-up time
Apparent current (ru.15)
Delay current
Speed (ru.07)
Magnetising current
Frequency (ru.03)
time
7
Attention: The torque display (ru.12) is invalid after change-over to frequency controlled operation!
ASCL / reversing
If one wants to run the drive through zero speed without stopping to change (reverse) the direction of rotation,
switching to the frequency controlled mode can be disruptive.
Therefore, this switching can be deactivated by setting bit 2 in parameter "model adjustment" (dS.18).
To utilise the open-loop mode for stopping, but, on the other hand, avoid negative effects during reversal, the
inverter must be programmed by way that stopping of the motor always follows in the same set.
Then, one can let the switch to the open-loop mode be activated for this set (the stop-set) (dS.18 = 0), and avoid
interfering effects during reversal for other sets with dS.18 = 4.
The, it is only necessary to ascertain that the range of low frequencies is traversed quickly.
This can be achieved by suitably setting parameter "ASCL start ramp time" (dS.22) and parameter "ASCL start
ramp speed" (dS.21), which apply to acceleration as well as deceleration.
If the motor still rotates when switching the modulation (e.g. "coast down" after malfunction) the calculation of
the actual speed can become unstable due to the motor model.
Therefore, if there is a risk that the motor has not reached speed 0 for the start, there are two alternative starting
methods:
speed search (Pn.26) or DC braking (Pn.28 / Pn.33)
During speed search, the drive attempts to determine the current speed via its mathematical model. The ope-
ration corresponding to the setpoint settings is re-established starting at this speed. For many standard motors,
this type of addition can be used.
For some motors or applications, e.g., for spindles, application of the speed search will be unsuccessful. In
these cases, speed is calculated incorrectly, the drive can vibrate, or the inverter can malfunction.
In these cases, the motor must be stopped by DC braking before the drive can be restarted. During DC braking,
a DC voltage is connected to the motor´s clamps. The small braking torque while the motor is still running at
high speed is a disadvantage.
For more (appropriate parameters, settings, etc.) see in 7.13.4 speed search and 7.15.1 DC braking, respec-
tively.
In some cases, the one-time current offset measurement (either with modulation switched off or via test signals
during motor identification) is insufficient since operation-dependent effects (like temperature) are not consi-
dered. With this "residual offset", a vibration is created with a frequency equal to the output frequency. The
current offset adaption can reduce this effect.
Attention: Is the simple vibration not caused by the current offset, the adaption is behaving incorrectly. There-
fore, this function must be activated with caution, or only to prove a current offset existence and to utilise its
value.
The stator resistance can stabilise the model at low output frequencies, particularly in generatoric operation.
At low motor rating, the effect of the stator resistance in this range is quite large. Due to the motor warming,
changes of up to 40% compared to the resistance calibrated in the cold state are possible. The stator resistance
adaption can compensate for this change.
Under certain operating conditions, (e.g., high dynamic) the adaption diminishes the operational performance
of the drive. Therefore, this function should only be activated when problems with breaking and stopping may
occur for motors with small power (< 5 kW).
For the current control, either the measured currents or those calculated from the model can be used as actual
values. As a standard, the measured currents are used for control since only this assures direct control over
the real currents.
Using the calculated currents is advantageous only in high frequency applications: The delay (detection of the
actual current until the output of the voltages as response to the current measurement) is noticeable in these
applications. For control based on calculated current, this time is minimised.
The observer causes an equalisation between the measured currents and the currents calculated from the
motor model. This is useful for some high frequency applications.
The reciprocal of amplification of the observer is set with the parameter "observer effect / motor model" (ds.23).
At high output frequencies, the voltage vector must be calculated and output in a shorter time pattern. This is
possible only at 8 and 16 kHz. Important for high frequency applications
7.5.2.4.4 KP/KI Speed calc. ASCL (dS.14, 15) and speed PT1-time ASCL (dS.17)
The KP (dS.14) and the KI (dS.15) of the speed calculation controller are calculated automatically during the
identification of the motor parameters and may not be changed.
Only the parameter dS.17 "ASCL speed PT1 time" can be adapted to a specific application. In non-dynamic
applications, a higher PT1 time (up to 32ms for large motors) leads to a steadier calculated speed, without
degradation of the control characteristics of the drive.
In contrast, a lower speed frequently permits a more dynamic setting of the speed control parameters.
If parameter dS.17 "ASCL speed PT1 time" is changed, a previously conducted adaption of the speed controller
must be checked.
If the automatic calculation of the speed control parameters is used, it must be reactivated.
In speed control with speed feedback, the motor model can be used to adapt the rotor time constant. The rotor
time constant is dependent on the rotor resistance, among others. Due to the temperature change of the motor
rotor, the rotor resistance can change significantly compared to the identified value. This also changes the rotor
time constant. This change leads to a less accurate torque display and an inferior performance of the drive.
The rotor adaption compensates for the temperature deviations of the resistance. It is activated by bit 1 in pa-
rameters dS.04 "Flux-/Rotor adaption mode".
Bit 2 determines whether the drive stores the rotor adaption value on modulation switch-off. If memory ist ac-
tivated (memory: yes), the inverter starts with the last value obtained during operation after reactivation of the
modulation. If memory ist deactivated (memory: no), the inverter starts with the value 100%. After"net on",the
inverter always starts with the value 100%.
In parameter ru.59 "factor rotor adaption", the status of the rotor adaption can be read: 100% means that the
drive is working with the identified values. Values unequal 100% mean that the actual rotor resistance is = ru.59
* dr.08 (DASM rotor resistance).
Page 7.5 - 34
-
ru07 iu
ru.18
Transformation
Precontrol and and dead time Volt V Power unit iv
decoupling ASM
compensation
Flux controller isd controller 3ph
iw
7.5.3 Block diagram
Maximum voltage
Imr set
controller and field
weakening
isq reg
ru.87
Selection of the
current feedback Transformation
ru.17
© KEB, 2012-10
isq controller
ru02 Speed controller isq ref
and torque
limiting Usq Volt U
- ru.18
ru07 - iu
© KEB, 2012-10
and decou- and dead time Power unit
ASM
Flux isd controller
controller pling compensation iw
3ph
- IG
ru.03
ru.87
Transformation
ru.17
Ec-Parameter
Motor model and
Encoder eva-
luation
Page 7.5 - 35
7
Motor data and controller adjustments of the asynchronous motor
n F out trans-
f F out formation
V out
F ramp SMM V mag
slip compensation
F calc act autoboost
V
+
transformation
F slip I mag angle
F V/f +
f
V/f characteristic
uf.00 - uf.03
F out
(last cycle)
Iu
SMM V out (last cycle)
I app ^2 trans- Iv
motor model
formation
I act
n
f
omega_1
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp presetting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
Speed-controlled operation of synchronous motors without encoder feedback SCL (sensorless closed loop)
is only possible if the electrical characteristic data of the motor are known. The rotor position is emulated by
means of a mathematical model of the synchronous motor. Speed control is based on a speed calculated from
the rotor position rather than on the encoder feedback.
Standard version F5A-S does not contain operating mode SCL. It needs the special software F5E–S.
For SCL, also F5S eventually. Apart from the minimum value in dr.31 also the maximum value can be preset
from version 4.2. This parameter is also preset by the identification of the winding inductance.
First, it is only required for the calculation of the controller for the RF injection. (Normally the higher value of
inductance should be set in the q-component in dr.31. Motor model and current controller are however better
preset with the lower value. Servo motor manufacturer, as Wittur also indicate the lower value of inductance
as luv).
dr.26 = EMKeff x √ 2
No decimal places can be entered in parameter dr.26 for the EMC. The voltage per 1000 rpm at high-frequency
motors is partly very low, that an integer setting falsifies the EMC value. Parameter dr.63 (DSM EMC HR) can
be used for higher accuracy.
In order that there is a downward compatibility to older parameter lists, the parameter can be deactivated with
value "0: off".
The maximum permissible speed which is displayed in ru.79 (abs. speed [EMC]) is also calculated from the
EMC. The maximum DC link voltage, UZKmax, can be found in the power circuit manual.
Parameter cS.00 must set to value 4: "speed control" for closed-loop operation.
The actual value source for speed control must be selected in parameter cS.01 .
Possible values for
drives with speed encoder are 0 (speed measurement via encoder interface channel 1) or
1 (speed measurement via encoder interface channel 2).
Description of the correct parameter setting of the encoder interfaces is made in chapter 7.11 "Speed measure-
ment". cS.01 = 2 (calculated actual value) must be selected at operation without speed encoder (only SCL).
With activation of the system inversion it is reached that the motor with selected rotation direction "forward"
(e.g. by setpoint- or rotation setting) has the physically direction "reverse" respectively at setting "reverse" the
physical rotation "forward". Precondition is a correct wiring of motor and speed feedback (if available).
Fr.10 = 2 (for some applications Fr.10 = 1 /explanation see below) must be entered once after input of the motor
data.
The parameter can only be written in „nop“ status !
The calculation at Fr.10 = 1 is depending on the voltage entered in parameter uF.09 "Voltage stabilisation". If
this parameter displays "off" (standard adjustment), then the voltage class of the frequency inverter (400V or
230V) is used.
The current DC link voltage of the frequency inverter, which is proportional to the supply input voltage, is con-
sidered for the calculations at Fr.10 = 2.
However this only applies uF.09 is on "off".
Thus the following parameter are pre-charged dependent on the motor and inverter data:
Current controller
- dS.00 Kp current
- dS.01 Ki current
7
Torque limits:
- cS.19 Absolute torque reference
- cS.20...23 Torque limits clockwise- counter clockwise rotation/ motoring- generating
- Pn.61 Abnormal stopping torque limit
- dr.33 DSM max. torque
A reluctance torque can only be generated if there are differences in the inductance of the d/q axis. The in-
ductance (dr. 64) is indicated as Lsq and the inductance (dr. 31) as Lsd. This is particularly pronounced for IPM
motors (LSQ> Lsd). A current in the D-axis must be provided depending on the set torque in order to use the re-
luctance torque. The function IdRef = f (MRef) is too complex to be calculated in the inverter. It is calculated with
a KEB excel tool and stored as download in table cS.35. Please contact KEB to obtain the tool. If it is operated
in SCL mode (operation without encoder feedback) the presetting may not be activated. Here the magnetizing
current is automatically adjusted optimal by an angle deviation error in the estimation.
"
7
$
"
Diagram of the controller structure for operation with encoder feedback, see chapter 7.6.4.
The system position acquires the mechanical misalignment between rotor and zero position of the mounted
encoder system. This system position is preset at standard KEB motors in factory setting.
In order to operate a customer motor with encoder system it is necessary to make the automatically calibration
to detect the system position.
The following steps must be done:
The current system position is written into the respective parameter (Ec.02/ Ec.12).
If a S4 system shall be replaced by a F5-S, the system position for the F5 inverter can be calculated from the
data of S4:
49000 x 3
Intermediate value = ————— = 2.24304
65536
Adjustments shall be made in the Ec parameters in order to operate the servo system (depending on the used
encoder system).
See chapter 7.11 „speed measurement“.
With this software the speed of the motor can be calculated by the measured currents and the motor data (by
means of a model). This calculated speed can be used as feedback for the speed controller. The necessary
motor data for the model can be identified by the KEB COMBIVERT itselfs. Static operation with small frequen-
cies must be avoided, because the model can become unstable. The usable frequency range is approx. 1:100.
At setpoint speed 0, the speed control is deactivated and the motor is aligned using a predefined DC current.
The software version 2.x is only ready to run on the new control hardware xA.F5.230-0018 or -0019.
No compatibility exists between the previous versions 1.x and the versions 2.x; Parameter
lists of the old versions must be adjusted accordingly!
The following adjustments are default values and must not be adjusted:
- The controller configuration cS.00 must set to value „4: speed control“.
- The actual source cS.01 must set to value „2: calculated actual value“.
- The break handling Pn.34 must be activated (default value = 2: without display)
- The motor model nn.00 must be set to value "191".
The required equivalent circuit data for the motor model can be determined by the KEB COMBIVERT itselfs.
First, the motor data must be entered according to chapter 7.6.1, and the motor adaption must be executed.
- Writing of parameter dr.48 in inverter state "stop (LS)", measurement is starting automati-
cally.
- Writing of parameter dr.48 in inverter state „no control release (noP)“ with subsequent con-
trol release.
The measured values can be invalid in case of strong overdimensioning of the inverter. The rated current of the
motor should be at least 1/3 of the maximum short time current limit. The short time current limit is determined
by the overload characteristics and can be taken from the power circuit manual or parameter In.18 (hardware
current).
The direction of rotation during identification of the EMC is always "clockwise rotation"!
Value 82 "calculate drive data / Cdd" is output during measurement in inverter state ru.00. After successful
measurement ru.00 = 127 "drive data calculated/Cddr" is displayed.
If the measurement is interrupted with an error, in ru.00 = 60 "error! drive data / E.Cdd) is displayed.No correct
operation can be ensured in case of an abort.
The actual state of the identification is displayed in parameter dr.62 "state motor ident." The control release
must be switched off in order to leave the identification mode.
Parameter dr.48 must be written again in order to start a new measurement.
For safety reasons the output signal "brake release" is not set during measurement, since the motor cannot
generate a defined torque in this time.
Since the identification in the automatic mode is very reliable and for the user the most pleasant method it is
recommended to use generally this method according to chapter 7.6.3.3.1.
192: 15.625Hz
224: 7.8125Hz
* at dr.48 = 8 auto-identification
7.6.3.3.1 Auto-identification
The automatic identification can be carried out with rotation (dr.48=8) or without rotation (dr.48 = 7) (see table
dr.48). The measurement of the dead time compensation characteristics as well as stator resistance and lea-
kage inductance occurs during standstill.
For EMC identification it is necessary to accelerate the motor onto 60% of its rated speed. For this case an
additional ramp of dr.49 "Lh ident. acc/dec time" is effective". Calculation of the ramp can be taken from chapter
7.6.3.4.
For large synchronous motors, it leads to significant mechanical vibration and thus to noise development during
measurement of the leakage inductance with motor rated current. Here it is reasonable to reduce the measu-
ring current to 10..30% of the rated current.
The measured inductance is depending on the current level (saturation effects)!
The same function as for inductance is used for the current offset detection. Thus the parameter also has in-
fluence here on the current level.
A reduction is not necessary for the asynchronous machine. No flux is build-up, thus there is also no movement.
The speed controller should be parameterized with small Kp-, Ki values before the motor can be accelerated.
The speed controller can be preset optimally if the motor mass-moment of inertia is known (see chapter 7.7.1.2).
Identification at encoder operation can only be done with value 7: „automatic operation without rotation" or as
single identification as described in the following, because the motor model is not active.
Inductance (dr.48 = 2)
Measurement of dr.31"winding inductance" occurs with high-frequency AC current in standstill. The measure-
ment is started with dr.48 = 2. Measurement current is DSM rated current dr.23.
The frequency of the measurement signal is adjustable via bits 5... 7 in parameter dr.48.If the measurement
current cannot be reached with 1kHz, then the identification reduces the measuring frequency automatically.
Therefore the frequency value should not be changed.
The inductance value is automatically written in dr.31 after identification.
The EMC can be roughly calculated from the entered motor data like rated current and rated torque. dr.48 = 1
„calculation of the EMC“ must be written for this.
Mn x 90
EMC = ————
In
Resistance (dr.48 = 3)
The current controller parameters are calculated from the pre-identified equivalent circuit data with the adjust-
ment of dr.48 = 5. Is not identifed in the automatic mode if this calculation should occur before the identification
of the EMC.
The drive accelerates to a rated speed of 60% for the identification of the EMC. The ramp of dr.49 (Lh.ident.
acc/dec time) is used for the acceleration. The general speed limits of the oP parameters are valid! (see chapter
7.4 setpoint setting)
This measurement is only possible if the EMC adaptation of nn.00 (motor model adjustment) is activated (de-
fault setting!).
The value is written in dr.26 (DSM EMC peak value) and additionally in dr.63 (DSM EMC HR) if the identification
is successful executed.
Parameter dr.63 has a higher resolution and is suitable for applications with high frequencies.
The deadtime detection works only as single identification if the stator resistance is correct entered/identified.
The measured values can be read out via In.39 "deadtime selector" and In.40 "deadtime".
The measured deadtime compensation characteristics are effective during operation, if uF.18 "deadtime comp.
mode" is adjusted to value 3: "automatically" The characteristics are not cleared by Fr.01 "load default set" . 7
Torque detection (dr.48 = 14…18)
This should be executed only if the application really requires increased torque accuracy. The displayed idling
torque in ru.12 (actual torque) is subtracted during operation, so that the real shaft torque is displayed.
This residual torque is partly caused by switching frequency-dependent losses in the inverter and also by me-
ans of friction losses.
The torque offset of the complete drive for the different switching frequencies is measured with dr.48 = 14...18.
Thereby the drive accelerates in 16 steps with the adjusted ramp in dr.49 to maximum 1,3-fold synchronous
speed. The general speed limits of the op parameters are effective.
The measured residual torque is stored and interpolated as correction characteristic.
The torque offset characteristic can be read out with parameters dr.58 "torque offset selectort" and dr.59 "torque
offset".
The characteristics are deleted by Fr.01 -4 "copy parameter set" and also by Fr.10 "load motor dependent
data.".
The current offset is caused by tolerances of the components in the test circuit and as standard automatically
synchronized in non-energized state (inverter state "nop"). It is necessary in some cases to execute the ad-
justment in power on status by means of current-dependent tolerances in the current detection. For this adjust
parameter dr.48 = 19 and a high frequency AC current is output by the inverter. The rated current of the motor
is injected with a starting frequency of 1kHz. The frequency is automatically reduced if this is not possible.
Furthermore the automatic measurement is deactivated when the modulation is switched off, so the identified
offset remains permanently.
The drive has also measured the dead time compensation characteristic during automatic identification. The
calibrated characteristic must be activated for the control with motor model by the setting "dead time compen-
sation mode" (uF.18) = 3: „automatic“.
Further available kinds of the dead time compensation are only required for special applications (applications
with high frequencies, some special motors) or in other operating modes (e.g. V/f characteristics controlled).
The dead time compensation can be switched off via a digital input. The digital input is selected with parameter
uF.21. This disconnection is only required for special applications with high frequency.
It must be secured that the rotor is in a defined position after switching on of the control release ST. Therefore
a DC current is injected at standstill. Then the rotor rotates into its origin position.
The standstill current is ½ of the rated current and can be adapted in parameter nn.10 in default setting after
operation of Fr.10.
The times (Pn.35 and Pn.36) of the brake handling are active for standstill operation. In order that the rotor does
not vibrate after setting the control release, the current reaches the setpoint value in a half of the time adjusted
in Pn.35 "premagenetising time". (see picture 7.6.3.4a)
The half current-dependent load torque is acceptable as mechanical load (e.g. ¼ of the rated torque at ½ of
rated current at standstill).
Picture 7.6.3.4a
Current (A) Speed (rpm)
Current [A] Speed [rpm]
12 200
180
10
Standstill current 160
nn.10
140
8 Stabilisation
current Ramp output
nn.01 ru.02 1240
6 100
80
4
60 7
40
2
20
0 0
½Pn.35 ½Pn.35 Pn.36 nn.02 nn.03
Pn.35
Speed search
The rotor rotates at some applications when the modulation is switched on. The current speed can be deter-
mined with Pn.26 "speed search condition". (For further information seechapter 7.13.4 SSF)
In order to leave the critical range of small speed at starting and stopping there is an additional ramp for this
range.
The ramp is defined by parameter nn.08 nn.08 "startup speed" which indicates the speed range and parameter
nn.09 „startup time" which indicates the appropriate acceleration-/ deceleration time.
Picture 7.6.3.4b
Drehzahl / min-1
160
140
120
100 Rampenausgang
ru.02
nn.08 80
60
40
20
0
0 0,25 0,5 0,75 1 1,25 1,5
nn.09 Zeit / s
Example:
The open loop operation is activated with bit 9 of nn.00 "motor model select" and is only active during
Start ramp active. Condition:start ramp is parameterized.
The current of nn.01 "stabilisation current" must be regarded as maximum active current. The current ramp of
nn.02and nn.03 must be parametrized by such way (see also chapter 7.6.3.5 "Low speed") that the lowering of
current (nn.03) is upside the deactivation of the open-loop operation (nn.03 > nn.08).
The critical speed range (typically below 1% of the rated speed) is stabilized by reactive current. This current
adjustable in nn.01 "stabilisation current" is linear reduced depending on the actual speed in ru.07 from speed
nn.02 "min speed for current" to nn.03 "max. speed for current".
Picture 7.6.3.5a
180
10 Ausrichtstrom 160
nn.10
140
8
Stabilisierungsstrom
120
nn.01
Rampenausgang
6 100
ru.02
80
4
60
40
2
20
0 0
nn.02 nn.03
It is necessary to adapt the current or the ramp if there are vibrations during steady state.
0: measured cur-
rent *
6 Current control with Current control to model currents
64: estimated
currents
0: off
7 EMC adaption Adapts the EMC at upper speed
128: On *
Current offset adap- 0: Off *
8 Adapts the current offset during operation
tion 256: An
0: Off *
9 Controlled operation Switching off the model during start ramp
512: An
0: Off * Activates the harmonic absorber for operation with sine-
10 Band-stop filter
1024: An wave filter
0: Off *
11 Deviation controller Deviation of model currents to measured currents
2048: An
Voltage output for HF 0: Off *
12 Activates double voltage output
applications 4096: An
0: Off *
13 HF detection Determines rotor and system position when switching on.
8192: An
0: Off *
14 HF injection Determines the rotor position continuously at low speed.
16348: An
* Default values
The currents nn.01 "stabilisation current" and nn.10 "standstill current" can be switched off with bit 0 of nn.00.
The starting phase with activated currents runs more steady. In such a way this adjustment should not be
changed!
The values are limited to ½ of the HSR current In.18 if the rated motor current is higher than the inverter rated
current.
The stator resistance changing by temperature influences can affect the behavior at low speed as well as the
start. The RS adaptation adjusts the stator resistance and stabilizes the motor model therefore.
The I-part of the adaptation can be adjusted with nn.06 "rs adaption factor". The rs adaptation becomes active
with ru.17 "active current" > nn.01.
EMC adaption
The EMC changing by load and temperature influences is adjusted at upper speed.
The adaption becomes active at actual speed ru.07 > ¼ of the rated speed dr.24 and improves the accuracy
of the actual torque display ru.12.
Observer
The observer amplifies the influence of the measured currents in the model. The most effects become notice-
able in the upper speed range.
The value must be increased if current oscillations occur at e.g. applications with high frequency.
The observer factor can be adjusted with nn.07 "observer factor".
It is necessary for applications with high frequency to activate the double voltage output with bit 12 of nn.00.
Speed estimation
The speed estimate controller is calculated by writing on Fr.10 and cannot be changed. The speed estimate
controller estimates a speed from the currents of the motor model. Parameter nn.04 "time speed calculation"
determines the scan time of the speed estimate controller. This time should not be changed.
Parameter nn.05 "filter speed calculation " determines the smoothing time at the output of the controller. Oscil-
lations are reduced when the value is increased, but the drive becomes more non-dynamic.
At special applications the drive has to rotate only into oneway direction. The respective direction of rotation can
be locked with oP.40/ oP.41 "max. output val. for/rev" by writing the parameter value to "0" and thus the speed
estimation is limited. To limit the speed estimate controller see also chapter 7.5.2.4.3.
The general speed control settings can be adjusted according to chapter 7.7.1 "speed control".
Diagram of the controller structure for operation without encoder feedback, see chapter 7.6.4.
For the operation with sine-wave filter it is necessary to filter the resonance frequency with a band-stop filter. The
resonance frequency of the sine-wave filter and the corresponding filter parameters can be determined with the tool
sine-wave filter exe (www.keb.de). The equivalent circuit data of the motor and sine-wave filter must be entered
in order to generate a parameter list. Then this parameter list must be loaded to the frequency inverter. The filter
parameters are stored in the fh parameter group.
The resonance frequency is filtered of the estimated currents with software filter, in order that there is no reaction.
The band-stop filter must be activated in nn.00 "motor model select" bit 10 (band-stop filter). Also it must be con- 7
trolled to the estimated currents nn.00 bit 6 (current control). The deviation controller should be switched on with
bit 11 of nn.00 in order to avoid possible effects on wrong estimation. The deviation controller adjusts the estimated
currents to the measured currents with the scan time of nn.12 "deviation control time". This time can be increased
in case of current oscillations.
The inverter current is mostly higher than the motor current because there is a current through the capacitor of
the sine-wave filter. The single-phase capacitor value must be entered in nn.13 "C-filter [UF]" in order to clear
this error.
The increased current ripple and the capacitor current must be considered at the dimensioning of the inver-
ter!
The minimum inverter switching frequency must be greater or equal than the minimum switching frequency
of the sine-wave filter.
In order that these functions can be used, the following special requirements apply to the synchronous motor:
- The difference between Ld and Lq (minimum and maximum value) should be greater than 30%.
- The curve of inductance with a circulation of rho should be sinusoidal.
These criteria can be checked during the identification (dr.48 = 2) by recording dr.31 and EC.40.
dS.32 KI HF detection
Illustration of nn.16 in dS parameters
This parameter is used to optimize the first operating mode dS.31 = 0
The current is internally limited to maximum In.01 rated inverter current. For this the inductance in dr.31 must
be correctly measured or adjusted.
COMBIVIS 5 Scope - H:\CvKebData\F5\SCL\Identifikation\Luv\Luv_Messung_24SM000340F.sc5 - Node 1 COMBIVIS 5 Scope - H:\CvKebData\F5\SCL\Identifikation\Luv\BMP_Luv_1000Hz.sc5 - Node 1
COMBIVIS 5 Version 5.6 Registered to: :Horn COMBIVIS 5 Version 5.6 Registered to: :Horn
Printdate: 19.12.2007 14:59:48 Filedate: 22.12.2004 11:38:30 Printdate: 19.12.2007 14:51:41 Filedate: 19.12.2007 11:52:00
1017: F5E-S/V2.20 8000rpm 1017: F5E-S/V2.20 8000rpm
X (us/Division): 80000 X (us/Division): 512000
CH Parameter Set Y-Scaler (n/Division) Y-Zero at CH Parameter Set Y-Scaler (n/Division) Y-Zero at
A Ec40 act.absolute pos. el. 0 7282 32768 A dr31 DSM inductance 0 0,25 mH 14,82 mH
B dr31 DSM inductance 0 1,00 mH 5,66 mH B AA52 int. data 2 0 7235 170
C Off C Off
D Off D Off
B
A
Cursor I Cursor II Difference:
Ec40 act.absolute
[38415]
pos.
38415
el. [26824] 26824 [-11591] -11591 Cursor I Cursor II Difference:
dr31 DSM inductance
[644] 6,44 mH [484] 4,84 mH [-160] -1,60 mH dr31 DSM inductance
[1537] 15,37 mH [1425] 14,25 mH [-112] -1,12 mH
AA52 int. data[15969]
2 15969 [-16219] -16219 [-32188] -32188
The curve of rho is sinusoidal. The curve of rho is not sinusoidal and has constant
ranges.
The HF signals can be used optimally with this motor. The HF signals cannot be used with this motor.
HF detection
With this function the rotor position is determined first with each switching on of the modulation. The amplitude
of the test signal is determined by the winding inductance dr.31.
B
A
C
D Cursor I:16822us Cursor II:946us Differenz:-15876us
AA51 int. data[-466]
1 -466 [14] 14 [480] 480
AA52 int. data[1]2 1 [1] 1 [0] 0
AA53 int. data[5088]
3 5088 [11512] 11512 [6424] 6424
AA54 int. data[631]
4 631 [7199] 7199 [6568] 6568
If this function is activated, one can deactivate the general start with alignment current and brake handling.
The system position is also determined with this function. The determined position is stored in Ec.02 or Ec.12
depending on cS.01. Thus the system position of a motor at standstill can be determined. The operation of a
synchronous motor is still possible only with incremental encoder.
HF supply
The rotor position can be determined continuously with a test signal for operation at standstill and low speed.
HF supply is only possible with a switching frequency of 8 or 16 kHz. The power unit must supply the required
current also at 0 Hz and 8 or 16 kHz.
The internal FoH filter is used for HF supply. The default values of parameters fh.01... fh.09 are preset to 1 kHz
with quality factor 5. If the HF injection is used the speed controller must be adjusted a little softer.
Since the HF supply has only advantages in the lower speed range, it is switched off above nn.03. When falling
below nn.02 the HF supply is switched on again. The HF supply is advantageous only in lower speed range if
there are problems with the motor model.
The signal amplitude of the HF supply can be modified here. A higher value means more information, but also
higher losses and a higher noise level.
The default value is selected by way that 1/8 of the rated motor current is used. But maximally 25% of the ma-
ximum output voltage.
The Pi-controller for speed determination at HF injection can be modified here. This Pi-controller is paramete-
rized by dr.31, dr.64, nn.14 and nn.15.
The speed limit for the speed limit open-loop operation is now separately adjustable from the additional ramp via nn.17. In
order that all remains downward compatible, nn.08 is still active for the speed limit open-loop operation, if nn.17 is set to 0.
iu
Precontrol and Transformation
© KEB, 2012-10
decoupling and dead time Power unit
Volt V iv SM
compensation
isd controller 3ph
1/ru.18 iw
Usd
Volt W
-
isq reg
Block diagram
isd_ref
isd reg
Figure 7.6.4.a Block diagram SCL
isq model
ru.07 EC.40
Page 7.6 - 23
7
Page 7.6 - 24
nn.00 current control
isq
isq reg
nn.00 harmonic absorber
nn.00
Current controller adaption
isq model
Figure 7.6.4.b SCL Current feedback
isd modell
Band-stop
filter
isd
fh.01 ... fh.09
Motor data and controller adjustments of the synchronous motor
© KEB, 2012-10
© KEB, 2012-10
cS.06...cS.12 speed controller parameters
KP factor KI factor
cS.8 cS.9+cS.10
ru.07 cS.7
cS.6 cs.9
xd nact.
Xd(n) act. speed
cS.11 cS.12
Page 7.6 - 25
7
Motor data and controller adjustments of the synchronous motor
dr.33
ds.03.current/torque mode (Uzk-depending shifting)
nact. dr.40
dr.42
dr.44
dr.46
DC-link voltage
n
dr.39 dr.41 dr.43 dr.45 dr.46
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp presetting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
7.7.1.2 Automatic adjustment of the speed controller (only at operating with motor model)
Kp (cS.06) and Ki (cS.09) of the speed controller can be preset by the inverter. For this the mass-moment of
inertia of the complete system (motor + rigidly coupled load) must be entered in cS.25 „inertia“.
After the input of the motor data once parameter Fr.10 "Reset field-oriented control parameter" = 1 or 2 must
be written. Thus dependent on the adjusted rated power (dr.03) the mass-moment of inertia was pre-charged
for a standard asynchronous motor in cS.25. The value of cS.25 has the right dimension for 50Hz standard
motors, because at some applications the ratio of the load inertia is in a range of 0.5...2 x motor inertia.
Better results can be realized if the total moment of inertia is exactly preset. If the value is unknown it can be
determined.as described in chapter 7.7.2. 7
Parameter cS.26 „optimisation" determines the control characteristic which should be achieved by the calcu-
lated parameters.
Parameters for a dynamic, hard speed controller adjustment are calculated with cS.26 = 2. Interference fac-
tors, such as torsion or tolerance of the load coupling can intensify vibrations, so that a higher value must be
entered in cS.26.
Parameters for a very soft and slow speed controller adjustment are calculated with cS.26 = 15. Which value
between 2 and 15 is most suitable for the application is depending on the oscillation-grade of the total system.
An oscillation of the estimated speed is a possible disturbance at encoderless operation of asynchronous mo-
tors (ASCL). Extension of parameter "ASCL speed PT1 time" (ds.17) often enables a dynamic speed controller
adjustment, i.e. a smaller value for cS.26.
The precharging of the speed controller parameters can be deactivated with the adjustment of value "19 = off"
in cS.26.
The speed controller parameters are overwritten when the value for cS.26 is changed.
The speed controller parameters (KP, KI) can be influenced depending on the actual speed with this function.
LowGain = 100%
KP speed gain / peak % = cS.07
KP speed limit / gain % = cS.08
Speed for square function = cS.14
Max. speed for square function = cS.13
K1 = -(LowGain – PKGain) / Speed^2
K2 = -(HiGain – PKGain) / (Speed^2 – 2*Speed*MaxSpeed+MaxSpeed^2)
n = 0…MaxSpeed
Gain(n) = if [n<Speed, LowGain + K1*n^2;hiGain + K2 * (n – MaxSpeed)^2]
Gain is internally limited to 0…800%
KP = cS.06 * Gain / 100%
Ki = cS.09 * Gain / 100%
If the square mode is adjusted, the variable KI and KP have no function. The mode cS.11=-1 = brake release
is possible.
120
100
Verstärkung [%]
80
60
40
20
0
0 500 1000 1500 2000
n [U/min]
250
200
Verstärkung [%]
150
100
50
0
0 500 1000 1500 2000
n [U/min]
120
100
Verstärkung [%]
80
60
7
40
20
0
0 500 1000 1500 2000
n [U/min]
120
100
Verstärkung [%]
80
60
40
20
0
0 500 1000 1500 2000
n [U/min]
160
140
120
100
Gain
80
Inv. Imr
gain [%]
60
40
20
0
0 500 1000 1500 2000
n [rpm]
Picture.5: cS.07=100; cS.08=200; cS.12=1300; cS.13=3700, „field weakening“ (linear field weakening)
The following parameters serve for the "fine tuning" of the speed controller and may not be changed in many
applications.
Xd(n)
variable integral factor KI KI factor
Parameters cS.09...cS.12 determine the inte-
gral factor of the speed controller.
The KI-factor can be varied speed-dependent in cS.09+
order to reach a better speed rigidity at small cS. 10
speeds and in standstill.
- cS.09 forms the base value.
- the maximum value for the integral factor
is cS.09 + cS.10. cS.09 7
- the two corner speeds cS.11 and
cS.12 determine the speed range in which the
KI value is changed.
A special function can be activated in parameter max. speed for max. KI (CS.11) by setting -1: brake release
which works only in connection with the brake control.
An enormous speed rigidity is required for load transfer with hoist drives or lifts,
in order that the brake release and the load transfer are not significant by the inverter.
This controller adjustment is not to be used for normal operation, since the speed controller oscillates too much
at this adjustment.
The solution is to enter a high value in parameter "KI offset“ (cS.10) in order make the controller rigidy. If cS.11
indicates the value "- 1: brake release", this "KI offset" is set immediately to 0 at the end of the brake release
time, not reduced during operating in a speed range.
Set speed
calculated actual speed
Delta t = 0.26s
acceleration torque
Torque
Delta M = 662 Nm
time
The knowledge of the mass moment of inertia of the system (= motor + rigidly coupled load) is required for
the automatic calculation of the speed controller parameters as well as for the pre-control of the acceleration
torque.
If this mass moment of inertia is unknown, it can be determined by an acceleration test.
For this the system must be accelerated with defined, constant torque. It must be guaranteed that no significant
and acceleration-independent load torque occurs by the application.
∆t
J = 95493 x ∆M x ———
∆n
0.26 s
J = 95493 x 662 Nm —————— = 40886 kgcm2
402 rpm
In order to eliminate the effect of friction from the calculation, you can determine the mass moment of inertia a
second time in similar manner, however by deceleration test. The average value of both inertia, which is deter-
mined at ramp-up or deceleration must be entered in parameter cS.25 „inertia (kg cm^2)“.
active current
setpoint
PT1
cS.06...cS.12 cS. 29
High frequency oscillations (caused by spring elements in the mechanics of the drive train) can be filtered by
this way from the active current setpoint signal.
The filter time must be adjusted in parameter cS.29 „act. curr. ref. PT1-time“ (cS.29). A longer filter time causes
a stronger smoothing of the active current signal, but also less dynamic control characteristic and increased
oscillation inclination.
An adaptation of the speed controller is required on changing the Pt1-time. This filter is used e.g. for spindles
in order to avoid jumps in the current setpoint at fast load changes.
If the mass moment of inertia of a drive is known it can be calculated which torque is required to accelerate the
drive. This function is activated, if a value unequal 0 is entered in parameter cS.28 „pretorq. speed fact.%/". This
parameter must be set to 100% for a complete pre-control.
7
Picture 7.7.4 Acceleration dependent pre-control
pretorque speed
Inertia cS.25 fact. % cS.28
+
Speed controller
For some applications it is not necessary to pre-control the complete acceleration torque (cS.28 = 100%) see
the following reasons:
- a different torque is required with the same acceleration at motoring or generating (e.g. due to friction)
- the speed setpoint setting (e.g. by external control) is made in steps, so torque jumps can occur
- the (analog) speed setpoint setting is superimposed by a noise, which must be damped for the pre-
control
The influence of the pre-control can be damped with parameter cS.28 „pretorq. speed fact. %" for these appli-
cations
Torque peaks, which are caused by a speed setpoint setting in steps, can be reduced by means of a low
pass filter. Here also valid: the higher the time in parameter cS.27 „pretorq. speed PT1 time“ the better the
smoothing, the undynamic but also the precontrol.
For applications, when new setpoints are preset by an external control within fixed time base there is one
additional function for the acceleration torque pre-control: the setpoint smoothening.
Sollwertvorgabe von
externer Steuerung = Sollwertvorgabe von
Drehzahlsollwert vor Rampe externer Steuerung
Drehzahlsollwert
vor Rampe (ru.01)
with reference splitting and variable clock time of the external control
The longest clock time (in ms) must be entered in parameter oP.74 "reference splitting" for optimal precon-
trol at non-constant cycle time of the external control.
This causes a short delay of the reference value, but also a smoother precontrol value.
op.74 = Zeit für kürzesten op.74 = Zeit für längsten 7
Sollwertvorgabe-Zyklus Sollwertvorgabe-Zyklus
Drehzahlsollwert Drehzahlsollwert
vor Rampe (ru.01) vor Rampe (ru.01)
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp presetting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
The inverter always requires a voltage control reserve to adjust the current. The maximum voltage controller
limits the output voltage if the output voltage is too high (higher dS.10 "Umax modulation limit"). The maximum
voltage controller is activated by the input of values 8 or 24 under point „maximum voltage controller“ of para-
meter dS.04 "flux/rotor adaption mode".
The controller is switched off at value 0 or 16.
controller on,
16: off, max. 100%
Maximum voltage control- max. modulation factor = ds.10 + 2%
3, 4
ler
16: off, max. 100% controller off, max. modulation factor=100%
The voltage range for which a modulation factor > 100% is needed is designated as overmodulation range.
The voltages in this area are no longer sinusoidal, which causes distortions in the phase currents, turbulent
speed estimation in encoderless operation and a worse torque accuracy.
A better output voltage is contrary to these disadvantages.
Overmodulation is not allowed with selection "max. 100%" (value 16 and 24). This adjustment should be se-
lected if the drive shall be operated in a mode with motor model (with or without speed feedback).
For the selection "max. 110%" (value 0 or 8), the available voltage increases due to exploitation of the non-
sinusoidal overmodulation range.
Value 0 should not be used because there are important negative effects.
The negative effects are minimized at value 8 by limiting the overmodulation range to „Umax modulation
limit“dS.10 + 2%. i.e. the maximum modulation factor is 105%, if dS.10 = 103% is selected. This limitation is
only valid for the overmodulation range.
The values 0 and 8 should only be used after careful testing.
The controller is adapted by parameters dS.08 „KP Umax“, dS.09 „KI Umax“, dS.10 „Umax modulation limit“.
dS.08 has only a minor effect and can remain on value 0.
dS.09 determines the dynamics of the controller. If this parameter is adjusted too small, the drive can arrive vol-
tage limitation. If the parameter is adjusted too high, the drive starts to vibrate. If the modulation factor becomes
much noisier due to an increase of dS.09, it indicates that the controller setting is too high.
A temporary reaching of the voltage limitation is normally not a problem.
The modulation factor to be controlled is determined with parameter dS.10 . The closer this is to 100%, the
better the inverter voltage is utilised, but also the lower are the control reserves useable for the dynamic.
The default value of 97% is usually a good agreement.
For the asynchronous machine, the voltage limitation occurs by flux reduction.
The motor flow can be reduced by the controller to ¼ of the value which is required according to the magneti-
sation characteristic.
For the synchronous machine, the voltage limitation is done by setting a negative magnetising current. The
maximum value of this current is determined with parameter dS.13 „magn. current limit“. (see chapter 7.8.3
physical torque limits of the synchronous motor about the influence and adjustment of dS.13).
The rated torque (calculated from rated power and rated speed) of the motor is displayed in parameter dr.14.
The maximum torque (limited by the maximum current of the inverter) is displayed in dr.15.
At activated hardware current limit (uF.15 = 1 or 2) the maximum current is equal to the hardware current level
(In.18) less safety reserve of 5% of the inverter rated current.
At deactivated hardware current lim. (uF.15 = 0) the maximum current is equal to the overcurrent error limit less
safety reserve of 10%.
Additionally the motor current can be limited by software with parameter dr.37 „max. current“ (see chapter
7.10.2). This limitation also affects the maximum torque but is not displayed in dr.15.
Also the active current is limited by torque limitation in the base speed range. But nevertheless the current limit
of the inverter can be exceeded by the additional magnetizing current. Therefore, also the software current limit
should be activated.
When the motor overloads, i.e., when a torque upwards of its torque limit is demanded from it, the maximum
voltage controller reduces the flux too much and thereby also reduces the maximum achievable torque.
Therefore the maximum permissible torque must be reduced in the field weakening range.
The torque limiting characteristic is defined with parameters dr.15...dr.18.
Moment [Nm]
dr.15
dr.16
1 / x Absenkung
ds.03 "Feldschwächkennlinie" = 0: aus
0,5 * dr.16
The "max. torque FI" (dr.15) is depending on the maximum inverter current and can not be changed.
At default setting the maximum torque in the field weakening is reduced to a 1/x function - because of flux
reduction.
The physical breakdown torque characteristic of the motor is a quare characteristic, i.e. also the maximum
active power in the field weakening range must become smaller.
The square limiting characteristic must be activated if the motor shall be driven to its limits. This is done with 7
value 2 in point "field weakening characteristic" of parameter dS.03 „current/torque mode“.
The limiting characteristic is adapted to the motor with parameter dr.16 „DASM max torque corn. sp“ .
dr.16 = breakdown torque of the motor (at speed dr.18) - safety reserve
Example:
a motor shall have the following rated data:
rated speed: 1470 rpm rated frequency = 50Hz
rated torque 36Nm Mrated / Mbreakdown = 2.5
selected value for DASM field weakening speed (dr.18): 1500 rpm
Data sheet value for breakdown torque of the motor at rated 2.5 * 36Nm = 90Nm
frequency:
Safety reserve 25% = 22.5 Nm
The value for dr.16 can be higher than the value in dr.15, since the breakdown torque of the motor can be higher
than the maximum torque of the inverter.
The safety reserve is necessary because the limiting characteristic must be sufficiently far from the physical
breakdown torque of the motor.
Moment [Nm]
quadratische Kippmomentkennlinie
des Motors
dr.15
dr.16
quadratische Absenkung
ds.03 "Felschwächkennlinie" = 1: ein
0,25 * dr.16
The rated torque of the synchronous motor must be entered according to the name plate in parameter dr.27.
The maximum torque (limited by the maximum current of the inverter) is displayed in dr.15.
At activated hardware current limit (uF.15 = 1 or 2) the maximum current is equal to the hardware current level
(In.18) less safety reserve of 5% of the inverter rated current.
At deactivated hardware current limit (uF.15 = 0) the maximum current is equal to the overcurrent error limit less
safety reserve of 10%.
Attention:
There are several disadvantages contrary to the advantage of higher maximum speed:
- the drive is more prone to vibrations in the base speed range
- not all motors are suitable for field weakening operation
- a higher current is required for the same torque due to the magnetizing current demand
- the rotor position information must be very exactly. A system position error (e.g. in case of inaccurate
encoder mounting) can make the drive uncontrollable.
There is a specific 'ideal' magnetization current limit for each motor. The available field-weakening range is very
small if the limit is too small.
The following picture shows the connexion between the maximum reachable torque and the magnetizing cur-
rent limit dS.13.
Moment (dS.13 = 0)
Feldschwächbereich
Grunddrehzahlbereich
Drehzahl [U/min]
If the magnetizing current limit is selected too high, the available torque becomes smaller again. In addition,
too high value for dS.13 can cause the maximum voltage controller to "hang". That means: for setting the ma-
gnetising current, more voltage is used than is gained from the field weakening. The voltage remains too high.
A typical value for dS.13 is the motor rated current. In the field weakening range, the current needed to set a
defined torque increases.
Beginn des
"Feldschwächbereiches"
Motorstrom [A]
Momentengrenzkennlinie
Drehzahl [U/m
Attention:
To assure that the speed controller can control the drive, an active current must always be available that should
not fall below 0.5 x dS.13.
Pay attention to appropriate adjustment of the torque limit and the maximum current!
Starting at a certain speed, the drive cannot provide the same torque in field weakening operation that it provi-
des in the base speed range.
If the drive is to accelerate at a constant torque limit (e.g., double the rated torque), the motor is (despite field
weakening) physically unable to provide this torque.
The set torque can not be adjusted anymore and the drive ´hangs´ in the voltage limit (modulation factor ru.42
= 100%).Therefore, a limiting characteristic that mirrors the physical limits of the drive must be given. This limit
is depending on dS.13 „magnetization current limit“.
If no limiting characteristic is given, the user must insure that the motor is not asked to deliver an inadmissibly
high torque by choosing suitable acceleration /deceleration ramps and by appropriate selection of the load.
Parameters dr.33 and dr.39...47 serve for setting the limiting charcteristic.
Attention:
Value 0 should not be used for the torque values of the limiting characteristic. Also, the torque at the highest
speed (i.e., the last point on the characteristic) should minimally be set to the following value:
This value may not be fallen below for the following reason:
A potential error in the position sensing leads to the magnetising current creating a torque in the field weake-
ning range. An error of 20° electrical causes an unwanted torque from the magnetising current of maximally:
If this torque error cannot be compensated for due to the limiting characteristic, the drive becomes uncontrolla-
ble.
All other torque values must be selected higher accordingly.
Parameters dr.33, 40, 42, 44, 46 contain the maximum torque for the speeds in dr.39, 41, 43, 45, 47. Linear
interpolation between these points.
dr.33
measured limiting
characteristic
dr.40
Torque
dr.42
The physical torque limiting characteristic of the motor is depending on the maximum output voltage of the
inverter. This is determined by the DC link voltage, which is depending on the mains input voltage and the
inverter load.
Therefore different modes can be selected in dS.03 for the programmed limiting characteristic.
The value 0 ("off") can be used if the limiting characteristic for the mains input voltage is programmed, the ma-
chine is operated with it, and this voltage is relatively constant.
The advantage (e.g., during ramp-up at the torque limit) is that the continuous, load-dependent fluctuations of
the intermediate circuit cannot cause any torque fluctuations.
If, however, the mains input voltage is variable (e.g., affected by other users) or if the mains voltage at the lo-
cation of the machine is unknown, dS.03 equal 4, 8 or 12 must be selected.
Then the programmed limiting characteristic is always valid for the inverter rated voltage (400V or 230V) and is
proportional adjusted to the voltage.
Drehzahl [U/min]
The limiting characteristic must always be programmed higher than the speed range where the motor shall be
operated. Otherwise, the drive operates in an undefined range at lower DC link voltage values due to the shift
Therefore preferable setting is 8 („>Un(FI) = off, abnormal stopping = off“). Only the necessary physical shifting
of the characteristic due to low DC link voltage is executed here.
That means the characteristic is only shifted if the DC link voltage is lower than the rated DC link voltage (= √
2 * inverter rated voltage).
There is no shifting if the DC link voltage is higher than the rated voltage.
Value 12 („>Un(FI)= off, abnormal stopping= on“) can be selected if the maximum achievable torque shall
be available for emergency stop. In this mode the characteristic at higher DC link voltage is shifted to higher
speeds only during emergency stop operation. Generally value "8" should be used if possible.
The total power of the motor in the field weakening range consists of active current and magnetizing current.
The maximum torque limits only the active current.
For some motors there is a maximum current specified in the data sheet. This is valid for both components
together. Therefore the total motor current can be limited by this parameter.
The magnetization current has priority if both components together exceed the current limit.
Attention:
To ensure that the speed controller can control the drive, active current flow must always be able. Therefore the 7
magnetizing current limit (dS.13) must be significant lower than the maximum current (dr.37). Maximum dS.13
should be 0.75 x dr.37.
The max. current limit dr.37 is activated via bit 0 of parameter dS.03.
A possibility to limit the maximum torque in the field weakening is via the characteristic dr.39 .. dr47 and the ad-
justments in ds.03 bit1..3. This has the disadvantage that the characteristic must be determined, either abstract
calculated with the "known motor data" or measured at the test stand. Also the consideration of the DC link
voltage by parallel shifting of the corner can not optimally reflect the reality.
A further possibility of limitation can occur by the evaluation of voltage equation. The equation is converted to
isq and the max. permissible voltage is preset for the voltage component usd.
Rs * isd - usdmax
isqmax =
2 * π * f * Lsq
Since it is calculated with equivalent circuit diagrams in the formula, deviations can occur in
case of different temperatures or as a function of the saturation (Lsq) of the inductance.
For some applications it is not required to provide the maximum possible torque, but the application requires
other process-related limitations (e.g. to protect mechanical components).
These can be adjusted via parameters cS.19...cS.23. The torque limiting characteristic which is defined by
maximum current and available voltage remains always active as superior limit.
If only one limit is required for all operating ranges (forward, reverse, motorized and regenerative), parameter
"absolute torque setpoint" (cS.19) can be used therefore. All other limits (cS.20...cS.23) must be set to value
„-1:off“.
If different torque limits are required, these must be entered in parameters cS.20...cS.23 (= torque limit for the
different operating ranges).
The torque limits can be adjusted during operation for special applications by multiplying with a factor of
0...100%.
Parameter " torque reference source" (cS.15) determines how this factor is built for the adjusted torque limits
(cS.19...cS.23).
the output value of the PID controller (see section 7.15) is the factor for the torque limits
5: external PID
(cS.19...cS.23)
output (ru.57)
The output value can be read out in ru.57
Analog input value AN2 is the factor for the torque limits (cS.19...cS.23). The analog input
is scannned and processed in faster steps with this adjustment. To realize this faster pro-
6: AN2 direct
cessing, the following parameters do not have any function: „AN2 noise filter“ (An.11), „AN2
(+/- 10V)
offset Y“ (An.17), „AN2 zero clamp“ (An.14), „AN2 save mode“ (An.12). The value of AN2 is
limited as multiplier to 100%.
The parameter value of cS.19…cS.23 can be changed in 1 ms step with An.54 and e.g. an analog input. This
provides the possibility e.g. to change the motor limit analog and to keep the regenerative limit constant. 7
cS.19 is the reference torque if the actual torque is output via an analog output. Here it makes sense to keep
cS.19 constant and to change cS.20…cS.23.
Since the selection in standstill if regenerative or motorized operating mode is depending on the actual speed,
different setting can cause torque jumps when the torque limit is reached.
The torque in [%] referring to parameter "absolute torque reference" (cS.19) is displayed in ru.73 and ru.74.
The effective limits for the actual direction of rotation can be read in parameters ru.47 "act. torque limit mot."
and ru.48 "act. torque limit regen.". Parameters ru.47 and ru.48 are depending on the programmed torque li-
mits, the limiting characteristic and current limits (e.g. hardware current limit or dr.37 "max. current).
120
Istmoment (ru.12) 80
Moment [Nm]
40
Feldschw.- 20
drehzahl
(dr.18)
0
Drehzahl [U/min]
The maximum thermally permissible torque - i.e., in the base speed range, the rated torque, and the range
higher than the rated speed, the rated torque attenuated following a 1/x-function - is taken as 100% utilization
of the motor.
The programmed speed-torque characteristic is accepted as 100% utilization of the inverter. This characteristic
consists of the torque limits in the cS-Parameters (e.g. cS.19) and the limiting characteristic in the dr-Parame-
ters (e.g. dr.15...dr.18).
The adjusted value in parameter "torque reference level“ (LE.27) corresponds to 100% utilization in the applica-
tion. This could be e.g. the torque which is permanent permissible for the mounted worm gears or the mounted
gear box.
The smallest of the 3 values indicates the torque with which the whole drive can be loaded permanently at
the corresponding speed. This torque is the reference torque for the calculation of parameter "max torque in
percent" (ru.90).
Istmoment (ru.12)
Feldschw.- Nenn-
drehzahl drehzahl
(dr.18) (dr.01)
100
max. Drehmoment [%] (ru.90)
0
Drehzahl [U/min]
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp presetting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
The set torque is calculated from the value in parameter cS.19 multiplied with a factor (0 .. 100%) which can
be preset by different sources (analog inputs, motorpoti, etc.). The selection of the torque reference source is
made by parameter cS.15.
The superior torque limitations, such as "max. torque FI "(dr.15) still remain effective.
The rate of change of the torque reference can be limited with cS.16 .
The setpoint speed after the ramp generator (ru.02) serves for speed limiting.
The setpoint speed is formed (with exception of the rotation direction) in the same manner as in speed-con-
trolled operation or open-loop operation. The direction of rotation results from the sign of the torque reference.
Without limiting the speed, the drive would accelerate to indefinitely high speeds if the counter torque disap-
pears.
Since the limiting is based on the speed at the ramp generator output, the acceleration/ deceleration ramps
should be set to 0 s for this operating mode.
There are 2 different modes for torque-controlled operation. They can be selected by cS.00 = 5 or cS.00 = 6 .
The speed controller is not active until the drive does not exceed the maximum speed for torque-controlled
operation (= set speed ru.02) .
This has the advantage that the parameterization of the speed controller has no influence to the set torque.
The change to speed-controlled operation occurs only when the speed limit is reached. Caused by the change-
over the control characteristic is not optimal here. Overshoots can occur.
ru.07
"Istwert Anzeige"
(Istdrehzahl)
Drehzahlregler
Soll-
Soll- moment
drehzahl -
Betrag von Parameter ru.02
"Anzeige Rampenausgang"
Istdrehzahl
CS.15 größer
Vorzeichen
Quelle Solldrehzahl Momenten-
Momenten- sollwert
sollwert
Analog Ref 0 Drehzahl-
Analog Aux regler in der
1
+100% Begrenzung
digital absolut (CS.19) 2
Rampenzeit
"Sollmoment
digital % (CS.18) 3 -100% Momentensollwert
Momentregler"
CS.16
Motorpoti (ru.37) 4 ru.49
externer PID Ausgang (ru.57) 5
AN2 direkt (schnelle Abtastung) 6
CS.19
"absoluter Momentensollwert"
The speed controller is permanent active, but the limitation of the controller is always set equal to the torque
reference.
As long as the drive does not exceed the maximum speed for torque-controlled operation (setpoint speed = ru.
2), the speed controller is in limitation, i.e. its output signal is equal to the torque reference.
This mode has the advantage that the speed controller is always activated and therefore the behavior on rea-
ching the maximum speed is much better.
The disadvantage is that with an unfavorable parametrisation of the speed controller (e.g., very small ampli-
fication chosen), the torque reference can be further delayed by the controller. I.e., even if the ramp time is
cS.16 = 0:off, the speed controller must first run to the new limit value after an increase of the torque reference.
ru.07
"Istwert Anzeige"
(Istdrehzahl)
Drehzahlregler
Soll-
Soll- moment Momenten-
drehzahl - sollwert
Betrag von Parameter ru.02
"Anzeige Rampenausgang"
CS.15
Vorzeichen
Quelle
Momenten-
sollwert
Analog Ref 0
Analog Aux 1
+100% 7
digital absolut (CS.19) 2
Rampenzeit
"Sollmoment
digital % (CS.18) 3 -100% Momentensollwert
Momentregler"
CS.16
Motorpoti (ru.37) 4 ru.49
externer PID Ausgang (ru.57) 5
AN2 direkt (schnelle Abtastung) 6
CS.19
"absoluter Momentensollwert"
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp presetting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
The current controller (dS.00 „KP current“, dS.01 „KI current“) are automatically precharged by actuation of
Fr.10 by means of the equivalent circuit data.
The controller parameters are calculated from the equivalent circuit data.
The current decoupling must be activated in dS.02 for an optimal control characteristic. For the asynchronous
machine it is differentiated between "1: on" and "2: on, without main inductance“.
Mode 2 (without main inductance) must be used, if there are high DC link voltage fluctuations (e.g. at low mains
or spindle motors). The complete decoupling can lead to boosted current oscillation.
Otherwise mode „1: on“ for synchronous and asynchronous motor must be selected.
Exception: The controller parameters are only calculated according to the rating plate data at speed-controlled
operation of an asynchronous motor without motor model. These adjustments are default values for standard
motors and they are not suitable for special motors (e.g. high-and medium-frequency motors). A manual adap-
tion must be made here.
A current decoupling is also not possible in cases where the equivalent circuit data are unknown. Parameter 7
ds.02 must set to value 0.
A change of bit 4 in parameter dS.03 is normally not necessary and should be done
Attention:
only by authorized KEB service personnel.
The priority of the active current controller can be assigned in regenerative operation with bit 4 of parameter
dS.03 . In special applications this is advantageous for the quality of the current control
The hardware current limit becomes active if the phase current exceeds the value of In.18 „hardware current“.
Through short-time power shutdown the current limit can eliminate short current peaks at low speeds, e.g.
when starting the motor.
However if the current level is exceeded at high speed under load, disconnection of the voltage leads to a red-
uction of the breakdown torque of the motor and thus to a "fall back" of the motor. Additionally the motor model
is falsified. Therefore this function should be switched off for controlled drives.
Attention: The hardware current limitation limits the current at the maximum and triggers no error. This can lead
to torque sags at the motor shaft. This function is very critical especially during operation "lifting and lowering".
The drive may sag due to missing torque without brake engage.
The only exception: in speed-controlled operation with encoder feedback without motor model, current
overshoots can occur during the start if the current controller are not optimally adjusted. They can be compen-
sated by uF.15 = 1: „single-phase mode“. The hardware current limiting makes sense also at controlled drives.
The software current limiting should be used instead of the hardware protective function. The maximum permis-
sible current must be entered in parameter dr.37.
It is reasonable to enter the hardware current level (In.18) here if the application does not require another value.
The function is activated by setting "current/torque mode = 1:on" in parameter (ds.03) "current/torque mode.
The desired switching frequency can be selected in parameter uF.11. The higher the switching frequency the
smaller the noise level and the smaller the current ripple and the losses in the motor involved. Simultaneously
the losses in the inverter and also the isolation straining of the motor increase caused by switching edges.
The current ripple is a harmonic current which superimposes the sine-wave output current.
It is generated by the clocked output voltage of the frequency inverter. This ripple increases the maximum value
of the current and thus may lead to triggering of the overcurrent error or the hardware current limit, although the
displayed apparent current (ru.15) or utilization value (ru.13) is significant below this limit.
20
15
10
-5
-15
-20
0.0955 0.096 0.0965 0.097 0.0975 0.098 0.0985 0.099 0.0995
The size of the current ripple is depending on the switching frequency and the motor inductance. The current
ripple is usually insignificant for standard motors with a power < 50kW and a rated switching frequency of the
unit of min. 4 kHz.
The smaller the leakage inductance (ASM) and/or the winding inductance (SM) the higher the ripple. This is
particularly the case for motors with high power or spindle motors. Therefore the carrier frequency must be
selected as high as possible for these motors.
Attention: Generally the switching frequency should be at least 10 times higher than the maximum occurring
output frequency of the inverter.
The maximum switching frequency can be read in parameter In.03. The frequency inverter can be operated
permanently only with its rated carrier frequency (In.04) (independent on temperature and utilization).
If a carrier frequency is selected in parameter uF.11 that is higher than the rated value, an automatic "derating",
i.e. reduction of the switching frequency is carried out depending on the temperature, output frequency and
utilization of the motor. This carrier frequency change-over is generally not good for the control characteristic
of the drive. Therefore the carrier frequency uF.11 should be preferably equal to the rated switching frequency.
However the effects of the deratings can be neglected in many applications.
Introduction of a minimum voltage reserve (dS.39, default 0%) for the non-prioritized current controller. Since
the current controller in the D-axis has priority by default, a reserve is assigned to the current controller in the
Q-axis. This ensures that the drive remains controllable in the field weakening if the max. permissible Id current
(dS.13) is selected too high.
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp setting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
7
W
arn
un
g:
Di it b erst als a me
e K etr rom l e ni
ze ehl ist nah
on Ìgt
de 5 m schu
Ein halt tzm
nsa in
sc chu sig!
F er aB
tore .!
S lÌs
ntl
zu
X3B X3A
X3B X3A
Signal X3A Description
Uvar 11 Supply voltage for encoder
+5 V 12 Supply voltage for encoder
0 V 13 Reference potential
A 8 Signal input A
_
A 3 Signal input A inverted
B 9 Signal input B
4 Signal input B inverted
n 15 Reference marking input N
14 Reference marking input N inverted
Shield housing shielding
Inputs
The signal and reference marking inputs can be triggered with rectangular pulses. The signal inputs must ge-
nerally be connected. The reference marking signals are only needed for the reference point approach in the
positioning operation (F5M/S).
Signal X3A KEB servo motor Description
SIN - 3 1 Sinus signal cable inverted
SIN+ 8 10 Sinus signal cable
REF- 5 5 Reference signal inverted
REF+ 10 7 Reference signal
COS- 4 2 Cosinus signal cable inverted
COS+ 9 11 Cosinus signal cable
GND 14 - Shielding of the signal cables
Shield housing housing shielding of the hole cable
7 6
8 12
5
9 11
1 10 4 7
2 3
Fig. 7.11.2.c Resolver cable
housing housing Core color
14 GND
SIN- 1 3 SIN - red
SIN+ 10 8 SIN+ blue
REF- 5 5 REF- yellow
REF+ 7 10 REF+ green
COS- 2 4 COS- pink
COS+ 11 9 COS+ gray
Channel 2 can be equipped with different interfaces. To avoid the connection of a wrong encoder, the installed
interface is indicated in ec.10.
Channel 2 can be equipped with different interfaces. To avoid the connection of a wrong encoder, the installed
interface is indicated in ec.10.
In synchronous operation the second incremental encoder serves as input of the master drive. A second posi-
tion encoder can be connected for positioning operation.
The signal inputs of the second encoder interface support only rectangular signals.
The incremental encoder output gives out the signals recorded at the encoder interface 1:1 in RS422-specifi-
cation over the second channel (e.g. master drive in synchronous operation).
1 Channel 2 Function
0 Incremental encoder input 7
1 Incremental encoder output
0 Input with terminating resistor
2 Terminating resistor at channel 2
2 Input without terminating resistor
X3A
Uvar
X2A 5,2V +5,2V
0V
0V 22
X3B
+20...30VDC 21 Uvar
+5,2V
0V
+24VDC 0V
Uvar , +5 V
Uvar is a non-stabilized voltage, which is made available from the power circuit of the KEB COMBIVERT. De-
pending on the unit size and the load it can be 15...30 V DC. Uvar can be loaded at X3A and X3B altogether with
max.170 mA. If higher currents are required for the supply of the incremental encoder, the FI must be supplied
with an external voltage.
The +5,2 V voltage is a stabilised voltage, which at X3A and X3B is loadable with altogether 500 mA. Since the
+5,2 V are generated from Uvar, the current from Uvar decreases in accordance with following formula:
5,2 V x I+5V
Ivar = 170 mA - –————
Uvar
A condition for good control characteristics of a drive is depending on the selection and correct connection of
the encoder (see following table):
„Abs. value
„Motor data
and rotor Number
„Speed Multi- storable in „Max. line
Encoder Resolution position of con-
feedback“ Turn the enco- length“
encoder for ductors
der“
DSM“
„Incremental TTL
with 10 - 30V high yes no no no 8 high
supply“
„Incremental TTL
high yes no no no 8 Standard
with 5V supply“
Incremental HTL high yes no no no 8 very high
„Incremental HTL
without invers Standard yes no no no 5 Standard
signals“
Resolver Standard yes yes no no 6 high
ENDAT very high yes yes no yes 10 high
EnDat Low cost high yes yes no yes 10 high
Endat Multiturn very high yes yes yes yes 10 Standard
EnDat 2.2 / BISS very high yes yes no yes 6 Standard
Hiperface very high yes yes no yes 6 high
Hiperface multiturn very high yes yes yes yes 6 Standard 7
Sin/Cos very high yes no no no 8 Standard
SIN/COS with
very high yes yes no no 12 Standard
absolute track
SIN/COS with SSI very high yes yes no no 10 Standard
only abso-
SSI Standard limited1 no no 6 Standard
lute
only abso-
SSI multiturn Standard limited1 yes no 6 Standard
lute
1
SSI encoder often have a high time constant for the internal position detection.
Thus they are not suitable for speed measurement of dynamic systems.
Detailed data can be taken from the instruction manual of the respective encoder interface and from the data
sheet of the encoder.
In case of an invalid encoder identification, error „E.Hyb“ is indicated and the measured value is displayed
inverted in Ec.00/Ec.10.
On changing the encoder interface the error „E.HybC“ is indicated. The change is confirmed by writing on pa-
rameter Ec.00 or Ec.10 and the default values for the new interface are loaded. From version 4.2 the message
is acknowledged with error reset (terminal ST or RST).
With this parameter the encoder line number is adjusted to the connected encoder within a range of 1…16383.
- Ec.01 for encoder interface 1
- Ec.11 for encoder interface 2
Incremental pulses are generated by the evaluation block for resolver which does not supply incremental
pulses. The number of the generated pulses is displyed in Ec.01 in this case.
The resolution of the resolver interface is adjusted via this parameter. A change is only possible when the mo-
dulation is switched off. Simultaneously with the resolution also the increments per revolution for emulation on
channel 2 changes.
Due to the high resolution the maximum speed is limited at channel 1. The following limit values are valid:
7
Resolution / bit Maximum speed / rpm
10 150000
12 60000
14 30000
16 7500
Typically the rated frequency of the used resolver is adjusted here, usually 4, 5, or 10 kHz.
The encoder signals are measured with value 0: auto and the threshold is set to the half measured value.
This parameter defines the time over which the speed average value is determined. Thereby the resolution of
the speed measurement is defined simultaneously:
A rotation change for encoder input 1 can be executed with Ec.06 bit 0...1 and for encoder input 2 with Ec.16.
A system inversion can be activated with bit 4 (value 16). With the system inversion it is possible to run the
motor in reverse direction with positive setting at the shaft, without changing the hardware.
Rotary encoder: The control detects the whole revolutions and outputs a continuous position value to
Ec.28
Linear encoder: The controller detects the position value and outputs it toEc.28
Adjustment:
Ec.10 = 30 = „Position via Ec.28“. Independent of connection to channel 2.
Increments per revolution Ec.11 (128….16384, default 1024 inc)
Multiturn resolution Ec.21 ( 32….32, default 32 )
Resolution Ec.17 ( 2….9, default 9) 7
Position format:
The position value in a range of -2^31….2^31.1 is described in Ec.28.
The display standardization can not be changed with Ec.41. The overflow detection after power-on is not
supported!
7.11.8.1 Definition
The gear factor (ratio drive speed tooutput speed) is defined by two parameters: gear factor numerator and
gear factor denominator
Counter
Gear factor = ––––––––
Numerator
For every encoder channel, a gear factor can be given. Ec.04 / 05 or Ec.56 / 57 defines the gear factor for
channel 1. Ec.14 / 15 or Ec.58 / 59 defines the gear factor for channel 2.
The gear factor can be preset with a higher resolution and a higher value range in the second parameter pair
(Ec.56 / 57 or Ec.58 / 59).
Parameter "gear factor counter long" (Ec.56 for channel 1 or Ec.58 for channel 2) determines the parameter
pair that defines the gear factor.
If this parameter contains a value not equal to "0:off" for the respective channel, the "long" gear factors apply.
Overview of the parameters for gear factor setting:
The main uses of the gear factor occur in the operating modes positioning and synchronous running. The effect
of the gear factor and the correct settings for the various mechanical set-ups are described in more detail in
chapters 7.12.3.4 and 7.12.4.4.2 .
Examples for the special case of encoder adaption via the gear factor are listed below:
Parameter Ec.39 must be set to: „1: motor encoder" for the operation of encoders that are not directly attached
at the motor or for the operation of resolvers with a pole-pair number > 1.
Ec.39 must be set to 2 for this special operation (splitting the real increments per revolution into gear factor
denominator and increments per revolution).
This special operation is only available for channel 1.
The gear factor numerator (Ec.04 and Ec.14, respectively) can be changed via the analog parameter setting
(see chapter 7.15.9).
Example:
The target is to be able to adjust the gear factor for encoder channel 2 between 0.9 and 1.1.
Gear factor denominator is chosen as 1000.
The gear factor numerator must also be settable to between 900 and 1100.
The target of the setting is Ec.14 gear factor channel 2 numerator (bus address 100Eh)
=> An.54 Analog parameter setting destination = 100Eh
For an analog value of 0%, one should have gear factor numerator = 1000
=> An.55 Analog parameter setting offset = 1000
For an analog value of 100%, the gear factor numerator should be 1100
=> An.56 Analog parameter setting max. value = 1100
With this setting, a gear factor of 0.9 to 1.1 can be set with an Aux value of -100%...100%.
Example:
In set 0, the gear factor should have the value 0.5, in set 1, the value 1, and in set 2, the value 1.5. Gear factor
denominator is chosen as 1000. The gear factor numerator must therefore be: in set 0 = 500, in set 1 = 1000,
and in set 2 = 1500.
The target of the setting is Ec.14 gear factor channel 2 numerator (bus address 100E hex)
=> An.54 Analog parameter setting destination = 100Eh
The maximum value for the gear factor numerator should be 1500
=> An.56 Analog parameter setting max. value = 1500
The set-dependent gear factors are now realised through the different values for oP.52 "motor potentiometer
value". For that purpose, the following setting have to be made :
→ To use the function of the zero pulse emulation, only the following
increments per revolution can be adjusted: 256/512/1024/2048 inc.
→ The increments per revolution of encoder source channel 1 must be
a multiple of two in order to use the zero pulse emulation.
→ If the output switching frequency is switched to 12kHz and back,
error increments are theoretically possible.
0: 256 inc
4: 512 inc Number of increments per revolution which are output over the enco-
2, 3 Actual value
8: 1024 inc der emulation channel with the adjustment „source = 2: actual value“.
12: 2048 inc
The increments of encoder channel 1 are directly output via the en-
coder emulation.
0: direct
Use this adjustment always if „source = 2: actual value“ is parame-
terized.
16: 2
The increments of encoder channel 1 are divided by the selected
4...7 Division 32: 4
factor (2, 4, 8,…).
48: 8
64: 16 Attention: The zero signal is not divided.
80: 32 It is output once per revolution. Also the pulse duration of the zero
signal is not changed compared to the direct output. Thus it is shorter
96: 64
than the divided trace A and B signals.
112: 128
see chapter
Encoder gear factor
ec.6
ch 1
Z/N
-
ec.4/ec.5 or ec.56/57
ec.29 ec.60 see chapter
Posi / Syn
ec.33
ru.54
see chapter
Encoder gear factor
ec.16
ch 2
Z/N
-
ec.14/ec.15 or ec.58/59
ec.30 ec.61
ec.34
The system position results from the position value - system offset
(Ec.60 = Ec.31 - Ec.33; Ec.61 = Ec.32 - Ec.34)
7.11.11 Further parameters / encoder
The following parameters are required only for specific encoder interfaces and are explained more closely in
the appropriate documentation.
„Intelligent interface“ is a general term for all encoder interface types which contain their own micro-controller
for the encoder evaluation.
Parameters Ec.36...Ec.38 are only supported with these interface types.
Some encoder contain a so-called „electronic type plate“ if an intelligent interface is necessary for evaluation.
That means: the most important motor and encoder data can be stored in the encoder. Then this data can be
read out and accepted by the inverter at the start-up (for more details see "Ec.38 encoder 1r/w)
Parameter Ec.37 "encoder 1 status" indicates the actual status of the encoder and encoder interface 1.
The encoder error is only displayed in the inverter status ru.00, if the control release is set.
The encoder status is always displayed in Ec.37.
If the error message: „35: ERROR! Encoder change“ (E.EncC) the exact error message can be read in Ec.37
in inverter state ru.00.
Ec.37: Encoder 1status
Value Explanation Error
0: no communication to the no communication between interface and control board E.Hyb
interface
16: transmit position Position values are transmitted, encoder and interface are correct no
64: encoder not defined Encoder is unknown and is not supported E.EncC
68: no communication to The signals of the absolute track are incorrect. The absolute track E.EncC
the encoder at Endat, Hiperface and SSI-SinCos is digital, at SinCos it is analog.
69: increase error counter Position deviation too large. The position, determined from the in- E.EncC
cremental signals and the absolute position (absolute track, zero
signal or serial read out) do not agree or they cannot be corrected.
70: Ec.01 unequal to the en- Adjusted increments per revolution of the inverter does not agree E.EncC
coder type with the encoder increments per revolution.
71: interface identification Interface type is unknown: Interface was not identified. E.EncC
Ec.37 enc. 1 encoder status
Value Explanation Error
75: encoder temperature Encoder temperature too high (message from encoder) E.EncC
76: speed too high Speed is too high (message from encoder) E.EncC
7: int. encoder signals too low Encoder signals are out of the specification (message from encoder) E.EncC
78: int. encoder defect Encoder has an internal error (message from encoder) E.EncC
92: format encoder Encoder is formatted. When writing on an encoder, whose storage E.EncC
structure does not correspond to the KEB definition, the storage
areas are reorganised, so they can be written on. This procedure
can take several seconds, depending on the available memory
structure.
96: new encoder identifier New value recognized, because another encoder was connected E.EncC
98: damaged interface Interface is busy E.EncC
97: invalid data KEB-identification is undefined. Memory structure in the encoder E.Enc1
is not corresponding to the KEB definition and consequently the
data can not be read.
255: no communication to no communication between interface and control board E.Hyb
the interface
An error due to wrong encoder increments per revolution (value 70) is immediately reset, if the correct number
of increments per revolution is adjusted.
Attention: if the control release is still set, the drive starts to run automatically.
• Error encoder 1
If no data can be read out from the encoder (value 97), „32: ERROR! encoder 1“ (E.Enc1) is released. At F5-S
the error can be reset either by writing of a system position in Ec.02 or by system position trimming.
Data can be stored and read out in some absolute encoders (e.g. Endat, Hiperface). Thus an „electronic name
plate“ can be stored for the motor/encoder system.
If an inverter is connected the first time with an encoder which contains an electronic name plate, this is auto-
matically read out, if „data load when switching on = 4: automatic “ is adjusted in Ec.38 (factory setting at F5-S).
KEB servo motors with electronic name plates already contain the complete motor data.
The stored parameters in the encoder are divided in two groups: system and application parameter.
The data are different for synchronous motors (F5-S) and asynchronous motors (F5-M).
The following table gives an overview of the data:
F5-S F5-M
dr.23 DSM rated current dr.00 DASM rated current
dr.24 DSM rated speed dr.01 DASM rated speed
dr.25 DSM rated frequency dr.02 DASM rated voltage
dr.26 DSM EMK [Vpk*1000RPM] dr.03 DASM rated power
dr.27 DSM rated torque dr.04 DASM rated cos(phi)
dr.28 DSM curr. f. zero speed dr.05 DASM rated frequency
System
dr.30 DSM winding resistance dr.06 DASM stator resistance
dr.31 DSM winding inductance dr.07 DASM sigma-inductance
dr.32 DSM rated power ---
dr.33 DSM max. torque ---
Ec.01 encoder (inc/r) 1 Ec.01 encoder (inc/r) 1
Ec.02 absolute pos. enc.1 ---
cS.19 abs. torque ref
Application
Ec.03 time 1 for speed calc.
Read data:
If another encoder (other serial number) is connected to the same inverter, error message „35: ERROR! enco-
der change“ (E.EncC) is released first. Ec.37 „encoder 1 status“ displays the value „96: new encoder identifier“.
If this error is reset, the inverter reads automatically the data of the new encoder (according to parameterizing)
and stores them.
If the encoder does not contain electronic rating plate data, „32: ERROR! encoder 1“ (E.Enc1) is released.
If the data shall be read later again from the encoder, this must be triggered manually by setting of bit 0 „reading
activated “in Ec.38.
Which data of the encoder shall be read can be selected with bit 3 and 4 „data group/ selection“.
If the stored data cannot be read or loaded into the inverter, error message „32: ERROR! encoder 1“ (E.Enc1)
is released and parameter Ec.37 displays value 97: "invalid data".
Attention: The controller adaption is automatically triggered after read out of the data (corresponds to the set-
ting of Fr.10 = 2) and Pn.61 „absolute torque limit“ = cS.19 „absolute torque reference“ is set.
Data storage:
The data must be entered to the inverter if an encoder does not contain an electronic type plate. Next the data
can be stored in the encoder.
For this the supervisor password must be entered and Ec.38 = „2: memory activated“ must be written.
If an error occurs during storage, Ec.37 displays the status „68: no communication to the encoder“. In case of
successful storage, bit „storage activated“ is automatically reset.
The system and application data are always stored!
This parameter allows the operation of encoders that either are not mounted directly to the motor (output, belt),
whose detection have a superior position evaluation (e.g., pole-pair resolver) or whose increments per revolu-
tion cannot be set in ec.01.
As standard the delay time for resolver and incremental encoder detection is set to "0 off" in order not to affect
existing applications. Parameter dS.38 must be set to "1 on" to activate the encoder delay.
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp presetting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
The inputs which are assigned in di.11... 22 with the functions "32: forward" (limit swich right) and "64: reverse
(limit switch left) serve as hardware limit switch. Therefore, the rotation setting via terminals (oP.01 "source
of rotation direction" = 2...6) may not be used if the limit switch function is to be used. To protect against cable
breakage, an unconnected input means that the drive has run onto the limit switch.
Attention: Only the limit switch for the current direction of rotation is ever evaluated, i.e., for clockwise
rotation, only the right limit switch is considered and the left limit switch is ignored. This applies
accordingly to counter clockwise rotation. Therefore, the limit switch can act only if the drive
runs in the correct sense of rotation and the connections of the limit switches are not interchan-
ged. Furthermore, one must ensure that the drive stops at the limit switch. If the limit switch is
overrun, a new positioning in the disabled direction can be carried out.
The response to the error (the run-on to the limit switch) is set in parameter Pn.07 "proh. rot. stopping
mode". Possible responses are, e.g., triggering of an error or emergency stop (see chapter 7.15 "Protection
functions").
Note: If a function with "AutoRestart" (automatic restart) is selected as response, status "ABN.STOP
prot.rot.“ is displayed only during deceleration (display in ru.00 "inverter state" or by digital
output). Afterwards, the status changes to "ready for positioning" again.
If a function without AutoRestart is chosen, the error-/ warning- message remains displayed un-
til reset. Afterwards the status changes to "ready for positioning" again, even if the drive is still
in the direction of the limit switch. The error-/ warning- message is set again only at the next
"start positioning" command.
The software limit switches complement the function of the hardware limit switch.
They are active only after an approach to reference point or the setting of reference points, respectively (see
chapter 7.12.2 approach to reference point). In contrast to hardware limit switches, the software limit switches
can lose their protective function by, e.g., a faulty approach to reference point or a faulty position correction.
Their advantage is that they cannot be overrun.
For a positioning whose target lies outside of the permissible range, the "start positioning"-commands are
ignored. The permissible range lies between PS.15 "software limit switch left" and PS.16 "software limit switch
right". The software limit switches are active in the vector controlled operation, the synchronous mode, the po-
sitioning mode, and the contouring mode.
The response to the error (running onto the limit switch) is specified in Pn.66 "response software limit switch".
Possible responses are, e.g., triggering of an error or emergency stop (see chapter 7.13 "protection functions").
Note: If a function with "AutoRestart" (automatic restart) is selected as response, neither "disabled direction of
rotation" is displayed in status "warning! nor switching condition "quick stop/ error" is set. Cause: As soon as
the setpoint speed is equal to zero, the drive does not run in a disabled direction of rotation anymore, and the
malfunction is reset automatically. The drive also displays "ready for positioning", but does not react to "start
positioning" commands anymore as long as the target position lies outside the permissible range.
If a function without AutoRestart is chosen, the error-/ warning- message remains displayed until reset. After-
wards, the status changes to "ready for positioning" again.
● program and connect an input as reference point switch(PS.18). The same input can also be used
as a limit switch. Since the limit switches are "zero active" for protection against cable breakage, the
reference switch, in this case, is also "zero active". If the reference switch is connected to its own
input, it is "one active".
● Define an input for the start of the approach to reference point (with PS.19/ only necessary in mode 1
approach to reference point).
● connect the limit switches (forward = right limit switch / reverse = left limit switch) to the inputs which
are programmed in di.11...12 with the function "32: forward" and "64: reverse". If the limit switches
shall be omitted (e.g. at rotary table applications), no input may be assigned with the function "for-
ward" or "backward".
● The approach to reference point must be activated in "approach to reference point mode" of parame-
ter PS.14.
In mode 1 and 2, the approach to reference point is started on the rising edge of input "start approach to refe-
rence point" (mode 1) and "start positioning" (mode 2), respectively.
The approach to reference point starts with the speed adjusted in PS.21 "reference speed". The direction of
rotation which is used first for the reference point search (the preferred direction f rotation) is set by the sign
of PS.21 . A positive sign means the drive first looks for the reference point switch in the clockwise direction of
rotation.
The acceleration / deceleration ramps during the approach to reference point are not defined via the OP para-
meters but via PS.20 "reference acc./dec. time".
Attention: the ramp time and the approach to reference point speed must be chosen so that the drive can stop
and reverse as long as the reference point switch is active. Otherwise, faulty referencing can occur (e.g., stop
on the wrong side of the reference point).
To achieve the most precise referencing, an "approach to reference point free drive-speed" can be programmed
in PS.22. for free driving of the reference switch. If this parameter is set to "0:off", the free drive-speed is taken
as ¼ of the approach to reference point speed (PS.21) .
The actual position is overwritten with the value of PS.17 "reference point" at the reference point.
The following two pictures show an exemplary approach to reference point process. The further possibililties
programmable in PS.14 are shown later in this chapter.
Reference
point search
Start completed
PS.21: Approach to
reference point Speed = 5/6 2/3
positive 1
1 2 3 4 5 6
ru.02: Set
speed
ru.54: actual
position
Approach to
reference point
completed
Target window
reached
Reference point
switch
Start approach
7
to reference
point
1. PS.21 = positive, i.e., the drive accelerates with the ramp from PS.20 and seeks in forward direction for
the reference switch
2. Stopping at the reference switch
3. Free driving of the reference switch with free drive-speed (PS.21 / PS.22)
4. Stopping of the drive with ramp from PS.20
Setting of the signal "target window reached"
5. Wait for the damping period of 100ms
6. Overwriting the current actual position (ru.54) with the reference point position (PS.17)
Resetting of the signal "target window reached"
Setting of the signal "approach to reference point completed"
Stopping of the drive left to the reference point (programmable via PS.14)
PS.14 determines on which side of the reference point switch the drive is positioned after the approach to re-
ference point. Even if, after the free driving of the reference switch, positioning is to occur based on the zero
signal, "stopping point " determines if the first null signal is to be driven on at the right or left of the reference
switch.
The adjustment is only considered if the reference point switch serves not simultaneously as limit switch.
An approach to reference point that depends only on the initiator signal of the reference point switch is insuf-
ficiently precise for many applications. Therefore, the possibility exists to couple the reference point with the
marker pulse of the encoder.
To that end, positioning after the free driving of the reference switch is done on the marker pulse of the encoder
and the current actual position=marker pulse is then overwritten with the reference point value.
With "stop at null signal= 4: yes" this function is activated.
Additionally, two monitoring functions can be switched in with bit 4 "error if no zero signal" and bit 8 "Verify zero
signal". These are only active if "stop at null signal= yes" is programmed.
Apart from the two modes "stop at null signal" or "stop after driving free" of the reference switch, there is a
third reference point drive mode:
When the drive reaches the hardware limit switch for the direction of rotation, it automatically reverses and
begins searching
the reference point in the other direction of rotation.
If no reference point switch is found, the drive continuously shuttles between the two hardware limit switches.
Note: During the approach to reference point, the limit switch function works differently than in usual operation.
If in Pn.07 "proh. rot. stopping mode" value: "6: function switched off" is programmed, the drive reverses with
the defined acceleration and deceleration times of parameter PS.20 "reference acc/dec. time".
For all other value of Pn.07, the setpoint speed without ramp is set to zero. The drive stops and then accele-
rates in the other direction of rotation with the ramp from PS.20. No normal quick stop is executed, the quick
stop parameters (Pn.60 / Pn.61 / Pn.67) have no function.
Reference
7
point search
completed Start
1 2 3 4 5 6 7
ru.02: Set
speed
ru.54: actual
position
Limit switch
right
Approach to
reference point
completed
Target window
reached
Reference
point switch
Start approach
to reference
point
1. PS.21 positive
Drive accelerates with ramp from PS.20 and seeks in forward direction for the reference switch
2. Run-on to the limit switch
Reverse and seek in the other direction of rotation
3. Overdriving of the reference switch
(because stopping the drive left to the reference switch is selected in PS.14, the switch must be ap-
proached from the right)
4. Reversing and running onto the reference switch in direction of rotation clockwise
5. Reversing on the reference switch and driving free with drive free speed (PS.21/ PS.22)
6. Stopping of the drive with the ramp from PS.20
Setting of the signal "target window reached"
Wait for the damping period of 100ms
7. Overwriting the current actual position (ru.54) with the reference point position(PS.17)
Resetting of the signal "target window reached"
Setting of the signal "approach to reference point completed"
Stopping of the drive left of the reference point (programmable via PS.14)
If no reference point switch is provided in the application, the drive can also be manually referenced:
PS.14: Mode of position reference
Bit Meaning Value Explanation
0: off No manual setting
The drive is approached in inching mode to reference point and then
6 Manual setting "manual setting = on" (bit 6) is set. The reference point position (PS.17)
64: on is taken as actual position (ru.54). The switching condition "approach to
reference point completed” (do.00...07, value 29) is set, the software limit
switch function can be used.
Independent of PS.14 "mode of position reference" or PS.00 "pos/syn mode", the actual position ru.54 can be
overwritten with the value of PS.17 "reference point" by setting a digital input.
TO that end, an input must be selected in PS.13 "set reference point input selection".
(assignment of a digital input see chapter 7.3)
In order for the software limit switch function to be useable, an approach to reference point must be executed
prior to the positioning. In some cases (e.g., when using an absolute encoder), an approach to reference point
is, however, not required. By activation of bit 7 "the captured position is valid = yes", the drive is informed that
no approach to reference point is necessary.
By setting bit 5 (stop at index 0 = 32: on), it can be programmed that the drive after completion approach to
reference point moves automatically (i.e. without „start positioning“ signal) to the position of index 0.
PS.20 "approach to reference point ramp time" specifies the acceleration / deceleration values for the posi-
tioning to index 0. The maximum profile speed for positioning is determined for Index 0 by the value of PS.25
"index speed".
The drive remains at that Position. The setting "continue profile = yes” from index 0 is ignored.
The following figure shows an approach to reference point with stop at the null signal left of the reference point
switch and automatic positioning to index 0:
1 2 3 4 5 6 7 8
In order that approach to reference point is more precise, positioning to the zero position of the encoder can be
done after driving free of the reference switch. That this always happens to the same position, it is necessary
to adjust the zero position of the encoder (zero signal) mechanically in such a way that it occurs a half motor
revolution after reference switch. The distance from reference switch to zero position is displayed in ru.69.
Adjustment by software is significant more comfortable compared to the mechanical adjustment and is done via
parameter PS.60. Parameter PS.60 „zero puls offset“ specifies the position offset of the zero signal:
reasonable value range of PS.60 : - increments per revolution / 2……. + increments per revolution /2
PS.60 = +/- encoder increments per revolution/2 + position value – zero signal position of the encoder
PS.14 bit12 must be deactivated after the calculation. The position of the reference switch to the zero signal
can be monitored with parameter PS.14 Bit8 in a range 1/4… 2/3 revolution (ru.59).
7
7.12.2.10 Approach to reference point with overdriving of the reference limit switch
The reference limit switch is overdriven at first with setting PS.14 = Bit13. Driving free occurs when the refe-
rence limit switch with constant direction of rotation is no longer active.
If the reference limit switch is already active at the start of the approach to reference point, it is driven free at
first against to the preferred direction. Subsequently the drive reverses and drives to the reference limit switch
again.
Reading and the position can be reset to "0" via parameter Ec.38. Value 33 must be entered in parameter Ec.38
to execute the reset. If the reset is successfully completed, parameter Ec.38 is set to 0.
The reset can only be carried out if the encoder / encoder interface supports this function.
This function is supported upto encoder interface software date 11.05.2011.
Only Hiperface encoder are supported.
The synchronous module realises an angle / speed synchronous control of a master drive (control drive) to one
or more slave drives. The control drive must not be closed-loop.
The master position is passed on to the slave. The master must therefore be equipped with an encoder inter-
face with incremental encoder output, and every slave with a second incremental encoder input.
Alternatively, the master can also be operated uncontrolled and the encoder signals of the master drive can be
connected directly to the slave.
The speed ratios are adjustable individually. The gear ratio is adjusted via the numerator / denominator ratio. If the direc-
tions of rotation have to be different, a negative gear ratio has to be set.
For activated position controller, the slave is driven angular-synchronous, for deactivated position controller
(PS.06 = 0), speed-synchronous to the master drive.
The synchronous module contains other variants for synchronisation (constant acceleration ramp or constant
synchronisation path) and a programmable angle adjustment.
The following mapping shows the general behaviour of synchronous control (without synchronisation phases):
oP.14/oP.15
ru.10 absolute
speed
Actual speed maximum
Ec.11 calculation
encoder 2 setpoint
encoder ru.02
2 (inc/r) + display
output display
(setpoint speed for
Ec.01 + speed controller)
encoder 1
(inc/r)
PS.09
Increments pos
ru.56
of encoder Posi/synchronous
Set position
channel 2
Ec.14
Gear 2 nume- + PS.06 to
rator
PS.08
Position
Ec.15 - controller
Gear factor
2 denominator
Increments ru.54
of encoder actual
channel 1 position
For the synchronous module, the incremental signals from the encoder of the master drive must be passed on
to the slave.
START
START START START
ENTER FUNC.
ENTER FUNC. ENTER FUNC. ENTER FUNC.
F/R SPEED
F/R SPEED F/R SPEED F/R SPEED
STOP
STOP STOP STOP
ANTRIEBSTECHNI K
ANTRIEBSTECHNI K ANTRIEBSTECHNI K ANTRIEBSTECHNI K
3 3 3 3
~ ~ ~ ~
If more than one slave is connected, there are two different variations for assembling the master-slave-chain:
direct transfer of the signals from the output of the master encoder interface to all slaves.
7
START
START START START
ENTER FUNC.
ENTER FUNC. ENTER FUNC. ENTER FUNC.
F/R SPEED
F/R SPEED F/R SPEED F/R SPEED
STOP
STOP STOP STOP
ANTRIEBSTECHNI K
ANTRIEBSTECHNI K ANTRIEBSTECHNI K ANTRIEBSTECHNI K
3 3 3 3
~ ~ ~ ~
Ec.20 encoder operating mode = 1 Ec.20 encoder operating mode = 0
(input + without terminating resistor) (input + with terminating resistor)
Disadvantages:
START
START START START
ENTER FUNC.
ENTER FUNC. ENTER FUNC. ENTER FUNC.
F/R SPEED
F/R SPEED F/R SPEED F/R SPEED
STOP
STOP STOP STOP
ANTRIEBSTECHNI K
ANTRIEBSTECHNI K ANTRIEBSTECHNI K ANTRIEBSTECHNI K
3 3 3 3
~ ~ ~ ~
Advantages:
- The incremental signals are processed. Thus, no limitation of the number of the connected slaves
- off-the-shelf cables available that guarantee EMC compliant assembly. Further information on availa-
ble components can be found at wow.keb.de => Service & Downloads
- Error control for processed master signal integrated in repeater
The channel where the slave receives the master position is selected via parameter PS.01 actual position sour-
ce. This must be encoder channel 2 for most applications. (Off-the-shelf cables and a terminating resistor that
can be switched off exist only for channel 2 ).
chines or
STOP
STOP
rolling machines.
ANTRIEBSTECHNI K
ANTRIEBSTECHNI K
3 3
~ ~
i= 5,25 i= 15
7
Load Load
The slave position (i.e. the number of increments of the slave motor) is displayed in parameter ru.54 "actual
position". In ru.54, one revolution of the slave load corresponds to:
The master position is displayed in parameter ru.56 "set position". The display occurs in increments and is con-
verted to the slave position. The conversion considers the ratio of the increments per revolution of the encoder
and the ratio of the two gear factors. If the master is connected to encoder channel 2, the gear factor of the
slave must be entered in parameter Ec.14 "gear factor 2 numerator" and the gear factor of the master must be
entered in parameter Ec.15 "gear factor 2 denominator" for conversion of the gear ratios.
Since only integer values can be preset, the gear factors must be extended correspondingly
(15 : 5.25 becomes 1500 : 525). Display in ru.56 (master position converted to slave units):
The inverse direction of rotation of the slave to the master drive is reached by a negative value for
Ec.14.
"Normal" speed-controlled operation is programmed in the master, the synchronous module is not activated.
For the slave, encoder channel 1 serves as speed feedback and encoder channel 2 as master position infor-
mation. Both loads shall be moved angular-synchronous, but in opposite direction of rotation.
Typically, an approach to reference point is executed for the slave drive before starting synchronous running,
in order to get the reference between position display of the slave drive and the mechanics of the application.
The reference between master and slave position is done only with activation of the synchronous module.
During activation the master position (= ru.56 "set position") is set equal to the slave position (= ru.54 "actual
position").
The operating mode synchronous mode is selected via parameters PS.00 bit 0...3 or via the control word (Sy.43
or Sy.50)
PS.00 can only be written if the modulation is switched off, Sy.50 can always be written. The synchronous mo-
dule must be activated by an input. Which input is to be used is determined via parameter PS.02 "posi / sync
input selection".
7.12.3.5.1 Principle
With the activation of the synchronous module, the relation between master position and slave position is es-
tablished.
The synchronous module is not deactivated by switching of the modulation. The angle difference is continuous-
ly calculated, and, after again switching on the modulation, a synchronisation with ramps is always carried out
(independent of the type of initial synchronisation).
The synchronisation type at activation of the synchronous running is determined by the adjustment of "syn-
chronous running / starting ramp (oP.28)" in parameter PS.00 "pos/syn mode" and by parameter PS.05 "start
offset" .
The starting ramp must be deactivated for the synchronization at torque limit.
For the synchronisation within a constant path, the starting ramp must be deactivated.
The distance the master is made during synchronisation is entered in parameter PS.05 "start offset". The slave
drive calculates internally the acceleration / deceleration times on which it reaches the master speed within the
adjusted distance. , The master position is set to the slave position if the master has travelled the programmed
distance.
Example:
The master speed is 1500 rpm. The encoder type is an incremental encoder with 2500 pulses. Value "2: 4-fold"
is adjusted for "multiple evaluation".
If the value 250000 increments is adjusted in PS.05, the slave must accelerate to the master speed in 1s.
The disadvantages of this method of start-synchronization are described as follows:
- The synchronisation path is set via parameter PS.05 "start offset", making it difficult to realise an offset
between master and slave at the start.
- Checking whether the synchronisation was successful is not possible. If the slave drive cannot follow the
calculated ramp (e.g., due to reaching the torque limit), the master position is still set to the slave position.
The angle synchronicity is lost thereby (for the example above, the connection to the position of the switch
"activate synchronous running" would be lost). The switching condition "drive running synchronously" is
also still set in spite of the angle error.
- Even if the slave drive can generally follow, system deviations can distort the accuracy of the angle-
synchronous running.
6
7 7
t
Synchronization with ramp is the most comfortable method for the initial synchronization. It is always used for
synchronization after interruption of the synchronous running due to switching off the modulation.
With activation of the synchronous module, the master position (set point position ru.56) is set to the slave
position (current position ru.54).
The slave accelerates with the predefined ramps, to follow the master.
Because of the different speeds of master and slave, an angle difference occurs. This missing distance to reach
the master position is made up for by increasing the slave speed beyond the master speed. The slave calcu-
lates a setpoint speed profile, which allows it to make up for the angle difference.
Maximum speed for this setpoint profile is the maximum setpoint oP.10 / oP.11. If the drive cannot follow the
setpoint speed profile, the remaining angle difference is eliminated by the position controller. Therefore, the
maximum speed during the synchronisation phase is oP.10 / oP.11 + position controller limit PS.09. This value
is still limited by the absolute maximum setpoints (oP.14 / oP.15).
As soon as the slave reaches the last phase of the synchronization(that means: the last deceleration or acceleration
to the master speed), the ramp can deviate from the programmed values. This is the case if the master speed is not
constant, i.e., if adjustments still have to be made during the running-in. Adjustments of the values for acceleration or
deceleration are not accepted anymore during this phase.
Additionally, an offset can be entered in parameter PS.05 "start offset" to run the master offset to the slave . The
master position is set to the value slave position – PS.05 upon activation of the synchronous running.
That means: ru.56 = ru.54 – PS.05 (at the time of activation)
Action start
(drilling) 1 cm
0 10 20 30 40 2
50 60 70 80
Action end
(drilling)
1 cm
0 10 20 30 40 50 602 70 80
The master drive is a material carrier (e.g. a conveyor belt) which is used to transport material (e.g. boards)
with variable speed.
The leading edge of the material crosses an indicator and thereby activates the synchronous running of the
slave drive. 7
The slave is a tool carrier (transporting, e.g., a drill drive). As long as there is no board, it remains at a defined
resting position (20 cm).
The hole should be drilled 5cm from the front edge while the conveyor is running.
The slave must run absolutely angular-synchronously to the master during the drilling.
If the board has reached the second switch, the drilling process must be completed safely. The synchronous
running is deactivated and now the slave can (e.g., in positioning operation) run back to the starting position.
From the activation of the synchronous running to the start of the drilling, the master must travel 50cm and the
slave 30cm.
In parameter PS.05 "start offset", therefore, an offset of 20cm, converted to increments, must be entered.
For the example - 10cm = 3 motor rotations => 20cm = 6 motor rotations
above:
- 2500 encoder / 4-fold evaluation => 10000 increments per revolution
- PS.05 = 6 * 10000 = 60000 increments
During the acceleration phase to the master speed, the slave runs with average speed:
In our example:
The master speed converted to the slave standardisation therefore is 1500 rpm.
The slave speed at the start is zero. Let the acceleration time be 0,2s (per 1000 rpm).
For the acceleration from 0 to 1500 rpm the slave therefore needs 0,3s. The mean speed in the acceleration
phase is 750 rpm = 12.5 U/s. Each revolution corresponds to 10,000 increments.
The master (converted to slave standardisation) constantly runs with 1500 rpm = 75000 increments in 0,3s. The
difference between master and slave is 37500 increments. PS.05 is 60000 increments.
The slave has to wait until the master has still travelled 22500 increments and then synchronises itself without
overshooting according to the adjusted ramp.
Master and slave are running at the correct offset angle after 37500 increments = 3.75 revolution = 12.5 cm of
the slave. They are synchronous starting at the drill head position 32.5cm.
The gear factor between master and slave is entered in the parameters for the encoder channel connected to
the master position. Normally, this is encoder channel 2. Therefore, the gear factor must be entered in Ec.14 (or
Ec.58) "gear factor 2 numerator" (= gear factor of the slave) and Ec.15 (or Ec.59) "gear factor 2 denominator"
(= gear factor of the master).
The gear factor is not set-programmable. If it is to be adjusted set-dependently, this can be implemented by ap-
propriately setting the analog parameters (see chapter 7.15.9). Value „1: motorpoti" must be selected in An.53
as source. The motorpoti value (OP.52) is set-programmable.
Control by means of an analog channel is also possible via the analog parameter setting.
The new gear factor during the active synchronous operation changes (at equal actual speed of the master
drive) the master speed expressed in the scale of the slave. Caused by the different
speed of master and slave now there is an angle difference and the slave must be synchronized again.
If the gear factor change is smaller than 0.5%, the change is applied without ramp.
An angle deviation between master and slave can be generated or eliminated in synchronous operation with
parameter PS.04.
With the positive edge of the input selected in parameter PS.03 "shift. slave input selection", a positive adjust-
ment is triggered.
A negative adjustment is triggered with the positive edge of the input selected in parameter PS.10 "shift slave
inv. input selection".
The adjustment is always made with the synchronisation via ramps (see item 7.12.3.5.4) to avoid torque sur-
ges in the drive. The treatment of the switching condition "drive running synchronously" corresponds to the
behaviour during initial synchronisation with ramps.
The angle adjustment can, e.g., be used to align master and slave after the approach to reference point in the
inching mode.
An input can be defined via the parameter "reset master/slave difference input selection" (ps.11) that sets the
current angle difference between master and slave to zero.
At the rising edge of the input, the master position (= ru.56 "set point position") is set equal to the slave position
(= ru.54 "actual position"). Resetting the angle adjustment is done without ramps. Switching condition "drive
running synchronously" remains set.
3 3
~ ~
i= 5,25 i= 15
Load Load
The positioning module must be activated by an input. The input is selected in parameter PS.02 "Posi/Sync.
input selection".
In the posi mode, the drive can approach a single position or sequences of positions can be programmed that
are reached consecutively or passed through with a defined speed.
Up to 32 positions can be stored in the inverter. For every position, a maximum profile speed can be program-
med.
To be able to report various operating condition (e.g., positioning active, target reached) to an overriding con-
trol, specific progress messages and switching conditions exist for the digital outputs.
The drive can be adapted very flexible to the application, since different reactions are programmable, e.g. at a
new target setting during running positioning it can be selected between:
At each positioning, the inverter calculates in 1ms-cycle a speed and position that the drive
should have at that time, to reach the target in compliance with all settings. This is the so-called speed / position
profile.
Settings
Maximum acceleration /
defined by acceleration / deceleration time (oP.28...oP.31)
deceleration
Maximum jerk defined by S-curve times (oP.32...oP.35 and oP.70...oP.73)
= ru.63 "profile speed" + PS.09 "pos/syn position limit.
The profile speed is either PS.25 "index / speed" or PS.31 "max. speed %" *
Maximum speed during oP.10 "max. reference forward" (dependent on PS.00/ bit 4). The speed limits
positioning oP.10 / oP.11 "max. reference" do not act as setpoint limits anymore. oP.14 /
oP.15 "abs. max. reference" remains operative. The error "speed limit exceeded"
is triggered when exceeding the trigger level oP.40 / oP41.
That results in the following example behaviour of position, speed, acceleration, and jolt:
ru.61 target
position
Profile speed
acceleration/
deceleration
jerk
time
If the drive cannot follow the position profile (e.g., due to reaching the torque limits), the position controller
intervenes and changes the setpoint speed with respect to the profile speed. Thereby it is possible that the
programmed values for maximum acceleration / deceleration and maximum jolt are exceeded.
oP.28..31
Acceleration /
deceleration time
ru.56 PS.06..PS.08
Set position + Position
oP.32..35 and oP.70..73
- controller
S-curve times
(jerk limitation)
ru.54
actual position
Note: after activation of the positioning module, the drive remains in vector controlled operation until the first
"start positioning" command has been executed. The behavior during this phase is determined with PS.00 bit
13.
Parameter ru.00 shows, with the progress message "121: ready for positioning", that the positioning mode has
been activated. But the drive is in position controlled operation only after the first "start positioning". The positi-
on controlled operation is ended as soon as the positioning module is deactivated.
The resolution of the position display/ setting is done in increments and depends on the used encoder system.
Position control is based on the motor position encoder. I.e., the position values refer to the motor position.
The number of increments per motor revolution amounts to "encoder increments per revolution" x 2 "multiple
evaluation".
If encoder interface 1 (X3A) is used, Ec.01 "encoder 1 (inc/r)" and Ec.07 "enc.1 trigger" must be used for the
calculation.
If encoder interface 2 (X3B) is used, the number of increments per motor revolution must be calculated corre-
sponding of Ec.11 "encoder 2 (inc/r)" and Ec.17 "enc.2 trigger".
If the position control is done directly on the motor encoder, the same encoder channel must be selected in
PS.01 "master source" and cS.01 "act. source".
Value 1 (i.e. gear factor numerator = gear factor denominator) must be selected for the gear factors .
Example:
KEB COMBIVERT
F5A-S
Let an incremental encoder with 2500 increments be
or
M connected to encoder channel 1.
F5A-S 3~ The motor shall travel 5.5 revolutions
Revolution in increments:
cS.01 = PS.01 = 0: Channel 1
X3B X3A Ec.04 = EC.05 = 1000 default value
Encoder Encoder Ec.01 = 2500 increments per revolution
channel 2 channel 1
Ec.07 = 2: 4-fold evaluation
→ 2500 x 22 x 5.5 = 55000 increments
The position control is done directly on the value of the output encoder. I.e., the position values refer to
the position of the load.
Number of increments per load revolution = "encoder increments per revolution" (output encoder) x 2 "multiple
evaluation". Typically, the encoder interface 1 (X3A) is used for the motor position encoder and the encoder
interface 2 (X3B) for the output encoder.
To allow calculation of the speed precontrol profile for the speed control, the gear factor
between motor and load must be known, to convert the precontrol profile to the motor speed.
The speed limits and the values for maximum profile speed (PS.25) and maximum position control effect
(PS.09) refer to the motor speed.
Example: 7
Encoder channel 1: Incremental encoder with 2500 increments per revolution
Encoder channel 2: SSI encoder multiturn with 12bit resolution per revolution and 12bit multiturn
Gear ratio: 3 motor revolutions cause 1 load revolution
The load shall travel 5.5 revolutions: 1024 x 22 x 5.5 = 22,528 increments
7.12.4.4.3 Speed and position control by motor encoder/ encoder mounting after gear
Mounting of the encoder for the speed control via a gear box is not ideal, since the tolerance of the gearbox and
the ratio of gear 1 affect the control quality and dynamics of the speed controller
(and thus also the superimposed position control).
Two reasons can make this set-up necessary:
Application 1:
Gear
KEB COMBIVERT
2
Directly mounting the encoder
F5A-S to the motor is not possible for
or M mechanical reasons.
F5A-S
3~
Referencing the load position is
Last not possible.
Gear
X3B X3A 1
Encoder Encoder
channel 2 channel 1
Motor
encoder
Application 2:
KEB COMBIVERT Gear Output
encoder
One wants to run with only one
F5A-S encoder which must be installed
or Last
F5A-S behind the gear for reasons of
M positioning accuracy.
3~
X3B X3A
Encoder Encoder
channel 2 channel 1
If synchronous motors are to be operated in this set-up, it must be ensured that the gear ratio is < 1 and the
value of pole-pair number x gear factor is integer.
Example:
3 encoder revolutions correspond to one motor revolution
Pole-pair number = 15
"encoder increments per revolution" * 2 "multiple evaluation" x "gear factor denominator" / "gear factor nume-
rator"
The number of increments per load revolution for application 2 is equal to:
The encoder should always be connected to channel 1, since the software for this channel optimally supports
the motor encoder connection via gear. Parameter Ec.39 „enc. 1 over transmission“ must be set to value „1:
motor encoder“. (For further functions and settings of Ec.39 see chapter 7.11)
Encoder channel 1: SSI encoder multiturn with 12bit resolution per revolution and 12bit multiturn
Gear ratio: Motor to encoder = 5 motor revolutions correspond to motor revolution = 5 to 1
The load shall travel 5.5 revolutions: → 1024 x 22 x 5.5 = 22628 increments
The operation of synchronous motors is not supported and the position normalisation is as follows:
The number of increments per motor revolution is calculated as:
"encoder increments per revolution" x 2 "multiple evaluation" x "gear factor numerator" / "gear factor denomi-
nator"
With Ec.39 = 0 (=off), therefore, the meaning of the position display and setting (the position normalisation) is
changed.
The encoder interface for the feedback of the position control is adjusted via PS.01.
*) Exception: If PS.00 is selected in "synchronous" mode, cS.01 determines the source for the position value
channel. The source for the master position is defined in PS.01 bit 1...3.
**) Positioning is not possible in torque operation, actual position source remains active.
***) If cs.01 <> ps01 (in „posi“ mode), the gear ratio is bridged for the position evaluation in the selected channel
of ps.01 bit0..3.
This factor is considered in the calculation of the speed profile at positioning.
If the positioning mode is selected in parameter PS.00 „pos/syn mode", the actual position (ru.54) is taken from
the encoder channel adjusted in „act. master source“ PS.01. This also applies if the positioning module is not
activated (i.e., the selected input in PS.02 is not set).
If the positioning mode is deactivated in PS.00 , the actual position of the encoder channel set in "actual value
source" cS.01 is evaluated.
Approach to reference point must be executed (see chapter 7.12.2) to obtain a reference point for the actual
position at speed encoders without absolute position information (e.g. incremental encoder).
This determines to which mechanical position value 0 (e.g.) shall be assigned in parameter ru.54 "actual posi-
tion".
In posi mode there are two parameters that provide information about the set position:
Parameter ru.61 "target position" shows the target position for the running positioning, i.e., the position the drive
should have reached at the end of the positioning.
Parameter ru.56 "set position" displays the position the drive should have reached currently.
This position is the setpoint for the position controller. It is calculated by the inverter in 1ms cycles, dependent
on the adjusted ramp times and the permitted positioning speed.
Special function for position detection systems with high deceleration (e.g., some opto-electronic distance
measurement systems):
In parameter Ec.46 "PT1-time channel 1", or Ec.47 "PT1-time channel 2", one can enter the time by which the
position information from the measurement system is delayed.
If a PT1-time is defined for the encoder channel entered in PS.01 as feedback for the position control, the set
position ru.56 is also delayed by that time. Thereby, the position controller does not respond to the position
difference caused by the time delay of the measurement system.Since these position differences do not really
exist, their masking improves the control characteristic of the position controller.
The difference between the set point position ru.56 and the actual position ru.54 is displayed in parameter ru.58
"angle difference".
To carry out a single positioning, the following initial settings must be met:
Now one can adjust the maximum profile speed via the AUX-input (see also chapter 7.15.9). The parameter
PS.25 must be set to the value "0" , so that the drive stops at the target. If PS.25 contains a value unequal to 0,
the drive reaches this speed at the target position and continues running constantly at that speed.
time
PS.00 bit 4 = 16
For an application, the drive shall always run with a lower profile speed between 2 positions
(e.g., joint in traversing rail). By switching the set, parameter oP.10 is decreased in this range.
After reaching this range, the drive decelerates according to the adjusted deceleration and jolt values to the
new maximum profile speed. This insures that during every positioning to an arbitrary position, the maximum
speed for this range is observed without the need for intervention by a superior control.
target
position
actual
position
Set
speed
time
With the sequential- or index-positioning, it is possible to run to several positions consecutively and, respec-
tively, traverse these with a defined speed.
Sequential positioning is meant, if several target positions are defined in the inverter
which are to be processed in a fixed sequence.
A possible example for a sequential positioning would be a drive that lowers a drill head. The drilling process
shall consist of 5 positioning steps:
● fast lowering of the drill with speed A from the starting position "0" up to position "1" ("1" = position
just before the material surface)
● slow penetration into the material (position "1" to "2") at speed B
● somewhat faster lowering (at speed C) during drilling of the material up to position "3"
● withdrawal of drill from the material at speed D, back to position "1"
● return to starting position "0" at speed A and stop there
numbered).
Maximally 32 blocks can be programmed. There are two possibililties to define the maximum profile speed:
There are 2 possibililties for the positioning process that are distinguished by PS.27 bit 0:
In parameter PS.28 "starting index new profile", the block with which the sequence begins is defined.
The position selection via terminal strip can only be selected indirectly via parameter sets 0 ... 7 with PS.58.
With PS.56 and 57 it is possible to reach all 32 posi indices via the terminal (or IA, IB, IC, ID). Settings > 31 are
limited to 31 without error.
To clearly explain the sequential positioning, four examples are listed in the following:
● Sequential positioning with automatic continuation and definition of the profile speed by PS.25 (drill
head positioning).
● Sequential positioning with stop between the separate positioning steps. New start pulse by external
control required for each step. Definition of the profile speed by PS.25 (positioning of a workpiece for
various processing steps).
● Example 3 is a variation of example 2. The profile speed is defined by PS.31 instead of PS.25 . PS.31
is defined via an analog input.
● Sequential positioning with stop between the positioning steps. The length of pause is adjustable.
The new start pulse is generated automatically. Definition of the profile speed by PS.25. Utilisation of
the timer functionality and of the input / output handling of the inverter for generation of an automatic
sequence control (positioning of a workpiece for various processing steps)
If these detailed examples are not required, continue reading in chapter 7.12.4.10 positioning mode/ round
table.
Sequential positioning with automatically continuation and definition of the profile speed by PS.25.
Settings:
● Let the position at which the positioning process starts have the value 0.
● Driving to this position (whether after "power on“, after error or as part of the positioning process) can
always occur at maximum speed = 1500 rpm.
● Let the position just before the material surface have the value 95,000. The speed at that point is to be
250 rpm.
● The penetration is completed at position 100,000. Here, the drive may again have accelerated to 500
rpm. At that speed, drilling is continued up to position 150,000.
● The withdrawal from the material back to position 95,000 shall occur at 700 rpm.
● Thereafter, the drill head returns to the starting position at 1500 rpm.
● The drilling process is followed automatically.
To solve this problem, several approaches are possible. An exact description of the available parameters and
alternative settings follows later in this chapter.
Adjustments:
All other bits can remain at the factory settings for this example and are explained in the following chapters.
ru.61: tar-
get position
automatic operation of the drilling profile
ru.56: Set
position
act. positi-
on index
Start posi-
tioning
In the example above, the drive did not stop after each step in the drilling process, instead, the target position
of the individual steps was crossed already at the speed set for the next drilling step. I.e., the parameter PS.25
"index/ speed" defines the positioning speed for a block, while the value of PS.25 of the following block deter-
mined the speed at which the target position is crossed.
For example: the "penetration block" is block 2. The speedduring penetration (positioning speed) is the value
of PS.25 in block 2 = 250 rpm.
The drilling is to continue with 500 rpm, so the drive already accelerates at the end of the penetration to the
drilling speed of 500 rpm, i.e., the value of PS.25 in block 3. The speed at which the target of block 2 is passed
(= the target speed) is also determined by block 3 (the following block).
If the drive has to reverse to reach the next target (change the direction of rotation) or if the next target shall
not be driven to automatically (PS.27: "continuation of the profile processing" = no), the target speed of a block
automatically becomes 0 (standstill at target).
Example 2: Positioning of a workpiece for various processing steps / sequence control by external
control
Sequential positioning with stop between the positioning steps and definition of the profile speed by PS.25.
Settings:
● The drive shall stop at each position to allow processing of the workpiece, until the external control
gives the signal to continue, i.e., "start positioning". The external signal occurs via input I3. The drive
shall signal the control through an output that it has reached target with an accuracy of 10 increments,
so that the processing can begin.
● Let the position at which the workpiece starts have the value 0.
● Driving to this position (whether after "power on“, after error or as part of the positioning process) can
always occur at maximum speed = 1500 rpm.
● The first stop shall be at position 100,000. The profile speed up to that point shall be 1000 rpm.
● The second stop shall be at position 200,000. The profile speed up to that position shall again be 1000
rpm.
● After that, the drive shall return to its starting position with maximally 1500 rpm.
Adjustments:
● PS.00:
All other bits can remain at the factory settings for this example and are explained in the following chapters. 7
● Block 0 defines the start position → PS.28 = 0
Example 3: Positioning of a workpiece for various processing steps / sequence control by external
control / analog setting of the maximum profile speed
Sequential positioning with stop between the positioning steps and definition of the profile speed by PS.31 /
oP.10.
Settings:
● The drive shall stop at each position to allow processing of the workpiece, until the external control
gives the signal to continue, i.e., "start positioning". The external signal occurs via input I3. The drive
shall signal the control through an output that it has reached target with an accuracy of 10 increments,
so that the processing can begin.
The maximum profile speed shall be set via the analog input AN2 (X2A.3 / X2A.4):
● Let the position at which the workpiece starts have the value 0.
● The first stop shall be at position 100,000.
● The second stop shall be at position 200,000.
● After that, the drive shall return to its starting position
Adjustments:
● PS.00:
All other bits can remain at the factory settings for this example and are explained in the following chapters.
An.56 defines the maximum value (unnormalised) for parameter PS.31. Then the normalised value for
PS.31 is calculated as follows:
100% AUX value = maximum value (An.56) x resolution of the parameter PS.31 (0.1%) =
1000 x 0.1% = 100%
no automatic start, i.e., "continuation of the profile processing" = 0: no and "position setting"
= 0: absolute → PS.27 = 0: no + absolute
ru.61: target
position
ru.54: actual
position
ru.63/ -ru.63
profile speed
I3 (X2A.12) start
positioning
ru.60: act.
position index
time
Change of the maximum profile speed is possible at any time. The speed / position profile for each positioning
step is adjusted permanently, thus the drive (in compliance with the acceleration and jerk settings) is positioned
with the maximum permitted speed.
Example 4: Positioning of a workpiece for various processing steps / sequence control by timer-func-
tionality and the input / output handling of the inverter
Sequential positioning with stop between the positioning steps and definition of the profile speed by PS.25.
Note: this example requires detailed knowledge of the timer functionality and of the input / output handling.
These chapters must therefore be read prior to programming an internal sequence control system. If no internal
control is to be implemented, this example can be skipped.
Settings:
Adjustments:
● PS.00:
7
PS.00: Posi / synchronous mode
Bit Meaning Value Explanation
Posi / synchro-
0...2 5: Posi mode Activation of the positioning mode
nous mode
Positioning / 0: PS.25 /
4 The maximum profile speed is set via PS.25 "Index / speed".
target speed PS.25
All other bits can remain at the factory settings for this example and are explained in the following chapters.
The start of the timer must be triggered by a software input. Input IA was chosen here.
Output OA must also be set if the switching condition act.. index = 1 or act.. index = 2 and, simultaneously, the
condition "target reached" are met
→ do.00: Condition 0 = 54: target window reached (positioning) M0 = SB0 = target reached
→ do.16: Condition selection for flag 0 = 1: SB0
→ do.01: Condition 1 = 72: act. position index = level SB1: Index = 1
→ LE.01: Comparison level 1 = 1,00
→ do.02: Condition 2 = 72: act. position index = level SB2: Index = 2
→ LE.02: Comparison level 2 = 2,00
→ do.19: Condition selection for flag 3 = 6: SB1+SB2 M3 = SB1 or SB2
→ do.37: Flag selection for OA = 9: M0+M3 OA = M0 and M3
→ do.41: AND conn. for outputs = 16: OA
When the timer has counted 1500ms (for index 1) or 1200ms (for index 2), a "start positioning" signal shall
be generated.
As input for the "start positioning" signal, software input IB is used.
→ PS.29 = 576: I3 (X2A.12) + IB
Output OB must therefore be set, when the switching conditions index = 1 (SB1) and timer 1 > 500ms, or
index = 2 (SB2) and timer 1 > 1200ms, are met.
→ do.03: Condition 3 = 37: Timer 1 > level SB3: Timer > 500 ms
→ LE.03: Comparison level 1 = 0,50
→ do.04: Condition 4 = 37: Timer 1 > level SB4: Timer > 1200 ms
→ LE.02: Comparison level 2 = 1,20
→ do.17: Condition selection for flag 1 = 10: SB1+ SB3 M1 = SB1 and SB3
→ do.18: Condition selection for flag 2 = 20: SB2+ SB4 M2 = SB2 and SB4
→ do.24: SB AND/OR conjunction = 6: M1+M2
→ do38: Flag selection for OB = 6: M1+M2 OB = M1 or M2
The resetting of the timer must be triggered by a software input. Input IC was chosen here.
→ LE.19: timer 1 start inp. sel. = 1024: IC
Output OC must therefore be set when the condition "target window reached" is not set.
automatic generated
„start
positioning“ signal (IB)
In the positioning blocks, only target position, profile speed, traversal manner, and sequence of the positionings
are stored. To reach the acceleration /, deceleration time and the S-curves, one must use various sets.
Before the start of a positioning, that set must be activated which contains the desired acceleration and jolt
values.
If positions are to be connected directly with certain profiles, the set-programmability of the parameter PS.28
can be utilised.
I.e., in the 4 sets used, the required acceleration and S-curve times are programmed. In each set, a different
index is parametrised as the starting index.
Example:
The motor must execute 21 revolutions for the round table to complete one full revolution. Let the increments
per revolution of the incremental encoder on the motor be 2500 increments (parameter Ec.01) and in parameter
Ec.07 "multiple evaluation 1", the value 2 is: 4-fold programmed.
This results in:
The actual position and the set point position vary only in a range of 0 to (PS.39 - 1).
Attention: Only positions from 0 to (PS.39 - 1) may be set as target position (PS.24).
Attention: The difference between set position and actual position may never be greater than PS.39/2, i.e., the
drive may not be blocked!
The different round table modes were created for different applications.
inverter Gear
F5A-S
control terminal block X2A
or
F5A-M
M
3~
Rotary table
position 7
Motor position
encoder
Rotary table
encoder
X3B X3A
Encoder Encoder
channel channel 1
2
This mode is particularly suitable for rotary table applications where a second encoder is used for the rotary
table position. Here, the gear backlash cannot cause a position error and the target position can be approached
precisely from both directions of rotation.
Here, the mode 8 "round table with path optimisation" is optimal since the shortest positioning times can be
achieved in this mode.
inverter Gear
F5A-S
or
F5A-M
M
control terminal block X2A
3~
Rotary
Incremental
encoder
This mode is particularly suitable for round table applications where only one encoder is used for the motor po-
sition. The gear between the motor and the round table can cause the position of the round table to be different
for identical motor positions, depending on the direction of rotation from which the position was approached.
To avoid these problem for applications where the gear backlash cannot be ignored, the target must always
be approached from the same direction of rotation. The direction from which a position is approached is deter-
mined by the sign of PS.24 "index / position": Positive values mean the position is approached from the right,
negative values lead to an approach from the left.
7
ru.02: ramp output display (setpoint speed controller)
In this example, all position setpoint in PS.24 are positive. The position values 0 and PS.39 are identical, there-
fore, the value 0 as well as the value PS.39 can be displayed during traversal of the round table range.
Thereby, apparent jumps can occur in the position values if the display changes between the value 0 and
PS.39.
Example:
The gear ratio between motor and rotary table is 50 : 3
encoder increments per revolution = 2500 and multiple evaluation = 2: 4-fold
In PS.39, however, only integer values can be entered. Thereby, one obtains (for predominant motion in one
direction of rotation) an error that increases with each round table revolution.
To correct this error, there is the possibility of flying referencing for round table applications.
For that purpose, an initiator is connected to a digital input which generates a pulse at a fixed round table posi-
tion. Whenever this pulse is recognised, the actual position must be equal to the position of the initiator. If this
is not the case, the actual position is set to the initiator position. The setpoint- and target position are corrected
by the same value as the actual position.
Example:
The drive rotates clockwise.
The initiator provides a signal at position 1000.
The drive starts at position 0.
The actual position (ru.54) at the time the edge of the initiator is 999.
The set point position (ru.56) is 1002, the target position is 5000 increments.
The value range round table (PS.39) is 10,000 increments.
The actual position is set to 1000, i.e., corrected by +1.
The set point position is set accordingly to 1003 and the target position to 5001.
ref.-point (PS.17)
ru.65: Set position
edge
Start positioning
For the positioning for which corrections are made, the value of PS.24 "index/ position", therefore, does not
match ru.61 "target position" anymore, instead of the value 5000, positioning is done to 5001.
The next position is again approached corresponding to PS.24, i.e., the value of target position (ru.61) and
index / position (PS.24) match. Thereby, the error caused by the noninteger gear ratio is compensated.
Since the initiator signal is longer than one increment, the same point of the initiator must always be used for
the adjustment. Therefore, adjustments for clockwise direction of rotation is made as soon as the initiator is
reached (positive edge). For counter clockwise direction of rotation, adjustments are made when the initiator is
left (negative edge).
To have the adjustment executed always at position 5000, adjustments must be executed at the positive edge
for clockwise rotation and at the negative edge for counter clockwise rotation.
Initiatior signal
Initiatior
3000 signal 7000 7000 6000 5000 4000 3000
An important point for flying referencing is the suppression of interference pulses that can trigger a referencing
at the wrong position.
Basic requirement is the EMC conform installation. The programming of a digital filter in the di-parameters is
unsuitable for the flying referencing since the time delay caused by the filter distorts the referencing.
Therefore, there is the parameter PS.40 "reference point window". Only an initiator pulse within the position
window of +/- PS.40 around the reference point PS.17 triggers an adjustment.
Example:
In some applications, a drive shall stop within one revolution at a defined position during vector controlled ope-
ration. For such applications, the mode "position specification relative to the null signal" was created.
In this application, the functions "position / synchronous activation" and "start positioning" are connected to the
same input. If the signal at this input is not active, the drive runs vector controlled.
If the input is activated, the drive correspondingly delays the defined acceleration-, deceleration- and S-curve-
times. Thereby, it positions to the distance to the marker pulse defined in PS.24. How many revolutions it still
travels during the delay dependent on the speed and the adjusted ramps. Only the position within one revolu-
tion of the position encoder at which the drive stops is defined.
Only values from 0 to encoder (inc/r) * 2 enc. trigger may be preset in this mode in PS.24 „index/ position“.
The direction of rotation from which the position is approached is always the direction of rotation of the drive
before activation of the positioning.
Attention: If the modulation is switched off (switch-off the control release, error) while the drive rests at the stop-
ping position, the drive will be in speed-controlled operation again after switching on the modulation.
One application would be, e.g., a drill for which the chuck key must always have a defined position during
standstill.
The remaining distance positioning is similar to the mode defined stop. Only that here, positioning is done to a
set distance from a marker, rather than to a defined position within one revolution.
The drive leaves the speed-controlled operation and runs, starting at a marker, the adjusted residual distance.
To that end, the value 2: relative must be set in PS.27.
After activation of the positioning mode by a digital input, the drive remains in vector controlled operation,
until the marker triggers a "start positioning" command. The target position for the positioning is the actual po-
sition ru.54 at the time of the positive edge + PS.24 "index/position".
PS.24
PS.24
Start positioning
During the round table positioning with only one encoder for motor and round table position, a flying referencing
for correcting a gear factor error exists. For other applications using only one encoder, one needs compensati-
on for slip (undercarriages) or cable stretch (hoists).
An undercarriage shall, e.g., be moved 1m, corresponding to 10,000 increments. During starting, the powered
wheels slip on the steel rail, however, so that after 10,000 increments of encoder revolution, the undercarriage
has only travelled 0.95m.
To compensate for this error, the slip-afflicted system can be resynchronised via reference markers. There re-
ference markers indicate the real position of the drive. An adjustment value is calculated from this information.
The adjustment is still carried out within one active positioning, to reach the target at the predefined position.
To be able to approach the target from both directions, two reference markers that can be located at different
positions must be supported.
Dependent on the direction of rotation, the positive edge of the reference marker is expected at a distance
to target of PS.46 "relative adjustment switch clockwise rotation" (direction of rotation clockwise) and PS.47
"relative adjustment switch counter clockwise rotation" (direction of rotation counter clockwise), respectively.
7
In parameter ru.69 "distance ref.-zero point", the adjustment value is displayed. It is calculated as:
Clockwise rotation: ru.69 = PS.46 - (ru.61: target position - ru.56: set position)
Counter clockwise rotation: ru.69 = (ru.61: target position – ru.56: set position) - PS.47
Example:
Un undercarriageshall drive to the position 5m (= 5,000 increments).
The undercarriage loses 02m (= 200 increments) during acceleration due to slip.
The right reference marker is situated at position 4...4.3m => PS.46 = 1m = 1,000.
The left reference marker is situated at position 5.2...5.5m => PS.47 = 0.5m = 500.
Ref.-marker
Start positioning
No reference point setting of the actual position takes place. This has the following con-
sequence: If the position controller is not active (PS.06 = 0), or if the drive cannot follow
the position setpoint due to torque limits or controller settings, errors in the actual
position can add up.
If the referencing marker is recognized only during the deceleration ramp when reaching the target, the target
position cannot be reached with the adjusted ramps anymore.
During the constant running phase, too, the adjustment of the position via the reference markers can lead to
the target position not being reachable anymore. This is possible every time the reference marker shows that
the drive is already nearer the target than expected.
To be able to directly drive to the target in these cases as well, the ramp times must be changed.
In parameter PS.44 "adjustment ramp limit %", a factor between 25 and 100% can be set. 25% means that the
deceleration times may be reduced to maximally 25% and increased by up to a factor of 4. The value 33% in
PS.44 would correspondingly allow a change in the deceleration times of between 33% and a factor of 3 of the
values set in the oP-parameters.
If the adjustment pulse is received when the drive is already in the deceleration phase, adjustment occurs only
during the last S-curve before reaching the target.
If the change of the ramp times by the adjusted correction factor PS.44 is insufficient, the drive enters the status
"position inaccessible".
- with correction
more sharply.
Speed (with correction /
- with correction
If no successful adjustment can be executed during the running positioning, there are two different response
options which can be selected via PS.00:
The two reference markers belong to the block (index) that defines a positioning step. (For the description of
the positioning indexes see chapter 7.12.4.8 "Sequential positioning"). In connection with the reference marker,
it is useful to define index blocks even for single positionings.
Example:
PS.46 = 3500
PS.47 = 1000 PS.46 = 2500
PS.46 = 1750
PS.46 = 1000
From a starting position, an undercarriage shall drive to 4 different positions . The positions of target 1...4 are
always approached from the left, the start always from the right. For the targets, therefore, reference markers
are always defined for clockwise rotation, and for the start only a reference marker for counter clockwise rota-
tion. The values for the other reference markers are set to zero (= deactivated)
The profile speed definition is given via parameter PS.31 "max. speed %". That means value „16: PS.31 /
PS.25“ must be entered in PS.00 / bit 5 "positioning / target speed".
The drive shall stop at the target, value 0 must be entered for all blocks in parameter PS.25 "index / speed".
This is a single positioning, PS.26 „index/ next“ must always be set to „-1: PS.28“, since there is no next po-
sition. The positions are preset as absolute values and "continuation of the profile processing" is deactivated.
Therefore parameter PS.27 must be set to value 0.
If the drive shall now travel from the start to target 2, value 2 must be entered in parameter PS.28 "start index
new profil", and the "start positioning" command must be given then.
In the process, the undercarriage also crosses reference marker 1, which can trigger an adjustment only during
the positioning to the starting point. For the drive to target 2, marker 1 must be ignored. For that purpose, the 7
parameter PS.40 "reference point window" is used. Only an initiator pulse within the position window of +/-
PS.40 around the programmed value for the reference marker triggers an adjustment.
Example target 2:
PS.46 = 1750 increments / PS.24 = 6250 → the reference marker is expected at 6250 – 1750 = 4500.
If the reference point window is set to, e.g., 300 increments, the initiator signal is accepted only if the drive
rotates in clockwise direction of rotation and as long as the actual position ru.54 is in the range of 4200 to 4800
increments.
The size of the reference point window depends on the maximum expected slip. If one assumes, that maximally
150 increments „are lost“ due to the slipping of the wheels (i.e. are not converted to propulsion), a value > 150
increments must be adjusted in PS.40.
If the drive is approaching target 2 from the left, reference marker 1 generates a positive edge at an actual po-
sition of 700...850 (depending on the level of slippage) and therefore outside of the permitted window.
This marker is therefore ignored. Reference marker 2 generates its pulse within the reference window and is
evaluated for the adjustment.
PS.47 = 1000 increments / PS.24 = 0 → the reference marker is expected at 0 + 1000 = 1000,
The initiator signal is evaluated only if the direction of rotation is counter clockwise and the actual position lies
within a range of700...1300 increments. With this, the reference marker 2 is masked for the return path.
The following figure illustrates the connection between direction of rotation, target, the values of PS.46 / 47 and
the reference point window PS.40.
Reference marker
(300 increments)
The parameter PS.45 "adjustment index selection" is identical to parameter PS.23 "index/selection". It has
been inserted here a second time only to simplify the operation.
Starting a new positioning is only possible if bit 10 has temporarily been set to 0.
To perform a positioning, the inputs wired as limit switches (inputs with functions "forward" and "reverse") must
be active. If the hardware limit switches are not to be used, the protection functions in Pn.07 "limit switch error
response" must be deactivated (value 6: function switched off).
If the target position can still be changed during a running positioning is determined by PS.00 bit 3.
If the drive leaves the speed-controlled operation (first "start positioning" after activation of the posi mode), or if
during an active positioning a new "start positioning" command is given, the new target is possibly inaccessible
with the adjusted ramp and jerk times.
Parameter ru.84 "accessible rel. posi." displays the distance that the target must have to the actual position
ru.54 at the time of the "start positioning" pulse to be accessible with the programmed ramp and S-curve times.
Example:
Start Starting of the decelerati-
on of the target appoach At the "start positioning command", ru.84
positioning
indicates that the target must be at least
ru.84 200000 increments away from the current
ru.84: accessible rel. position
programmed ramps.
ru.54: actual position
If the target is inaccessible with the ramps from the oP-parameters, the ramp times can be changes online in
the event of a new "start positioning" pulse. For that purpose, there is the parameter PS.32 "limit acc/dec red-
ucing %".
Note: For the case of a start from the vector controlled operation, the online adaption is not active. A factor bet-
ween 25 and 100% can be adjusted in parameter PS.32. 25% means that the ramp and S-curve times may be
reduced to maximally 25% and increased by up to a factor of 4. The value 33% in PS.32 would correspondingly
allow a change of between 33% upto factor 3 of the values adjusted in the OP-parameters.
In this example
the same values are set for the
ru.54: actual position
The new "start positioning" command must be made within the positioning, in the acceleration or constant run-
ning phase. In the deceleration phase to the final target approach, the new starting pulse leads to the status
"position inaccessible".
If the new starting pulse is received in the correct phase, but the target is inaccessible despite ramp adjustment
, the drive goes into the state "position inaccessible" with the default ramp/jerk times.
The parameter PS.32 is similar to the parameter PS.44 "limit adjustment ramp %" in its mode of action. PS.44,
however, works only for ramp changes necessitated by flying referencing with adjustment. PS.32, on the other
hand, is responsible for adjustments based on new target settings.
Furthermore, the ramp can also be changed during the acceleration phase by PS.32 if the new target setting
makes it necessary.
With the parameter PS.53 "distance for no abort", unwanted "start positioning" commands can be masked.
Since the last "start positioning" command, the drive must have travelled a greater distance than PS.53 in order
that a new starting command is accepted during a running positioning.
7
Example:
At the "start positioning command", ru.84
indicates that the target must be at least
actual position
PS.53: max. termi- The target is far away and the drive
nation distance positions to the target.
The deceleration phase starts, if the amount
of ru.84 ahead of the target is reached.
Speed
An "inaccessible position" is a target position that cannot be reached with the programmed acceleration / de-
celeration and jerk times.
● Transition from vector controlled to position controlled operation due to the first "start positioning" after
activation of the positioning mode
● Change of the target position during a running positioning by a new "start positioning" signal
● during sequential positioning, when no stop is scheduled at the target
● due to a change in the actual and set point positions by flying referencing with adjustment
● if the drive is running at the time of the "start positioning" command and a rotation change is required for
the positioning
I.e., every time the drive is running at the time of the "start positioning" command, inaccessible positions can
occur.
Also, if a new "start positioning" pulse is given during the deceleration phase to the original target, and the S-
curves are switched off, the message "inaccessible position" is generated because an adjustment during the
final approach to target is only executed during the lower S-curve.
Position not reachable, because new start Position not reachable, because new target is too close
positioning during deceleration ramp
ru.61: target position
ru.56: Set position
ru.61: target position
ru.56: Set position
ru.02: ramp output display
Start positioning
Ignore new
target
Target not
reachable
Start posi-
tioning
A deactivation of the input occupied with the function "positioning /synchronous activation" concludes the po-
sitioning mode. A running positioning that was aborted by the deactivation cannot be resumed after switching
the input back on.
If the input remains activated, an active positioning can be interrupted by quick stopping, deactivation of the
control release, triggering of the „power off“ function or occurence of an error (e.g.: E.OC, E.OP, etc.).
After the drive is "ready for positioning" again, an interrupted positioning must be restarted with a "start posi-
tioning" command.
With parameter PS.00, one can select how the drive should behave after an error occurs (i.e., after interruption
of an active positioning):
Note: a sequential positioning is aborted only if the error (the interruption) occurred during an active positioning.
If the drive is stopped on target (even if it is only an intermediate target of the sequential positioning) e.g. swit-
ching off the control release does not lead to an interruption of the sequential positioning.
An active positioning can also be interrupted by setting bit 11 in control word Sy.43 or Sy.50
(field 2048: „activate interruption“).
In contrast to a termination due to abnormal stopping, the ramps from the OP-parameters are used and started
S-curves are not interrupted in a termination initiated by the control word. The modulation remains on.
If bit 11 "termination" has again been deactivated in the control word, there are two possibililties for the drive
to continue:
By means of the analog parameter setting, the parameter PS.24 "index / position" can also be adjusted. (For
more detailed information on the analog parameter setting, see chapter 7.15.9 and for the adaption of the ana-
log channels, see chapter 7.15).
To set a position setpoint in PS.24 "index / position" by analog input, the following adjustments have to be made:
● Select the index in PS.23, where the set position should be written
● Select PS.24 as the target for the analog parameter setting An.54: Analog para. setting / target =
1318h (bus address PS.24)
● Select in An.53 which analog channel shall preset the set position
Example: an analog value of -100%..+100% shall permit setting position values of 100000 to 300000
increments. Then An.55 and An.56 must be parametrised as follows: An.55 = 200000 An.56 =
300000
Via the analog outputs ANOUT, position values can also be issued. (For further information on adaption of the
analog outputs, see chapter 7.15).
For the analog setting of actual position (ru.54) or set point position (ru.56), the following adjustments must be
made:
● for an analog output (ANOUT1 or 2), choose actual position (An.31 / An.36 = 27) or setpoint position
(An.31/ An.36 = 28) as the output value
● configure the conversion of the position value to an analog value:
Position while 0% analog value is issued: PS.41 "reference position 0%"
Position while 100% analog value is issued: PS.42 "Reference position 100%"
Example: for position values in the range of 100.000 to 300.000 increments, an analog value of 7
-100% to 100% is to be issued. Then PS.41 and PS.42 must be parametrised as follows: PS.41 =
200.000, PS.42 = 300.000
● Parametrise analog output (amplification, offset)
The target window range is specified in parameter PS.30 "target window“. The target window has been reached
if the actual position is in the range of +/- PS.30 / 2 around the target position.
7.12.4.20 Backlash
This parameter is used to compensate a possibly available backlash. Each positioning is approached with the
gearless adjusted preferred direction. If the positioning is started against this preferred direction, the target po-
sition is initially overdriven by the gearless. After this the positioning occurs in the target position with preferred
direction.
PS.55 Backlash
Default value 0: off
Upper limit 1073741824 inc.
Lower limit -1073741824 inc.
With parameter PS.37 "pos. scan index inp. sel." or via the digital input function (di.24...di.35) "scan position",
an input can be defined for scanning the actual position (ru.54) with its positive edge.
The position scan occurs only in status "positioning active". The scanned actual position value (ru.54) is displa-
yed in parameter ru.71 "teach/ scan position display". Each additional edge overwrites the old scan position.
A digital output can be set dependent on the scanned position. Switching condition „75: amount actual position-
scan position > level“ must be selected for this.
Note:
Switching condition As long as the switching condition remains set, the
0
teach-function (see chapter 7.12.4.21) cannot be
Scan position used.
positioning active
Start positioning
With parameter PS.36 "teach index input selection", or via the digital input function (di.24...di.35) "store position
(teach)", an input can be defined for scanning the actual position (ru.54) with the positive edge.
The scanned value is displayed in parameter ru.71 "teach / scan position display" and stored in PS.24 as the
target position. Parameter PS.35 "teaching mode" determines which positioning block (which index) the target
position is stored in.
●
ru.54: actual position
This chapter provides a compressed summary of the possibililties for control and visualisation of the positioning
mode. Some of the parameters and functions are described in more detail in the corresponding chapters.
Display parameters
Parameter Function
Display of the actual position (calculated from the position informa-
ru.54 actual position
tion of the encoder interface, selected in PS.01).
The position that the drive is supposed to have reached currently,
ru.56 Set position according to the calculated profile (in contrast to ru.61 "target posi-
tion", to be reached at the end of the positioning).
ru.58 Angle difference Difference between set and actual position.
ru.60 Act. position index Number of the positioning block currently being processed.
The target which the drive should reach at the end of the positio-
ru.61 target position
ning.
Maximum value of the speed profile the inverter calculates for opti-
ru.63 profile speed
mal target approach.
The value by which the actual position is adjusted during the "flying
ru.69 Distance ref.-zeropoint
referencing with adjustment".
Position value scanned by a digital input (used for digital output and
ru.71 Teach / scan position
teaching).
If the drive rotates at the time of the "start positioning" command
(vector controlled operation or new pulse during a running positio-
ning), this parameter shows what distance to the actual position
ru.84 accessible rel. position
the target must be to be reachable with the programmed ramp and
S-curve times. The ramp adjustment (PS.32) is not considered (see
chapter 7.12.4.13 "Start positioning").
72
Actual position index = For sequential positioning: the actual position index is equal to the swit- 7
level ching level (scaling factor: values of 0.51...1.5 count as index 1 etc.).
Amount act. position – Actual position (ru.54) - teach/scan position (ru.71) > comparison level
75
scan position > level (LE.00...LE.07)
The output is set if the switching condition "target window reached" is
Actual Position = position
77 met (see No. 54) and the "actual position index" (ru.60) is equal to the
index PS.28
"start index new profile" (PS.28).
Only an initiator pulse within the position window of +/- PS.40 around the
reference point PS.17 may trigger an adjustment. If a pulse is received
Rotary table reference
78 outside of this window, it is interpreted as an interference pulse and is
invalid
ignored. With this output switching condition, the user can recognize the
presence of invalid pulses.
The output is set if a "start positioning" command is ignored because the
Ignore position not reach- new target position is "inaccessible" (see chapter 7.12.4.14 "inaccessible
79
able positions"). The output is reset by a new "start positioning" command or
by deactivation of the positioning mode.
The switching level for the switching conditions are set in LE.00...LE.07. Since the switching level can be used
for very different quantities (current, voltage, speed, position, etc.), they have the following scaling factors for
comparisons with position values:
LE.00...LE.07 = 1,00 ⇒ comparison level is 100 increments
For the contouring control mode, the bus-synchronous operation must be activated. Bus-synchronous operati-
on means that a control sends telegrams in a constant time pattern and that all connected inverters synchronise
to this pattern. This allows angular-synchronous and multi-axis operation, respectively.
To implement the bus-synchronous operation, one requires either a fast fieldbus system with the associated
KEB-operator or a fast control supporting the HSP5-protocol (e.g., a drive control COMBICONTROL C5).
The following bus systems are supported: CAN, SERCOS, EtherCAT, Powerlink, HSP5.
Fieldbus
KEB inver-
(e.g. CAN, SERCOS,
EtherCat, Powerlink) COMBICONTROL ter
KEB inverter F5A-S
C5 or
optional
fieldbus Drive F5A-M
interface interface
F5A-S
or RS232
Parametrisation
F5A-M digital
HSP5 I/O Ethernet
Bus Parametrisation
Visualisation
X3B X3A
Encoder Encoder
KEB channel 2 channel 1
Operator
X3B X3A
Customer Encoder Encoder
control channel 2 channel 1
How the bus-synchronous operation is realised and initialised depends on the fieldbus system used and must
be looked up in the instructions for the corresponding operator (CAN-operator, SERCOS operator, etc.).
The inverter enters the mode "bus-synchronous operation" if a value unequal to zero is set in parameter Sy.08
"bus synchronisation time" .
Note: As soon as the control sends cyclic telegrams in the pattern set in Sy.08, the bit 9 "HSP5 bus synchro-
nous" is set in the status word (Sy.51). The synchronous communication can be monitored with this bit. The
control release may only be given if the assembly of the bus-synchronous operation is completed.
All parameters can be specified in bus-synchronous operation. Typically, this operating mode is used for the
contouring control mode. The activation of this mode is done in parameter PS.00 or via the control word.
To activate the contouring control mode, the digital input assigned with the function "posi / synchronous" must
be set.
In contouring control mode, the drive is positioned via the setting of a position setpoint value in the bus-syn-
chronous grid. The control, therefore, does not give the final target position but the set point position for each
individual cycle.
The inverter calculates the speed required to reach the position setpoint in one bus cycle. The setting of the
position setpoint can be done via PS.24 "index position" or via PS.34 "ctm position".
PS.33 "contouring mode setpoint source" determines which parameter provides the setpoint position.
For the bus-synchronous operation, the read and write data that are to be transmitted with each bus cycle must
be defined. In the parameters Sy.24, Sy.26 and Sy.28, the bus addresses of the parameters that are to be pro-
vided bus-synchronously by the customer control must be set.Only 3 parameters can be selected (one 32 bit
parameter and two 16 bit parameters).
The set position (32 bit parameter) must always be preset bus-synchronously for the contouring control mode.
To that end, the bus address of PS.34 (=1322h or 4898) or of PS.24 (= 1318h or 4888) must be entered in Sy.24
"write data 1 definition" (dependent on setting of PS.33).
The second parameter is typically occupied with the control word Sy.50 to allow driving the inverter as com-
pletely as possible via the bus.
Parameter Sy.43 "control word (long)" cannot be utilised because only a 16 bit parameter may be used. Para-
meter oP.03 "reference setting" can be used as third parameter if change-over from contouring control mode to
speed-controlled operation shall be possible.
With activation of the countouring module, the current position value (ru.54) must be read once and then be
transmitted three times as the setpoint. This is necessary to initialise the drive-internal speed pilot control.
In the same manner the same setpoint must also be transmitted three times to stop the drive, e.g., before en-
ding the bus-synchronous operation. This sets the speed pilot control to zero.
When setting the position setpoints, one must ensure that the drive will be able to follow them with the maxi-
mum permissible speed. Only oP.14 "absolute maximum reference for" and oP.15 "absolute maximum refe-
rence rev" function as speed limits. 7
7.12.5.5 Contouring control mode / watchdog
A breakdown of the bus system is particularly critical during bus-synchronous operation. Therefore, bus moni-
toring should always be activated. To that end, a value unequal to 6 must be programmed in parameter Pn.05
"watchdog response" .
In Sy.09 "HSP5 watchdog time", the monitoring time for the HSP5 connection between inverter and operator
can be adjusted. Value "0:off" means that the watchdog is not active. The monitoring time of the fieldbus con-
nection is parametrized in Pn.06 "watchdog time". The actual monitoring of the fieldbus is carried out by the
operator. With the parameter Pn.05 "watchdog response", one selects how the drive responds (e.g., immediate
error message or quick stop) to the occurence of a bus breakdown (HSP5 or fieldbus).
From software 4.3 the contouring control was reveised as follows with PS.61 and Sy.76:
The interpolated pre-profile is temporal optimal calculated to the adjustment of the cycle time and the smoothing
time of the speed measurement.
For this PS.61 must be activated. The half value of Ec.03 should be adjusted in cS.30.
Synchronization to the time pattern of the master of the contouring control is internally done no longer hard by
compensation of the error up to the next ms interrupt, but via adjustable low pass filter with Sy.76.
The initialization is improved. The contouring mode can be activated at any time. Not several cycles must be
await before release.
If PS.61 is set to 1: on, the precontrol profile in the contouring mode is corrected with a PT1 element. This func-
tion is only active in the contouring mode. The adjustment occurs automatically by the value written in Sy.08.
A new Pt1 element is introduced with Sy.76 whereby the internal time pattern is synchronized to the external
cycle time. Advantageous is this if the superior control can not keep the cycle. The PT1 element is only active
if the synchronous operation is achieved.
Parameter list:
The position controller is constructed as P controller. The increments per revolution of the encoder and the
resolution of the speed is considered in the controller. The speed controller setting can be assumed when ex-
changing the encoder (e.g., from 1024 => 2500 increments) or changing the speed range (e.g., from the 4000
to the 8000 speed mode by changing the parameter ud.02 "control type").
The Kp of the position controller can be changed speed-dependently. Thereby, one can, e.g., choose a
very hard setting for the load transfer and the approach to the final position. For the remaining positioning,
the Kp is lowered then to achieve a smooth running of the drive and to dampen the effects of mechanical distur-
bances (e.g. welds in the linear rails or similar).
The base value of the controller is set in PS.06 "KP for positioning / synchronous". Parameter PS.08 "limit
speed for PS.07" determines the limit for speed reduction and parameter PS.07 "KP speed limit reduction"
determines the percentage value the Kp shall still have in relation to its base value at the speed PS.08.
positioning module:
ru.63 Profile speed
Synchronous module:
ru.10 Encoder 2 speed
ru.56 KP
Set position +
ru.58
Angle difference
ru.54 -
actual position
PS.09
pos
pos/syn-position limit
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp setting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
The protective functions protect the inverter against switch off caused by overcurrent, overvoltage as well as
thermal overheating. Furthermore, you can restart the drive after an error automatically (Keep-On-Running).
For diagnostic purposes, the inverter displays various malfunction- and error messages. Errors are all those
events that trigger an immediate switch-off of the modulation, malfunctions allow a defined response (shutdown
of the drive by abnormal stopping).
For some events (ext. error, bus monitor response, the drive hitting a limit switch, etc.), one can decide in the
programming whether this is an error or a malfunction.
For some errors, e.g., the overload error, a pre-warning can be generated. This pre-warning is treated like a
malfunction, i.e., the appropriate response to the pre-warning is programmable.
Example 1 (error):
The inverter detects overcurrent and raises the error.Display in parameter ru.00: "Error! Overcurrent" (E. OC).
Since this error cannot be predicted, there is no possibililty of a pre-warning. The modulation is switched off
immediately and the drive spins down.
The reaction of the bus monitor ("watchdog") shall trigger an error. Programming Pn.05: "Watchdog response"
= 0 (error / no auto restart).Display in parameter ru.00: "Error! Watchdog" (E. buS). If a digital output is program-
med on a fault signalling relay, the relay switches.
Hitting a hardware limit switch shall be treated as a malfunction. Desired response: Abnormal stopping, modu-
lation switch-off after reaching standstill, no automatic restart.
Programming Pn.07: "Limit switch error response" = 1 (Stop / modulation off / no auto restart)
Display in parameter ru.00: "Warning! disabled direction of rotation clockwise" (A.PrF) or: "Warning! disabled
direction of rotation counter clockwise" (A.Prr)
If a digital output is programmed on a fault signalling relay, the relay does not switch by default.
(If the digital output shall also respond to malfunctions, switching condition 6 "abnormal stopping / error" must
be used. Alternatively, Pn.65 can be adjusted by way that a malfunction in relation to the status displays and
digital outputs is treated as error. See chapter 7.13.12)
Example 4 (pre-warning):
When the heat sink temperature exceeds a limit (dependent on the inverter type), the modulation is switched
off, the inverter raises an error. With Pn.11 "heat sink overtemperature warning level" a temperature can be set
at which a pre-warning is generated.
Desired response: when exceeding the temperature of Pn.11, the inverter executes an abnormal stop and
switches off the modulation. When the heat sink temperature decreases again, an automatic restart shall occur.
Programming Pn.10 "heat sink overtemperature response" = 4 (stop/modulation off/ auto restart).
Display in parameter ru.00: "Warning! Heat sink temperature" (A. OH)
If the temperature decreases due to the abnormal stopping, the inverter executes an automatic restart. If,
however, the heat sink temperature continues to rise and exceeds the error limit, the inverter raises an "Error!
Heat sink temperature" (E. OH).
7.13.1.1 underpotential
"Error! Undervoltage" (E.UP) is triggered if the DC link voltage drops due to brownouts or a generally too weak
power grid. For this error, the automatic restart can be activated.
The following function is activated with this parameter (value <> 0):
With inactive control release E.UP is no ERROR (status word), ru.00 displays nevertheless E.UP
With inactive control release no PU is no ERROR (status word), ru.00 displays nevertheless NO_PU
Other malfunctions (e.g. E.EF, E.dOH) are triggered furthermore
If the control release is activated, E.UP or no PU is triggered after the adjusted time is up, if the condition is still
met (DC link too low or LT_OK signal not active).
To ensure a safe operation Pn.77 should be adjusted smaller max. by the hysteresis than the internal upper
limit .
7.13.1.2 Overvoltage
"Error! Overvoltage" is triggered if the DC link voltage increases beyond the overvoltage level due to energy
recovery in regenerative operation.
7.13.1.3 overcurrent
The "Error! Overcurrent" (E.OC) is trigger when the "OC-tripping current" (see technical data in the instruction
manual power circuit F5) is exceeded.
If this error occurs permanently, either the connected motor (short circuit or ground fault) or the inverter itself
is defective.
The "maximum short time current limit" is below the overcurrent limit. If it is exceeded, the hardware current
limit can be triggered with uF.15. The response of this function is not considered an error or malfunction, and
the corresponding switching conditions are not set. If the function is active, the status "80:hardware current
limitation active" (HCL) is displayed.
For current regulated drives, this function should be deactivated since it may have negative effects for the motor
model calculation and the behaviour of the drive.
7.13.1.4 overload
The inverter-overload protection is a function that triggers an error for which, however, a pre-warning can be
generated.
There are two overload protection functions: one for the range of standstill and low frequencies (overload at
standstill/ OL2) and one for the remaining frequency range (overload/ OL).
With Pn.09 "Overload warning level" a value between 0...100 % can be adjusted, when the "Warning! Overload"
or the "Warning! Overload during standstill", is set. The response to the overload warning is set with Pn.08
"warning OL stop. mode".
The implementation of the function "19: Overload during standstill" is described in chapter 2.1.9 "Overload
protection in the lower speed range". The motor current is guided via a PT1 link with a time constant of 280 ms.
If this delayed current exceeds the OL2 limit, "Error! Overload during standstill" (E.OL2) is triggered. If the dela-
yed current decreases to 0 again, the inverter enters the status "20: Overload during standstill fixed" (E.nOL2).
The error can now be reset.
Overload (OL)
The implementation of the general overload protection is described in chapter 2.1.8 "Overload characteristics".
If the 100% load factor of the inverter is exceeded by 5 %, the internal overload counter starts to count forward.
If the load factor falls below 100 %, the counter counts backward. The current counter content can be read in
parameter ru.39. Upon reaching 100 % the inverter switches off with error message "E.OL" and the counter
counts backward. If it reaches 0 %, the status changes to E.nOL and the error can be reset.
The heat sink temperature acquisition protects the power module from thermal overload. The temperature at
which the inverter switches off with error message "8: ERROR! Overtemperature" (E.OH) depends on the po-
wer circuit (generally 90°C).
After cooling period the status changes from "Error! Overtemperature" to "36: Heat sink temperature normal
again" (E.nOH) and is therefore resettable.
A level between 0° C and 90 °C, when the pre-warning is triggered can be adjusted with Pn.11 "OH warning
level". The response to the warning message is set with Pn.10 "warning OH stop. mode".
Internal overtemperature
The interior temperature monitoring protects the inverter against malfunctions caused by too high temperature
in the interior of the inverter. Upon exceeding a unit-specific temperature the interior fan is activated. If the tem-
perature is still too high after about 10 min., the disconnecting time set with Pn.17 "internal overtemperature
disconnecting time" (0...120s) starts.
With the start of the disconnecting time, switching condition "11: Warning internal overheating" is met and the
response to the warning message set in Pn.16 "warning OHI stop. mode" is executed.
After expiration of the disconnecting time, "6: ERROR! Overtemperature interior" (E.OHI) is triggered.
When the interior temperature has dropped again, the inverter state changes again to "7: interior temperature
back to normal" (E.nOHI). The error can now be reset.
With Pn.04 "Input selection external error", one or more digital inputs can be programmed which can trigger the
error "31: ERROR! External input" (E.EF).
With Pn.03 "Response to external error" , the response of the inverter to the digital input is defined. With Pn.65/ 7
bit 1 "2:Pn.04 = E.UP", the function of Pn.04 can be changed and the triggering of an error via a digital input
can be deactivated.
The inverter contains two watchdogs that monitor the communication between an external bus, operator, and
inverter control.
With parameter Pn.05 "Response to E.bus", the response to a watchdog error is defined. Dependent on the
chosen adjustment, either "Error! Watchdog" (E.buS) or "Warning! Watchdog error" (A.buS), is issued or a war-
ning message via a digital output is generated.
This watchdog monitors the communication at the operator interface. With activated watchdog, the response
set in Pn.05 is triggered after expiration of an adjustable time (0.01...40 s) without incoming telegrams.
The function is deactivated by setting the value "0: off".
The HSP5 watchdog function monitors the communication of the HSP5 interface (control card - operator; or
control card - PC). After expiration of an adjustable time (0.01...10 s) without incoming telegrams, the response
adjusted in Pn.05 is triggered. Value „0: off“ deactivates the function.
The inputs assigned with the functions „32: forward“ (limit switch right) and „64: reverse“ (limit switch left) serve
as hardware limit switches. Therefore, the rotation setting via terminals (oP.01 "source of rotation direction" =
2...6) may not be used if the limit switch function is to be used.
To protect against cable breakage, an unconnected input means that the drive has run onto the limit switch.
Depending on the setting of parameter Pn.07 "Limit switch error
response", the response to the hardware limit switches can be a malfunction.
Hitting a limit switch with clockwise direction of rotation is indicated by status "46: ERROR! disabled direction of
rotation clockwise" (A.PrF) and:94 ABN. disabled direction of rotation clockwise" (A.PrF), respectively. The cor-
responding messages for counter clockwise direction of rotation are "47: ERROR! disabled direction of rotation
counter clockwise" (A.PrF) and:95 ABN. disabled direction of rotation counter clockwise" (A.Prr), respectively.
Attention: Only the limit switch for the current direction of rotation is ever evaluated, i.e., for clockwise rotation,
only the right limit switch is considered and the left limit switch is ignored. The analog applies to counter clock-
wise rotation.
Furthermore, one must ensure that the drive stops at the limit switch. If the limit switch is overrun, the drive can
subsequently continue to run in the disabled direction.
The software limit switches complement the function of the hardware limit switch.
They are active only after an approach to reference point or the setting of reference points, respectively (see
chapter 7.12.2 approach to reference point).
In contrast to hardware limit switches, the software limit switches can lose their protective function by, e.g., a
faulty approach to reference point. Their advantage is that they cannot be overrun.
The permissible range of the actual position ru.54 is between PS.15 "limit switch left" and PS.16 "limit switch
right".
The software limit switches are active in the vector controlled operation, in synchronous mode, in positioning
mode, or in contouring mode.
The response to the software limit switch is set in parameter Pn.66 "soft. limit stopping mode". In the factory
setting, the software limit switches are deactivated.
Reaching the limit switch with clockwise direction of rotation is indicated by status "44: ERROR! Software limit
switch clockwise rotation" (E.SLF) and "104: ABN. Software limit switch clockwise rotation" (A.SLF), respec-
tively. The corresponding messages for counter clockwise direction of rotation are "45: ERROR! Software limit
switch clockwise rotation" (E.SLF) and "105: Software limit switch counter clockwise rotation" (A.SLr), respec-
tively.
Parameter In.17 displays the temperature evaluation installed in the inverter. A reversible KTY / PTC detection
is installed if in In.17 = 5x hex is displayed. This can be adjusted with Pn.72 to the corresponding sensor (0 =
KTY; 1 = PTC). If no reversible detection is installed, Pn.72 has no function.
If PTC or KTY message an overtemperature, the adjusted delay time in Pn.13 "E.dOH delay time" starts. The
switching condition "9: Pre-warning motor overheating is set and the adjusted response to the pre-warning in
Pn.12 "warning dOH stop. mode" is executed. If a value of 1...5 is selected in Pn.12, the inverter displays the
malfunction „Warning! A.dOH
After expiration of the disconnecting time Pn.13, the error "Error! Motor overheating" (E.dOH) is triggered.
If the overtemperature condition is past, the message "All clear! Motor overheating" (A.ndOH) or "Motor tem-
perature back to normal" (E.ndOH) is issued. Only then the error can be reset or the automatic restart can be
carried out.
PTC
A temperature sensor integrated into the motor winding is connected to the terminals T1/T2 of the inverter. If a
resistance of 1650...4000 Ohm is exceeded, motor overtemperature is detected. If the resistance drops below
750...1650 Ohm, the state motor overtemperature is reset.
KTY
In Pn.62 "dOH warning level", a temperature in the range of 0...200 °C is defined which, when exceeded,
causes a motor overtemperature message.
The current temperature is displayed in ru.46 "motor temperature". With a standard power circuit Pn.62 has no
function. In the motor temperature display ru.46 only T1-T2 closed or T1-T2 open is displayed.
Additionally to the monitoring of the motor with a temperature sensor, a motor protection can be realised also
by monitoring the motor current.
The monitoring function is implemented differently for asynchronous and synchronous motors.
The functional description (times, current level, etc.) are found in chapter 7.13.9 "electronic motor protection".
The response to the triggering of the electronic motor protection relay can be defined with Pn.14 "warning
OH2 stop. mode". Dependent on the programming, the inverter raises "30: ERROR! Motor protection function"
(E.dOH2) or "97: ABN. Motor protection function" (A.OH2). 7
After the cooling period, the error or the malfunction, respectively, can be reset.
The functional description (times, current level, etc.) are found in chapter 7.13.9 "electronic motor protection".
When the protection function triggers, the error "30: ERROR! Motor protection function" (E.dOH) is triggered.
The error is resettable after approximately 100 ms.
A level of 0...100 % (100% = triggering time of the error) can be adjusted with Pn.15 "OH2 warning level" which
generates a pre-warning.
The response to the pre-warning is set with Pn.14 "Motor protection function response". With this, an abnormal
stopping can be executed before the drive raises an error. During the abnormal stopping, the inverter has status
"97: ABN. Motor protection function" (A.OH2). The switching condition "10: Motor protection relay function" is
met.
Sets can be locked with Fr.03 "parameter set lock". If a disabled set is selected, the inverter remains in the old
set, i.e., no set change occurs.
The response to the selection of a disabled set is set via Pn.18 "Set selection error response". In the factory
setting, the error "39: ERROR! Parameter set selection" (E.Set) is triggered. For Pn.18 = 1...5 a malfunction
"102: ABN. Set selection error" (A.Set) is generated. For Pn.18 = „6: Function disabled", the drive continues
running in the old set without message.
Encoder error
The progress message "32: ERROR! encoder 1 (E.EnC1)“ or „34: ERROR! Encoder 2 (E.EnC2)" is trigger if:
• a defective track is identified for an incremental encoder interface with monitoring of the incremental
tracks
• for resolver interfaces, signals outside of the specifications are identified
• for encoder types that permit the storage of data (e.g., motor data, system position, etc.) in the encoder,
an invalid identifier is read, and therefore the stored data cannot be interpreted.
For "intelligent" encoder interfaces (e.g., absolute encoder, Sin-/Cos encoder), "35: ERROR! Encoder change"
(E.EnCC) is trigger if:
• the encoder type or the interface type of the current software of the control board are not supported
• the signals of the absolute track or the signals of the incremental track are defective
• the identified position deviation between incremental position and absolute position is too large
• the encoder transmits an error message
• the encoder (for encoder types with data storage in the encoder) was swapped
• Adjusted increments per revolution of the inverter does not agree with the encoder increments per
revolution
The error E.EncC can be reset via parameter Ec.00 and from version 4.2 also via hardware or bus reset.
Exception! An error due to incorrect encoder increments per revolution is immediately (without reset!) reset if
the correct encoder increments per revolution is set.
Not all monitoring functions are available for all interface types. A more detailed description of the encoder error
can be found in chapter 7.11 "Speed measurement".
The status "58: ERROR! Overspeed“ (E.OS) is triggered, if ru.07 "actual value display" exceeds either the va-
lue of oP.40 / oP.41 "max. output val." or the value of ru.79 „abs. speed EMC“ (only for synchronous motors).
With oP.40 / oP.41, the user defines limits that may not be exceeded by the application under any circum-
stances.
ru.79 shows the maximum speed for a synchronous motor which, if exceeded, leads to an EMC of the motor
high enough to damage the DC-intermediate circuit of the inverter.
Reason for the occurence of excessive speed can be too small a distance between the maximum setpoint and
the speed limit, so that overshoots can trigger the error. Other causes can be (e.g., caused by EMC) malfunc-
tions in the speed measurement or a noisy, insufficiently smoothed speed estimate in the encoderless control
(SCL or ASCL).
Pn.75 "response to error E.SCL" determines how the output should respond if the speed controller reaches the
limit, i.e., if the set torque reaches the maximum possible value. In the factory setting, this operating condition
can be applied to a digital output (switching condition "53: Speed control at the limit"). With Pn.75, however, it is
also possible to execute an abnormal stopping on reaching the torque limit (Status "107: ABN. Speed controller
limit"/ A.SCL) or to trigger an error (Status "25: ERROR! speed controller limit“ / E.SCL)
The maximum permissible acceleration is defined with Pn.79 "Acceleration limit 1/s^2".
Pn.80 "Acceleration scan time" determines the time period used for acceleration averaging.
The change of the actual speed (ru.07) in this time period, divided by the scan time (Pn.80) is the actual acce-
leration. The speed difference must be converted from 1/min to 1/s for the calculation of the acceleration.
7
Speed change during scan time
Acceleration = —————————————————————
60 x acceleration scan time (in seconds)
If the acceleration exceeds the limit (Pn.79), the response defined by Pn.81 "warning acc. stop mode" is trig-
gered.
The drive, dependent on the programming, enters the status "24: ERROR! Maximum acceleration" (E.Acc) or
"106: ABN. Maximum acceleration" (A.Acc)
Monitors for the internal hardware (e.g. fans) are integrated on some inverter types. If one of these monitoring
circuits reports an error, "12: general power circuit error" (E. PU) is triggered.
"Error! Phase loss" (E. UPh) is identified indirectly via the ripples in the DC link voltage.
If one power phase is missing, the waviness in the DC link is considerably increased under load. In no-load
operation or at small load, the error of the power phase is, however, not recognised. For this error, an automatic
restart cannot be programmed.
If one of the output phases is interrupted during operation, this should be detected by the inverter.
Function:
The function is active if all following conditions are met:
Pn.74 = 1 (on)
F5-G or F5-M vvc (cs.00 Bit 0-2 = 0..2)
Output frequency ru.03 > 4 Hz
Utilization ru.13 > Utilization lower limit
Since checking of the phase currents is done with a constant ADC level, the result of the utilization lower limit
is dependent on the value of the overload characteristic:
The release time sets itself load and frequency-dependent between 250 and approx. 400 ms. The cause there-
fore is the strong fluctuating load after disconnecting a phase. The load can fall temporary below the utilization
lower limit, whereby the release time is extended.
Abnormal stopping (i.e., automatic shutdown of the drive) is possible for all errors that do not enforce immediate
shutdown of the modulation or for which pre-warnings can be generated.
If abnormal stopping is not sensible in the application, the possibility to set a digital output is available for many
malfunctions.
The motor protection functions can be deactivated. If they are to be used, a pre-warning can be generated here
as well prior to the triggering of an error, providing time to shut down the drive.
- motor protection Pn.14 Motor protect. function response
- Motor overtemperature Pn.12 Motor overtemperature response
The descriptions of the errors and the corresponding pre-warning signals are contained in chapter 8.1 "Error
assistance".
The following responses can be used for all malfunctions and errors, respectively:
Pn.03, Pn.05, Pn.07, Pn.08, Pn.10, Pn.12, Pn.14, Pn.16, Pn.18, Pn.66, Pn.75, Pn.81: Response
Value Explanation
the malfunction turns into the error (Status: E.xx), immediate shutdown of the
0: error / no auto restart
modulation, restart only after RESET
Deceleration at the abnormal stopping-ramp or the torque- and current limit,
1: Stop / modulation off/
respectively, shutdown of the modulation after reaching speed 0, restart only
no auto restart
after RESET
2: Stop / modulation on / Deceleration at the abnormal stopping-ramp or the torque- and current limit,
no auto restart respectively, holding torque after reaching speed 0, restart only after RESET
3: Modulation off / auto Immediate shutdown of the modulation, automatic restart as soon as the mal-
restart function is resolved
Deceleration at the abnormal stopping-ramp or the torque- and current limit,
4: Stop / modulation off/
respectively, shutdown of the modulation after reaching speed 0, automatic
auto restart
restart, as soon as the malfunction has been resolved
5: Stop / modulation on /
Deceleration at the abnormal stopping-ramp or the torque- and current limit, 7
respectively, holding torque after reaching speed 0, automatic restart, as soon
auto restart
as the malfunction has been resolved
The response to the malfunction message limit switch error (hardware or software) and set selection error
can be switched off completely.
For the malfunction "internal overtemperature", there are 2 response options as well:
The abnormal stopping function is different for speed-controlled systems (CS.00 = 4, 5, 6) and for systems with
v/f-characteristic control.
For abnormal stopping with closed-loop systems, the drive is decelerated with the adjusted ramp time, and at
the torque limit, respectively.
For the abnormal stopping, the "normal" torque limitations of the application often do not apply since the auto-
matic shutdown is always a malfunction response. To permit a quicker deceleration with a greater torque here,
there is a unique torque limit for abnormal stopping.
The torque limitation superimposed by the limiting characteristic and the available current remain in effect.
For asynchronous motors, the maximum cutoff torque for abnormal stopping can also be increased to make
more torque available for braking, even in the field weakening range.
For abnormal stopping with v/f-characteristic control, the drive is decelerated with the adjusted ramp time, and
at the torque limit, respectively. Whether braking occurs at the ramp or at the current limit is defined in para-
meter Pn.58.
If no abnormal stopping shall be executed, there are different possibililties for the response, depending on the
type of malfunction.
Additionally, for most malfunctions, issuing the value 6 = warning via a digital output is possible.
Thereby, the inverter does not automatically execute an abnormal stopping. With the warning via a digital out-
put, however, an external control is given the opportunity to respond to the malfunction as is appropriate for
the application. To issue the warning message, a digital output must be programmed with the corresponding
switching condition (see chapter 7.3 Programming of the digital outputs).
For abnormal stopping at the ramp generator, parameter Pn.60 is the ramp time for the deceleration ramp.
At abnormal stopping with differential controller, this ramp is modified that the drive decelerates preferably at a
current limit.
This current limit is set in Pn.59 "abnormal stopping level".
With Pn.58 bit 1 can be selected if the inverter shall decelerate at the active current limit or at the apparent
current limit.
Wirkstrom (ru.17) 1 +
Umrichternennstrom (In.01)
For safety, a maximum time for the abnormal stopping-function can be programmed.
If the inverter is still in the malfunction state (A.XX) after this time (no RESET or automatic restart was exe-
cuted), the inverter switches off the modulation and changes to the corresponding error state (A.XX => E.XX).
Abnormal stopping can also be triggered via the control word (SY.43 and SY.50, respectively ). The parameter
Pn.58 abnormal stopping mode determines the behaviour of the abnormal stopping via control word.
With the automatic restart, the inverter error automatically resets or automatically terminates the abnormal
stopping caused by a malfunction or pre-warning.
The function can be activated separately for the various errors and malfunctions with the Pn-parameters.
The automatic restart only makes sense if the error can be expected based on the application. Normally, the
cause of the error must first be investigated and eliminated before the drive can be put back in operation by
executing the reset.
Therefore, it must be selected after which errors an automatic restart should be executed
7
7.13.3.1 Undervoltage error (E.UP)
In Pn.00 "automatic restart E.UP", the automatic restart for the undervoltage error is activated in the factory
setting.
A typical application for the automatic restart E.UP (Pn.00) is operation on a bad power grid where sporadic
brownouts are to be expected. With this function, the application continues running as soon as the mains vol-
tage is sufficiently high again.
For the undervoltage error, a time can be defined within which the automatic restart is permissible.
The error overvoltage occurs mostly at high speed. By activation of Pn.01 "automatic restart E.OP", it can be
avoided that the drive "spins down" for a long time after this error. This function makes sense only in combina-
tion with the speed search (see chapter 7.15).
The base-block time (bbL) is at least 1 second, even if the value in uF.12"base-block time" is smaller. Further-
more, the base-block time before the restart is always observed, even if uF.13 "motor de-excitation lower limit"
is undershot.
The automatic restart after occurence of an overcurrent error is activated with Pn.02 "automatic restart E.OC". It
can be used if impact overloads of the FI, e.g. due to blocking of the motor, can be expected in v/f-characteristic
operation.
The base-block time is treated as in overvoltage errors.
After 10 restart attempts, the inverter state must be unequal to the base-block time or the overcurrent error for
at least one second, otherwise the restart is aborted.
A malfunction response with automatic restart is selected in parameters Pn.03, Pn.05, Pn.07, Pn.08, Pn.10,
Pn.12, Pn.14, Pn.16, Pn.18, Pn.66, Pn.75 and Pn.81 with the values3...5.
The base-block time is observed only if the drive is above uF.13 "motor de-excitation lower limit".
After shutdown of the modulation (e.g., when opening the control release or if an error occurs), one must wait
for the time shown in uF.12 "base-block time" before the modulation can be switched on again. During this pha-
se, ru.00 displays the status "motor de-excitation" and the display shows "bbL", respectively.
If ru.42 "modulation grade" is below uF.13 "base block voltage level" when switching off the modulation, there
is no base-block time. Even at low frequencies, the base-block time is not observed.
Exception: After overvoltage- or overcurrent-error, a minimum base-block time of 1s is inserted.
The parameters uF.12 and uF.13 are dependent on the power circuit and serve only as information for the user
on which minimum switch-off times to expect in the application.
In parameter Pn.65 / bit 8 "256: bbL is not displayed", the status message "motor de-excitation" can be sup-
pressed so that the event caused by the modulation switch-off becomes visible immediately.
In v/f-characteristic operation, one can choose in Pn.58 "quick stop mode" between ramp generator and diffe-
rential controller. For the differential controller, the time constant is set in Pn.60. The setpoint of the differential
controller is preset in Pn.59: "quick stop level", Pn.58 selects the actual value from either apparent current or
active current.
The abnormal stopping can be activated by malfunction as well as via the control word (Sy.50 Bit 8). The func-
tionality is the same in both cases, but for state "79: abnormal stopping" (StOP) is always displayed.
For all modes, one can choose whether the abnormal stopping-bit is reset in the status word (Sy.51 or Sy.44 Bit
For abnormal stopping with v/f-characteristic control, the drive is decelerated with the adjusted ramp time, and
with the differential controller, respectively.
The current limit for the differential control is set in Pn.59 "abnormal stopping level".
Pn.59: Quick stop level
Value Explanation
0...200 % Current limit for difference control = 0...200% rated inverter current (In.01)
-
Active current (ru.17) 1
Rated inverter current (In.01)
For abnormal stopping with closed-loop systems, the drive is decelerated with the adjusted ramp time, and at
the torque limit, respectively.
For the abnormal stopping, the "normal" torque limitations of the application often do not apply since the auto-
matic shutdown is always a malfunction response. To permit a quicker deceleration with a greater torque here,
there is a unique torque limit for abnormal stopping.
The torque limitation superimposed by the limiting characteristic and the available current remain in effect.
For asynchronous motors, the maximum cutoff torque for abnormal stopping can also be increased to make
more torque available for braking, even in the field weakening range.
For safety, a maximum time for the abnormal stopping-function can be programmed.
If the inverter is still in the malfunction state (A.XX) after this time (no RESET or automatic restart was exe-
cuted), the inverter switches off the modulation and changes to the corresponding error state (A.XX => E.XX).
Abnormal stopping can also be triggered via the control word (SY.43 and SY.50, respectively ). Then, the status
displays "79: abnormal stopping" (StOP). The behaviour for abnormal stopping via control word can be defined
in parameter Pn.58 „quick stop mode“.
Abnormal stopping mode determines the behaviour for abnormal stopping via control word.
The speed search permits a relatively smooth engagement of the frequency inverter onto a running motor.
Without activation of the speed search, the motor is always slowed down first. In closed-loop operation with
encoder, this occurs at the torque limit, in closed-loop operation without encoder, the motor must be stopped
with DC-current braking.
On activation of the speed search, however, the current speed is determined and the drive is accelerated or
decelerated from this starting point to the setpoint speed, according to the adjusted ramps.
Parameter Pn.26 "Speed search starting condition" determines after which events the speed search is to be
executed.
The speed search mode determines the frequency and voltage jumps as well as the maximum load factor with
which the function works. Higher values let the function work faster, lower values make the function "softer".
In closed-loop operation with activated speed search, the ramp output value is set to the current actual speed.
After the motor flux has been built up, the drive runs up to the setpoint speed.
In closed-loop operation without encoder, the current actual speed must be estimated from the motor model.
For special motors (e.g., high frequency spindles) or applications (e.g., operation in very high field weakening
range), this estimate for the engagement onto a running motor may not work. The speed is then calculated
incorrectly and the drive vibrates or the inverter raises a malfunction.
In these cases, the motor must be stopped by DC braking (see chapter 7.15.) before the drive can be restarted.
Generally, however, the speed search is the most jerk-free and quickest path to switch to a running motor.
If the determined speed after speed search is below the limit in pn.90, speed = 0 rpm is preset.
Advantage: The drive is not in "delay" state, thus the motor model remains always active from the start.
The ramp stop function essentially fulfils two tasks. It reduces the risk of:
- Overcurrent errors (E.OC) during the acceleration or deceleration phase (only for v/f-characteristic
operation)
- Overvoltage error (E.OP) during the deceleration phase (in all operating modes)
By stopping the ramp if Pn.24 "ramp stop current level" or Pn.25 ramp stop DC-link voltage level" is exceeded.
Moreover, the ramp stop function can be activated by a digital input.
Pn.22 selects which of the ramps (acceleration, deceleration or both ramps) can be stopped.
In the operating modes positioning or synchronous running, this function is not active. If a ramp time is entered
in Pn.60 for abnormal stopping, the deceleration stop is active.
If the acceleration stop in speed-controlled operation is active in order to use the interruption of the ramp via
digital input, the current level in Pn.25 must be set to 200% to avoid negative effects.
7
The LD-Stop (U) function can be used to prevent overvoltage errors during deceleration.
During deceleration energy is refeed into the frequency inverter, which causes a rise of the DC-link voltage.
If too much energy is recovered, the inverter can switch to overvoltage (OP) error.
If the LD-Stop (U) function is activated with Pn.22, the DEC-ramp is stopped when the current DC link voltage
(ru.18) exceeds the adjusted LD voltage (Pn.25).
Pn.25: LD voltage
Value Explanation
200V...1200V DC-link voltage level at which the ramp is stopped
Overvoltage errors cannot always be securely prevented with this protection function, because after setting
the ramps and the speed controller, further deceleration can occur despite stopping the ramp. If the drive
decelerates, e.g., at the torque limit, and can therefore not follow the ramp, it does not help to stop this ramp.
An undershoot of the speed controller due to a sudden termination of the ramp can also lead to further energy
recovery in the DC link.
Generally, the deceleration process is slowed down by this protection function. The use of a braking resistor is
necessary for a dynamic deceleration.
With Pn.23 "ramp stop input selection", a digital input can be selected for triggering the ramp stop.This input is
only active if the stop is permitted in Pn.22 for the corresponding ramp.
Whether the output frequency must be increased or decreased depends on the torque characteristic of the
application. For a fan, e.g., the load factor increases with the speed, and the output frequency must be reduced
during overload. For a drilling machine, the load factor decreases with the speed, and the drive must therefore
be accelerated during overload.
When falling below the maximal constant current the inverter accelerates / decelerates again with the normal
ramp times.
The stall function is active until the original setpoint speed is reached.
This protection function is active only for F5A-M in open loop operation (cS.00 = off).
The stall level is adjusted in parameter Pn.20. When exceeding this limit, the inverter increases or decreases
automatically the output frequency (depending on the adjustment in Pn.19) in order to reduce the load.
The rate of change of the output frequency is dependent on Pn.21. Depending on the setting of Pn.19, the ramp
time of the stall function or the time constant of the controller is adjusted here.
Picture 7.13.8.1 Function of the current limit at constant run (default setting)
adjusted
constant current limit
Actual frequency ru.03
In the operating modes positioning or synchronous running, this function is not active. If a quick stop ramp time
is entered in Pn.60 for abnormal stopping, the deceleration stop is active.
If the acceleration stop in speed-controlled operation is active in order to use the interruption of the ramp via
digital input, the current level in Pn.25 must be set to 200% to avoid negative effects.
The LD-Stop (U) function can be used to prevent overvoltage errors during deceleration.
During deceleration energy is refeed into the frequency inverter, which causes a rise of the DC-link voltage.
If too much energy is recovered, the inverter can switch to overvoltage (OP) error.
If the LD-Stop (U) function is activated with Pn.22, the DEC ramp is stopped if the actual DC link voltage (ru.18)
exceeds the adjusted LD voltage (Pn.25).
Pn.25: LD voltage
Value Explanation
200V...1200V DC-link voltage level at which the ramp is stopped
Overvoltage errors cannot always be securely prevented with this protection function, because after setting
the ramps and the speed controller, further deceleration can occur despite stopping the ramp. If the drive
decelerates, e.g., at the torque limit, and can therefore not follow the ramp, it does not help to stop this ramp.
An undershoot of the speed controller due to a sudden termination of the ramp can also lead to further energy
recovery in the DC link.
Generally, the deceleration process is slowed down by this protection function. The use of a braking resistor is
necessary for a dynamic deceleration.
With Pn.23 "ramp stop input selection", a digital input can be selected for triggering the ramp stop.
This input is only active if the stop for the corresponding ramp is permitted in Pn.22.
The motor protection function protects the connected motor against thermal destruction caused by high cur-
rents. The function corresponds largely to mechanical motor protection components, additionally the influence
of the motor speed on the cooling of the motor is taken into consideration. The load of the motor is calculated
from the measured apparent current (ru.15) and the adjusted rated motor current (dr.12).
The following tripping times (VDE 0660, Part 104) apply for motors with separately driven fan or at rated fre-
quency of a self-ventilated motor:
1,2 • In ⇒ 2 hours
1,5 • In ⇒ 2 minutes
2 • In ⇒ 1 minute
8 • In ⇒ 5 seconds
This parameter specifies for each set the rated current (= 100% utilisation) for the motor protective function.
The motor protection-load is calculated as follows:
Pn.14 specifies the performance of the drive on activation of the motor protective function.
120 min f = 0 Hz
60 min
30 min
Motor current
2 min
1 min
30 sec 120
7
5 sec
%
50 100 150 200 400 600 800
For self-ventilated motors the tripping times decrease with the frequency of the motor (see picture). The motor
protection function acts integrating, i.e. times with overload on the motor are added, times with underload are
subtracted. After triggering the motor protection function, the new tripping time is reduced to 1/4 of the specified
value, if the motor has not been operated for an appropriate time with underload.
In some applications, several motors are operated alternatively at one inverter. The change-over between the
motors is done synchronously with the set changeover.
The motor protection function must then be able to distinguish which of the motors is currently being supplied.
For that purpose there is the parameter Fr.08 "motor set classification". Each motor is assigned a number from
0 to maximally 7 and this value is entered in parameter Fr.08 in all sets where the respective motor is supplied.
Example:
The motor protection function is calculated separately for all motors, i.e., for each individual motor, a separate
overload counter is running.
If one of the counters reaches the limit of 100%, the behaviour programmed in Pn.14 "motor protection function
response" is triggered.
The motor protection function is activated if the ratio of apparent current (ru.15) to continuous current (Is/Id)
exceeds the value of dr.50 "motor protection min Is/Id". The release time for this point is set in dr.34 "motor
protection time min Is/Id". The release time for maximum current is set in dr.35 "motor protection time Imax". If
a higher value is programmed in dr.35 than in dr.34, the time dr.34 is valid in the whole range.
The maximum current is defined by dr.33 "DSM max. torque" or dr.15 "max. torque FU". The smaller of the two
values determines the maximum current.
The continuous current is speed-dependent. At speed 0, it is equal to dr.28 "DSM stand still current" and at
dr.24 "DSM rated speed" it reaches the value dr.23 "DSM rated current".
dr.28
DSM Stillstandsdauerstrom
Dauerstrom (Id)
dr. 23
DSM Nennstrom
dr.24
DSM Nenndrehzahl
Drehzahl
The tripping time is the time required by the internal counter to count from 0 to 100%. On reaching 100%, error
"30: ERROR! Motor protection function" (E.dOH) is triggered.
Auslösezeit
dr.34
Motorschutzzeit
min Is/Id
dr.35
Motorschutzzeit Motorstrom / Dauerstrom
bei Imax ( Is / Id )
dr.50 dr.33
Motorschutz Dauerstrom
min Is/Id
A warning level can be adjusted in Pn.15 "OH2 warning level". If the counter reaches this level, the adjusted
response in Pn.14 "Waning OH2 stop. mode" is carried out.
The counter is reduced if the ratio apparent current to continuous current is lower than dr.50. The recovery time
dr.36 "mot.prot. recovery time" is the time the counter requires to count from 100% to 0% (after triggering of
the error, i.e., without current flow). The error triggered by the motor protection function can already be reset
before the recovery time expires.
The Power-Off function has to provide a controlled deceleration of the drive until standstill in case of undervol-
tage (e.g. due to power failure). The kinetic energy of the rotating drive is used to support the inverter DC-link
voltage. As a result the inverter remains in operation and can decelerate the drive in a controlled manner.
Especially in the case of parallel running drives (e.g.textile machines) the uncontrolled running down of the
7
motors and the consequences resulting from it (thread breakage) can be avoided.
For the various operating modes, the amount of available functions differs.
For the vector controlled modes, some parameters have no function or are even hidden.
Here is an overview:
The parameter Power-Off-Mode (Pn.44) switches on the function and determines the basic behaviour:
The power-off function starts when the DC-link voltage drops below a certain value (start voltage). The start
voltage can be set automatically or manually depending on Pn.44 Bit 1.
Pn.44 bit 2
Starting jump 2: calculated actual
Slip calculation by the active value
current 0: Active current IW
Motor data
Slip calculation 4: Utilisation IS
Utilisation
By means of the jump factor the automatically determined starting jump can be adapted to the respective ap-
plication.
In case the jump factor is too small, the inverter trips to "2 Error! Undervoltage" (E.UP)!
For too great a jump factor, the inverter runs into the hardware current limitation (status value 80 "HCL"). The
control cannot work correctly, thus causing a wrong calculation of the active current!
Power off controller (not for vector controlled operation with DSM)
There are two controllers: the DC-link voltage controller and the active current controller.
In v/f-characteristic operation the active current controller is subsequently connected to the DC link voltage
controller.
In vector controlled operation, the output of the DC link voltage controller is used as the torque limit.
If the braking torque is selected as the setpoint (Pn.44 bit 6...7 = 3), the DC link voltage controller is deactivated.
The initial voltage value selected with Pn.44 bit 1 is the setpoint of the DC link voltage controller (see also:
tripping of Power-Off
This setting is used to bridge voltage drops. At power recovery, the drive restarts if the output speed is not yet
fallen below the "power off restart level" (Pn.48).
Setpoint = braking torque (Pn.47; Pn.44 Bit 6...7 = 3) (only v/f characteristic operation)
The braking torque is used as setpoint source, if the drive must be stopped as fast as possible in case of po-
wer failure. The DC link voltage controller is switched off in this case. Since the DC link voltage increases very
strongly in this mode, a braking module is required.
Since the hardware current limit should not be reached with active current control, the setpoint value is limited
internally which can lead to oscillations. In that case the setpoint value can be reduced, which leads to prolon-
gation of the delay. If the voltage stabilization is switched on (Pn.44 bit 8 = "1") and uF.09 = rated voltage is
adjusted, the voltage is not very high and the deceleration is more evenly.
Setpoint = voltage setpoint (Pn.50; Pn.44 Bit 6...7 = 1 or 2) (only v/f characteristic operation)
7
The setpoint DC-link voltage can be preset in the range of 200...1200 V. The internal value is limited down in
order to ensure a safe operation. The value of the DC-link voltage in normal operation plus approx. 50V adjusts
itself as minimum value. If a braking resistor is connected, the adjusted value may not lie above the threshold
of the braking resistor, else the controller cannot work (threshold 200V-class: 380V; 400V-class: 740V; 600V-
class: 1140V).
If Pn.44 Bit 6...7 = 1, the voltage setpoint is used as setpoint immediately after the triggering of power off. A
restart is only possible in this mode after reaching the standstill.
In mode Pn.44 Bit 6...7 = 2, initial voltage is adjusted to first after triggering. On undershooting the restart value
(Pn.48), the setpoint is increased to the voltage setpoint (Pn.50) with a ramp. This ensures that the drive upon
reaching the standstill in critical applications has enough energy for braking.
To be able to adapt the drive individually to the application, the proportional factor of the DC link voltage con-
troller can be adjusted with Pn.51 and the integral factor with Pn.57, in the range of 0...32767. In most cases
the default setting will achieve sufficient results. If, however, overshoots or fall backs of the motor occur, the
values must be lowered.
Pn.53, Pn.54 and Pn.55 are the controller parameters of the active current controller
A D-part usually has a positive effect in the control. Pn.55 should have about 10-times the value of Pn.53.
The following parameters effect the behaviour of the inverter if the mains voltage returns during the power-off
function.
Depending on the application, it can be reasonable to execute the restart on power recovery only above a de-
fined output value. This restart value is adjusted in Pn.48.
Dependent on the setpoint value source (Pn.44 bit 6 ..7) following conditions occur:
2. Control to voltage setpoint Pn.50 for a base value smaller than the restart value (Pn.44 bit 6...7 = 2):
As long as the base value is greater than the restart value, the inverter behaves as in item 1.Below the restart
value the voltage setpoint value of Pn.50 is increased and with active current control (without speed detection)
the control parameters of the active current control are reduced linearly with the output value. At power recove-
ry a restart is only possible after reaching the standstill.
3. Control to voltage setpoint Pn.50 or braking torque Pn.47 (Pn.44 bit 6.,.7 = 1 or 3):
The controller parameters of the active current controller (without speed detection) are reduced below the re-
start value linearly with the output value.
At power recovery a restart is only possible after reaching the standstill.
This parameter has principally the same function as Pn.48 (power off restart threshold).
The drive does not restart at system recovery if the adjusted time is up but it is decelerated with quick stop to
standstill.
If both parameters are adjusted (Pn.48 and Pn.96) the condition that is reached earlier is valid.
Example: Pn.48 = 25 Hz; Pn.96 = 10 s.
a) 25 Hz fall below after 8 s - no restart after 8 s
a) 25 Hz fall below after 12 s - no restart after 10 s
Pay attention:
The maximum time contains Pn.52 (restart delay).
Example: Pn.52 = 1 s; Pn.96 = 10 s.
a)System recovery after 8.5 s - restart after 9.5 s
b) System recovery after 9.5 s - no restart because Pn.96 is exceeded.
Pn.44 bit 3..4 determines how the drive behaves on reaching standstill.
function active" (POFF). Attention:Motor warming! A reset is necessary for the restart.
Pn.44 bit 3...4 = 1:
The inverter modulates independent of a set direction or rotation with the adjusted boost and is in status "78:
power off function active" (POFF). After expiration of the restart delay Pn.52 (if adjusted) the inverter restarts
automatically when the mains voltage is returned.
Pn.44 bit 3...4 = 2:
The inverter switches off the modulation and is in status "84: no direction of rotation after power off" (PLS). A
reset is necessary for the restart.
If a restart is allowed the restart delay time is kept constant after power recovery of the output value. It is ad-
justable within the range of 0...100 s (Default 0 s). After expiration of this time it is accelerated again onto the
current setpoint.
Only hardware inputs can be adjusted with this parameter, since these inputs are scanned in the same cycle
where the power off control is active. A setting via the control word or di.01/02 is not possible.
To better illustrate the context, the operating modes are explained in more detail in the following section.
Bridging of mains gaps (not for vector controlled operation with DSM)
Setpoint source: Starting voltage (Pn.44 bit 6..7 = 0)
Voltage setpoint Pn.50, if base value < Pn.48 (Pn.44 bit 6..7 = 2)
In this mode the motor shall be operated almost in no-load operation and only the energy, the inverter requires
for operation should be regenerated. The starting voltage is simultaneously the setpoint of the DC-link voltage 7
controller. In v/f-characteristic operation, the set value is the setpoint of the active current controller, and in
vector controlled operation, the torque limit of the speed controller.
In case of weak supply systems it is recommended to choose the automatic starting voltage, as in this case the
starting voltage value is adapted to slow voltage fluctuations.
In the first cycle, a speed jump is issued in v/f-characteristic operation, and in vector controlled operation, the
limit of the speed controller is set to the measured slip so that the drive is put in no-load operation.
To safely decelerate the drive in v/f-characteristic operation, the controller parameters of the active current
controller are lowered to below of the restart value, linearly with the base value.
An immediate restart is not executed below the restart value (Pn.48). The drive decelerates with the quick stop
function (Pn.60..61) and behaves according to the adjustment in Pn.44 bit 3...4 or after expiration of Pn.96.
To have more energy available for braking the inertia mass on reaching standstill in v/f-characteristic operation,
the voltage setpoint can be increased to the voltage setpoint (Pn.50) (Pn.44 Bit 6...7 = 2) (if below the restart
level (Pn.48)).
In this case the control remains active with the increased setpoint value. A restart is possible then only after
reaching the standstill.Then the behaviour of Pn.44 Bit 3...4 is defined.
In this mode, too, the lowering of the controller parameter is executed.
Emergency stop with braking module (not for vector controlled with DSM)
Setpoint source: Brake torque (Pn.44 bit 6...7 = 3)
In this mode the drive is to be stopped as fast as possible. Since the recovered energy can be very high, a
braking resistor is necessary.
The DC-link voltage controller is not active. In v/f-characteristic operation, the setpoint of the active current
controller is the braking torque (Pn.47). In vector controlled operation, the drive decelerates with the abnormal
stopping-function (Pn.60, 61, 67; see chapter 7.13.5) and behaves dependent on the setting in Pn.44 Bit 3...4.
The drive supplies no energy anymore at low speed in v/f-characteristic operation. In this case, the control must
be very soft, to prevent fall back.
It is possible to adjust the restart (Pn.48). The controller parameters of the active current controller are lowered
to below this value, linearly with the base value.
Emergency stop without braking module (only V/f characteristic operation)
Setpoint source: voltage setpoint Pn.50 (Pn.44 Bit 6...7 = 1)
In some cases one can do without a braking module with the Emergency-stop function, if the losses in the motor
are very high at high DC-link voltage.
The voltage stabilization should be switched off in this case. This can be done with Pn.44 Bit 8 = 1 during
Power-Off.
The DC-link voltage control is active. Deceleration is always to standstill. Then the behaviour results from the
setting of Pn.44 bit 3...4.
At small speeds the drive supplies no more energy. In this case, the control must be very soft, to prevent fall
back.
It is possible to adjust the restart value (Pn.48). The controller parameters of the active current controller are
lowered to below this value, linearly with the base value.
Only the emergency-stop function with braking module can be activated in this operating mode. After triggering
of power off, the drive decelerates with the abnormal stopping-function (Pn.60, 61; see chapter 7.13.5) and
behaves dependent on the setting in Pn.44 Bit 3...4.
7.13.11 GTR7-control
An input for activation of the GTR7 can be defined with Pn.64. Thereby, the activation of the braking transistors
of all inverters in a DC interconnection of several drives can be synchronised and the occuring braking energy
is distributed to all inverters.
The GTR7 controls the braking resistor, in this case, independent of the inverter state and the DC link voltage,
as soon as the input is active.
Exception: On opening the control release (noP), the GTR7 is always switched off for safety reasons.
I.e., as soon as a digital input is selected for activation of the braking transistor, the adjustments in Pn.65 con-
cerning the GTR7 and the parameter Pn.69 are without function.
The DC link voltage threshold where the braking transistor becomes active can be adjusted with Pn.69 "GTR7
voltage".
The settings range is 300..1500V. This value is internally limited down in the inverter: the braking transistor
becomes active no earlier than ru.68 "DC link voltage rated value" * 1.0625.
The DC link voltage rated value is the DC link voltage measured at "Power on".
In the factory settings, the braking transistor is only active if the modulation is also enabled.
The reason for it is that for "standard" asynchronous motors, the recovery of energy into the inverter also ends
with shutdown of the modulation.
When using synchronous machines in the field weakening range, or using sine filters at the inverter output,
recovery may continue despite the modulation being switched off.
In that case, Pn.65 should be changed.
In which cases the braking transistor remains active even if modulation is switched off depends of the specific
application.
The electrical work that is converted via the GTR7 resistance is displayed with parameter ru.91. The correct
input of the resistance in Pn.82 is required therefor. On reaching the maximum value of 99999 kWh the counter
is limited to this value. The limitation is given by the operator display (5 digits).
Parameter ru. 91 is writable. It is set to its default value by new-initialization and writing on power on counter
(ru.40).
In these parameters, many different functions for adapting the inverter behaviour to special applications are
pooled.
*1 to bit 0, 3, 5: With the GTR7 (braking transistor), a braking resistor can be connected to the DC link
that absorbs recovered energy when the motor is working as a generator.
By default, the GTR7 is off when the modulation is switched off.
For some applications (e.g., synchronous motor operated in the field weakening range)
it is sensible to leave the GTR7 active for switched off modulation, so that the braking
resistor can be added when the DC link voltage exceeds the value of Pn.69 "GTR7 DC
link voltage level".
By setting this bit, the GTR7 function is available for the respective inverter state. 7
*2 to bit 11: If a drive is controlled via a bus system, and the control release is furthermore switched
via the control word, the two watchdogs (Operator watchdog and HSP5 watchdog)
should be activated so that the drive is stopped on failure of the bus system. However,
the drive can then not be repositioned by hand anymore, since – as long as the bus is
down - the malfunction- or error message of the watchdog remains in force.
With parameter di.39 "Shutdown ST input selection", an input can already be selected
with which the digital setting of the control release (i.e., setting via di - parameter or the
control word) can be deactivated. Only terminal ST (X2A.16) is effective and the control
of the control release can be done via the digital input.
If this bit is set, the two watchdogs are also deactivated with the selected (in di.39) input.
If a response with automatic restart is now selected for the watchdog error, the malfunc-
tion automatically resets and the drive can be used in manual operation.
The flow control with valve control and flow monitor is adjusted with this function.
Pn.91: Flow control mode
Bit Meaning Explanation
-No.
0: off (default)
0 Flow control mode
1: on
Flow control mode off/on
0: Operation (Default)
1 Mode drive active
2: bereits + ST
The function is active and the mode for the status "drive active" is selected. If the function is not active
(Pn.91 bit 0 = 0) no error and warning is triggered.
There are two adjustments for the status „drive active“:
operation The drive is active if the modulation is released.
ready + ST The drive is active, if all following conditions are met:
- the power unit is supplied (no NO_PU)
- there is no malfunction
- the control release is active (no NOP)
Pn.92: Valve control output selection
Name Function Decimal values
no no 0 (default)
O1 Transistor output 1
O2 Transistor output 2
R1 Relay output 4
R2 Relay output 8
OA Internal output 16
OB Internal output 32
OC Internal output 64
OD Internal output 128
no flow
ru.25 Drive active
Pn.93
& &
Pn.94
ru.22
Flow state Pn.94
0: no flow Pn.91
1: Flow
ru.38
<
Pn.95
Temperature state
0: OK
1: Condensation risk
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp presetting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
The KEB COMBIVERT contains 8 parameter sets (0...7), i.e. all programmable parameters are available 8
times in the inverter and independent of each other they can be assigned with different values. As a lot of
parameters in the parameter sets contain the same value, it would be relatively complicated to change every
parameter in each set individually. In this section it is described, how one copies whole parameter sets, locks
them, selects them and reinitializes the inverter.
Certain parameters are not set-programmable, since its value must be equal in all sets (e.g. bus address or
baud rate). For an easy identification of these parameters the parameter set number is missing in the parameter
identification.
For all non-programmable parameters the same value is valid independent of the selected parameter
set!
Sy-Parameter Pn.00...18/ 23/ 27/ 29/ 42/ 44...60/ 62...66/ 68/ 69/ 74…81
ru-Parameter uF.08/ 12...15/ 18 (uF.09 at F5-S) / 18…23
Ec parameters ud.01...17/ ud.22…31 (all at F5-S)
AA-Parameter Fr.02...04/ 07/ 09/ 11 (Fr.10 at F5-S)
di-Parameter An.00...04/ 10...14/ 20...24/ 41...56
In-Parameter (exception: In.25) LE.16-27
dr-Parameter (not at F5-S) cn.03/ 11...13
oP.19/ 20/ 50/ 53...58/ 60…63/ 65…68/ 74 dS.00...01 (only F5-S) 7
PS.02...04/ 10...27/ 29...31
The security parameters contain the Baud rate, inverter address, hours/meter, control type, serial-/customer
number, trimming values and error diagnosis. They are not overwritten while copying parameter sets from the
default set.
dr-Parameter Pn.61/ 67
cS.00...19...22 dS.00...01/ 13
Ec.01...07/ 11-27/ 36...38 Fr.10
At indirect set-addressing the parameter values are displayed and edited based on the set indicator (Fr.09).
The direct set-addressing enables the display or writing of a parameter value independent of the set indicator
directly into one or several parameter sets. The direct set programming is only possible with bus operation.
With the keys UP/Down and at flashing The source set is adjusted with the
point the target set 0...7 is adjusted in keys UP/Down. The copying process
addition to the parameter set number. is started with „Enter“. Copying is only
The active (A) parameter set must not possible with opened control release or
be adjusted as target set while copying. error, otherwise „I_oPE“ appears in the
If the target set is > 0, only the program- display (invalid operation).
mable parameters are overwritten! After copying „PASS“ appears in the dis-
play and could be erase by „ENTER“.
Two parameters are responsible at indirect set-addressing for copying of parameter sets via bus. Fr.09 defines
the target set. Fr.01 defines the source parameter set and starts the copying process. The source set (Fr. 01)
is copied in the selected parameter sets at direct set-programming. The following copying actions can be prac-
tised:
By loading the factory setting all definitions defined by the mechanical engineer are reset! This can comprise
the terminal assignment, set changeover or operating states. Before loading the default set it is to be ensured
that no unintended operating states occur.
7
The values -5 to -8 are corresponding to the previous values -1 to -4 referring to the copied parameters and
target sets. They differ only in the default value source.
Value -9 enables storing of the current parameter setting as custom-specific default values. The values of all
customer and system parameters are stored in all sets thereby.
Bit 27 is set in characteristics 2 for parameters, which contains only the KEB default value. These are among
others all security parameters and all write protected parameters.
During loading the specific default values (fr.01 = -5..-8) these parameters are loaded with KEB default values
if necessary.
The indicator parameter (first parameter of an indirect addressed group) has not a customer default value, be-
cause the parameter was set to 0 with Power-On-Reset. The parameters belonging to the group have a default
value for each value of the indicator.
A source table is generated. For this one byte is reserved for each parameter in the sequence of the bus
addresses. This byte contains the information for each set, whether the default value is determined from the
parameter definition (= 0, KEB default value) or if the value is stored in the custom-specific storage area (= 1).
This information is determined by comparison to the KEB default value.
For indirect addressed parameters the number of reserved bytes for each group member is equal to the number
of valid values of the indicator. 36 byte are reserved for ud.16 and ud.17, ud.15 = 1...36. 16 byte are reserved
for ps.24...27, ps.23 = 0...15. The custom-specific default values are stored in the sequence of the bus addres-
ses (ascending) set depending (set 7..0).
The custom-specific default values for indirect addressed parameters are stored first to bus address (ascen-
ding), then to indicator value (max. ... min.), then set-dependent (set 7..0).
Example: Default value ud.09, default values ud.16 for ud.15 = 36..1, default values ud.17 for ud.15 = 36..1,
default values ud.18 for set 7..0, etc.
With the bits in the source table the default value for each set is determined either from the parameter definition
for each parameter in the sequence of the bus address or read-out from the custom-specific memory area and
written into the parameter.
Parameters only with KEB default value are loaded with the KEB default value in this case.
The default values are reset for all parameters to KEB default values in the following cases:
- The power unit identifiction dependent KEB default values are adapted.
- If necessary uf.11 is limited in all sets to the maximal switching frequency (in.03).
- If a custom-specific default value of uf.11 is not within the value range (0..in.03)uF.11 is loaded during
default value loading in the corresponding set with the KEB default value.
- If the write value of sy.03 is unequal to the reading power unit identification, all customer and system
parameters are loaded with KEB default values (corresponding Fr.01 = -4).
Memory management
The length of the source table (in byte) and the length of the memory area of the customer default values (in
byte) are stored in one word at the end of the external RAM.
The source table for the custom-specific default value range is in front of these two cells. The length is de-
pendent on the number of permitted parameters of the adjusted control type (ud.02 bit 2+3)
7
The memory area for the custom-specific default values begins next to the source table. The length depends
on the number of values stored here. Only the values which are different to the KEB default values are sto-
red. The default values are stored in descending order of memory addresses.
The off-line memory includes the period between the temporary variables and the memory area for the cu-
stom-specific default values. The size of the off-line memory is depending on the number of custom-specific
default values.
Fr.01 = -10 (customer default memory is full) is set if the memory area is completely filled with custom-
specific default values, all values could not be stored.
That means only one part of the parameter setting (with low bus addresses) contain custom-specific default
values, further settings only contain KEB default values.
This restriction should not occur since enough memory is available.
22 21 20 765 4 3 2 1 765 4 3 2 1
digital
set selection deacti- Fr.4 via terminal strip terminal strip SY.50 with
vated keyboard/bus terminal strip binary coded binary coded control word
set 0 always active binary coded ST>RST>...>ID ID>IC>...>ST (only bus)
2 3
1 4
0 5 ATTENTION! The valency at Fr.2
Fr.2 = 2...4 is generally ID > IC > ...>ST.
The sequence above refers to the
Parameter set 0...7 priority.
Is
Fr.3 Parameter set lock selected set yes Error message
Set 0...7 (Value 0...255) locked ? E.SEt
no
As shown in Fig. 7.14.7 it is defined with Fr.02 whether the parameter set selection is enabled or disabled via
keyboard/bus (Fr.04), terminal strip or via control word (SY.50). The selection is activated with „Enter".
This parameter can be written by bus as well as by keyboard. The desired parameter set (0...7) is preadjusted
directly as value and activated with „Enter“.
The setting via terminal strip can be made binary-coded or input-coded. The inputs are defined with parameter
Fr.07. With binary-coded set selection maximally 3 inputs should be programmed to set selection to avoid set
selection errors.
Example
I1, I2 and F are defined for set selection at input-coded set selection (Fr.02=3).In this case F = set 1; I1 2 =
and I2 3 = would be acticated as the valence is (I2>I1>F). If I1 and I2 are triggered simulateously the inverter
switches into set2 since the priority is F>I1>I2 at Fr.02=3.
Example 1: With 3 inputs (F, I1 and I4) set 0...7 shall be selected
1.) Adjust parameter Fr. 07 to value „148“
2.) Adjust Fr.02 to value „2“ (set selection binary-coded via terminal strip)
I4 I1 F Input
I4
2² 21 20 set
0 0 0 0 I1
0 0 1 1 F
0 2 0 2
0 2 1 3 Set 7
Set 6
4 0 0 4 Set 5
4 0 1 5 Set 4
Set 3
4 2 0 6 Set 2
4 2 1 7 Set 1
Set 0 t
Example 1: With 5 inputs (I1, I2, I4, IB and ID) set 0...5 shall be selected.
Id IB I4 I2 I1 set set
Fr.02 = 3 4
0 0 0 0 0 0 0 ID
0 0 0 0 1 1 1
0 0 0 2 0 2 2 IB
0 0 3 0 0 3 3
0 4 0 0 0 4 4 I4
5 0 0 0 0 5 5
I2
5 0 3 0 0 3 5
5 0 3 0 1 1 5
I1
set 5
set 4
set 3
set 2
set 1
set 0
The parameter Fr.11 defines an input, with which one can switch independently of the current parameter set in
to parameter set 0. This function is only active at Fr.02 = 0...4.
- with static input assignment the inverter remains in set 0 as long as the input is set. 7
- with edge-triggered inputs set 0 is always activated with the 1st edge. With the 2 nd edge the set activa-
ted by the other inputs is selected again.
Parameter Fr.12 adjusts the behavior at set change. A motor set change without parameter set changes is only
possible when the modulation is switched off.
If the set change is disabled and the modulation is switched on, then a planned set change releases the errror
‘set selection error‘ (E.SET, A.SET). The set change is done, as soon as the modulation is switched off.
Bit 1 has no function at F5-S, since only one motor set is available here.
Parameter sets, that shall not and must not be selected, can be locked with Fr.03. If one of the locked sets is
selected, the adjusted response in Pn.18 is executed (default: set selection error (E.SEt).
The OFF time of the old set and ON time of the new set are added at set changeover.
Set 3
Set 2
Set 2 3
Adjusted set
Set 1
4
Set 0
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp presetting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
The following section should facilitate the adjustment and programming of special functions.
7.15.1 DC-braking
- in software type F5-A (standard software), for v/f characteristic control of asynchronous motors (control
type F5-M and cS.00/ control mode < 4)
- in software type F5-H in speed-controlled operation of asynchronous motors without encoder feedback
(cS.01/ actual value source = "2: calculated actual value") and
- at v/f characteristic control (cS.00/ control mode < 4)
During the DC-braking the motor is not decelerated over the ramp. The braking is done with a DC voltage and
a DC current, respectively, that is applied to the motor winding.
After activation of the DC braking, the modulation is switched off and the base-block time (base-block time,
duration dependent on the power circuit) waited for until the DC value is applied to the motor.
With Pn.28 it is adjusted whereby the DC brake is triggered. According to the adjusted mode, the speed when
the DC brake triggers can be preset with Pn.32.
Pn.30 „DC braking time“ determines the braking time (0..100,00 s).
Pn.29 is bit-coded and defines the inputs which trigger DC braking.
1
The braking time is dependent on the actual frequency (ru.03), not from the actual speed (ru.07). The
reference value for the calculation of the braking time, however, is a speed (dependent on ud.02 "control
type", for "4: F5-M / 4000rpm“ is the reference value 1000 rpm). To calculate the braking time, the actual
frequency (ru.03) must be converted to a speed according to the following formula:
ru.03 * 60
———————————
pole-pair number of the
motor
2
Actual braking time = Pn.30 * ru.03 * 60 / pole-pair number of the motor / reference value (The reference
value is dependent on ud.02 "control type". The reverence value in 4000 rpm mode is 1000 rpm and 2000
rpm in 8000 rpm mode etc.)
3
Actual braking time = Pn.30 * Pn.32 / reference value (The reference value is dependent on ud.02 "con-
trol type". The reverence value in 4000 rpm mode is 1000 rpm and 2000 rpm in 8000 rpm mode etc.)
4
These adjustment are operative only if in Bit 0...3 "DC-brake mode" the value "10: conditions" is chosen.
If the same condition is also set for speed search, DC-brake has priority.
In v/f-characteristic control, a DC voltage is applied to the motor. The max. braking voltage is set with Pn.31
"DC braking max voltage".
The current is limited only by the inverter. If the inverter is oversized compared to the motor, the maximum bra-
king voltage (Pn.31) must be decreased to avoid overheating of the motor.
At large ratings the maximum braking voltage can lead to overcurrent errors (E.OC). In that case reduce it with
Pn31.
After completion of the DC braking function, the rated flux of the machine must flow before the motor is started.
To that end, "Wait for magnetisation = 128: on" (Bit 7 = 1) must be programmed in parameter dS.04. The tor-
que display is not valid in the DC braking (display always = 0 Nm).
The energy saving function allows the lowering or raising of the current output voltage. In accordance with the
activation conditions defined in uF.6, the voltage valid according to the V/Hz-characteristic is changed in per-
cent onto the energy saving level (uF.07).
However, the maximal output voltage cannot be higher than the input voltage even if the value is > 100 %. The
function is used for example in cyclic executed load/no-load applications. During the no-load phase the speed
is maintained, but energy is saved as a result of the voltage reduction.
0 Default time
16 Default time / 2
32 Default time / 4
48 Default time / 8
64 Default time / 16
factual ; nactual; UA
Hysteresis
fset ; nset
uF.7
This function simulates a mechanic motor potentiometer. Over two inputs the motor potentiometer value can
be increased or decreased.
100%
oP.54
If the motorpoti function
is used for setpoint set- oP.55
ting, 100% correspond to
the maximum setpoint. oP.59 oP.59
0% t
The target parameter de- oP.59 oP.59
termines the 100% value
at analog setting of the
parameter values. oP.53
7
-100%
oP.56 Increase motor poti va-
lue input selection
First two inputs must be defined which increases or decreases the motorpoti value. To that end, one input is as-
signed to parameters oP.56 and oP.57 according to the input table. If both inputs are triggered simultaneously,
the potentiometer value is decreased.
Input table
The fundamental function of the motorpoti is defined with oP.50. The parameter is bit-coded.
This parameter defines the time for the motorpoti in order to run from 0...100%. The time is adjustable between
0...50000 s.
The control range is limited by parameters oP.53 "motorpoti min. value" and oP.54 "motorpoti max. value" (see
picture 7.15.3).
This parameter shows the current value of the motor potentiometer in percent.
A percentage value can be adjusted directly by operator or bus via this parameter. The ramp time remains
unconsidered at this setting.
The parameter value is limited by oP.53 / oP.54. If a digital input is set for increasing or decreasing the motorpoti
value, the value of oP.52 changes.
Two timers are incorporated in the COMBIVERT. As long as one of the adjustable starting conditions (LE.18/23)
or a programmable input (LE.17/22) is set, the timer counts until reaching the final value range. If one of the re-
set conditions (LE.20/25) is fulfilled or one programmable input (LE.19/24) is set, the timer jumps back to zero.
The clock source and the counting direction is adjusted with LE.21/26. It can be counted in seconds, hours or 7
by a special programmed input for that. The current timer content is displayed in ru.43/44. Switching condition
37/38 is set with reaching an adjustable switching level (LE.00...07). It can be used to set an output.
continues to count
no
Is a
Timer counts up to maximum
reset condition no yes Switching condition
LE.21 Timer 1 / value; down to 0, if a condition
I3 of LE.20 or LE.19 Timer value > Level 37 „Timer 1 > Level A“
Mode of LE.18 is fulfilled or an input
fulfilled ? LE.0...LE.7 ? is being set!
of LE.17 is set
ru.43 timer 1 display
yes
Resetting to zero
continues to count
no
Is a
Timer counts up to maximum
Is a reset condition no yes Switching condition
LE.26 Timer 2 / value; down to 0, if a condition Timer value > Level
I4 of LE.25 or LE.24 38 „Timer 2 > Level A“
Mode of LE.23 is fulfilled or an input
fulfilled ? LE.0...LE.7 ? is being set!
of LE.22 is set
ru.44 timer 2 display
yes
Resetting to zero
LE.21 and LE.26 determine the clock source and the counting direction of timer 1 and 2. Clock pulse source can
be the time counter in 0.01s or 0.01h grid, pulses from a digital input, or revolutions of the encoder on encoder
channel 1. The timer runs generally as long as a starting condition is active. After a reset the timer starts again
at zero. Following clock sources can be selected:
The timer start conditions can be selected from the following table. The individual conditions are OR-operated
with the timer start input selection (LE.17/LE.22).
Additionally the timer can be activated by one or several inputs. The sum of the valences is to be entered, if
the timer shall be started by different inputs The individual inputs are OR-operated. The start input selection is
OR-operated with the timer / starting condition (LE.18/LE.22).
According to the following table one can define the conditions to reset the timer additionally to the inputs. The
individual conditions are OR-operated.
ru.43 / ru.44 display the actual counter reading depending on the selected clock source (LE.21 / 26). The coun-
ter can be set to a value by writing on ru.43 / 44. If the clock source is changed at run time, the counter reading
remains but is interpreted according to the new clock source.
LE.00...LE.07 define the level for the switching conditions 37 / 38 ("Timer > Level").The switching condition is
set if the timer exceeds the adjusted value. A level in the range of -10.737.418,24 to 10.737.418,23 can be
adjusted.But only values of 0...655,35 are sensible for the counter.
For applications in the areas lifting and lowering, or other applications requiring the use of a brake, the control
of the brake can be taken over by the KEB frequency inverter.
For this the brake control must be activated in parameter Pn.34 "brake control mode" and a transistor or relay
output must be assigned with the function "18: brake control". The output becomes active if the brake shall
be released.
The status display during brake handling can also be defined with Pn.34 and a monitoring function can be
activated.
The brake control is set-programmable.
Further brake control can be activated with Pn.43 "min. load brake control".
For the monitoring of the utilization acceptance through the inverter a minimal utilization level can be adjusted
in this parameter.
If the brake is to be opened on start at the end of the pre-magnetising time (Pn.35), the load factor may not be
smaller than the adjusted level. Otherwise error E. br is triggered. Reaching the hardware current limit during
this phase also triggers error E.br. The current is monitored only at this time (directly before the opening of the
brake).
The monitoring is deactivated when Pn.43 is set to 0.
The brake must always be closed between the end of the brake closing time (Pn.40) and the beginning of the
break release time (Pn.36). E.br is triggered if the input becomes (or is) active during this phase.
Similarly, from the end of the brake opening period (Pn.36) to the end of the brake delay time (Pn.39), the brake
must always remain ventilated. If the input becomes (or is) inactive in this phase, E.br is also triggered.
With this input, e.g., a protection monitoring could be executed.
The sequence of the brake control is defined by five times, two for the opening and three for the closing of the
brake.
open brake
The opening of the brake is started when the control release is closed and the command to start the drive is
received.
In vector controlled operation, this is the activation of the direction of rotation, the setpoint speed has no effect.
That means: the brake is opened as well on setting the speed setpoint value = 0.
During positioning, the opening of the brake is triggered, e.g., by a "Start positioning" or a "Start approach to
reference point" command
close brake
The closing of the brake is triggered by disabling the direction of rotation (speed control), reaching of the target
position (positioning), or switching off the modulation (opening of the control release or error).
If the modulation is switched off, the break control output is immediately deactivated so that the brake closes.
In all other cases, the sequence is as follows:
The following figure shows the sequence of the brake control without fadeout time. In vector controlled system,
the start- and the stop-value (Pn.37 / Pn.41) must be set to 0 rpm.
Reference
(ru.01)
Actual speed
Stop value
(Pn.41)
Start value
(Pn.37)
7
Set direction
of rotation for
opened
brake
closed
Since, independent of the selected deceleration ramp, the brake closing time only starts when the actual speed
reaches the value of the stop speed (= 0 rpm), no brake delay time has to be awaited.
In some applications, however, the brake delay time is used to save time.
If the output for brake control has been deactivated once, the complete brake handling (brake closing time +
fadeout time + brake opening time) must be executed for a new start of the drive.
By setting a brake delay time, the collapse of the brake can be suppressed for a quick succession of starts (e.g.,
for positioning). Only when the drive remains stationary for a longer period, the brake is closed.
Example: Let a lift be equipped with a counterweight so that for a half-loaded cabin, no holding torque
must be expended.
For an empty cabin, the load weighing setup provides a signal of 0%.
Since the settings range of the drive is limited during operation without encoder, no brake handling should be
used for SCL as well as for ASCL.
For SCL, parameters "premagnetizing time" (Pn.35) and "brake release time" (Pn.36) are used for the align-
ment of the motor with DC current (see chapter 7.6.3.4 SCL / standstill and starting phase)
Start- and, respectively, (synchronous speed– rated speed) x required holding torque
——————————————————————————————
stop value = Rated torque
Based on these value, an adaption to the particular application must be made since other values, e.g., the
boost, also have an effect on the behaviour during load transfer.
Example: a 4-pole motor has a rated frequency of 50 Hz and a rated speed of 1460 rpm.
The synchronous speed of the motor thereby is = 1500 rpm and the slip speed is 1500 – 1460 = 60 rpm at rated
torque and nominal voltage.
The drive should be able to provide rated torque at brake release if a start value (Pn.37) of 60 rpm is preset.
So that a torque can be built up, the flux in the motor must have been built up. The motor is energized with
starting the premagnetizing time. This time must be long enough for the motor to build up its flux.
Depending on the motor, this time can be between approximately 100 ms (small power) and fractions of a se-
cond (motors with large power).
In v/f-characteristic controlled operation, the speed follows the predefined deceleration ramp not quite exactly.
After completion of the deceleration ramp a delay time must be await to mask dynamic effects.
The sweep generator enables in period and amplitude changeable sawtooth process of the setpoint. It is acti-
vated with the parameter oP.44 Bit 0...3 = „1“.
1
oP.44 Reference
Bit 4...7 2
AN1 (ru.28) 0 oP.46
1 Amplitude
AN2 (ru.30)
AN3 (ru.32) 2 Diameter signal
ru.02
digital (oP.45) 3
oP.49
Diameter correction
Parameter oP.44 determines which source presets the amplitude height of the sweep function.
The sweep amplitude can be preset also via oP.45 „ext. function digital source“ in the range of 0… 100%, be-
side to the analog setting via AN 1, AN 2, AN 3 or AUX.
The maximum rate of change of the sweep amplitude is limited with parameter oP.46 „ext. function acc/dec.
time“.
Parameter oP.46 defines a time between 0.00...20.00 s, inside the sweep amplitude can increase or decrease.
The specified value refers to a sweep amplitude change of 100 %.
The acceleration time is parameterized with oP.47 „sweep-gen. acceleration time“, the deceleration time of the
sweep signal with oP.48 „sweep-gen. deceleration time“.
Both times are adjustable within the range of 0… 20.00 s.
Together the two parameters result in the period duration of the wobbel period.
7
Operating principle of the wobble function
The following picture shows the setpoint process, which is generated by the wobble function:
Ampl.
nset +----------
2
nset
Ampl.
nset -----------
2
oP.47 oP.48
Through the use of the diameter correction the tool path feedrate of a winding product can be kept constant at
changing diameter of the reel bale.
1
oP.44 Reference
Bit 4...7 2
AN1 (ru.28) 0 oP.46
1 Amplitude
AN2 (ru.30)
AN3 (ru.32) 2 Diameter signal
ru.02
digital (oP.45) 3
oP.49
Diameter correction
The diameter signal is evaluated within the range of 0 % to 100%. Values < 0% are set to 0%, values > 100%
are limited to 100%.
The diameter signal of 0% corresponds to the minimum diameter of the reel bale (dmin).
The output speed of the ramp generator (ru.02) is not changed in this case.
A diameter signal of 100% corresponds to the maximum diameter of the reel bale (dmax).
The ratio of minimum to maximum diameter (dmin/dmax) must be known in order to be able to calculate the
required speed change.
The ratio of minimum to maximum diameter (dmin/dmax) is preset by way of oP.49 and can be adjusted within
the range of 0,010...0,990 with a resolution of 0,001.
fn_Ramp
fn_Setting: = –––––––––––––––
1+DS • (1/oP.49-1)
The rate of change of the diameter signal can be limited by a ramp generator.
Parameter oP.46 „ext. funct. acc/dec time“ defines a time within the range of 0.00...20.00 s, which is required
for a change of the diameter signal of 0...100%.
With this function it is possible to preset parameter values analog. The AUX-function or the motor-poti function
can be adjusted as source.
This parameter determines whether the analog parameter setting occurs via the motor-poti or the aux-function.
An.53: Analog parameter setting source
Value Function
0 AUX
1 Motorpoti function
The bus address of the parameter, that is to be adjusted in analog mode, is adjusted here (see chapter 11.1).
Following parameters can be adjusted.
uF.01 / 07
cn. 04 / 05 / 06
An.32 / 37 / 42 / 48
LE.00 / 01 / 02 / 03 / 04 / 05 / 06 / 07
cS.06 / 09 / 19 / 20 / 21 / 22 / 23
Ec.04 / 14
PS.31 / 33
In case an invalid parameter address is selected, message "IdAtA" (or "data invalid" via COMBIVIS) is output
and the setting is ignored.
Defines the parameter value, that adjusts itself at 0 % analog parameter setting. The parameter value must be
entered with the internal standardization of the target parameter.
Defines the parameter value, that adjusts itself at 100 % analog parameter setting. The parameter value must
be entered with the internal standardization of the target parameter.
An.57 determines the parameter set which edited the selected parameter. If a programmable parameter is ad-
justed as target parameter, the adjusted set in An.57 is edited.
It is possible in synchronous operation that master and slave are additionally synchronized to two reference
signals. This reference signals are e.g. in form of approximation initiators at the master and slave axis. The gear
factor is adapted by the register function in order to adjust the time per cycle. Dependent on parameter „pos/syn
mode “ps.00 bit11 it can be adjusted if the gear change is executed in consideration of the ramp times (op.28…
33). An adjustable angle deviation can be used to synchronize the two reference signals.
Rotation direction
Initiator Initiator
I1
I2 Angle deviation > 0
F5 Encoder emulation F5
Master Slave
i = 9,8 M M i = 23
Encoder Encoder
2500 Inc/ rev. 2500 Inc/ rev.
Both initiator signals must release two times after activation of the register function before the register function
triggers an activity.
register distance master (rG.14), number of increments from main pulse to main pulse
register distance slave (rG.15), number of increments from main pulse to main pulse
numerator, gear factor (channel1(ec.56) or channel2 (ec.58))
denominator, gear factor (channel1(ec.57) or channel2 (ec.59))
numerator = i * denominator
If it is switched in a parameter set with rG.00 = Bit 0....1=2, the first calculated angle deviation between master
and slave is stored in parameter „register angle level 1“ (ps.64) (teachen). The max. angle correction per pe-
riod (rG.02) must be adjusted in order to activate the angle deviation.
The direction of the angle deviation can be adjusted via rG.00 bit 2… 3.
Two interpolation points where the angle deviation via interpolation can be preset dependent on the master
speed can be defined via parameters rG.08, rG.09 and rG.10, rG.11.The angle deviation is constantly defined
via rG.08 with rG.10 = 0 = off. The function of the angle setting is not active at speed left to rG.09 and right to
rG.11.
0° / 360° / -360° The angle deviation can be preset optionally via rG.08…
0 / rG.15 / -rG.15 rG.11, but it is internal limited to +-slave difference (rG.15).
Additionally a renormalization takes place, thus the an-
gle deviation setpoint ranges between –rG.15/2
…0….+rG.15/2.
-90° 90°
-rG.15/ 4 rG.15/ 4
+ / 180°
+ / -rG.15/ 2
Reg. master / slave diffe- -2^31…2^31-1 Difference between the main pulses of master
rG.16
rence inc. and slave
0…100.000 ms
rG.17 Reg. Register time master
(0.125 res.)
0…100.000 ms
rG.18 Reg. Register time slave
(0.125 res.)
Reg. master / slave 0…100.000 ms Time difference between the period of master
rG.19
Period difference (0.125 res.) and slave
Angle deviation setpoint [inc] Angle deviation setpoint [inc]
rG.10
rG.15
Reference 7
rG.15/ 2 rG.15/ 2
rG.08 n_master
rG.09 rG.11 rG.09 rG.11 rG.10 [rpm]
-rG.15/ 2
Reference
-rG.15
rG.08
off off off off
The KEB COMBIVERT is equipped with an universal programmable technology controller, which is able to
create pressure control, temperature or dancer position control.
The technology controller consists of a setpoint/actual value comparator, which transmits the system deviation
to the PID controller. The P- I- and D-component is adjusted with cn.04, 05 and 06. Parameters cn.07 and cn.08
limit the max. manipulated variable of the controller.
The controller ratio is defined from 0… 100% with the PID fading time (cn.09). Parameter cn.14 adjusts the fre-
quency ratio in Hz/% (only F5-G/B). The PID controller, the I-component separately and/or the controller fading
can be reset with parameters cn.11, 12 and 13. A PID reset condition can be adjusted with cn.10.
PID controller
Digital inputs
The maximum positive manipulated variable in the range of -400.0...400.0 % is specified with cn.07, the maxi-
mum negative manipulated variable in the range of -400.0...400.0 % with cn.08.
Thereby the controller intervention can be linerarly increased at the start or linearly decreased at the reset of
fading. The time refers to 100% controller output value. If an input is programmed to „fade in reset (cn.13)“, the
fade in is counted down when the input is active and counted up when the input is inactive.
The function is only active if the technology controller is used as process controller (cS.00 Bit 0...2 = 1). The
fading time is 0 if it is used as setpoint controller. 7
PID reset condition (cn.10)
The reset condition of the PID controller can be preset via cn.10. Simple speed controls for both directions of
rotation can be realized thereby.
Value „2“ must be adjusted for speed control in order that the I-component of the controller is reset at LS or nOP.
Value „1“serves primary for the start-up, in order to reset the controller manually.
The total controller, the I-component, as well as the controller fading can be reset via the digital inputs. The
fade-in time is valid at reset of fading. For that the decimal value of the corresponding inputs must be entered
This block converts the proportional controller output value to frequency. The adjustment of cn.14 determines,
which frequency is output at 100% controller output value. A frequency of -400.0...400.0 Hz can be adjusted
(dependent on ud.02). The output value forms the output frequency (ru.03) at cS.00 Bit 0… 1 = 1 added with
the ramp output frequency (ru.02).
This block describes the PID controller setpoint. The PID setpoint consists of the absolute reference (cn.01)
and an additional reference source adjustable with cn.00. The two values are added and form the PID controller
setpoint.
off 0
The setpoint of the PID controller is preset in percentage with cn.01 within the range of -400.0…400.0%. The
parameter is set-programmable.
Parameter cn.00 determines which input supplies the additional setpoint. The following options are available:
If one of the analog channels is adjusted, the signals can be adapted with the analog amplifiers individually to
the requirements (see chapter 6.2).
This block describes the PID controller actual value. The actual value is selected with the PID actual value
source (cn.02).
off 0
The PID actual value source (cn.02) defines wherefrom the PID controller receives the actual value signal.
Following signals are available:
cn.02: PID actual value source
Va- Signal Explanation
lue
0 AN1 Signal of the analog input 1 (see chapter 7.2)
1 AN2 Signal of the analog input 2 (see chapter 7.2)
2 AN3 Signal of the analog input 3 (see chapter 7.2)
3 AUX Signal of the Aux input (see chapter 7.2 )
PID absolute actual value is preset with cn.03 within the range of
4 cn. 03
-400.0...400.0 %
The displayed active current -200...200 % in parameter ru.17 is used as actual
5 Active current
value signal (100 % = Irated)
The displayed utilization 0...255 % in parameter ru.13 is used as actual value
6 Utilization
signal (100 % = 100 %)
DC link vol- The displayed DC link voltage 0...1000 V (1000 V = 100 %) in parameter ru.18 is
7
tage used as actual value signal.
active power
8 Reference value: 2 * Default dr.03 100%
(ru.81)
Some application examples of the PID controller are given in the following part.
PID controller without precontrol (e.g. for pressure-, temperature-, liquid level closed-loop control)
Digital reference
cn. 01
Ramp generator
Characteristic amplifier PID controller oP-Parameter
Analog reference
ru.27 ru.28 cn. 07 ru.01
AN1 ru.02
cn. 08 ru.52
cn.00 = 1 Ramp output
value
cn.04...06 oP.00 = 6
cn.10
An-Parameter
Actual value
ru.29 ru.30
AN2
cn.00 = 2
Characteristic amplifier
An-Parameter
ru.02
Ramp output
value
cn.04...06
An-Parameter cn.10 An.31 = 9/10 ru.29 ru.30
ANOUT 2
Actual value
Characteristic amplifier
ru.29 ru.30
AN2
cn.02 = 1
Characteristic amplifier
cn. 01
Analog reference
+ cn. 07 ru.02
ru.27 ru.28 ru.01
AN1
- cn. 08 ru.52
cn.00 = 1
cn.04...06 oP.00 = 6
An-Parameter cn.10
Active current-
actual value
ru.17
cn.02 = 5
An-Parameter
Digital current
reference
cn. 01 7
PID controller
Analog current cn.14
reference +
ru.29 ru.30 + cn. 07 ru.52 %
AN1 +
- cn. 08 Hz cs.04
cn.00 = 2
Active current- cn.04...06
actual value cn.10
ru.17 Ramp output
cn.02 = 5 value
2. Summary
7.3 Digital in- and outputs
3. Hardware
7.4 Setpoint-, rotation- and ramp presetting
6. Initial Start-up
7.8 Torque display and -limiting
8. Error Assistance
7.11 Speed measurement
9. Project Design
7.12 Posi- and synchronous operation
12. Annex
7.16 CP-Parameter definition
7.16.1 Overview
Read / Read
Application Mode Service Mode Write only
Drive Mode
ud.15 CP-Mode
Address
ud.16
1 Set / Norm ru.3 CP.1
ud.17
Address
ud.16
2
Set / Norm
ru. 1 CP.2
ud.17 7
Address
ud.16
36
Set / Norm
An.4 CP.36
ud.17
With ud.15 the CP-Parameter that is to be edited is determined. With ud.16 and ud.17 the CP-parameter is
defined through its address and the respective set. Depending on the adjusted password (CP.0 or ud.1)
Parameter CP.0 is not programmable, it always contains the password input. If the inverter is in the Application
Mode or Service Mode ud.1 is used for the password input.
Parameters which are not permitted as CP-Parameter (ud.15...17 and Fr.1) are acknowledged with „data in-
valid". When entering an invalid parameter address the parameter is set to „oFF“ (-1). The appropriate CP-
parameter is not displayed at this setting.
CP selector (ud.15)
The CP-Parameter to be programmed is set in the range of 1...36 with ud.15. CP.0 is not adjustable.
CP address (ud.16)
ud.16 determines the parameter address (see chapter 11) of the parameter to be displayed:
ud.17 determines the set, the addressing and the standardization of the parameter to be displayed. The para-
meter is bit-coded. The individual bits are decoded as follows:
Bit 0...7 determines the set selection for direct set programming, i.e. all selected sets contain the same value,
which is defined by the CP-parameter. If direct set programming (Bit 8, 9) is selected at least one set must be
selected as otherwise an error message is triggered in the cp mode.
Bit
7 6 5 4 3 2 1 0 Value Set
0 0 0 0 0 0 0 0 0 no -> Data invalid, if Bit 8 and 9 = 0
0 0 0 0 0 0 0 1 1 0
0 0 0 0 0 0 1 0 2 1
0 0 0 0 0 1 0 0 3 0+1
... ... ...
1 1 1 1 1 1 1 1 255 All
Bit
8 9 Value Function
0 0 0 direct set-addressing; the sets determined by Bit 0...7 are valid
0 1 256 current set; the current set is displayed / edited
512 indirect set addressing, the parameter set determined with the set pointer Fr.9
1 0
is displayed / edited
1 1 768 reserved
Display norm
Bit 10...12 determine how the defined parameter value is displayed. Up to seven different user norms (see
below in this chapter) can be determined with parameters ud.18...21.
Bit
12 11 10 Value Function
0 0 0 0 Use standard standardization of the parameter
0 0 1 1024 Display norm of parameters ud.18...21 from set 1
0 1 0 2048 Display norm of parameters ud.18...21 from set 2
... ...
1 1 1 7168 Display norm of parameters ud.18...21 from set 7
7.16.3 Example
The KEB COMBIVERT gives the user the possibility to define his own norms (e.g. km/h or bottles/min) in the
CP mode. Parameters ud.18...20 are used for conversion, ud.21 for specifying the calculation method, decimal
places as well as the units indicated in KEB COMBIVIS.
ud.19 numerator
± 32767
ud.19 numerator
± 32767
The unstandardized value or the standardized value/resolution is always used for the „selected parameter“ !
Adjusts the divisor in the range of ±32767 (default 1). The parameter is set-programmable.
Adjusts the multiplier in the range of ±32767 (default 1). The parameter is set-programmable.
Adjusts the offset in the range of ±32767 (default 0). The parameter is set-programmable.
The calculation mode, the decimal places as well as the units indicated in KEB COMBIVIS are set with ud.21.
The parameter is bit-coded and set-programmable. It is adjustable in the range of 0...1791.
Value Function
0 ud.19
= CP.xx
(selected parameter + ud.20) x ———
ud.18
64 ud.19
——————————————————— = CP.xx
(selected parameter + ud.20) x ud.18
- free
Value Representation
0 0 decimal places
256 1 decimal place
512 2 decimal places
768 3 decimal places
1024 4 decimal places
1280 variable decimal places
1536 Hexadecimal
- free
Example
The actual frequency shall be displayed in CP.1 in rpm. Display standardization from set 4.
ud.15 = 1 ; CP.1
ud.16 = 0203h ; Actual frequency ru.3
ud.17 = 4352 ; Display in the actual set, display norm from set 4
Set 4 ud.18 = 80 ; Conversion from 1/80 Hz into rpm without pole-pair number
Set 4 ud.19 = 60
Set 4 ud.20 = 0 ; ; no Offset
Set 4 ud.21 = 17 ; Unit rpm; direct calculation mode;no decimal place
Target of these parameters is to allocate a set of parameter addresses to the control. By this way arbitrary in-
verter parameters with self-specified standardizations are addressed.
Required parameters
The following configuration parameters must be available for one programmable parameter.
· Target address
· Characteristics
The configuration parameters are inserted in the Ud.group and indirect addressed like the configuration para-
meters of the CP parameters over a selector.
Example
Reading of the prog. parameters
The values of the source parameter in the selected sets are compared. If all values are equal then this value
is displayed, otherwise „data invalid“ is displayed. If no source parameter is defined, „data invalid“ is displayed.
The write value is written into all selected sets of the target parameter.
The following characteristics of the target parameter are checked:
Exceeding the limits: „invalid data“
Generally write protection: „write protected parameters“
Wwrite protection at switch on modulation: „operation not possible“
Write protection in the active set: „invalid set“
Password: ‘Password invalid‘ is only displayed at parameters with supervisor-password
„Data invalid“ is always displayed if no source parameter is defined.
Some parameters cannot be adjusted as target/source parameter in ud.23. This means all parameters, which
are not permissible as CP-Parameters (characteristics 2 bits 15 = 1) or process date (characteristics 1 bit 28 =
1), as well as the prog. parameters. Explanation:
The prog. parameters can be used as process data. Restrictions occur only if a prog. parameter is assigned
with a process date invalid parameter. In this case the process date is switched off and the adjusted address
in the corresponding sy parameter is negated, in order to mark this process date as switched off. This applies
also if the prog. parameter is switched off (ud.23 = -1).
A prog. parameter is additionally inadmissible as process writing date when the target parameter is read-only
(generally at activated modulation in active set).
The set definition of the process date is always valid as set source for process data (e.g. sy.17 for process read
date 1). The adjustment in ud.24 is without meaning.
The prog. parameters can be used as scope data. If the selected prog. parameter is switched off (ud.23 = -1)
the scope date is switched off and the adjusted address in the appropriate SY-parameter is negated in order to
mark this scope date switched off.
Since the prog.parameters have the type LONG they
1. Introduction
2. Summary
3. Hardware
4. Operation
Selection of Operating
5.
Mode
6. Initial Start-up
7. Functions
9. Project Design
10. Networks
12. Annex
8. Error Assistance
The following chapter shall help you to avoid errors as well as help you to determine and remove the cause
of errors on your own. The error messages of all KEB COMBIVERT F5 are represented, although depending
upon device and design some are missing.
8.1 Troubleshooting
8.1.1 General
If error messages or malfunctions occur repeatedly during operation, the first thing to do is to pinpoint the exact
error. To do that go through the following checklist:
- Does the error occur during a certain operating phase (e.g. always during acceleration)?
If so, consult the error messages and remove the causes listed there.
At KEB COMBIVERT error messages are always represented with an "E." and the appropriate error in the
display.Error messages cause the immediate deactivation of the modulation. Restart possible only after reset.
Malfunction are represented with an " A." and the appropriate message. Reactions to malfunctions can vary.
Status messages have no addition. The status message shows the current operating status of the inverter (e.g.
forward constant run, standstill etc.).
The display and their cause are described in the following: 8
1. Introduction
2. Summary
3. Hardware
4. Operation
Selection of Operating
5.
Mode
6. Initial Start-up
8. Error Assistance
9. Project Design 9
10. Networks
12. Annex
9. Project Design
The following chapter shall assist you in the planning stage of applications.
F4 F4 30
F5 F5
100
Warm air
outlet
KEB
COMBIVERT
9
Control cabinet surface
Calculation of control cabinet surface: Air flow rate with fan cooling:
PV 3.1 • PV
A= ———— [m2] V= ———— [m3/h]
∆T • K ∆T
The KEB COMBIVERT fitted with an external braking resistor or an external braking option is suitable for a
limited 4-quadrant operation. The braking energy, refeed into the DC-bus at generatoric operation, is dissipated
over the braking transistor to the braking resistor.
The braking resistor heats up during the braking process. If it is installed in a control cabinet sufficient cooling
of the control cabinet interior and sufficient distance to the KEB COMBIVERT must be observed.
Different braking resistors are available for the KEB COMBIVERT. Please refer to the next page for the corre-
sponding formula and restrictions (valid range).
3. If the desired braking time shall be smaller than the calculated time, it is necessary to use a braking resistor.
(tB < tBmin)
4. Calculate braking torque (MB). Take the load torque into account at the calculation.
5. Calculate peak braking power (PB). The peak braking power must always be calculated for the worst case
(nmax to standstill).
The braking resistors may be used only for the listed unit sizes. The maximum cyclic duration of a braking re-
sistor shall not be exceeded.
For a longer cyclic duration time special designed braking resistors are necessary. The continuous output of the
braking transistor must be taken into consideration.
7. Check, whether the desired braking time is attained with the braking resistor (tBmin).
Restriction:
Under consideration of the rating of the braking resistor and the brake power of the motor, the braking torque
may not exceed 1.5times of the rating torque of the motor (see formula).
When utilizing the maximum possible braking torque the frequency inverter must be dimensioned for the higher
current.
The braking time DEC is adjusted at the frequency inverter. If it is chosen too small the KEB COMBIVERT
switches off automatically and the error message OP or OC appears. The approximate braking time can be
determined according to following formula.
Formula
PR • 9.55
Condition: —————— < MN • (1.5 -K)
(n1 - n2)
f < 70 Hz
PB < PR
K= 0.25 for motors upto 1.5 kW JM = mass moment of inertia motor [kgm2]
0,20 for motors 2.2 4 kW JL = mass moment of inertia load [kgm2]
upto
0.15 for motors 5.5 11 kW n1 = motor speed prior to deceleration [rpm]
upto
9
0.08 for motors 15 45 kW n2 = motor speed after deceleration [rpm]
upto (standstill = 0 rpm)
0.05 for motors > 45 kW nN = rated motor speed [rpm]
MN = rated motor torque [Nm]
MB = Braking torque (necessary) [Nm]
ML = load torque [Nm]
tB = braking torque (necessary) [s]
tmin = minimum braking time [s]
tZ = cycle time [s]
PB = peak braking power [W]
PR = peak power of braking resistor [W]
Cyclic duration factor for cycle time tZ < 120 s Cyclic duration factor for cycle time tZ > 120 s
tB tB
ED = ——— • 100 % ED = ——— • 100 %
tZ 120 s
t
tB
tZ
By means of this section you can check whether you can still optimize your machine with regard to the material
usage. The specifications are derived for the DIN VDE 0298 Part 4. The values apply approximately and only
for the intended operation. In marginal cases it must be always proceed according to the standard described
obove.
The following table shows the current capability of 3 and/or 5 core PVC cables (i.e. 2 and/or 3 loaded cores)
in dependence with the ambient temperature. The current is to be laid out to the input current of the frequency
inverter.
The use of special cables or the way of laying the cables allows even higher currents (see DIN VDE 0298 Part
4). The motor cable must correspond to the cross-section of the mains cable.
If in case of long lines (>30m) still maximum torque is required at the motor shaft, the cable should be dimensi-
oned for the next larger cross-section in order to reduce line resistances.
Mains fuses are to be designed for the rated input current of the inverter. The current/time-characteristic of the
fuse must be slow-acting in order to avoid premature tripping when the power reserves of the inverter are used.
1. Introduction
2. Summary
3. Hardware
4. Operation
Selection of Operating
5.
Mode
7. Functions
8. Error Assistance
9. Project Design
10. Networks 10
12. Annex
10. Networks
The KEB COMBIVERT F5 can be easily integrated into different networks. For that purpose the inverter is
fitted with an operator that is appropriate for the respective bus system. Following hardware components are
available:
The cable of 3m length is used for the direct RS232 connection between PC (9-pole SUB-D connector) and
operator.
2 2 10
3 3
5 7
Housing (PE)
PC F5-Operator
The RS232 cable is suitable exclusively for the communication between PC and operator. If the cable is
plugged in directly onto the control board, it can lead to the desctruction of the interface of the PC.
The HSP5-cable is used for the direct connection between PC and control board. The necessary conversion to
TTL-level occurs in the cable.
HSP5
TTL
PC
PC F5-Control board
5 4 3 2 1
RS232/RS485
9 8 7 6
The PROFIBUS-DP interface module realizes a passive user (Slave). This means that the PROFIBUS-DP
interface module only transmits, if it receives an enquiry for that from the master.
The PROFIBUS-DP protocol defines different operating conditions, that must be executed first, before the ac-
tual user data can be exchanged. The responsible DP master must first parameterize and then configure his
slaves. If these two functions are successfully completed, the cyclic exchange of user data begins.
PDOUT F/R
STOP
SPEED Diag: Diagnostic interface to the
PDIN PC
PBS1: PROFIBUS-DP interface
(socket connector)
Diag PBS2: PROFIBUS-DP interface
(pin-connector)
ANTRIEBSTECHNIK
ANTRIEBSTECHNIK
PBS1 PBS2
10
The InterBus operator F5 is a slip-on operator with interbus 2-wire remote bus connection for KEB COM-
BIVERT F5. The voltage supply occurs via the inverter, for an independent supply it can also be fed in externally
over the control terminal strip of the inverter. Over the PCP channel 0, 1, 2 or 3 interbus register words can
be configured for the process data channel. Parallel to the field bus operation the operation via the integrated
display/keyboard as well as a further serial interface for diagnosis/parameterization (COMBIVIS) is possible.
BA (green)
On: Interbus runs
Blinking: Interbus was stopped by
the host
Off: Remote bus cable not available
or defective/host not in operation or
COM (green) defective
Lights when communication via Inter-
Bus PCP or diagnostic interface E (red)
START
On: Ready for operation
ENTER FUNC. Blinking: Inverter failure
RD (red)
F/R SPEED Off: No supply voltage
On: the remote bus interface (IB_out) STOP
was turned off by the host
RC (green)
On: Remote bus ready for operation
Diag
Off: Remote bus cable not available
Diagnostic interface to the PC
or defective/host not in operation or
The diagnostic interface is connected
defective
to the PC via an adaptor and a HS5P
cable. By way of the PC software COM-
BIVIS one has now normal access to all
inverter parameters. The internal ope-
rator parameters like interbus process
data length and occupation can be read
and adjusted or parameterized through
download. Alternatively a monitoring of
the InterBus PCP as well as the process
data channels can be carried out with
the PC-software HSP5-monitor.
IB_in IB_out
1 2 3 4 5 5 4 3 2 1
6 7 8 9 9 8 7 6
IB_in IB_out
Remote bus input Remote bus output
(D-Sub 9 connector) (D-Sub 9 socket)
CAN is a Multi-Master-System. This means every node has access to the BUS and can send telegrams. In
order to prevent problems when two nodes simultaneously access the BUS, the CAN-BUS has an arbitration
phase which determines who may continue to send his telegram. When there is a conflict in accessing BUS the
user with the lowest telegram number (identifier) has priority. This user then can completely send his telegram
without repeating the first part. All other nodes go into receiving status and stop sending their telegram. Thus it
is determined that lower telegram numbers have automatically priority. The available telegram numbers in the
CAN version 2.0A are limited to 2032 identifiers (0...2031).
PDOUT F/R
STOP
SPEED
CAN f: CAN interface (socket-con-
PDIN nector)
CAN m: CAN interface
(pin-connector)
Diag
ANTRIEBSTECHNIK
ANTRIEBSTECHNIK
CAN f CAN m 10
The herein described unit is a plugable operator with SERCOS interface for the frequency inverter or servo
KEB COMBIVERT F5. As far as possible the hard and software were developed taking the DIN/EN 61491 into
consideration. The voltage supply occurs via the inverter, for an independent supply it can also be fed in exter-
nally over the control terminal strip of the inverter. The SERCOS interface is designed as optical fibre ring for
plastic (POF) or fibre glas cable (HCS) with F-SMA plugs. The SERCOS service channel as well as cyclic data
transfer are available. Operation via integrated display/keyboard and an additional serial interface for diagno-
sis/parameterization (KEB COMBIVIS) is possible parallel to SERCOS operation (depending on the operation
mode it may be disabled). SERCOS operation parameters like slave address, transmitting power etc. can be
adjusted via the keyboard.
ANTRIEBSTECHNIK
S-IN S-OUT
In SY.06 can be adjusted, if the inverter shall be responded via "COMBIVIS" or another control. Values between
0 and 239 are possible, the default value is 1. If several inverters are operated on the bus simultaneously, it is
absolutely necessary to assign different addresses to them, since otherwise it leads to communication failu-
res, because several inverters may answer at the same time. The description of the DIN 66019II protocol (C0.
F5.01I-K001) contains further information to this. SY.06 is not reset on loading the default parameters.
Following values for the baud rate of the serial interface are possible:
If the value for the baud rate is changed via the serial interface, it can be changed again only via keyboard or
after adapting the baud rate of the master, because no communication is possible with different baud rates of
master and slave.
Should problems occur at the data transmission choose a transfer rate of maximal 38400 baud.
The transmission speed between operator and inverter is determined with the internal baud rate. The following
values are possible (unit-dependent):
For continuous control of the communication at the operator interface it is possible to trigger an error message
of the inverter, without incoming telegrams after expiration of an adjustable time (0.01… 10 s). The function can
be deactivated by setting the value "off".
This parameter determines the response to a watchdog error. The message E.buS or A.buS is output depen-
dent on the selected adjustment.
The HSP5 watchdog function monitors the communication of the HSP5 interface (control card - operator; or
control card - PC). The adjusted response in Pn.05 is released if no telegrams after expiration of an adjustable
time (0,01...10 s) are received. The value „off“ deactivates the function.
Generally the internal RAM of the microcontroller is used. Thereby there are no waiting periods or other diffe-
rences to other F5 units. The non-volatile storage is however not done immediately, but carefully to the life of
the EEPROM adapted algorithm.
The approx. 10 Kbyte parameter range is cyclically compared every 10 seconds with the EEPROM. The diffe-
rent memory cell is stored in the EEPROM at detected changes.
Although cyclic writing is automatically detected via the process data (e.g.) you should deactivate the automatic
storage in these cases.
If you want switch off the unit after download, ensure that any changes are stored in the EEPROM. This follow-
ing procedure is necessary.
Following the download switch ud.05 to off. Then read about 2 seconds status ud.04 until it is on stand by. Now
all changes are stored non-volatile.
If this procedure is not observed, it takes 6 minutes in worst case until the last
change is stored in the EEPROM. If the unit is switched off during this time, the
last changes are not stored.
With active input function, the status in ud.04 changes after about 3 seconds to "0: stand by" if all the parame-
ters are stored non-volatile in the EEPROM.
The control word is used for the status control of the inverter via bus. The actual state of the inverter can be
read out with the status word.
22 IC 64: IC
23 Id 128: Id
24 O1 256: O1 appropriate output is set via the control word or via the switching conditions.
25 O2 512: O2 Output signals O1, O2, R1 and R2 (visible in parameter ru.80) are OR ope-
rated with the appropriate bits of the control word. The connection occurs
26 R1 1024: R1
according di.42 „inverted outputs “(inverting level for the output signals) and
27 R2 2048: R2 before they are switched to the hardware outputs with do.51 „hardware output
allocation“.
28...31 reserved
The control word long (32 Bit) consists of SY.50 and SY.41.
The actual state of the inverter can be read out with the status word.
128: Actual value = Setpoint ru.07 „actual value display“ with a hysteresis of +/- LE.16 „freq./
7
value speed hysteresis“ has the same value as ru.01 „set value display“
8 256: Fast stop Fast stop is active
9 512: HSP5 bus-synchronous Inverter in bus-synchronous operation
1024: Approach to reference Approach to reference point was executed since the last power on
10
point completed (or the position was validate otherwise)
Display that the position profile is completed and the drive is within
11 2048: Position reached
the range of “+/- PS.30 „target window“ near ru.61 „target position“
0: Speed control Display of the operating mode selected via control word
4096: Synchronous running (only identical with the actual operating mode of the inverter, if in
12, 13
8192: positioning PS.00 „pos/syn mode“ in bits 0...2 value „7: via control word“ is
12288: Contouring control programmed)
It is necessary that the watchdogs are active at control via bus (operator and HSP5).
1. Introduction
2. Summary
3. Hardware
4. Operation
Selection of Operating
5.
Mode
7. Functions
8. error assistance
9. Project Design
10. Networks 10
12. Annex
Adjustment System para- Status display Inverter Parameter set Bus Interface User-defined
/ Assistant meter Information Programming and Keyboard Parameters
Parameter Configuration
AA-Parameter SY-Parameter ru-Parameter In-Parameter Fr-Parameter ud-Parameter CP - Parameter
Master
Synchronous mode drive
Posi mode Encoder
PS-Parameter interfaces PG
PS-Parameter Ec-
Motor model
Parameter
Protective and dr-Parameter
special functions nn-Parameter
ru-Parameter
Speed and PG
Pn-Parameter torque controller
drive-dep. control
Modula- M
Ramp generator cS/cn-Parameter dS-Parameter tor 3~
and s-curves
oP-Parameter
Setpoint calculation
U/f-characteristic
uF-Parameter
oP-Parameter
10
Terminal strip
Ec.08, 09
Ud.01, 05, 07
Legend
Parameter: Parameter group, number and name (ordered by parameter group and number)
Addr.: Parameter address in hex
R: Password level appl => application, ro => read only
P: p => set-programmable; np => not set-programmable
E: E => Enter-Parameter
Lower limit: Min. value (normalized); the non-normalized value results on division by the step range
Upper limit: Max. value (normalized); the non-normalized value results on division by the step range
Step: Step size, resolution
Default: Default value (normalized); the non-normalized value results on division by the step range
LTK => the default value is dependent on the power circuit identification
Unit: Unit
Reference: further information to this parameter on stated page (not chapter)
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
AA.16 Speed diff. filter 1210h appl np --- 0: off 1: on 0: off 1 ---
AA.59 modus isd_ref 123Bh appl np --- 0 2 0 1 ---
AA.60 PT1-Tau isd_ref 123Ch appl np --- 0 65535 1024 1 ---
AA.61 appc./act. torq. PT1-time 123Dh appl np --- 0 10 3 1 ---
AA.62 sel. int. data addr. 123Eh appl np --- 0 38 0 1 ---
AA.63 int. Data address 123Fh RO np --- 0 0FFFFh 0 1 hex
AA.64 Actual value PT1 time 1240h appl np --- 0 10 0 1 ---
An.00 AN1 interface selection 0A00h appl np E 0 2 0 1 --- 3.1-3, 7.1-9, 7.2-3, 7.2-4
An.01 AN1 noise filter 0A01h appl np E 0 4 0 1 --- 7.2-3, 7.2-5, 7.3-15, 7.4-5
7.2 - 3,7.2 - 5,7.2 - 6,7.3 - 17
An.02 AN1 save mode 0A02h appl np E 0 3 0 1 ---
An.03 AN1 save trig. inp. sel. 0A03h appl np E 0 4095 0 1 --- 7.2 - 3,7.2 - 6,7.3 - 11
An.04 AN1 zero point hysteresis 0A04h appl np --- -10.0 10.0 0.2 0.1 % 7.2-3, 7.2-7
An.05 AN1 gain 0A05h appl P --- -20.00 20.00 1.00 0.01 --- 7.2-3, 7.2-8, 7.4-5
An.06 AN1 offset X 0A06h appl P --- -100.0 100.0 0.0 0.1 % 7.2-3, 7.2-8, 7.2-9, 7.4-5
An.07 AN1 offset Y 0A07h appl P --- -100.0 100.0 0.0 0.1 % 7.2-3, 7.2-8, 7.2-9, 7.4-5
An.08 AN1 lower limit 0A08h appl P --- -400.0 400.0 -400.0 0.1 % 7.2-3, 7.2-9
An.09 AN1 upper limit 0A09h appl P --- -400.0 400.0 400.0 0.1 % 7.2-3, 7.2-9
3.1-3, 7.1-9, 7.1-10, 7.2-
An.10 AN2 interface selection 0A0Ah appl np E 0 2 0 1 ---
3,7.2-4
AN2 interference suppressi-
An.11 0A0Bh appl np E 0 4 0 1 --- 7.2-3, 7.8-13, 7.9-3
on filter
An.12 AN2 save mode 0A0Ch appl np E 0 3 0 1 --- 7.2-3, 7.2-5, 7.8-13, 7.9-3
An.13 AN2 save trig. inp. sel. 0A0Dh appl np E 0 4095 0 1 --- 7.2 - 3,7.2 - 6,7.3 - 10,7.3 - 11
An.14 AN2 zero point hysteresis 0A0Eh appl np --- -10.0 10.0 0.2 0.1 % 7.2-3, 7.2-7, 7.8-13
An.15 AN2 gain 0A0Fh appl P --- -20.00 20.00 1.00 0.01 --- 7.2-3, 7.2-8, 7.15-17
An.16 AN2 offset X 0A10h appl P --- -100.0 100.0 0.0 0.1 % 7.2-3, 7.2-8, 7.15-17
7.2-3, 7.2-8, 7.8-13, 7.9-3,
An.17 AN2 offset Y 0A11h appl P --- -100.0 100.0 0.0 0.1 %
7.15-17
An.18 AN2 lower limit 0A12h appl P --- -400.0 400.0 0.0 0.1 % 7.2-3, 7.2-9, 7.15-17
An.19 AN2 upper limit 0A13h appl P --- -400.0 400.0 400.0 0.1 % 7.2-3, 7.2-9, 7.15-17
An.20 AN3 interface selection 0A14h appl np E 0 1 0 1 --- 7.2-3, 7.2-5
An.21 AN3 noise filter 0A15h appl np E 0 4 0 1 --- 7.2-3, 7.2-5
An.22
An.23
AN3 save mode
AN3 save trig. inp. sel.
0A16h
0A17h
appl
appl
np
np
E
E
0
0
3
4095
0
0
1
1
---
---
7.2-3
7.2 - 3,7.2 - 6,7.3 - 10,7.3 - 11
10
An.24 AN3 zero point hysteresis 0A18h appl np --- -10.0 10.0 0.0 0.1 % 7.2-3, 7.2-7
An.25 AN3 amplification 0A19h appl P --- -20.00 20.00 1.00 0.01 --- 7.2-3, 7.2-8
An.26 AN3 offset X 0A1Ah appl P --- -100.0 100.0 0.0 0.1 % 7.2-3, 7.2-8
An.27 AN3 offset Y 0A1Bh appl P --- -100.0 100.0 0.0 0.1 % 7.2-3, 7.2-8
An.28
An.29
AN3 lower limit
AN3 upper limit
0A1Ch
0A1Dh
appl
appl
P
P
---
---
-400.0
-400.0
400.0
400.0
-400.0
400.0
0.1
0.1
%
%
7.2-3, 7.2-9
7.2-3, 7.2-9, 7.2-10
11
3.1-4, 7.1-12, 7.2-3, 7.2-
An.30 Sel. REF input./AUX function 0A1Eh appl P E 0 16383 2112 1 --- 10, 7.4-4, 7.8-13, 7.9-3,
7.15-30
3.1 - 3,7.2 - 11,7.2 - 13,7.12
An.31 ANOUT1 function 0A1Fh appl P E 0 29 2 1 --- - 70,7.15 - 31
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
An.33 ANOUT1 gain 0A21h appl P --- -20.00 20.00 1.00 0.01 --- 7.2-11, 7.2-14, 7.2-15
An.34 ANOUT1 offset X 0A22h appl P --- -100.0 100.0 0.0 0.1 % 7.2-14, 7.2-15
An.35 ANOUT1 offset Y 0A23h appl P --- -100.0 100.0 0.0 0.1 % 7.2-14
An.36 ANOUT2 function 0A24h appl P E 0 29 6 1 --- 7.2-13, 7.12-70
An.37 ANOUT2 value 0A25h appl P --- -100.0 100.0 0.0 0.1 % 7.2-15
An.38 ANOUT2 gain 0A26h appl P --- -20.00 20.00 1.00 0.01 --- 7.2-11, 7.2-14
An.39 ANOUT2 offset X 0A27h appl P --- -100.0 100.0 0.0 0.1 % 7.2-14
An.40 ANOUT2 offset Y 0A28h appl P --- -100.0 100.0 0.0 0.1 % 7.2-14
An.41 ANOUT3 function 0A29h appl np E 0 29 12 1 --- 7.2-11, 7.2-13
An.42 ANOUT3 value 0A2Ah appl np --- -100.0 100.0 0.0 0.1 % 7.2-15
An.43 ANOUT3 gain 0A2Bh appl np --- -20.00 20.00 1.00 0.01 --- 7.2-14
An.44 ANOUT3 offset X 0A2Ch appl np --- -100.0 100.0 0.0 0.1 % 7.2-14
An.45 ANOUT3 offset Y 0A2Dh appl np --- -100.0 100.0 0.0 0.1 % 7.2-14
An.46 ANOUT3 period 0A2Eh appl np E 1 240 1 1 s 7.2-11, 7.2-12, 7.3-19
An.47 ANOUT4 function 0A2Fh appl np E 0 29 12 1 --- 7.2-13
An.48 ANOUT4 value 0A30h appl np --- -100.0 100.0 0.0 0.1 % 7.2-15
An.49 ANOUT4 gain 0A31h appl np --- -20.00 20.00 1.00 0.01 --- 7.2-14
An.50 ANOUT4 offset X 0A32h appl np --- -100.0 100.0 0.0 0.1 % 7.2-14
An.51 ANOUT4 offset Y 0A33h appl np --- -100.0 100.0 0.0 0.1 % 7.2-14
An.52 ANOUT4 period 0A34h appl np E 1 240 1 1 s 7.2-12, 7.3-19
analog parameter setting
An.53 0A35h appl np E 0 5 0 1 --- 7.11-15, 7.15-22
max. value
an. parameter setting max.
An.54 0A36h appl np E -1: off 7FFFH -1: off 1 hex 7.11-15, 7.15-22
value
an. para setting offset para-
An.55 0A37h appl np --- -2^31 2^31-1 0 1 --- 7.11-15, 7.15-22
meter setting max. value
an. parameter setting max.
An.56 0A38h appl np --- -2^31 2^31-1 0 1 --- 7.11-16, 7.12-46, 7.12-70
value
An.57 an. para set. set pointer 0A39h appl np E -1: act set 7 0 1 --- 7.15-23
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
6.2-5, 7.5-9, 7.5-10, 7.5-
cs.06 KP speed 0F06h appl P --- 0 32767 300 1 --- 13, 7.7-3, 7.7-7,
7.7-9, 7.15-22
cS. 07 KP speed gain 0F07h appl P --- 0 32767 0 1 --- 7.7-7
cS. 08 KP speed limit 0F08h appl P --- 0 32767 0 1 --- 7.7-7
6.2 - 5,6.2 - 11,6.2 - 15,7.5 -
cs.09 KI speed 0F09h appl P --- 0 32767 100 1 --- 9,7.5 - 10,7.5 - 13,7.7 - 3,7.
7 - 7
cS. 10 KI offset 0F0Ah appl P --- 0 32767 0 1 --- 7.6-24, 7.7-7, 7.15-16
1; 5.1-4, 7.6-24, 7.7-7, 7.15-
cS. 11 max speed for max KI 0F0Bh appl P --- -1 ; -0,125 16000 ; 2000 10 ; 1.25 rpm
0.125 16
1;
cS. 12 min. speed for cS.09 0F0Ch appl P --- 0 16000 ; 2000 500 ; 62.5 rpm 5.1-4, 7.6-24, 7.7-7, 7.7-9
0.125
cS. 13 max. speed for quadr. Function 0F0D appl P --- 0 32000 32000 1 rpm
cs.15 torque reference source 0F0Fh appl P E 0 6 2 1 --- 7.8-13, 7.9-3
cS. 16 torque acc. time 0F10h appl P --- 0: off 60000 0: off 1 ms 7.9-3, 7.9-5
cs.18 torque ref. setting % 0F12h appl P --- -100.0 100.0 100.0 0.1 % 7.8-13, 7.9-3
7.1 - 14,7.5 - 12,7.6 - 5,7.8 -
cS.19 abs. torque ref. 0F13h appl P --- -32000.00 32000.00 LTK 0.01 Nm 13,7.8 - 14,7.8 - 15,7.9 - 3,7.
11 - 23,7.11 - 24,7.11 - 25
cS.20 torque limit for. mot. 0F14h appl P --- -0.01: off 32000.00 -0.01: off 0.01 Nm 7.5-12, 7.6-5, 7.8-13
cS. 21 torque limit rev. mot. 0F15h appl P --- -0.01: off 32000.00 -0.01: off 0.01 Nm 7.8-13
cs.22 torque limit for. gen. 0F16h appl P --- -0.01: off 32000.00 -0.01: off 0.01 Nm 7.8-13
cs.23 torque limit rev. gen. 0F17h appl P --- -0.01: off 32000.00 -0.01: off 0.01 Nm 7.5-12, 7.8-13, 7.9-3
cs.25 inertia (kg*cm^2) 0F19h appl P --- 0.00 10737418.23 0.00 0.01 --- 7.5-13, 7.7-3, 7.7-8, 7.7-9
cs.26 optimisation 0F1Ah appl P E 1.9: off 15.0 1.9: off 0.1 --- 7.5-13, 7.7-3
cs.27 pretorq. speed PT1-time 0F1Bh appl P --- 0 9 3 1 --- 7.7-9, 7.7-10
cS. 28 pretorq. speed fact. % 0F1Ch appl P --- 0.0 200.0 0.0 0.1 % 7.7-9, 7.7-10
cS. 29 act. curr. ref. PT1-time 0F1Dh appl P --- 0 9 0 1 --- 7.5-26, 7.6-24, 7.7-9
cS. 31 spline pret. PT1-time 0F1Fh appl np --- 0 65535 0 1 ms
cs.34 ref. torque isq table 0F22h appl np --- -32000 32000 0 1 --- 7.6-6
cs.35 ref. torque isd table 0F23h appl np --- -32000 32000 0 1 --- 7.6-6
di.00 PNP / NPN selection 0B00h appl np E 0: PNP SHR 0: PNP 1 --- 3.1-4, 7.3-4
7.3-3, 7.3-4, 7.3-5, 7.3-11,
di.01 select signal source 0B01h appl np E 0 4095 0 1 --- 7.3-12, 7.13-34, 10.1-10,
10.1-11
7.3-4, 7.3-5, 7.3-11,
di.02 digital input setting 0B02h appl np E 0 4095 0 1 ---
7.3-12, 10.1-10, 10.1-11
di.03 digital noise filter 0B03h appl np E 0 127 0 1 ms 7.3-6
di.04 input logic 0B04h appl np E 0 4095 0 1 --- 7.3-6
di.05 Input trigger 0B05h appl np E 0 4095 0 1 --- 7.3-7, 7.3-8
di.06 select strobe source 0B06h appl np E 0 4095 0 1 --- 7.3-8
di.07 strobe mode 0B07h appl np E 0 2 0 1 --- 7.3-8, 7.3-9
di.08 input strobe dependence 0B08h appl np E 0 4095 0 1 --- 7.3-8
di.09 reset input selection 0B09h appl np E 0 4095 3 1 --- 7.3-9, 7.3-10, 7.3-11
di.10 reset input slope sel. 0B0Ah appl np E 0 4095 3 1 --- 7.3-9
7.3 - 3,7.3 - 10,7.3 - 11,7.3 -
di.11 I1 functions 0B0Bh appl np E -2^31 2^31-1 1 1 hex 12,7.3 - 13,7.12 - 4,7.12 - 5,
7.12 - 63,7.12 - 74
7.3 - 10,7.3 - 11,7.3 - 12,7.3
di.22 ST functions 0B16h appl np E -2^31 2^31-1 128 1 hex
- 13,7.12 - 63,7.12 - 74
di.23 fast dig. noise filter 0B17h appl np E 0.00 31.75 0.00 0.25 ms 7.3-3, 7.3-6
di.24 I1 prog. function 0B18h appl np E 0 21 0 1 ---
di.25 I2 prog. function 0B19h appl np E 0 21 0 1 --- 10
di.26 I3 prog. function 0B1Ah appl np E 0 21 0 1 ---
di.27 I4 prog. function 0B1Bh appl np E 0 21 0 1 --- 7.3 - 10,7.3 - 12,7.12 - 35,7.
di 28 IA prog. function 0B1Ch appl np E 0 21 0 1 --- 12 - 39,7.12 - 71,7.12 - 73,7
di 29 IB prog. function 0B1Dh appl np E 0 21 0 1 --- .12 - 74
di 30 IC prog. function 0B1Eh appl np E 0 21 0 1 ---
di 31 ID prog. function 0B1Fh appl np E 0 21 0 1 --- 11
di.35 ST prog. function 0B23h appl np E 0 21 0 1 ---
7.3-10, 7.3-12, 7.3-13,
di.36 software ST input sel. 0B24h appl np E 0 4095 0 1 ---
7.13-39,7.13-40
di.37 ST lock input sel. 0B25h appl np E 0 4095 0 1 --- 7.3-10, 7.3-12, 7.3-13
di.38 turn off ST delay time 0B26h appl np --- 0 10.0 0 0.1 s 7.3-13
7.3-10, 7.3-12, 7.3-14,
di.39 disable dig. ST inp.sel. 0B27h appl np E 0 4095 0 1 ---
7.13-40
di.40 I1 activation delay 0B28h appl np – 0 32.00 0 0.01 s
di.41 I1 deactivation delay 0B29h appl np – 0 32.00 0 0.01 s
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
di.42 I2 activation delay 0B2Ah appl np – 0 32.00 0 0.01 s
di.43 I2 deactivation delay 0B2Bh appl np – 0 32.00 0 0.01 s
di.44 I3 activation delay 0B2Ch appl np – 0 32.00 0 0.01 s
di.45 I3 deactivation delay 0B2Dh appl np – 0 32.00 0 0.01 s
di.46 I4 activation delay 0B2Eh appl np – 0 32.00 0 0.01 s
di.47 I4 deactivation delay 0B2Fh appl np – 0 32.00 0 0.01 s
di.48 IA activation delay 0B30h appl np – 0 32.00 0 0.01 s
di.49 IA deactivation delay 0B31h appl np – 0 32.00 0 0.01 s
di.50 IB activation delay 0B32h appl np – 0 32.00 0 0.01 s
di.51 IB deactivation delay 0B33h appl np – 0 32.00 0 0.01 s
di.52 IC activation delay 0B34h appl np – 0 32.00 0 0.01 s
di.53 IC deactivation delay 0B35h appl np – 0 32.00 0 0.01 s
di.54 ID activation delay 0B36h appl np – 0 32.00 0 0.01 s
di.55 ID deactivation delay 0B37h appl np – 0 32.00 0 0.01 s
7.3 - 14,7.3 - 16,7.3 - 17,7.3
do.00 condition 0 0C00h appl P E 0 92 20 1 --- - 22,7.3 - 26,7.12 - 11,7.12 -
48,7.12 - 49,7.12 - 75
do.01 condition 1 0C01h appl P E 0 92 3 1 --- 7.3-23, 7.3-26, 7.12-49
do.02 condition 2 0C02h appl P E 0 92 4 1 --- 7.3-26, 7.12-49
do.03 condition 3 0C03h appl P E 0 92 2 1 --- 7.12-49
do.04 condition 4 0C04h appl P E 0 92 0 1 --- 7.5-26, 7.12-49
do.05 condition 5 0C05h appl P E 0 92 0 1 ---
do.06 condition 6 0C06h appl P E 0 92 0 1 ---
7.3-14, 7.3-16, 7.3-17, 7.3-
do.07 condition 7 0C07h appl P E 0 92 0 1 ---
22, 7.12-75
do.08 inv. cond. for flag 0 0C08h appl P E 0 255 0 1 --- 7.1-8, 7.3-16, 7.3-22
do.09 inv. cond. for flag 1 0C09h appl P E 0 255 0 1 --- 7.3-23
do.10 inv. cond. for flag 2 0C0Ah appl P E 0 255 0 1 --- 7.3-26
do.11 inv. cond. for flag 3 0C0Bh appl P E 0 255 0 1 ---
do.12 inv. cond. for flag 4 0C0Ch appl P E 0 255 0 1 ---
do.13 inv. cond. for flag 5 0C0Dh appl P E 0 255 0 1 ---
do.14 inv. cond. for flag 6 0C0Eh appl P E 0 255 0 1 ---
do.15 inv. cond. for flag 7 0C0Fh appl P E 0 255 0 1 --- 7.3-16, 7.3-22
7.3-16, 7.3-23, 7.3-26,
do.16 cond. select. for flag 0 0C10h appl P E 0 255 1 1 ---
7.12-48
do.17 cond. select. for flag 1 0C11h appl P E 0 255 2 1 --- 7.3-26, 7.12-49
do.18 cond. select. for flag 2 0C12h appl P E 0 255 4 1 --- 7.3-26, 7.12-49
do.19 cond. select. for flag 3 0C13h appl P E 0 255 8 1 --- 7.12-49
do.20 cond. select. for flag 4 0C14h appl P E 0 255 16 1 ---
do.21 cond. select. for flag 5 0C15h appl P E 0 255 32 1 ---
do.22 cond. select. for flag 6 0C16h appl P E 0 255 64 1 ---
do.23 cond. select. for flag 7 0C17h appl P E 0 255 128 1 --- 7.3-14, 7.3-20
7.3 - 16,7.3 - 23,7.3 - 26,7.1
do.24 AND/OR conn. for flags 0C18h appl P E 0 255 0 1 ---
2 - 49
do.25 inv. flags for O1 0C19h appl P E 0 255 0 1 --- 7.3-16, 7.3-24, 7.3-26
do.26 inv. flags for O2 0C1Ah appl P E 0 255 0 1 ---
do.27 inv. flags for R1 0C1Bh appl P E 0 255 0 1 --- 7.3-26
do.28 inv. flags for R2 0C1Ch appl P E 0 255 0 1 ---
do.29 inv. flags for OA 0C1Dh appl P E 0 255 0 1 ---
do.30 inv. flags for OB 0C1Eh appl P E 0 255 0 1 ---
do.31 inv. flags for OC 0C1Fh appl P E 0 255 0 1 --- 7.12-50
do.32 inv. flags for OD 0C20h appl P E 0 255 0 1 --- 7.3-24
7.3 - 16,7.3 - 24,7.3 - 26,7.1
do.33 flag select. for O1 0C21h appl P E 0 255 1 1 ---
2 - 48
do.34 flag select. for O2 0C22h appl P E 0 255 2 1 --- 7.3-26
do.35 flag select. for R1 0C23h appl P E 0 255 4 1 --- 7.3-26
do.36 flag select. for R2 0C24h appl P E 0 255 8 1 --- 7.3-4
do.37 flag select. for OA 0C25h appl P E 0 255 16 1 --- 7.12-49
do.38 flag select. for OB 0C26h appl P E 0 255 32 1 ---
do.39 flag select. for OC 0C27h appl P E 0 255 64 1 --- 7.12-50
do.40 flag select. for OD 0C28h appl P E 0 255 128 1 --- 7.3-24
7.3 - 4,7.3 - 16,7.3 - 24,7.3 -
do.41 AND conn. for outputs 0C29h appl P E 0 255 0 1 ---
26,7.12 - 49
do.42 inverted outputs 0C2Ah appl P E 0 255 0 1 --- 7.3-16, 7.3-24
do.43 cond. 0 filter time 0C2Bh appl P --- 0 1000 0 1 ms 7.3-16
do.44 cond. 1 filter time 0C2Ch appl P --- 0 1000 0 1 ms 7.3-16
7.1-14,7.3-16,7.3-25,7.3-
do.51 hardw. output allocation 0C33h appl P E 0 255 228 1 ---
26,10.1-11
6.2 - 5,6.2 - 7,6.2 - 10,6.2 - 1
dr.00 DASM rated current 0600h appl P --- 0.0 1100.0 LTK 0.1 A 4,7.5 - 8,7.5 - 12,7.5 - 16,7.5
- 23,7.11 - 24,7.15 - 5
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
1; 5.1 - 4,6.2 - 5,6.2 - 7,6.2 - 10,
dr.01 DASM rated speed 0601h appl P --- 0 64000 ; 8000 LTK rpm 6.2 - 14,7.5 - 8,7.5 - 12,7.5 -
0.125
25,7.11 - 24
6.2 - 5,6.2 - 7,6.2 - 10,6.2 - 1
dr.02 DASM rated voltage 0602h appl P --- 120 830 LTK 1 V 4,7.5 - 8,7.5 - 9,7.5 - 12,7.1
1 - 24
6.2 - 7,6.2 - 10,6.2 - 14,7.2 -
dr.03 DASM rated power 0603h appl P --- 0.10 1000.00 LTK 0.01 kW 13,7.5 - 8,7.5 - 12,7.7 - 3,7.
11 - 24
6.2 - 5,6.2 - 7,6.2 - 10,6.2 - 1
dr.04 DASM rated cos (phi) 0604h appl P --- 0.50 1.00 LTK 0.01 ---
4,7.5 - 8,7.5 - 16,7.11 - 24
6.2 - 5,6.2 - 7,6.2 - 10,6.2 - 1
dr.05 DASM rated frequency 0605h appl P --- 0.0 1600.0 LTK 0.1 Hz 4,7.5 - 8,7.5 - 9,7.5 - 12,7.1
1 - 24
6.2 - 5,6.2 - 7,6.2 - 10,6.2 - 1
dr.06 DASM stator resistance 0606h appl P E 0.000 250.000 LTK 0.001 Ohm 1,6.2 - 14,6.2 - 15,7.5 - 8,7.5
- 9,7.5 - 14,7.5 - 16,7.11 - 24
6.2-10, 6.2-14, 7.5-16, 7.5-
dr.07 DASM leakage inductance 0607h appl P --- 0.01 655.35 LTK 0.01 mH
19, 7.5-24, 7.5-27, 7.11-24
6.2-10, 6.2-14, 7.5-4, 7.5-
dr.08 DASM rotor resistance 0608h appl P --- 0.000 250.000 LTK 0.001 Ohm
16, 7.5-19
dr.09 breakdown factor 0609h appl P --- 0.5 4.0 2.5 0.1 --- 7.5-8, 7.5-9, 7.5-10
6.2-7, 6.2-10, 6.2-11, 6.2-
dr.10 DASM head-inductance 060Ah appl P --- 0.1 3276.7 LTK 0.1 mH
14, 6.2-15, 7.5-14, 7.5-16
dr.11 motorprotection mode 060Bh appl P --- 0 1 1 1 --- 7.13-27
dr.12 motorprot. rated current 060Ch appl P --- 0.0 1100.0 LTK 0.1 A 7.13-26, 7.13-27
dr.13 DASM magnetizing current 060Dh appl P --- 0.0 1100.0 0.0 0.1 A
dr.14 DASM rated torque 060Eh RO P --- 0.01 32000.00 0.01 Motdat 0.01 Nm 7.2-13, 7.5-26, 7.8-4
7.8-5, 7.8-6, 7.8-7,
dr.15 max. torque FU 060Fh RO P --- 0.01 32000.00 0.01 Motdat 0.01 Nm
7.8-15, 7.9-3, 7.13-29
7.5 - 12,7.5 - 25,7.5 - 26,7.8
dr.16 DASM Mmax at dr.18 0610h appl P --- 0.01 32000.00 0.01 Adpt 0.01 Nm
- 5,7.8 - 6,7.13 - 13,7.13 - 20
5.1-4, 6.2-11, 6.2-15, 7.5-
900 ; 112.5 1 ;
dr.17 DASM speed for max. torque 0611h appl P --- 1 ; 0.125 64000 ; 8000 rpm 13, 7.5-17, 7.5-18, 7.5-20,
Adpt 0.125
7.5-22
5.1 - 4,6.2 - 10,6.2 - 14,7.5
1; - 12,7.5 - 15,7.5 - 23,7.5 - 2
dr.18 DASM field weak speed 0612h appl P --- 0 64000 ; 8000 0 Adpt rpm
0.125 5,7.5 - 37,7.8 - 5,7.8 - 6,7.8
- 15
7.5 - 12,7.5 - 15,7.5 - 16,7.5
dr.19 flux adaption factor 0613h appl P --- 25 250 100 Adpt 1 %
- 19,7.5 - 23
dr.20 field weak. curve 0614h appl P --- 0.01 2.00 1.20 Adpt 0.01 --- 7.5-12, 7.5-15, 7.5-37
dr.21 no load voltage 0615h appl P --- 0.0 100.0 75.0 0.1 %
6.2 - 18,6.2 - 19,7.6 - 3,7.6 -
dr.23 DSM rated current 0617h appl np --- 0.0 1100.0 LTK 0.1 A
6,7.6 - 10,7.11 - 24,7.13 - 30
1;
dr.24 DSM rated speed 0618h appl np --- 0 64000 ; 8000 LTK rpm
0.125 5.1 - 4,6.2 - 18,6.2 - 19,7.6 -
1; 3,7.6 - 17,7.11 - 24,7.13 - 30
dr.24 DSM rated speed 0618h appl np --- 0 32000 ; 4000 LTK rpm
0.125
6.2-18, 6.2-19, 7.6-3, 7.11-
dr.25 DSM rated frequency 0619h appl np --- 0.0 1600.0 LTK 0.1 Hz
24
6.2-18, 6.2-19, 7.6-3, 7.6-
dr.26 DSM EMC [Vpk*1000RPM] 061Ah appl np --- 0 32000 LTK 1 ---
4, 7.6-11, 7.8-7, 7.11-24
10
5.1-4, 6.2-18, 6.2-19, 6.2-
20, 7.2-13, 7.6-3, 7.8-2,
dr.27 DSM rated torque 061Bh appl np --- 0.1 ; 1 6553.5 ; 65535 LTK 0.1 ; 1 Nm
7.8-7, 7.8-9, 7.8-10, 7.11-
24
6.2-18, 6.2-19, 7.6-3, 7.6-
11
dr.28 DSM curr. f. zero speed 061Ch appl np --- 0.0 1090.0 LTK 0.1 A
4, 7.11-24, 7.13-30
DSM motor winding resi- 6.2-18, 6.2-19, 7.6-3, 7.6-
dr.30 061Eh appl np --- 0.000 250.000 LTK 0.001 Ohm
stance 11, 7.11-24
DSM motor winding in- 6.2-18, 6.2-19, 7.6-3, 7.6-
dr.31 061Fh appl np --- 0.01 500.00 LTK 0.01 mH
ductance 10, 7.11-24
7.2-13, 7.8-9, 7.8-10, 7.11-
dr.32 DSM rated power 0620h RO np --- 0.01 1000.00 LTK 0.01 kW
24
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
5.1-4, 6.2-18,6.2-20, 7.6-5,
dr.33 DSM max. torque 0621h appl np --- 0.1 ; 1 6553.5 ; 65535 LTK 0.1 ; 1 Nm 7.6-22, 7.8-9, 7.8-10, 7.11-
24, 7.13-29
dr.34 mot. prot. time min. Is/Id 0622h appl np --- 0.1 25.5 8.0 0.1 s
7.13-29
dr.34 mot. prot. time min. Is/Id 0622h appl np --- 0.1 10.0 0.5 0.1 s
dr.35 mot. prot. time Imax 0623h appl np --- 0.1 10.0 0.2 0.1 s 7.13-29
dr.36 mot. prot. recovery time 0624h appl np --- 0.1 300.0 5.0 0.1 s 7.13-30
7.5-27, 7.8-12, 7.8-14,
dr.37 max. current 0625h appl np --- 0.0 1100.0 LTK 0.1 A
7.10-4, 7.15-5
32000 ; 1;
dr.39 DSM corner speed 1 0627h appl np --- 0 64000 ; 8000 rpm 5.1-4, 7.6-25, 7.8-9, 7.8-10
4000 0.125
dr.40 DSM corn. max. torque 2 0628h appl np --- 0.1 ; 1 6553.5 ; 65535 0.1 ; 1 0.1 ; 1 Nm 5.1-4, 7.6-25, 7.8-10
32000 ; 1;
dr.41 DSM corner speed 1 2 0629h appl np --- 0 64000 ; 8000 rpm 5.1-4, 7.6-25, 7.8-10
4000 0.125
dr.42 DSM corner max. torque 3 062Ah appl np --- 0.1 ; 1 6553.5 ; 65535 0.1 ; 1 0.1 ; 1 Nm 5.1-4, 7.6-22, 7.8-10
32000 ; 1;
dr.43 DSM corner speed 1 3 062Bh appl np --- 0 64000 ; 8000 rpm 5.1-4, 7.6-22, 7.8-10
4000 0.125
dr.44 DSM corn. max. torque 4 062Ch appl np --- 0.1 ; 1 6553.5 ; 65535 0.1 ; 1 0.1 ; 1 Nm 5.1-4, 7.6-22, 7.8-10
32000 ; 1;
dr.45 DSM corner speed 1 4 062Dh appl np --- 0 64000 ; 8000 rpm 5.1-4, 7.6-22, 7.8-10
4000 0.125
dr.46 DSM corner max. torque 5 062Eh appl np --- 0.1 ; 1 6553.5 ; 65535 0.1 ; 1 0.1 ; 1 Nm 5.1-4, 7.6-22, 7.8-10
32000 ; 1;
dr.47 DSM corner speed 1 5 062Fh appl np --- 0 64000 ; 8000 rpm 5.1-4, 7.8-10
4000 0.125
6.2 - 11,6.2 - 15,6.2 - 18,6.2
- 19,7.5 - 16,7.5 - 17,7.5 - 18
dr.48 motor identification 0630h appl np E 0 255 0 1 --- ,7.5 - 19,7.5 - 20,7.5 - 21,7.5
- 22,7.5 - 23,7.6 - 8,7.6 - 9,7.
6 - 10,7.6 - 11,7.6 - 12
5.1-5, 6.2-11, 6.2-15, 7.5-
dr.49 Lh ident. acc/dec time 0631h appl np --- 0.00 300.00 5.00 0.01 s 17, 7.5-20, 7.5-21, 7.6-11,
7.6-12
dr.50 mot. prot. min. Is/Id 0632h appl np --- 100 500 150 1 % 7.13-29, 7.13-30
dr.51 motortemp.for Rs corr. 0633h appl np --- 0 200 20 1 degree 7.3-21
dr.52 temperature coefficient 0634h appl np --- 0.0: off 25.0 0.0: off 0.1 ---
dr.53 Rs corr. delta temp. 0635h appl np --- 0: off 200 0: off 1 degree
dr.54 Rs corr. warning time 0636h appl np --- 240 16000 4000 1 s
dr.55 Rs corr. cooling time 0637h appl np --- 240 16000 4000 1 s
dr.56 Rs corr. max. temp. 0638h appl np --- 30 200 90 1 degree
dr.58 torque offset selector 063Ah appl np E 0 79 0 1 --- 7.5-21, 7.6-12
dr.59 torque offset 063Bh appl np --- -320.00 320.00 0.00 0.01 Nm 7.5-21, 7.6-12
dr.60 Rs corr auto temp. mode 063Ch appl np --- 0: off 1: on 0: off 1 ---
dr.61 Rs corr auto temp in.sel. 063Dh appl np E 0 4095 0 1 --- 7.3-10, 7.3-12
6.2-11, 6.2-15, 7.5-17, 7.6-
dr.62 state motor ident. 063Eh RO np --- 0 255 0 1 ---
9
DSM EMC HR
dr.63 063Fh appl np --- 0 255.996 0 0.004 --- 7.6-3, 7.6-4, 7.6-11
(Vpk/1000rpm)
DSM winding inductance ma-
dr.64 0640h appl np --- 0.01 500.00 LTK 0.01 mH 7.6-3
ximum
DASM head inductance 50%
dr.65 0641h appl P --- 99 305 99 0.006 %
flux
dr.66 motor identification error 0642h RO np --- 0 255 0 1 ---
current for Ls/loff identifica-
dr.67 0643h appl np --- 10 250 100 1 %
tion
dS.00 Kp current 1100h appl P --- 0 32767 1500 Adpt 1 --- 7.5-12, 7.6-5, 7.10-3
dS.01 Ki current 1101h appl np --- 0 32767 1500 Adpt 1 --- 7.5-12, 7.6-5, 7.10-3
dS.02 current decoupling 1102h appl np E 0: off 2 0: off 1 --- 6.2-11, 6.2-15, 6.2-18,
dS.02 current decoupling 1102h appl np E 0: off 1 0: off 1 --- 6.2-20, 7.10-3
7.5-25, 7.8-5, 7.8-10, 7.8-
dS.03 curr./torq. mode 1103h appl np E 0 63 0 1 --- 11, 7.8-12, 7.8-13, 7.10-3,
7.10-4, 7.15-5
6.2-7, 6.2-10, 6.2-14, 7.5-
16, 7.5-22, 7.5-23, 7.5-25,
dS.04 flux/rotor adaption mode 1104h appl P E 0 4095 0 1 ---
7.5-33, 7.5-37, 7.8-3, 7.15-
5, 7.15-16
dS.07 KI rotor adaption 1107h appl P --- 0 32767 1000 1 ---
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
dS.08 KP Umax 1108h appl np --- 0 32767 0 1 --- 7.5-37
dS.09 KI Umax 1109h appl np --- 0 32767 50 1 --- 7.5-37
dS.10 Umax modulation limit 110Ah appl P --- 0 110 97 1 % 7.5-37, 7.8-3
ds.11 KP Flux 110Bh appl P --- 0 32767 1000 1 --- 7.5-13, 7.5-22
ds.12 KI Flux 110Ch appl P --- 0 32767 300 1 --- 7.5-13, 7.5-22
7.5 - 13,7.5 - 22,7.8 - 7,7.8 -
dS.13 magn. current limit 110Dh appl np --- 0 1100.0 0 0.1 A
8,7.8 - 9,7.8 - 10,7.8 - 12
dS.14 Kp speed calc. ASCL 110Eh appl P --- 0 32767 1500 1 --- 7.5-13, 7.5-32
dS.15 Ki speed calc. ASCL 110Fh appl P --- 0 32767 1500 1 --- 7.5-13, 7.5-33
dS.17 speed PT1-time ASCL 1111h appl P --- 0 9 3 1 --- 7.5-32
7.5 - 24,7.5 - 25,7.5 - 30,7.5
dS.18 Model adaption 1112h appl P --- 0 2048 0 1 ---
- 31
1; 5.1-4, 7.5-13, 7.5-28, 7.5-
dS.19 limit uf-control dec. ASCL 1113h appl P --- 0 32000 ; 4000 0 rpm
0.125 29
dS.20 delay time uf-contrl. 1114h appl P --- -1 4000 0 1 ms 7.5-28, 7.5-29
n* 5.1-4, 7.5-24, 7.5-28, 7.5-
dS.21 startup speed 1115h appl P --- 0 n * 4000 0 rpm
0.125 30
7.5-24, 7.5-25, 7.5-28, 7.5-
dS.22 startup time 1116h appl P --- 0.00 300.00 5.00 0.01 s
30
dS.23 observer factor 1117h appl P --- 0 100 0.02 0.006 % 7.5-25, 7.5-32
dS.24 Ki current multiplier 1118h appl np --- 0 65535 65535 1 ---
dS.25 current decoupling time 1119h appl np --- 0.000 4095.938 0.000 0.063 ms
dS.26 wait for minimum flux 111Ah appl P --- 40 110 95 0.006 %
dS.30 rotor position detection 111Eh appl np E 0 15 0 1 --- 7.6-18
dS.31 rotor position mode 111Fh appl np E 0 1 0 1 --- 7.6-18, 7.6-19
dS.32 KI HF detection 1120h appl np --- 0 32767 1500 1 --- 7.6-18
dS.33 step current 1121h appl np --- 0 15000 0 1 A 7.6-18
dS.34 diff. Ld Lq Level 1122h appl np --- 0 1000 200 1 %
dS.35 diff. saturation Level 1123h appl np --- 0 1000 50 1 %
dS.36 diff. actual saturation 1124h appl np --- 0 1000 0 1 %
dS.37 usd max modulation ref. 1125h appl np --- 0 100 95 0.006 % 7.8 -12, 7.8-13
dS.38 encoder delay 1126h appl np --- 0 1 0 1 --- 7.11-25
dS.39 reserved modulation factor 1127h appl np --- 0 50 0 0.006 % 7.8-4, 7.10-4
dS.40 torque mode 1128h appl np --- 0 2 0 1 --- 7.9-6
7.11 - 10,7.11 - 11,7.11 - 18,
Ec.00 encoder 1 interface 1000h appl np E -127 127 GBK 1 --- 7.11 - 21,7.11 - 22,7.13 - 10,
7.16 - 12,8.1 - 5
6.2 - 7,6.2 - 9,6.2 - 17,7.6 - 6
,7.11 - 11,7.11 - 14,7.11 - 15
,7.11 - 17,7.11 - 22,7.11 - 2
Ec.01 encoder 1 (inc/r) 1001h appl np E 1 65535 GBK 1 inc
3,7.11 - 24,7.11 - 25,7.12 -
14,7.12 - 17,7.12 - 30,7.12
- 31,7.12 - 33,7.12 - 51
6.2 - 18,7.6 - 6,7.6 - 7,7.11 -
Ec.02 absolute pos. enc. 1 1002h appl np E 0 65535 57057 1 ---
23,7.11 - 24
Ec.03 time 1 for speed calc. 1 1003h appl np E 0 9 3 1 --- 7.11-11, 7.11-23, 7.11-24
7.11-13, 7.11-14, 7.11-15,
Ec.04 gear 1 numerator 1004h appl np --- -32000 32000 1000 1 ---
7.12-30, 7.12-33, 7.15-22
7.11 - 14,7.11 - 15,7.11 - 25,
Ec.05 gear 1 determinator 1005h appl np --- 1 32000 1000 1 ---
7.12 - 33
Ec.06 enc. 1 rotation 1006h appl np E 0 19 0 1 --- 7.5 - 14,7.11 - 12
Ec.07 enc. 1 trigger 1007h appl np E 0 13 GBK 1 ---
7.6 - 7,7.11 - 13,7.12 - 30,7.1 10
2 - 31,7.12 - 33,7.12 - 51
Ec.08 encoder 1 excitation 1008h appl np E -1.94 9.14 6.10 0.14 kHz
7.11 - 6,7.11 - 10,7.11 - 11,7.
Ec.10 encoder 2 interface 100Ah appl np E -127 127 GBK 1 ---
11 - 18,8.1 - 5
7.6 - 6,7.11 - 11,7.11 - 14,7.
Ec.11 encoder 2 (inc/r) 100Bh appl np E 1 65535 GBK 1 inc 12 - 14,7.12 - 17,7.12 - 30,7
11
.12 - 31
Ec.12 absolute position enc.2 100Ch appl np E 0 65535 57057 1 --- 7.6-6, 7.6-7
Ec.13 time 2 for speed calc 100Dh appl np E 0 9 3 1 --- 7.11-11
7.11 - 13,7.11 - 14,7.11 - 15,
7.11 - 16,7.12 - 14,7.12 - 17,
Ec.14 gear 2 numerator 100Eh appl np --- -32000 32000 1000 1 ---
7.12 - 18,7.12 - 24,7.12 - 26,
7.12 - 31
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
7.11 - 14,7.12 - 14,7.12 - 17,
Ec.15 gear 2 determinator 100Fh appl np --- 1 32000 1000 1 --- 7.12 - 18,7.12 - 24,7.12 - 26,
7.12 - 31
Ec.16 enc. 2 rotation 1010h appl np E 0 19 0 1 --- 7.11-12
Ec.17 enc.2 trigger 1011h appl np E 0 13 GBK 1 --- 7.11 - 13,7.12 - 30,7.12 - 31
7.11 - 7,7.11 - 10,7.12 - 15,7
Ec.20 enc. 2 operating mode 1014h appl np --- 0 3 GBK 1 ---
.12 - 16
Ec.21 SSI multiturn resolution 1015h appl np E 0 13 12 1 --- 7.11-19, 7.12-31
Ec.22 SSI clock frq. sel. 1016h appl np --- 0 1 0 1 --- 7.11-19
Ec.23 SSI data code 1017h appl np --- 0 1 1 1 --- 7.11-19
Ec.24 SSI power failure bit 1018h appl np --- 0: off 1: on 0: off 1 --- 7.11-19
1;
Ec.25 nominal tacho speed 1019h appl np --- 1 16000 ; 2000 1500 ; 187.5 rpm 5.1-4, 7.11-20
0.125
Ec.27 operation mode output 101Bh appl np E 0 127 0 1 --- 7.11-10, 7.11-16, 7.11-17
Ec.28 position ch2 over bus 101Ch appl np --- -2147483648 2147483647 0 1 inc 7.11-10, 7.11-13
Ec.29 position ch1 direct 101Dh RO np --- -2147483648 2147483647 0 1 inc
Ec.30 position ch2 direct 101Eh RO np --- -2147483648 2147483647 0 1 inc
Ec.31 position ch1 101Fh RO np --- -2147483648 2147483647 0 1 inc 7.3-12
Ec.32 position ch2 1020h RO np --- -2147483648 2147483647 0 1 inc 7.3-12
Ec.33 system offset ch1 1021h appl np E -2147483648 2147483647 0 1 inc 7.11-18
Ec.34 system offset ch2 1022h appl np E -2147483648 2147483647 0 1 inc 7.11-18
Ec.36 enc. 1 encoder type 1024h RO np --- GBK GBK GBK 1 --- 7.11-21
7.11-21, 7.11-22, 7.11-24,
Ec.37 enc. 1 encoder status 1025h RO np --- 0 255 0 1 ---
7.11-25
appl
(rd)
Ec.38 enc. 1 encoder r/w 1026h np E 0 30 4 1 ---
sup
(st) 7.11-21, 7.11-23, 7.11-24,
appl 7.11-25, 7.12-14
(rd)
Ec.38 enc. 1 encoder r/w 1026h np E 0 62 4 1 ---
sup
(st)
Ec.39 enc. 1 over transmission 1027h appl np E 0 5 0 1 --- 7.11-4, 7.11-15,
Ec.39 enc. 1 over transmission 1027h appl np E 0 4 0 1 --- 7.11-25, 7.12-33, 7.12-34
Ec.40 act. absolute pos. el. 1028h RO np --- 0 65535 0 1 --- 7.5-34, 7.5-35
Ec.41 mode disp. multiturn 1029h appl np E 0 15 0 1 --- 7.11-20
Ec.42 enc. alarm mode 102Ah appl np --- 0 15 0 1 ---
7.4-14, 7.11-7, 7.11-8
Ec.42 enc. alarm mode 102Ah appl np --- 0 15 1 1 ---
Ec.43 SSI data code channel 1 102Bh appl np E 0 1 0 1 --- 7.11-19
Ec.44 abs. Res. channel 1 102Ch appl np E 0 13 10 1 ---
Ec.45 UVW commutation per res. 102Dh appl np E 0 127 0 1 ---
Ec.46 PT1 time channel 1 102Eh appl np --- 0 256 0 1 ms 7.12-34
Ec.47 PT1 time channel 2 102Fh appl np --- 0 256 0 1 ms 7.12-34
scan channel 2 input selec-
Ec.48 1030h appl np E 0 4095 0 1 --- 7.3-10
tion
scan channel 1+ channel 2
Ec.49 1031h appl np E 0 4095 0 1 --- 7.3-10
input selection
Ec.50 scan position Ec.60 1032h RO np --- -2^31 2^31-1 0 1 inc 7.3-12
Ec.51 scan position Ec.61 1033h RO np --- -2^31 2^31-1 0 1 inc 7.3-12
encoder 1 SSI multiturn reso-
Ec.53 1035h appl np E 0 13 0 1 --- 7.11-18, 7.12-33
lution
Ec.54 encoder 1 SSI mode 1036h appl np E 0 3 0 1 --- 7.11-19
Ec.55 encoder 2 SSI mode 1037h appl np E 0 2 0 1 --- 7.11-20
Ec.56 gear 1 numerator 1038h appl np --- -2^30 2^30-1 0 1 --- 7.11-14
Ec.57 gear 1 denominator (long) 1039h appl np --- 1 2^30-1 1000 1 --- 7.11-14
Ec.58 gear 2 numerator 103Ah appl np --- -2^30 2^30-1 0 1 --- 7.11 - 13,7.11 - 14,7.12 - 24
Ec.59 gear 2 denominator (long) 103Bh appl np --- 1 2^30-1 1000 1 --- 7.11-14
Ec.60 system position channel 1 103Ch RO np --- -2^31 2^31-1 0 1 inc
Ec.61 system position channel 2 103Dh RO np --- -2^31 2^31-1 0 1 inc
Ec.63 encoder 2 over gear 103Fh appl np E 0 1 0 1 ---
6.2 - 4,6.2 - 6,6.2 - 9,6.2 - 13
Fr.01 copy parameter set 0901h appl P E -9 7 0 1 --- ,6.2 - 17,6.2 - 19,7.6 - 11,7.
6 - 12
Fr.02 parameter set source 0902h appl np E 0 5 0 1 --- 10.1 - 10
Fr.03 parameter set lock 0903h appl np E 0 255 0 1 --- 7.13-9
Fr.04 parameter set setting 0904h appl np E 0 7 0 1 ---
Fr.05 set activation delay 0905h appl P --- 0.00 32.00 0.00 0.01 s
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
Fr.06 set deactivation delay 0906h appl P --- 0.00 32.00 0.00 0.01 s
Fr.07 paraset input selection 0907h appl np E 0 4095 0 1 --- 7.3-10, 7.3-11
Fr.08 motor set classification 0908h appl P E 0 7 0 1 --- 7.13-28, 7.13-29
Fr.09 bus parameter set 0909h appl np --- -1: act set 7 0 1 ---
Fr.10 load mot. dependent para. 090Ah appl P E 1 3 1 1 --- 6.2 - 5,6.2 - 7,6.2 - 10,6.2 - 1
, 4,6.2 - 18,6.2 - 19,7.5 - 9,7.
5 - 12,7.5 - 13,7.5 - 22,7.5 - 2
Fr.10 load mot. dependent para. 090Ah appl np E 1 2 1 1 ---
9,7.6 - 5,7.6 - 12,7.6 - 13,7.6
- 17,7.7 - 3,7.10 - 3,7.11 - 25
Fr.11 reset set input sel. 090Bh appl np E 0 4095 0 1 --- 7.3-10, 7.3-11
Fr.12 set change mode mod. on 090Ch appl np E 0 3 2 1 ---
Fr.12 set change mode mod. on 090Ch appl np E 0 3 0 1 ---
LE.00 comparison level 0 0D00h appl P --- -10737418.24 10737418.23 0.00 0.01 ---
7.3-22, 7.12-75, 7.15-76, 10
7.15-9, 7.15-12, 7.15-22
LE.01 comparison level 1 0D01h appl P --- -10737418.24 10737418.23 0.00 0.01 --- 7.3-26, 7.12-49
LE.02 comparison level 2 0D02h appl P --- -10737418.24 10737418.23 100.00 0.01 --- 7.3-26, 7.12-49
LE.03 comparison level 3 0D03h appl P --- -10737418.24 10737418.23 4.00 0.01 --- 7.12-49
LE.04 comparison level 4 0D04h appl P --- -10737418.24 10737418.23 0.00 0.01 --- 7.5-26
LE.05 comparison level 5 0D05h appl P --- -10737418.24 10737418.23 0.00 0.01 --- 11
LE.06 comparison level 6 0D06h appl P --- -10737418.24 10737418.23 0.00 0.01 ---
7.3 - 22,7.12 - 75,7.12 - 76,7
LE.07 comparison level 7 0D07h appl P --- -10737418.24 10737418.23 0.00 0.01 ---
.15 - 12
LE.08 hysteresis 0 0D08h appl P --- 0.00 300.00 0.00 0.01 --- 7.3-22
LE.09 hysteresis 1 0D09h appl P --- 0.00 300.00 0.00 0.01 --- 7.3-22, 7.3-26
LE.10 hysteresis 2 0D0Ah appl P --- 0.00 300.00 5.00 0.01 --- 7.3-26
LE.11 hysteresis 3 0D0Bh appl P --- 0.00 300.00 0.50 0.01 ---
LE.12 hysteresis 4 0D0Ch appl P --- 0.00 300.00 0.00 0.01 --- 7.5-26
LE.13 hysteresis 5 0D0Dh appl P --- 0.00 300.00 0.00 0.01 ---
LE.14 hysteresis 6 0D0Eh appl P --- 0.00 300.00 0.00 0.01 ---
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
LE.15 hysteresis 7 0D0Fh appl P --- 0.00 300.00 0.00 0.01 --- 7.3-22
n* 7.3 - 18,7.3 - 22,7.13 - 41,7.
LE.16 freq/speed hysteresis 0D10h appl np --- 0 n * 200 n * 15 rpm
0.125 15 - 4,10.1 - 12
7.1 - 11,7.3 - 10,7.3 - 11,7.1
LE.17 timer 1 start inp. sel. 0D11h appl np E 0 4095 0 1 ---
5 - 9,7.15 - 10,7.15 - 11
LE.18 timer 1 start condition 0D12h appl np E 0 15 0 1 --- 7.15-9, 7.15-10, 7.15-11
7.3 - 10,7.3 - 11,7.12 - 50,7.1
LE.19 timer 1 start inp. sel. 0D13h appl np E 0 4095 0 1 ---
5 - 9,7.15 - 11
LE.20 timer 1 reset condition 0D14h appl np E 0 31 16 1 --- 7.15-9, 7.15-11
7.1 - 11,7.15 - 9,7.15 - 10,7.1
LE.21 timer 1 mode 0D15h appl np --- 0 63 0 1 ---
5 - 11,7.15 - 12
7.1 - 11,7.3 - 10,7.3 - 11,7.1
LE.22 timer 2 start inp. sel. 0D16h appl np E 0 4095 0 1 ---
5 - 9,7.15 - 10,7.15 - 11
LE.23 timer 2 start condition 0D17h appl np E 0 15 0 1 --- 7.15-9, 7.15-10
7.3-10, 7.3-11, 7.15-9,
LE.24 timer 2 start inp. sel. 0D18h appl np E 0 4095 0 1 ---
7.15-11
LE.25 timer 2 reset condition 0D19h appl np E 0 31 16 1 --- 7.15-9, 7.15-11
LE.26 timer 2 mode 0D1Ah appl np --- 0 63 0 1 --- 7.1-11, 7.15-9, 7.15-10
LE.27 reference torque 0D1Bh appl np --- 0.00 32000.00 0.00 0.01 Nm 7.8-15
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
n * -0.125: n * -0.125: n* 7.4-4, 7.4-9, 7.4-13, 7.4-14,
oP.11 max. reference reverse 030Bh appl P --- n * 4000 rpm
=For =For 0.125 7.4-15, 7.12-22, 7.12-28
7.4 - 5,7.4 - 13,7.4 - 14,7.12
n*
oP.14 abs. max. reference for 030Eh appl P --- 0 n * 4000 n * 4000 rpm - 14,7.12 - 22,7.12 - 28,7.12
0.125
- 29,7.12 - 78,7.12 - 81
7.4 - 13,7.4 - 14,7.12 - 14,7.
n * -0.125: n * -0.125: n*
oP.15 abs. max. reference rev 030Fh appl P --- n * 4000 rpm 12 - 22,7.12 - 28,7.12 - 78,7
=For =For 0.125
.12 - 81
oP.16 rotation delay time 0310h appl np --- 0 1000 0 1 s 7.4-10
oP.18 step value rot. source 0312h appl P E 0 10 7 1 --- 7.4-11, 7.4-12
7.3-10, 7.3-11, 7.4-11, 7.4-
oP.19 step value input sel. 1 0313h appl np E 0 4095 16 1 ---
12
7.3-10, 7.3-11, 7.4-11, 7.4-
oP.20 step value input sel. 2 0314h appl np E 0 4095 32 1 ---
12
n*
oP.21 step value 1 0315h appl P --- n * -4000 n * 4000 n * 100 rpm 7.4-11, 7.4-12, 7.16-6
0.125
n*
oP.22 step value 2 0316h appl P --- n * -4000 n * 4000 n * -100 rpm 7.4-11
0.125
n*
oP.23 Step value 3 0317h appl P --- n * -4000 n * 4000 n*0 rpm 7.4-11, 7.4-12
0.125
acceleration/deceleration
oP.27 031Bh appl P E 0 511 0 1 --- 7.4-16, 7.4-20, 7.4-22
mode
7.4-17, 7.4-18, 7.4-20, 7.4-
oP.28 acceleration time forward 031Ch appl P --- 0.00 300.00 5.00 0.01 s
22, 7.12-19, 7.16-6
oP.29 acc. time reverse 031Dh appl P --- -0.01: =For 300.00 -0.01: =For 0.01 s 7.4-17, 7.4-18
7.4-18, 7.4-20, 7.4-21,
oP.30 deceleration time forward 031Eh appl P --- -0.01: =Acc 300.00 5.00 0.01 s
7.16-6
7.4-17, 7.4-18, 7.4-20, 7.4-
oP.31 dec. time reverse 031Fh appl P --- -0.01: =For 300.00 -0.01: =For 0.01 s
21, 7.12-28
oP.32 s-curve time acc. for. 0320h appl P --- 0.00: off 5.00 0.00: off 0.01 s 7.4-17, 7.4-19, 7.12-28
oP.33 s-curve time acc. 0321h appl P --- -0.01: =For 5.00 -0.01: =For 0.01 s 7.4-19, 7.4-20
oP.34 s-curve time dec. for. 0322h appl P --- -0.01: =Acc 5.00 -0.01: =Acc 0.01 s 7.4-19, 7.4-20
oP.35 s-curve time dec. rev. 0323h appl P --- -0.01: =For 5.00 -0.01: =For 0.01 s 7.4-17, 7.4-19, 7.12-28
n* 7.4-14, 7.6-17, 7.12-28,
oP.40 max. output val. for. 0328h appl P --- 0 n * 4000 n * 4000 rpm
0.125 7.13-10, 7.13-24
n * -0.125: n * -0.125: n*
oP.41 max. output val. rev. 0329h appl P --- n * 4000 rpm 7.4-14, 7.6-17, 7.13-10
=For =For 0.125
oP.44 ext. function mode/source 032Ch appl P E 0 79 0 1 --- 7.15-19, 7.15-21
oP.45 ext. funct. dig. source 032Dh appl P --- 0.00 100.00 0.00 0.01 % 7.15-19, 7.15-21
oP.46 ext. funct. acc/dec time 032Eh appl P --- 0.00 20.00 10.00 0.01 s 7.15-19, 7.15-22
oP.47 sweep-gen. acc. time 032Fh appl P --- 0.00 20.00 10.00 0.01 s 7.15-19, 7.15-20
oP.48 sweep-gen. dec. time 0330h appl P --- 0.00 20.00 10.00 0.01 s 7.15-19, 7.15-20
oP.49 diam. corr. dmin/dmax 0331h appl P --- 0.010 0.990 0.500 0.001 --- 7.15-21
oP.50 motorpoti function 0332h appl np E 0 7 0 1 --- 7.1-11, 7.15-7, 7.15-8
7.4-4, 7.11-16, 7.12-24,
oP.52 motorpoti value 0334h appl P --- -100.00 100.00 0.00 0.01 %
7.15-7
oP.53 motorpoti min. value 0335h appl np --- -100.00 100.00 0.00 0.01 % 7.1-11, 7.11-16, 7.15-9
oP.54 motorpoti max. value 0336h appl np --- -100.00 100.00 100.00 0.01 % 7.11-16, 7.15-9
oP.55 motorpoti reset value 0337h appl np --- -100.00 100.00 0.00 0.01 % 7.15-7, 7.15-8
oP.56 mot.poti inc. input sel. 0338h appl np E 0 4095 0 1 --- 7.3-10, 7.3-11, 7.15-7
oP.57 mot.poti dec. input sel. 0339h appl np E 0 4095 0 1 --- 7.3-10, 7.3-11, 7.15-7
oP.58 mot.poti reset inp. sel. 033Ah appl np E 0 4095 0 1 ---
7.3-10, 7.3-11, 7.15-7, 10
7.15-8
motorpoti increase/decrease
oP.59 033Bh appl P --- 0.00 50000.00 66.00 0.01 s 7.1-11, 7.15-7, 7.15-8
time
oP.60 dir. forward input sel. 033Ch appl np E 0 4095 4 1 --- 7.3-10, 7.3-11, 7.4-8, 7.4-9
direction reverse input selec-
oP.61
tion
033Dh appl np E 0 4095 8 1 --- 7.3-10, 7.3-11, 7.4-8, 7.4-9 11
oP.62 acc/dec time factor 033Eh appl np E 0 4 0 1 --- 7.4-22
oP.63 ref. value high-res. 033Fh appl np --- -2^31 2^31-1 0 1 --- 7.4-4, 7.4-6, 7.4-7
n*
oP.64 rel. value high-res. 0340h appl P --- n * 600 n * 4000 n * 2100 rpm 7.4-4, 7.4-5, 7.4-6, 7.4-7
0.125
n*
oP.65 min. proh. reference 1 0341h appl np --- n * -4000 n * 4000 0 rpm 7.4-15, 7.5-30
0.125
n*
oP.66 max. proh. reference 1 0342h appl np --- n * -4000 n * 4000 0 rpm 7.4-15
0.125
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
n*
oP.67 min. proh. reference 2 0343h appl np --- n * -4000 n * 4000 0 rpm 7.4-15
0.125
n*
oP.68 max. proh. reference 2 0344h appl np --- n * -4000 n * 4000 0 rpm 7.4-15, 7.5-30
0.125
oP.69 motorpoti deceleration time 0345h appl P --- -0.01 50000.00 -0.01 0.01 s
7.4-17, 7.4-19, 7.4-20,
oP.70 s-c. up time acc. for. 0346h appl P --- -0.01: = low 5.00 -0.01: = low 0.01 s
7.12-28
oP.71 s-c. up time acc. rev. 0347h appl P --- -0.02: =For 5.00 -0.01: = low 0.01 s 7.4-19, 7.4-20
oP.72 s-c. up time dec. for. 0348h appl P --- -0.02: =Acc 5.00 -0.01: = low 0.01 s 7.4-19, 7.4-20
7.4-17, 7.4-19, 7.4-20,
oP.73 s-c. up time dec. rev. 0349h appl P --- -0.02: =Acc 5.00 -0.01: = low 0.01 s
7.12-28
oP.74 reference splitting 034Ah appl np --- 0 127 0 1 ms 7.7-10
Pn.00 auto. retry UP 0400h appl np --- 0: off 1: on 1: on 1 --- 7.3-21, 7.13-16
Pn.01 auto. retry OP 0401h appl np --- 0: off 1: on 0: off 1 --- 7.13-16
Pn.02 auto. retry OC 0402h appl np --- 0: off 1: on 0: off 1 --- 7.13-17
7.13-7, 7.13-11, 7.13-12,
Pn.03 E.EF stopping mode 0403h appl np --- 0 6 0 1 ---
7.13-17
7.3-10, 7.3-11, 7.13-7,
Pn.04 ext. fault input select 0404h appl np E 0 4095 64 1 ---
7.13-40
7.12 - 78,7.12 - 80,7.13 - 4,7
Pn.05 E.buS stopping mode 0405h appl np --- 0 6 6 1 --- .13 - 7,7.13 - 11,7.13 - 12,7.1
3 - 17,10.1 - 9,10.1 - 10
7.3-21, 7.12-80, 7.13-7,
Pn.06 watchdog time 0406h appl np E 0.00: off 40.00 0.00: off 0.01 s
10.1-9
7.12 - 4,7.12 - 9,7.12 - 29,7.
prohibit rotation stopping
Pn.07 0407h appl np --- 0 6 6 1 --- 12 - 64,7.13 - 4,7.13 - 7,7.13
mode
- 11,7.13 - 12,7.13 - 17
7.3 - 17,7.13 - 6,7.13 - 11,7.1
Pn.08 warning OL stop. mode 0408h appl np --- 0 6 6 1 ---
3 - 12,7.13 - 17
Pn.09 OL warning level 0409h appl np --- 0 100 80 1 % 7.3-17, 7.3-19, 7.13-6
warning OH stop. modeRes- 7.3 - 17,7.13 - 4,7.13 - 6,7.1
Pn.10 040Ah appl np --- 0 6 6 1 ---
ponse 3 - 11,7.13 - 12,7.13 - 17
Pn.11 OH warning level 040Bh appl np --- 0 90 70 1 degree 7.3-17, 7.13-4, 7.13-6
warning dOH stop. modeRe- 7.1-17, 7.3-17, 7.13-8,
Pn.12 040Ch appl np --- 0 8 6 1 ---
sponse 7.13-11, 7.13-12, 7.13-17
Pn.13 E.dOH delay time 040Dh appl np --- 0 120 0 1 s 7.1-17, 7.3-17, 7.13-8
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
7.3 - 17,7.13 - 8,7.13 - 9,7.1
warning OH2 stop. modeRe- 3 - 11,7.13 - 12,7.13 - 17,7.1
Pn.14 040Eh appl np --- 0 6 6 1 ---
sponse 3 - 27,7.13 - 28,7.13 - 29,7.
13 - 30
Pn.15 OH2 warning levellevel 040Fh appl np --- 0 100 100 1 % 7.3-17, 7.13-9, 7.13-30
7.3 - 17,7.13 - 6,7.13 - 11,7.1
Pn.16 warning OHI stop. mode 0410h appl np --- 0 7 7 1 ---
3 - 12,7.13 - 13,7.13 - 17
Pn.17 E.OHI delay time 0411h appl np --- 0 120 0 1 s 7.3-17, 7.13-6
7.13-9, 7.13-11, 7.13-12,
Pn.18 E.Set stopping mode 0412h appl np --- 0 6 0 1 ---
7.13-17
Pn.19 stall mode 0413h appl P E 0 255 0 1 --- 7.13-23, 7.13-24, 7.13-25
Pn.20 stall level 0414h appl P --- 0 200: off 200: off 1 % 7.3-17, 7.13-23, 7.13-24
stall acceleration/decelerati-
Pn.21 0415h appl P --- 0 300.00 2.00 0.01 s 5.1-5, 7.13-24, 7.13-25
on time
Pn.22 LAD stop function 0416h appl P E 0 7 0 1 --- 7.13-22, 7.13-23, 7.13-26
7.3-10, 7.3-11, 7.13-22,
Pn.23 LAD stop input selection 0417h appl np E 0 4095 0 1 ---
7.13-23, 7.13-26
Pn.24 LAD load level 0418h appl P --- 0 200 140 1 % 7.3-18, 7.13-22, 7.13-25
375 ; 720 ; 7.3-18, 7.13-22, 7.13-23,
Pn.25 LD voltage 0419h appl P --- 200 1200 1 V
1100 7.13-26
Pn.26 speed search condition 041Ah appl P E 0 31 8 1 --- 7.5-30, 7.6-13, 7.13-21
Pn.27 speed search mode 041Bh appl np E 0 255 88 1 ---
Pn.28 DC braking Mode 041Ch appl P E 0 506 7 1 --- 7.5-30, 7.15-4
Pn.29 DC brake input selection 041Dh appl np E 0 4095 0 1 --- 7.3-10, 7.3-11, 7.15-4
Pn.30 DC braking time 041Eh appl P --- 0.00 100.00 10.00 0.01 s 7.15-4, 7.15-5
Pn.31 DC braking max. voltage 041Fh appl P --- 0.0 25.5 25.5 0.1 % 7.15-5
n*
Pn.32 DC braking start level 0420h appl P --- 0 n * 4000 n * 120 rpm 5.1-4, 7.15-4, 7.15-5
0.125
DC braking maximum current
Pn.33 0421h appl P --- 0.0 400.0 100.0 0.1 % 7.5-30, 7.15-5
ASCL
Pn.34 brake control mode 0422h appl P E 0 4 0 1 ---
Pn.34 brake control mode 0422h appl P E 0 4 2 1 --- 7.6-8, 7.15-13
7.6 - 13,7.15 - 14,7.15 - 15,
Pn.35 premagnetizing time 0423h appl P --- 0.00 100.00 0.25 0.01 s
7.15 - 16,7.15 - 17,7.15 - 18
7.3 - 12,7.6 - 13,7.15 - 13,7.
Pn.36 brake release time 0424h appl P --- 0.00 100.00 0.25 0.01 s
15 - 14,7.15 - 15,7.15 - 17
n* 5.1 - 4,7.15 - 14,7.15 - 15,7.
Pn.37 brake ctrl. start ref. 0425h appl P --- n * -600 n * 600 0 rpm
0.125 15 - 16,7.15 - 17,7.15 - 18
Pn.38 brake fadeout time 0426h appl P --- 0.00 0.50 0.00 0.01 s 7.15-15
Pn.39 brake delay time 0427h appl P --- 0.00 100.00 0.25 0.01 s 7.15-14, 7.15-15, 7.15-18
7.3-12, 7.15-13, 7.15-14,
Pn.40 brake closing time 0428h appl P --- 0.00 100.00 0.25 0.01 s
7.15-15
n* 5.1-4, 7.15-15, 7.15-16,
Pn.41 brake ctrl. stop ref. 0429h appl P --- n * -600 n * 600 0 rpm
0.125 7.15-17
Pn.42 brake check input sel. 042Ah appl np E 0 4095 0 1 --- 7.3-12, 7.15-14
Pn.43 min. load brake ctrl. 042Bh appl P --- 0: off 100 0: off 1 % 7.15-13, 7.15-14, 8.1-5
7.13 - 31,7.13 - 32,7.13 - 33,
Pn.44 power off mode 042Ch appl np E 0 1023 0 1 --- 7.13 - 34,7.13 - 35,7.13 - 36,
7.13 - 37,7.13 - 38
290 ; 500 ;
Pn.45 power off start voltage 042Dh appl np --- 200 1200 1 V 7.13-31, 7.13-33
860
Pn.46 power off auto st. level 042Eh appl np --- 50 90 80 1 %
7.13-31, 7.13-32, 7.13-33
10
Pn.46 power off auto st. level 042Eh appl np --- 50 100 80 1 %
7.13-31, 7.13-32, 7.13-34,
Pn.47 power off brake torque 042Fh appl np --- 0.0 100.0 0.0 0.1 %
7.13-36, 7.13-37
5.1 - 4,7.13 - 31,7.13 - 32,7.
n*
Pn.48 power off restart level 0430h appl np --- 0 n * 4000 0
0.125
rpm 13 - 34,7.13 - 35,7.13 - 37,7 11
.13 - 38
Pn.49 power off start input selection 0431h appl np E 0 255 0 1 --- 7.13-31, 7.13-32, 7.13-37
290 ; 500 ; 7.13 - 31,7.13 - 32,7.13 - 35,
Pn.50 power off ref. DC volt. 0432h appl np --- 200 1200 1 V
860 7.13 - 36,7.13 - 37,7.13 - 38
Pn.51 power off KP DC voltage 0433h appl np --- 0 32767 128 1 --- 7.13 - 31,7.13 - 35
7.13 - 31,7.13 - 32,7.13 - 35,
Pn.52 power off restart delay 0434h appl np --- 0.00 100.00 0.00 0.01 s
7.13 - 36,7.13 - 37
Pn.53 power off KP act. curr. 0435h appl np --- 0 32767 800 1 --- 7.13-31, 7.13-35
Pn.54 power off KI act. curr. 0436h appl np --- 0 32767 800 1 --- 7.13-31, 7.13-35
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
Pn.55 power off KD act. curr. 0437h appl np --- 0 32767 0 1 --- 7.13-31, 7.13-35
Pn.56 power off jump factor 0438h appl np --- 0 800 100 1 % 7.13-31, 7.13-34, 8.1-8
Pn.57 power off KI DC voltage 0439h appl np --- 0 32767 5 1 --- 7.13-31, 7.13-35
7.13 - 14,7.13 - 15,7.13 - 18,
Pn.58 quick stop mode 043Ah appl np E 0 31 0 1 ---
7.13 - 19,7.13 - 20
Pn.59 quick stop level 043Bh appl np --- 0 200 200 1 % 7.13 - 14,7.13 - 18,7.13 - 19
5.1 - 5,7.12 - 9,7.13 - 13,7.1
3 - 14,7.13 - 18,7.13 - 19,7.
Pn.60 quick stop deceleration time 043Ch appl P --- 0 300.00 2.00 0.01 s
13 - 22,7.13 - 25,7.13 - 37,7
.13 - 38
7.5 - 12,7.6 - 5,7.11 - 25,7.1
Pn.61 quick stop torque limit 043Dh appl P --- 0 32000.00 0 Adpt 0.01 Nm 2 - 9,7.13 - 13,7.13 - 18,7.1
3 - 19
Pn.62 dOH warning level 043Eh appl np --- 0 200 100 1 degree 7.1-17, 7.13-17
Pn.64 act. GTR7 input selection 0440h appl np E 0 4095 0 1 --- 7.3-10, 7.3-11, 7.13-38
7.13 - 4,7.13 - 7,7.13 - 17,7.
13 - 38,7.13 - 39,7.13 - 40,7
Pn.65 special functions 0441h appl np E 0 32768 0 1 ---
.13 - 41,7.13 - 43,8.1 - 8,10.
1 - 12
Pn.66 software limit stopping mode 0442h appl np --- 0 6 6 1 --- 7.13-8, 7.13-11, 7.13-12
qucik stop max. torque corner 7.12 - 9,7.13 - 13,7.13 - 18,7
Pn.67 0443h appl P --- 0 32000.00 0 Adpt 0.01 Nm
speed .13 - 20
Pn.68 max. abn. stopping time 0444h appl np --- 0.00: off 100.00 0.00: off 0.01 s 7.13-15, 7.13-20
380 ; 740 ;
Pn.69 GTR7 voltage 0445h appl np --- 300 1500 1 V 7.13-38, 7.13-41
1140
Pn.70 brake pretorq. source 0446h appl P E 0 3 0 1 --- 7.15-16
Pn.71 pretorque ref. setting % 0447h appl P --- -400.0 400.0 100.0 0.1 % 7.15-16
Pn.72 set prog. spec. functions 0448h appl P --- 0 1 0 1 --- 7.1-17
Pn.74 out phase check mode 044Ah appl np --- 0 1 0 1 ---
Pn.75 E.SCL stopping mode 044Bh appl np --- 0 6 6 1 --- 7.13-10, 7.13-11, 7.13-12
Pn.76 max. E.UP warning time 044Ch appl np --- 0.00: off 32.00 0.00: off 0.01 s 7.3-21, 7.13-16
Pn.77 load-shunt activation voltage 044Dh appl np --- 0 1500 0 1 V
Pn.78 USV operation input selection 044Eh appl np E 0 4095 0 1 --- 7.3-12
Pn.79 acceleration limit 1/s^s 044Fh appl np --- 0.01 10737418.23 0.01 0.01 --- 7.3-21, 7.13-10
Pn.80 acc. scan time 0450h appl np --- 0 60000 0 1 ms 7.3-21, 7.13-10
warning acceleration stop
Pn.81 0451h appl np --- 0 6 6 1 --- 7.13-10, 7.13-11, 7.13-12
mode
Pn.82 GTR7 resistance 0452h appl np --- 0.000 5000.000 0.000 0.001 Ohm 7.13-39
Pn.83 quick stop s-curve time 0453h appl P --- 0 500 0 0 s
Pn.84 no Pu / E.UP delay time 0454h appl np --- 0 3200 0 1 s 7.13-5
Pn.90 speed limit (ASCL) 045Ah appl np --- -20 20 2.0 0.1 %
Pn.91 flow ctrl. mode 045Bh appl np E 0 3 0 1 --- 7.13-42 ff
Pn.92 valve ctrl. output select 045Ch appl np E 0 255 0 1 ---
Pn.93 flow switch input select 045Dh appl np E 0 4095 0 1 --- 7.13-42 ff
Pn.94 flow ctrl. warning delay 045Eh appl np --- 0 6000 0 1 s
Pn.95 flow ctrl. min. temp. 045Fh appl np --- 0 90 0 1 °C 7.13-42 ff
Pn.96 pow.off max. time f. rest. 0460h appl np --- 0 10000 0 1 s 7.13-36
PP.00 Prog. Parameter 00 3300h appl np --- ud. 31 ud.30 0 1 --- 7.16-12
PP.01 Prog. Parameter 01 3301h appl np --- ud. 31 ud.30 0 1 ---
PP.02 Prog. Parameter 02 3302h appl np --- ud. 31 ud.30 0 1 ---
PP.03 Prog. Parameter 03 3303h appl np --- ud. 31 ud.30 0 1 ---
PP.04 Prog. Parameter 04 3304h appl np --- ud. 31 ud.30 0 1 ---
PP.05 Prog. Parameter 05 3305h appl np --- ud. 31 ud.30 0 1 ---
PP.06 Prog. Parameter 06 3306h appl np --- ud. 31 ud.30 0 1 ---
PP.07 Prog. Parameter 07 3307h appl np --- ud. 31 ud.30 0 1 ---
PP.08 Prog. Parameter 08 3308h appl np --- ud. 31 ud.30 0 1 ---
PP.09 Prog. Parameter 09 3309h appl np --- ud. 31 ud.30 0 1 ---
PP.10 Prog. Parameter 10 330Ah appl np --- ud. 31 ud.30 0 1 ---
PP.11 Prog. Parameter 11 330Bh appl np --- ud. 31 ud.30 0 1 ---
PP.12 Prog. Parameter 12 330Ch appl np --- ud. 31 ud.30 0 1 ---
PP.13 Prog. Parameter 13 330Dh appl np --- ud. 31 ud.30 0 1 ---
PP.14 Prog. Parameter 14 330Eh appl np --- ud. 31 ud.30 0 1 ---
PP.15 Prog. Parameter 15 330Fh appl np --- ud. 31 ud.30 0 1 ---
PP.16 Prog. Parameter 16 3310h appl np --- ud. 31 ud.30 0 1 ---
PP.17 Prog. Parameter 17 3311h appl np --- ud. 31 ud.30 0 1 ---
PP.18 Prog. Parameter 18 3312h appl np --- ud. 31 ud.30 0 1 ---
PP.19 Prog. Parameter 19 3313h appl np --- ud. 31 ud.30 0 1 ---
PP.20 Prog. Parameter 20 3314h appl np --- ud. 31 ud.30 0 1 ---
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
PP.21 Prog. Parameter 21 3315h appl np --- ud. 31 ud.30 0 1 ---
PP.22 Prog. Parameter 22 3316h appl np --- ud. 31 ud.30 0 1 ---
PP.23 Prog. Parameter 23 3317h appl np --- ud. 31 ud.30 0 1 ---
PP.24 Prog. Parameter 24 3318h appl np --- ud. 31 ud.30 0 1 ---
PP.25 Prog. Parameter 25 3319h appl np --- ud. 31 ud.30 0 1 ---
PP.26 Prog. Parameter 26 331Ah appl np --- ud. 31 ud.30 0 1 ---
PP.27 Prog. Parameter 27 331Bh appl np --- ud. 31 ud.30 0 1 ---
PP.28 Prog. Parameter 28 331Ch appl np --- ud. 31 ud.30 0 1 ---
PP.29 Prog. Parameter 29 331Dh appl np --- ud. 31 ud.30 0 1 ---
PP.30 Prog. Parameter 30 331Eh appl np --- ud. 31 ud.30 0 1 ---
PP.31 Prog. Parameter 31 331Fh appl np --- ud. 31 ud.30 0 1 ---
PP.32 Prog. Parameter 32 3320h appl np --- ud. 31 ud.30 0 1 ---
PP.33 Prog. Parameter 33 3321h appl np --- ud. 31 ud.30 0 1 ---
PP.34 Prog. Parameter 34 3322h appl np --- ud. 31 ud.30 0 1 ---
PP.35 Prog. Parameter 35 3323h appl np --- ud. 31 ud.30 0 1 ---
PP.36 Prog. Parameter 36 3324h appl np --- ud. 31 ud.30 0 1 ---
PP.37 Prog. Parameter 37 3325h appl np --- ud. 31 ud.30 0 1 ---
PP.38 Prog. Parameter 38 3326h appl np --- ud. 31 ud.30 0 1 ---
PP.39 Prog. Parameter 39 3327h appl np --- ud. 31 ud.30 0 1 ---
PP.40 Prog. Parameter 40 3328h appl np --- ud. 31 ud.30 0 1 ---
PP.41 Prog. Parameter 41 3329h appl np --- ud. 31 ud.30 0 1 ---
PP.42 Prog. Parameter 42 332Ah appl np --- ud. 31 ud.30 0 1 ---
PP.43 Prog. Parameter 43 332Bh appl np --- ud. 31 ud.30 0 1 ---
PP.44 Prog. Parameter 44 332Ch appl np --- ud. 31 ud.30 0 1 ---
PP.45 Prog. Parameter 45 332Dh appl np --- ud. 31 ud.30 0 1 ---
PP.46 Prog. Parameter 46 332Eh appl np --- ud. 31 ud.30 0 1 ---
PP.47 Prog. Parameter 47 332Fh appl np --- ud. 31 ud.30 0 1 ---
7.12 - 11,7.12 - 18,7.12 - 19,
7.12 - 20,7.12 - 22,7.12 - 24,
7.12 - 26,7.12 - 27,7.12 - 28
,7.12 - 30,7.12 - 31,7.12 - 34
,7.12 - 35,7.12 - 37,7.12 - 3
8,7.12 - 39,7.12 - 40,7.12 - 4
PS.00 posi/synchronous mode 1300h appl P E 0 8127 0 1 ---
1,7.12 - 44,7.12 - 45,7.12 - 4
6,7.12 - 48,7.12 - 49,7.12 - 6
0,7.12 - 61,7.12 - 62,7.12 - 6
4,7.12 - 68,7.12 - 69,7.12 - 7
5,7.12 - 76,7.12 - 77,7.12 - 7
9,10.1 - 11,10.1 - 12
7.12 - 17,7.12 - 18,7.12 - 26,
PS.01 actual position source 1301h appl P --- 0 2 1 1 --- 7.12 - 29,7.12 - 30,7.12 - 33,
7.12 - 34,7.12 - 74,7.12 - 79,
7.3 - 10,7.3 - 11,7.12 - 19,7.1
PS.02 posi/synch input select 1302h appl np E 0 4095 0 1 --- 2 - 26,7.12 - 27,7.12 - 29,7.1
2 - 34,7.12 - 79
PS.03 shifting slave input selection 1303h appl np E 0 4095 0 1 --- 7.3-10, 7.3-11, 7.12-25
PS.04 shifting slave 1304h appl np --- -2^30 2^30-1 0 1 inc 7.12-25
7.12 - 19,7.12 - 20,7.12 - 21,
PS.05 start offset 1305h appl P --- -2^30 2^30-1 0 1 inc
7.12 - 22,7.12 - 23,7.12 - 24
7.12 - 4,7.12 - 14,7.12 - 18,7
PS.06 KP pos/syn 1306h appl P --- 0 32767 500 1 --- .12 - 26,7.12 - 29,7.12 - 59,7 10
.12 - 79,7.12 - 81
PS.07 KP speed limit reduction 1307h appl P --- 0.0 100.0 100.0 0.1 % 7.12-4, 7.12-81
n * -0.125: n* 5.1-4, 7.12-4, 7.12-14,
PS.08 speed limit for ps.07 1308h appl P --- n * 4000 n * 4000 rpm
off(ru.63) 0.125 7.12-29, 7.12-81
5.1 - 4,7.12 - 14,7.12 - 20,7.
PS.09 pos/syn position limit 1309h appl P --- 0 n * 4000 n * 250
n*
rpm 12 - 22,7.12 - 28,7.12 - 29,7.
11
0.125
12 - 32,7.12 - 81
PS.10 shift. slave inv. input sel. 130Ah appl np E 0 4095 0 1 --- 7.3-10, 7.3-11, 7.12-25
PS.11 reset m/s diff. inp.sel. 130Bh appl np E 0 4095 0 1 --- 7.3-12
PS.13 set ref. point inp. sel. 130Dh appl np E 0 4095 0 1 --- 7.3-12, 7.12-13
7.12 - 5,7.12 - 6,7.12 - 7,7.1
PS.14 mode of position reference 130Eh appl np E 0 32767 0 1 --- 2 - 8,7.12 - 9,7.12 - 10,7.12 -
11,7.12 - 12,7.12 - 13
PS.15 limit switch left 130Fh appl np --- -2^31 2^31-1 -2^30 1 inc 7.13-8
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
PS.16 limit switch right 1310h appl np --- -2^31 2^31-1 2^30-1 1 inc 7.13-8
7.3 - 12,7.12 - 7,7.12 - 10,7.
PS.17 reference point 1311h appl np E -2^31 2^31-1 0 1 inc 12 - 11,7.12 - 12,7.12 - 55,7.
12 - 56,7.12 - 75
reference switch input selec-
PS.18 1312h appl np E 0 4095 0 1 --- 7.3-10, 7.3-11, 7.12-5
tion
7.3-10, 7.3-11, 7.12-5,
PS.19 start reference input selection 1313h appl np E 0 4095 0 1 ---
7.12-6
7.12-6, 7.12-7, 7.12-9,
PS.20 reference acc/dec time 1314h appl np --- 0.00 300.00 0.50 0.01 s
7.12-10, 7.12-12
n* 5.1-4, 7.12-6, 7.12-7, 7.12-
PS.21 reference speed 1315h appl np --- n * -4000 n * 4000 n * 100 rpm
0.125 10
n* 5.1-4, 7.12-6, 7.12-7, 7.12-
PS.22 reference drive free speed 1316h appl np --- 0: off n * 4000 0: off rpm
0.125 10
7.12 - 35,7.12 - 38,7.12 - 41,
7.12 - 42,7.12 - 44,7.12 - 45,
PS.23 index selection 1317h appl np E 0 31 0 1 --- 7.12 - 46,7.12 - 47,7.12 - 50,
7.12 - 51,7.12 - 62,7.12 - 64,
7.12 - 70,7.12 - 72,7.12 - 73
7.3 - 12,7.12 - 29,7.12 - 35,
7.12 - 36,7.12 - 38,7.12 - 39
,7.12 - 41,7.12 - 42,7.12 - 44
,7.12 - 45,7.12 - 46,7.12 - 47
,7.12 - 49,7.12 - 50,7.12 - 51
PS.24 index position 1318h appl np --- -2^31 2^31-1 0 1 inc ,7.12 - 52,7.12 - 54,7.12 - 55
,7.12 - 56,7.12 - 57,7.12 - 58
,7.12 - 62,7.12 - 63,7.12 - 64
,7.12 - 68,7.12 - 69,7.12 - 70
,7.12 - 71,7.12 - 73,7.12 - 74
,7.12 - 75,7.12 - 77,7.12 - 78
5.1 - 4,7.12 - 12,7.12 - 28,7.
12 - 29,7.12 - 30,7.12 - 32,7
.12 - 35,7.12 - 36,7.12 - 37,7
n* .12 - 38,7.12 - 39,7.12 - 40,
PS.25 index speed 1319h appl np E n * -4000 n * 4000 0 rpm
0.125 7.12 - 41,7.12 - 42,7.12 - 43,
7.12 - 44,7.12 - 45,7.12 - 46,
7.12 - 47,7.12 - 48,7.12 - 50,
7.12 - 51,7.12 - 62,7.12 - 68
7.3 - 19,7.12 - 35,7.12 - 38,7
.12 - 39,7.12 - 41,7.12 - 42,7
PS.26 next index 131Ah appl np E -1: PS.28 31 -1: PS.28 1 --- .12 - 44,7.12 - 45,7.12 - 47,7
.12 - 50,7.12 - 51,7.12 - 62,7
.12 - 75
7.3 - 12,7.12 - 35,7.12 - 38,
7.12 - 39,7.12 - 40,7.12 - 41
,7.12 - 42,7.12 - 44,7.12 - 45
PS.27 index mode 131Bh appl np E 0 31 0 1 ---
,7.12 - 47,7.12 - 50,7.12 - 51
,7.12 - 52,7.12 - 56,7.12 - 57
,7.12 - 58,7.12 - 62,7.12 - 74
7.3 - 19,7.3 - 20,7.12 - 35,7.
12 - 39,7.12 - 40,7.12 - 41,7
.12 - 44,7.12 - 46,7.12 - 48,7
PS.28 start index new profile 131Ch appl P E 0 31 0 1 ---
.12 - 51,7.12 - 62,7.12 - 63,
7.12 - 64,7.12 - 69,7.12 - 72,
7.12 - 75
7.3 - 10,7.3 - 11,7.12 - 29,7.1
PS.29 start posi input selection 131Dh appl np E 0 4095 0 1 --- 2 - 44,7.12 - 46,7.12 - 48,7.1
2 - 49,7.12 - 63
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
7.3 - 19,7.12 - 48,7.12 - 71,7
PS.30 target window 131Eh appl np E 0 65535 1024 1 inc
.12 - 75,10.1 - 12
7.12 - 28,7.12 - 30,7.12 - 35,
7.12 - 36,7.12 - 38,7.12 - 39,
PS.31 max. speed setting % 131Fh appl np --- 0.0 100.0 100.0 0.1 %
7.12 - 40,7.12 - 46,7.12 - 48,
7.12 - 62,7.15 - 22
limit acceleration/decelerati-
PS.32 1320h appl np --- 25.0 100.0 100.0 0.1 % 7.12-65, 7.12-66, 7.12-74
on reducing %
source contouring mode po-
PS.33 1321h appl np E 0 7 0 1 --- 7.12-77, 7.12-78
sition
PS.34 contouring mode position 1322h appl np E -2^31 2^31-1 0 1 inc 7.12-77, 7.12-78, 7.12-79
PS.35 teach mode 1323h appl np --- 0 4 0 1 --- 7.12-72, 7.12-73
PS.36 teach input selection 1324h appl np E 0 4095 0 1 --- 7.3-10, 7.3-12
7.3-10, 7.3-12, 7.12-71,
PS.37 pos. scan index inp.sel. 1325h appl np E 0 4095 0 1 ---
7.12-72
7.3 - 10,7.3 - 12,7.12 - 35,7.
PS.38 rel. pos. f/r inp.sel. 1326h appl np E 0 4095 0 1 ---
12 - 39
7.12 - 51,7.12 - 52,7.12 - 53,
PS.39 position range 1327h appl np E 0 2^30-1 0 1 inc
7.12 - 54,7.12 - 55
7.3 - 20,7.12 - 56,7.12 - 62,
PS.40 reference point window 1328h appl np --- 0 2^30-1 0 1 inc
7.12 - 63,7.12 - 64,7.12 - 75
PS.41 reference position 0% 1329h appl np --- -2^30 2^30-1 0 1 inc 7.2-13, 7.12-70
PS.42 reference position 100% 132Ah appl np --- -2^30 2^30-1 -2^30 1 inc 7.2-13, 7.12-70
PS.43 corr. ref. point inp.sel. 132Bh appl np E 0 4095 0 1 --- 7.3-10, 7.3-12
limit acceleration/decelerati- 7.12 - 59,7.12 - 60,7.12 - 61,
PS.44 132Ch appl np --- 25.0 100.0 100.0 0.1 %
on correction % 7.12 - 66
PS.45 index selection correction 132Dh appl np E 0 31 0 1 --- 7.12-64
7.12 - 38,7.12 - 58,7.12 - 59,
PS.46 rel. corr. switch for 132Eh appl np E 0 2^30-1 0 1 inc
7.12 - 62,7.12 - 63,7.12 - 64
7.12 - 38,7.12 - 58,7.12 - 59,
PS.47 rel. corr. switch rev 132Fh appl np E 0 2^30-1 0 1 inc
7.12 - 62,7.12 - 63
Automatically execution posi-
PS.52 1334h appl np --- 0: off 1: on 0: off 1 --- 7.12-69
tioning after STOP
PS.53 distance for no abort 1335h appl P --- 0 2^30-1 0 1 inc 7.12-66
PS.55 play of gear 1337h appl P E -2^31 2^31-1 -2^30 1 inc 7.12
PS.56 position target source 1338h appl np E 0 5 0 1 ---
PS.57 positon target input selection 1339h appl np E 0 4095 0 1 ---
PS.58 teach index selection 133Ah appl np E 0 31 0 1 ---
PS.59 teach index position 133Bh appl np --- -2^31 2^31-1 0 1 inc
PS.61 pre speed corr.correction 133Dh appl np E 0 1 0 1 ---
PS.64 spline position PT1-time 1340h appl np --- 0 65535 0 1 ms
rG.00 register mode 1700h appl P E 0 255 0 1 --- 7.15-23, 7.15-24, 7.15-25
register max. gear change
rG.01 1701h appl np E 0.0 100.0 1.0 0.1 % 7.15-25
per pulse
register max. angle change
rG.02 1702h appl np E 0 2^30-1 0 1 inc 7.15-25
per pulse
register difference time angle
rG.03 1703h appl P E 0.000 (2^31-1)/8 5.000 0.125 ms 7.15-25
correction
register master input selec-
rG.04 1704h appl np E 0 4095 0 1 --- 7.15-25
tion
rG.05 register slave input selection 1705h appl np E 0 4095 0 1 --- 7.15-25 10
rG.06 register ratio master 1706h appl np E 0 15 1 1 --- 7.15-25
rG.07 register ration slave 1707h appl np E 0 15 1 1 --- 7.15-25
rG.08 register angle level 1 1708h appl np E -2^30 2^30-1 0 1 inc 7.15-24, 7.15-25
n*
rG.09 min. speed for level 1 1709h appl np --- n * -4000 n * 4000 0 rpm 7.15-24, 7.15-25
0.125
rG.10 register angle level 2 170Ah appl np E -2^30 2^30-1 0 1 inc 7.15-24, 7.15-25 11
n*
rG.11 min. speed for level 2 170Bh appl np --- n * -4000 n * 4000 0 rpm 7.15-24, 7.15-25
0.125
rG.14 register distance master 170Eh RO np --- -2^31 2^31-1 0 1 inc 7.15-24
rG.15 register distance slave 170Fh RO np --- -2^31 2^31-1 0 1 inc 7.15-24, 7.15-25
register difference distance
rG.16 1710h RO np --- -2^31 2^31-1 0 1 inc
m/s
rG.17 register time master 1711h RO np --- 0.000 12500 0.000 0.125 ms
rG.18 register time slave 1712h RO np --- 0.000 (2^31-1)/8 0.000 0.125 ms
rG.19 register difference time m/s 1713h RO np --- -2^31/8 (2^31-1)/8 0.000 0.125 ms
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
7.1-3, 7.1-7, 7.3-2, 7.3-3,
ru.22 internal input state 0216h RO np --- 0 4095 0 1 --- 7.3-5, 7.3-19, 7.3-20, 10.1-
13
ru.23 output condition state 0217h RO np --- 0 255 0 1 --- 7.1-3, 7.1-8, 7.3-15, 7.3-16
ru.24 state of output flags 0-7 0218h RO np --- 0 255 0 1 --- 7.1-3, 7.1-8, 7.3-15, 7.3-16
7.1 - 3,7.1 - 9,7.3 - 15,7.3 - 1
ru.25 output terminal state 0219h RO np --- 0 255 0 1 ---
6,7.3 - 25,10.1 - 13
7.1 - 3,7.1 - 9,7.13 - 29,7.1
ru.26 active parameter set 021Ah RO np --- 0 7 0 1 ---
5 - 8
7.1 - 3,7.1 - 9,7.2 - 3,7.2 - 11,
ru.27 AN1 pre amplifier disp. 021Bh RO np --- -100.0 100.0 0 0.1 % 7.2 - 13,7.15 - 30,7.15 - 31,7
.15 - 32
7.1 - 3,7.1 - 9,7.2 - 3,7.2 - 10,
7.2 - 11,7.2 - 13,7.15 - 19,7.1
ru.28 AN1 post amplifier disp. 021Ch RO np --- -400.0 400.0 0 0.1 %
5 - 21,7.15 - 28,7.15 - 29,7.1
5 - 30,7.15 - 31,7.15 - 32
7.1 - 3,7.1 - 9,7.2 - 3,7.2 - 11,
ru.29 AN2 pre amplifier disp. 021Dh RO np --- -100.0 100.0 0 0.1 % 7.2 - 13,7.15 - 30,7.15 - 31,7
.15 - 32
7.1 - 3,7.1 - 10,7.2 - 3,7.2 - 1
0,7.2 - 11,7.2 - 13,7.15 - 19,
ru.30 AN2 post amplifier disp. 021Eh RO np --- -400.0 400.0 0 0.1 %
7.15 - 21,7.15 - 28,7.15 - 29,
7.15 - 30,7.15 - 31,7.15 - 32
ru.31 AN3 pre amplifier disp. 021Fh RO np --- -100.0 100.0 0 0.1 % 7.1-3, 7.1-10, 7.2-3
7.1 - 3,7.1 - 10,7.2 - 3,7.2 - 1
ru.32 AN3 post amplifier disp. 0220h RO np --- -400.0 400.0 0 0.1 % 0,7.15 - 19,7.15 - 21,7.15 - 2
8,7.15 - 29
ru.33 ANOUT1 pre ampl. disp. 0221h RO np --- -400.0 400.0 0 0.1 % 7.1-3, 7.1-10, 7.2-12
7.1-3, 7.1-10, 7.2-10, 7.2-
ru.34 ANOUT1 post ampl. disp. 0222h RO np --- -115.0 115.0 0 0.1 %
12, 7.3-20
ru.35 ANOUT2 pre ampl. disp. 0223h RO np --- -400.0 400.0 0 0.1 % 7.1-3, 7.1-10, 7.2-12
7.1-3, 7.1-10, 7.2-10, 7.3-
ru.36 ANOUT2 post ampl. disp. 0224h RO np --- -115.0 115.0 0 0.1 %
20
7.1 - 3,7.1 - 11,7.2 - 10,7.4 -
ru.37 motorpoti actual value 0225h RO np --- -100.00 100.00 0 0.01 % 4,7.8 - 13,7.9 - 3,7.11 - 16,7.
15 - 7,7.15 - 9
7.1-3, 7.1-11, 7.2-11, 7.2-
ru.38 power module temperature 0226h RO np --- 0 150 0 1 degree
13, 7.3-19
7.1-3, 7.1-11, 7.3-17, 7.13-
ru.39 OL counter display 0227h RO np --- 0 100 0 1 %
6
ru.40 power on counter 0228h sup np --- 0 65535 0 1 h 7.1-3, 7.1-11
ru.41 modulation on counter 0229h sup np --- 0 ru.40 0 1 h 7.1-3, 7.1-11
6.2 - 7,7.1 - 3,7.1 - 11,7.5 - 4,
ru.42 modulation grade 022Ah RO np --- 0 110 0 1 % 7.5 - 5,7.5 - 15,7.5 - 23,7.8 -
9,7.13 - 17
7.1 - 3,7.1 - 11,7.3 - 18,7.12
ru.43 timer 1 display 022Bh appl np --- 0 655.35 0 0.01 --- - 50,7.15 - 9,7.15 - 11,7.15
- 12
7.1 - 3,7.1 - 11,7.3 - 18,7.15 10
ru.44 timer 2 display 022Ch appl np --- 0 655.35 0 0.01 ---
- 9,7.15 - 11,7.15 - 12
ru.45 act. switching frequency 022Dh RO np --- 0 4 0 1 --- 7.1-3, 7.1-12, 7.10-5
7.1 - 3,7.1 - 12,7.1 - 17,7.2 -
ru.46 motor temperature 022Eh RO np --- 0 255 0 1 degree 11,7.2 - 13,7.3 - 19,7.13 - 17
,7.13 - 18 11
ru.47 actual torque limit motor 022Fh RO np --- -32000.00 32000.00 0 0.01 Nm 7.1-3, 7.1-12, 7.8-14
ru.48 actual torque limit generator 0230h RO np --- -32000.00 32000.00 0 0.01 Nm 7.1-3, 7.1-12, 7.8-14
ru.49 actual reference torque 0231h RO np --- -32000.00 32000.00 0 0.01 Nm 7.1-3, 7.1-12
ru.51 power module temperature 0233h RO np --- -40 120 0 1 °C 7.1-3, 7.1-12,
7.1 - 3,7.1 - 12,7.2 - 10,7.2 -
ru.52 ext. PID out dislay 0234h RO np --- -400.0 400.0 0 0.1 % 11,7.2 - 13,7.4 - 4,7.15 - 26,
7.15 - 30,7.15 - 31,7.15 - 32
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
7.1 - 3,7.1 - 12,7.11 - 15,7.1
ru.53 AUX display 0235h RO np --- -400.0 400.0 0 0.1 % 2 - 46,7.15 - 19,7.15 - 21,7.1
5 - 28,7.15 - 29,7.15 - 30
7.1 - 3,7.1 - 13,7.2 - 11,7.2 -
13,7.3 - 12,7.3 - 19,7.3 - 20,
7.12 - 6,7.12 - 7,7.12 - 10,7.
12 - 11,7.12 - 12,7.12 - 14,7.
12 - 17,7.12 - 18,7.12 - 20,7.
12 - 22,7.12 - 23,7.12 - 26,7.
ru.54 actual position 0236h RO np --- -2^31 2^31-1 0 1 inc 12 - 29,7.12 - 34,7.12 - 35,7.
12 - 45,7.12 - 47,7.12 - 55,7.
12 - 56,7.12 - 57,7.12 - 58,7.
12 - 62,7.12 - 65,7.12 - 71,7.
12 - 72,7.12 - 73,7.12 - 74,7.
12 - 75,7.12 - 78,7.12 - 79,7.
12 - 80,7.12 - 81,7.13 - 8
7.1 - 3,7.2 - 11,7.2 - 13,7.3 -
12,7.3 - 19,7.12 - 14,7.12 - 1
7,7.12 - 18,7.12 - 20,7.12 -
22,7.12 - 23,7.12 - 25,7.12
- 26,7.12 - 28,7.12 - 29,7.1
ru.56 set position 0238h RO np --- -2^31 2^31-1 0 1 inc 2 - 34,7.12 - 35,7.12 - 36,7.1
2 - 43,7.12 - 50,7.12 - 53,7.1
2 - 54,7.12 - 55,7.12 - 58,7.1
2 - 59,7.12 - 60,7.12 - 61,7.1
2 - 67,7.12 - 71,7.12 - 74,7.1
2 - 75,7.12 - 76,7.12 - 81
7.1 - 3,7.1 - 13,7.2 - 11,7.2 -
ru.58 angle difference 023Ah RO np --- -2^31 2^31-1 0 1 inc 13,7.3 - 18,7.12 - 35,7.12 - 7
4,7.12 - 75,7.12 - 81
7.1-3, 7.1-13, 7.5-4, 7.5-
ru.59 factor rotor adaption 023Bh RO np --- 0 200 0 1 %
33, 7.12-13
7.1 - 3,7.1 - 13,7.3 - 20,7.12
ru.60 actual position index 023Ch RO np --- 0 255 0 1 --- - 45,7.12 - 47,7.12 - 50,7.12
- 74,7.12 - 75
7.1 - 3,7.1 - 13,7.3 - 19,7.12 -
28,7.12 - 29,7.12 - 34,7.12 - 3
6,7.12 - 43,7.12 - 45,7.12 - 47
,7.12 - 50,7.12 - 53,7.12 - 54,
ru.61 target position 023Dh RO np --- -2^31 2^31-1 0 1 inc
7.12 - 55,7.12 - 57,7.12 - 58,7
.12 - 59,7.12 - 61,7.12 - 67,7.
12 - 68,7.12 - 71,7.12 - 74,7.1
2 - 75,7.12 - 76,10.1 - 12
5.1 - 4,7.1 - 3,7.1 - 13,7.4 - 4,
n*
ru.63 profile speed 023Fh RO np --- n * -4000 n * 4000 0 rpm 7.12 - 28,7.12 - 46,7.12 - 47,
0.125
7.12 - 74,7.12 - 81
7.1-3, 7.1-13, 7.13-31,
ru.68 rated DC voltage 0244h RO np --- 0 1500 0 1 V
7.13-38
7.1 - 3,7.1 - 13,7.12 - 13,7.1
ru.69 distance ref.-zeropoint 0245h RO np --- -2^31 2^31-1 0 1 inc
2 - 58,7.12 - 59,7.12 - 74
7.1 - 3,7.1 - 13,7.3 - 12,7.3 -
ru.71 teach/scan position 0247h RO np --- -2^31 2^31-1 0 1 inc 20,7.12 - 72,7.12 - 73,7.12 -
74,7.12 - 75
ru.73 set torque in percent 0249h RO np --- -400.0 400.0 0 0.1 % 7.1-3, 7.1-14, 7.8-14
ru.74 actual torque in percent 024Ah RO np --- -400.0 400.0 0 0.1 % 7.1-3, 7.1-14, 7.8-14
actual value display in per-
ru.78 024Eh RO np --- -400.0 400.0 0 0.1 % 7.1-3, 7.1-14
cent
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
n* 5.1 - 4,7.1 - 3,7.1 - 14,7.4 - 1
ru.79 abs. speed value (EMK) 024Fh RO np --- n * -4000 n * 4000 0 rpm
0.125 4,7.6 - 4,7.8 - 7,7.13 - 10
7.1 - 3,7.1 - 14,7.3 - 15,7.3 -
ru.80 digital output state 0250h RO np --- 0 255 0 1 ---
16,7.3 - 25,10.1 - 11
7.1 - 3,7.1 - 14,7.2 - 11,7.2 -
ru.81 active power 0251h RO np --- -1000.00 1000.00 0.00 0.01 kW
13,7.3 - 20
ru.82 ramp. val. disp. high-res 0252h RO np --- -2^31 2^31-1 0 1 --- 7.1-3, 7.1-15, 7.4-7
ru.83 act. val. display high-res 0253h RO np --- -2^31 2^31-1 0 1 --- 7.1-3, 7.1-15
7.1-3, 7.1-15, 7.12-65,
ru.84 accessible rel. position 0254h RO np --- -2^31 2^31-1 0 1 inc
7.12-66, 7.12-74
n*
ru.85 peak encoder 1 speed 0255h appl np --- 0 n * 4095.875 0 rpm 5.1-4, 7.1-3, 7.1-15
0.125
n*
ru.86 peak encoder 2 speed 0256h appl np --- 0 n * 4095.875 0 rpm 5.1-4, 7.1-3, 7.1-15
0.125
7.1 - 3,7.1 - 15,7.5 - 25,7.5 -
ru.87 magnetizing current 0257h RO np --- -3276.7 3276.7 0 0.1 A
34,7.5 - 35
n*
ru.89 actual source speed 0259h RO np --- n * -4000 n * 4000 0 rpm 5.1-4, 7.1-3, 7.1-15
0.125
7.1 - 3,7.1 - 15,7.2 - 11,7.2 -
ru.90 max. torque in percent 025Ah RO np --- 0.00 400.00 0 0.01 % 13,7.8 - 3,7.8 - 14,7.8 - 15,7
.8 - 16
ru.91 energy over gtr 7 025Bh appl np --- 0 99999 0 1 KWh 7.1-15
ru.92 input power 025Ch RO np --- -1000.00 1000.00 0.00 0.01 kW 7.1-15
ru.93 power loss 025Dh RO np --- -1000.00 1000.00 0.00 0.01 kW 7.1-15
Sy.02 inverter identifier 0002h cp-ro np --- identifier identifier identifier 1 hex 7.1-4, 7.1-19
Sy.03 power unit code 0003h cp-ro np E 1 255 LTK 1 --- 7.1-4, 7.1-19, 8.1-7
Sy.04 cfg. data sel.data 0004h cp-ro np --- 0 24 0 1 --- 7.1-19
Sy.05 cfg. data 0005h RO np --- -32727 32767 0 1 --- 7.1-19
Sy.06 inverter address 0006h appl np E 0 239 1 1 --- 7.1-4, 7.1-19, 10.1-9
Sy.07 baud rate ext. bus 0007h appl np E 0 6 3 1 --- 7.1-4, 7.1-19, 10.1-9
7.1 - 4,7.1 - 20,7.12 - 77,7.1
Sy.08 bus synchronous time 0008h cp-ro np --- 0: off 65000 0: off 1 µs
2 - 80
7.1 - 4,7.1 - 20,7.3 - 21,7.12
Sy.09 HSP5 watchdog time 0009h cp-ro np E 0.00: off 10.00 0.00: off 0.01 s
- 80,7.13 - 7,10.1 - 10
Sy.10 F5-B; F5-G; F5-M 000Ah RO np --- 0 0 0 1 ---
Sy.11 baud rate int. bus 000Bh cp-ro np E 3 11 5 1 --- 7.1-4, 7.1-20, 10.1-9
Sy.12 message para. 1 defin. 000Ch cp-ro np --- -1: off 7FFFH -1: off 1 hex
Sy.13 message parameter 1 set 000Dh cp-ro np --- 1 128 1 1 ---
Sy.14 message para. 2 defin. 000Eh cp-ro np E -1: off 7FFFH -1: off 1 hex
Sy.15 message parameter 2 set 000Fh cp-ro np E 1 128 1 1 ---
Sy.16 proc. read data 1 defin. 0010h appl np E -1: off 7FFFH -1: off 1 hex 7.12-78, 7.12-80
Sy.17 proc. read data 1 set 0011h appl np E 1 128 1 1 --- 7.12-80 , 7.16-12
Sy.18 proc. read data 2 defin. 0012h appl np E -1: off 7FFFH -1: off 1 hex 7.12-78, 7.12-80
Sy.19 proc. read data 2 set 0013h appl np E 1 128 1 1 --- 7.12-80
Sy.20 proc. read data 3 defin. 0014h appl np E -1: off 7FFFH -1: off 1 hex 7.12-80
Sy.21 proc. read data 3 set 0015h appl np E 1 128 1 1 --- 7.12-80
Sy.22 proc. read data 4 defin. 0016h appl np E -1: off 7FFFH -1: off 1 hex
Sy.23 proc. read data 4 set 0017h appl np E 1 128 1 1 ---
Sy.24 proc. write data 1 def. 0018h appl np E -1: off 7FFFH -1: off 1 hex 7.12-80
Sy.25 proc. write data 1 set 0019h appl np E 1 255 255 1 --- 7.12-80
Sy.26 proc. write data 2 def. 001Ah appl np E -1: off 7FFFH -1: off 1 hex 7.12-80
Sy.27 proc. write data 2 set 001Bh appl np E 1 255 255 1 --- 7.12-80
Sy.28 proc. write data 3 def. 001Ch appl np E -1: off 7FFFH -1: off 1 hex 7.12-80 10
Sy.29 proc. write data 3 set 001Dh appl np E 1 255 255 1 --- 7.12-81
Sy.30 proc. write data 4 def. 001Eh appl np E -1: off 7FFFH -1: off 1 hex
Sy.31 proc. write data 4 set 001Fh appl np E 1 255 255 1 ---
Sy.32 scope timer 0020h RO np --- 0 65535 Sy.32 1 --- 7.1-4, 7.1-20
Sy.33 scope data 1 defin. 0021h cp-ro np --- -1: off 7FFFH -1: off 1 hex
SY.34
Sy.35
scope data 1 set
scope data 2 defin.
0022h
0023h
cp-ro
cp-ro
np
np
---
---
1
-1: off
128
7FFFH
1
-1: off
1
1
--- 7.16-4
hex
11
Sy.36 scope data 2 set 0024h cp-ro np --- 1 128 1 1 ---
Sy.37 scope data 3 defin. 0025h cp-ro np --- -1: off 7FFFH -1: off 1 hex
Sy.38 scope data 3 set 0026h cp-ro np --- 1 128 1 1 ---
Sy.39 scope data 4 defin. 0027h cp-ro np --- -1: off 7FFFH -1: off 1 hex
Sy.40 scope data 4 set 0028h cp-ro np --- 1 128 1 1 ---
SY.41 control word (high) 0029h appl np E 0 65535 0 1 hex 7.1-4, 7.1-21, 10.1-11
SY.42 status word (high) 002Ah RO np --- 0 65535 0 1 hex 7.1-4, 7.1-21, 10.1-13
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
7.1 - 4,7.1 - 21,7.3 - 12,7.3 -
14,7.12 - 18,7.12 - 27,7.12
SY.43 control word (long) 002Bh appl np E -2^31 2^311 0 1 hex
- 63,7.13 - 15,7.13 - 20,10.
1 - 11
7.1 - 4,7.1 - 21,7.13 - 16,7.1
SY.44 status word (long) 002Ch RO np --- -2^31 2^311 0 1 hex
3 - 18,10.1 - 13
n*
Sy.45 drive mode reference value 002Dh appl np --- 0 n * 4095 n * 1500 rpm 5.1-4
0.125
Sy.46 drive mode rotation 002Eh appl np --- 0 15 0 1 hex
7.1 - 4,7.1 - 21,7.3 - 12,7.3 -
14,7.4 - 7,7.4 - 11,7.4 - 12,7
.12 - 18,7.12 - 19,7.12 - 27,
SY.50 control word (low) 0032h appl np E 0 65535 0 1 hex
7.12 - 63,7.12 - 79,7.12 - 80
,7.13 - 14,7.13 - 15,7.13 - 1,
7.13 - 20,10.1 - 10,10.1 - 11
7.1-4, 7.1-20, 7.1-21,
Sy. 51 status word (low) 0033h RO np --- 0 65535 0 1 hex 7.3-19, 7.13-15, 7.13-17,
7.13-38, 10.1-12, 10.1-13
-32000;
32000; 64000; 5.1-4, 7.1-4, 7.1-21, 7.4-4,
SY.52 set speed value 0034h appl np --- -64000; 0 1; 2; 4 rpm
128000 10.1-13
-128000
-32000;
32000; 64000; 5.1-4, 7.1-4, 7.1-21, 10.1-
SY.53 actual speed value 0035h RO np --- -64000; 0 1; 2; 4 rpm
128000 13
-128000
Sy.54 message time stamps 0036h cp-ro np --- 0 255 0 1 hex
Sy.56 start display address 0038h cp-ro np E 0 7FFFH 0209h 1 hex 7.1-4, 7.1-22
Sy.57 watchdog time address 0039h cp-ro np --- -2 -1 -2 1 hex
Sy.58 proc. read data 5 defin. 003Ah appl np E -1 7FFFH -1 1 hex
Sy.59 proc. read data 5 set 003Bh appl np E 1 128 1 1 ---
Sy.60 proc. read data 6 defin. 003Ch appl np E -1 7FFFH -1 1 hex
Sy.61 proc. read data 6 set 003Dh appl np E 1 128 1 1 ---
Sy.62 proc. read data 7 defin. 003Eh appl np E -1 7FFFH -1 1 hex
SY.63 proc. read data 7 set 003Fh appl np E 1 128 1 1 ---
SY.64 proc. read data 8 defin. 0040h appl np E -1 7FFFH -1 1 hex
Sy.65 proc. read data 8 set 0041h appl np E 1 128 1 1 ---
Sy.66 proc. write data 5 def. 0042h appl np E -1 7FFFH -1 1 hex
Sy.67 Proc. write data 5 set 0043h appl np E 1 255 255 1 ---
Sy.68 proc. write data 6 def. 0044h appl np E -1 7FFFH -1 1 hex
Sy.69 Proc. write data 6 set 0045h appl np E 1 255 255 1 ---
Sy.70 proc. write data 7 def. 0046h appl np E -1 7FFFH -1 1 hex
Sy.71 Proc. write data 7 set 0047h appl np E 1 255 255 1 ---
Sy.72 proc. write data 8 def. 0048h appl np E -1 7FFFH -1 1 hex
Sy.73 Proc. write data 8 set 0049h appl np E 1 255 255 1 ---
Sy.74 Proc. data 1-4 size 004Ah appl np E 0 65535 0 1 hex
Sy.75 Proc. data 5-8 size 004Bh appl np E 0 65535 0 1 hex
SY.77 Control word S4 004Dh appl np E 0 65535 0 1 hex
SY.78 Status word S4 004Eh appl np --- 0 65535 0 1 hex
SY.79 Status word 1 PROFIdrive 004Fh appl np --- 0 65535 0 1 hex 7.1-22
SY.80 Status word 2 PROFIdrive 0050h appl np --- 0 65535 0 1 hex 7.1-22
ud. 01 Password 0801h cp-ro np o.P. 0 9999 Application 1 --- 6.2-3, 7.16-12
ud.02 Control type 0802h appl np E 0 15 0 1 --- 5.1 - 5,6.2 - 9,7.1 - 5,7.2 - 10
,7.2 - 11,7.2 - 12,7.4 - 4,7.4 -
5,7.4 - 6,7.4 - 18,7.4 - 21,7.4
ud.02 Control type 0802h appl np E 0 15 8 1 ---
- 22,7.4 - 24,7.12 - 81,7.15 -
5,7.15 - 28
ud. 04 Auto store state 0804h appl np --- 0: off 1: on 1: on 1 ---
ud. 05 auto store 0805h appl np --- 0: off 2 1: on 1 --- 10.1-10
ud. 09 drive-mode-control 0809h appl np --- 0 11 0 1 ---
7.16 - 3,7.16 - 4,7.16 - 6,7.1
ud.15 CP selector 080Fh appl np E 1 36 1 1 ---
6 - 9
7.16 - 3,7.16 - 4,7.16 - 6,7.1
ud.16 cp address 0810h appl np E -1: off 7FFFH CP def. 1 hex
6 - 9
ud.17 cp set norm 0811h appl np E 1 8191 1 1 --- 7.16-3, 7.16-6
further on next side
Parameter Addr. R P E Lower limit Upper limit Default Step Unit See on page
7.16 - 5,7.16 - 6,7.16 - 7,7.1
ud.18 divisor display norm 0812h appl P E -32767 32767 1 1 ---
6 - 8,7.16 - 9
ud.19 multiplier display norm 0813h appl P E -32767 32767 1 1 --- 7.16-7, 7.16-8, 7.16-9
ud.20 offset display norm 0814h appl P E -32767 32767 0 1 --- 7.16-7, 7.16-8, 7.16-9
ud.21 ctrl. display norm 0815h appl P E 0 1791 0 1 --- 7.16-6, 7.16-7, 7.16-8
ud. 22 PP selector 0816h appl np E 0 47 0 1 --- 7.16-12
ud.23 PP address 0817h appl np E -1: off 7FFFH -1: off 1 hex 7.16-12
ud.24 PP properties 0818h appl np E 1 2^20-1 1 1 --- 7.16-4, 7.16-12
ud. 25 PP write multiplier 0819h appl np --- -32767 32767 1 1 ---
ud. 26 PP write shifter 081Ah appl np --- 0 48 0 1 ---
ud. 27 PP read multiplier 081Bh appl np --- -32767 32767 1 1 ---
ud. 28 PP read shifter 081Ch appl np --- 0 48 0 1 ---
ud.29 PP offset 081Dh appl np --- -2^31+1 2^31-1 0 1 ---
ud.30 PP upper limit 081Eh appl np --- -2^31+1 2^31-1 1 1 ---
ud. 31 PP lower limit 081Fh appl np --- -2^31+1 2^31-1 0 1 ---
11
1. Introduction
2. Summary
3. Hardware
4. Operation
Selection of Operating
5.
Mode
6. Initial Start-up
8. Error Assistance
9. Project Design
10. Networks
12. Annex 12
PS.26 7.3 - 35, 7.12 - 35, 7.12 - 39, Ramp 7.5 - 26, 7.5 - 34, 7.5 - 35 rG.15 7.15 - 24, 7.15 - 25
7.12 - 40, 7.12 - 43, 7.12 - 45, calculation...............................7.4 - 3 Rotation................................ 7.11 - 12
7.12 - 46, 7.12 - 48, 7.12 - 51, generator...............................7.4 - 16 selection..................................7.4 - 3
7.12 - 52, 7.12 - 63, 7.12 - 75 setting 7.4 - 3 RS232/485.............................10.1 - 3
PS.27 7.3 - 28, 7.12 - 35, 7.12 - 36, Ramp output..............7.5 - 34, 7.5 - 35 RS485 interface......................10.1 - 4
7.12 - 39, 7.12 - 40, 7.12 - 41, display 4.1 - 5, 7.3 - 32, 7.3 - 34, ru-Parameter............................7.1 - 5
7.12 - 43, 7.12 - 44, 7.12 - 45, 7.3 - 36, 7.3 - 37, 7.4 - 3, ru.00 6.2 - 4, 6.2 - 6, 6.2 - 8,
7.12 - 46, 7.12 - 48, 7.12 - 51, 7.4 - 4, 7.4 - 14, 7.4 - 21, 6.2 - 10, 6.2 - 11, 6.2 - 14,
7.12 - 52, 7.12 - 53, 7.12 - 57, 7.6 - 9, 7.8 - 13, 7.13 - 4, 6.2 - 16, 6.2 - 18, 6.2 - 19,
7.12 - 58, 7.12 - 59, 7.12 - 63, 7.8 - 15, 7.13 - 4, 7.13 - 11, 7.1 - 3, 7.1 - 5, 7.6 - 8,
7.12 - 74 7.13 - 34, 7.13 - 37, 7.13 - 38, 7.6 - 10, 7.12 - 4, 7.12 - 30,
PS.28 7.3 - 35, 7.3 - 36, 7.12 - 35, 7.13 - 40, 7.13 - 41, 7.15 - 4, 7.12 - 37, 7.12 - 69, 7.12 - 76,
7.12 - 41, 7.12 - 43, 7.12 - 45, 7.16 - 5, 7.15 - 9, 7.15 - 10, 7.13 - 4, 7.13 - 17, 7.13 - 18,
7.12 - 47, 7.12 - 49, 7.12 - 52, 7.15 - 11, 7.15 - 12, 7.15 - 13, 7.13 - 26, 7.13 - 40
7.12 - 63, 7.12 - 64, 7.12 - 66, 7.16 - 5, 7.16 - 6, 7.16 - 8, ru.01 5.1 - 4, 7.1 - 3, 7.1 - 5, 7.1 - 6,
7.12 - 70, 7.12 - 73, 7.12 - 75 7.16 - 9, 8.1 - 3, 10.1 - 6, 7.2 - 14, 7.3 - 34, 7.4 - 3,
PS.29 7.3 - 26, 7.3 - 27, 7.12 - 29, 10.1 - 8, 10.1 - 12, 10.1 - 13, 7.4 - 21, 7.13 - 41, 7.15 - 4,
7.12 - 45, 7.12 - 47, 7.12 - 49, 7.6 - 23 7.15 - 14, 7.15 - 15, 7.15 - 31,
7.12 - 50, 7.12 - 64 Ramp time..............................7.5 - 26 7.15 - 32, 7.15 - 33, 10.1 - 13
PS.30 7.3 - 35, 7.12 - 49, 7.12 - 71, ratio 7.11 - 14, 7.11 - 15, 7.11 - 25 ru.02 5.1 - 4, 7.1 - 3, 7.1 - 5,
7.12 - 72, 7.12 - 75, 10.1 - 13 re 7.5 - 30 7.2 - 14, 7.3 - 34, 7.3 - 35,
PS.31 7.12 - 28, 7.12 - 30, 7.12 - 36, start 7.15 - 18 7.3 - 37, 7.4 - 3, 7.4 - 21,
7.12 - 39, 7.12 - 40, 7.12 - 42, reduction..........7.5 - 2, 7.5 - 7, 7.5 - 14, 7.6 - 15, 7.9 - 4, 7.12 - 7,
7.12 - 47, 7.12 - 48, 7.12 - 63, 7.5 - 15 7.12 - 10, 7.12 - 12, 7.12 - 14,
7.15 - 23 Reference 7.12 - 29, 7.12 - 44, 7.12 - 48,
PS.32 7.12 - 66, 7.12 - 67, 7.12 - 68, sources................................7.15 - 29 7.12 - 54, 7.12 - 55, 7.12 - 62,
7.12 - 74 Register function..................7.15 - 23 7.12 - 66, 7.12 - 67, 7.12 - 68,
PS.33 7.12 - 78, 7.12 - 79 reset 7.5 - 13, 7.5 - 29, 7.11 - 11, 7.12 - 69, 7.12 - 73, 7.12 - 82,
PS.34 7.12 - 78, 7.12 - 79, 7.12 - 80, 7.11 - 22, 7.11 - 23, 7.11 - 24, 7.13 - 23, 7.13 - 26, 7.15 - 4,
7.12 - 81 7.11 - 25 7.15 - 22, 7.15 - 27, 7.15 - 28,
PS.35 7.12 - 73, 7.12 - 74 Error messages................................ 7.15 - 31, 7.15 - 32, 7.15 - 33
PS.36 7.3 - 26, 7.3 - 28 4.1 - 5, 6.2 - 3, 6.2 - 8, 6.2 - 11, ru.03 5.1 - 4, 7.1 - 3, 7.1 - 6, 7.4 - 3,
PS.37 7.3 - 26, 7.3 - 28, 7.12 - 72 6.2 - 15, 6.2 - 18, 7.1 - 18, 7.13 - 25, 7.13 - 26, 7.15 - 4,
PS.38 7.3 - 26, 7.3 - 28, 7.12 - 35, 7.13 - 4, 8.1 - 3, 8.1 - 5 7.15 - 5, 7.15 - 28
7.12 - 40 peak values.............................4.1 - 5 ru.04 7.2 - 11
PS.39 7.12 - 52, 7.12 - 53, 7.12 - 54, Reset 3.1-3, 7.5 - 12 ru.05 7.2 - 11
7.12 - 55, 7.12 - 56 resistance................................7.5 - 2, ru.06 5.1 - 4
PS.40 7.3 - 36, 7.12 - 57, 7.12 - 63, 7.5 - 3, 7.5 - 8, 7.5 - 9, 7.5 - 15, ru.07 5.1 - 4, 7.1 - 3, 7.1 - 6,
7.12 - 64, 7.12 - 75 7.5 - 16, 7.5 - 17, 7.5 - 18, 7.1 - 14, 7.2 - 11, 7.2 - 14,
PS.41 7.2 - 14, 7.12 - 71 7.5 - 19, 7.5 - 20, 7.5 - 25, 7.3 - 34, 7.3 - 37, 7.4 - 14,
PS.42 7.2 - 14, 7.12 - 71 7.5 - 31, 7.5 - 32, 7.5 - 33, 7.6 - 17, 7.6 - 18, 7.13 - 11,
PS.43 7.3 - 26, 7.3 - 28 7.11 - 24 7.13 - 41, 7.15 - 5, 10.1 - 13
PS.44 7.12 - 60, 7.12 - 61, 7.12 - 62, rG-Parameter ru.09 5.1 - 4, 7.1 - 3, 7.1 - 6, 7.2 - 11,
7.12 - 67 rG.00 7.15 - 24, 7.15 - 25 7.3 - 37, 7.6 - 8
PS.45 7.12 - 65 rG.01 7.15 - 26 ru.10 5.1 - 4, 7.1 - 3, 7.1 - 6, 7.2 - 11,
PS.46 7.12 - 39, 7.12 - 59, 7.12 - 60, rG.02 7.15 - 24, 7.15 - 26 7.3 - 37, 7.6 - 8, 7.12 - 14,
7.12 - 63, 7.12 - 64 rG.03 7.15 - 26 7.12 - 82
PS.47 7.12 - 39, 7.12 - 59, 7.12 - 60, rG.04 7.15 - 26 ru.11 7.1 - 3, 7.1 - 6, 7.1 - 14,
7.12 - 63, 7.12 - 64 rG.05 7.15 - 26 7.2 - 12, 7.8 - 13
PS.52 7.12 - 70 rG.06 7.15 - 26 ru.12 7.1 - 3, 7.1 - 6, 7.1 - 14,
PS.53 7.12 - 67, 7.12 - 68 rG.07 7.15 - 26 7.2 - 12, 7.6 - 11, 7.6 - 13,
PTC 7.13 - 9 rG.08 7.15 - 24, 7.15 - 25, 7.15 - 26 7.6 - 18, 7.8 - 13, 7.8 - 14,
rG.09 7.15 - 24, 7.15 - 25, 7.15 - 26 7.8 - 15
Q rG.10 7.15 - 24, 7.15 - 25, 7.15 - 26 ru.13 7.1 - 3, 7.1 - 6, 7.3 - 34,
rG.11 7.15 - 24, 7.15 - 25, 7.15 - 26 7.10 - 5, 7.13 - 26, 7.15 - 30
R rG.14 7.15 - 24 ru.14 7.1 - 3, 7.1 - 6
ru.15 7.1 - 6, 7.2 - 12, 7.2 - 13, 7.9 - 3, 7.1 - 3, 7.15 - 7 ru.57 7.8 - 13, 7.9 - 3
7.2 - 14, 7.3 - 35, 7.13 - 20, ru.38 7.1 - 11, 7.2 - 12, 7.2 - 14, ru.58 7.2 - 12, 7.1 - 13, 7.2 - 13,
7.1 - 3, 7.13 - 25, 7.13 - 27, 7.3 - 34, 7.1 - 3, 7.3 - 34 7.2 - 14, 7.3 - 34, 7.1 - 3,
7.13 - 28, 7.13 - 30 ru.39 7.1 - 11, 7.13 - 6, 7.1 - 3 7.12 - 74, 7.12 - 75
ru.16 7.1 - 3, 7.1 - 6 ru.40 7.1 - 3, 7.1 - 11 ru.59 7.1 - 13, 7.1 - 3
ru.17 7.1 - 7, 7.2 - 12, 7.2 - 12, ru.41 7.1 - 3, 7.1 - 11 ru.60 7.1 - 13, 7.12 - 46, 7.1 - 3,
7.2 - 13, 7.2 - 14, 7.6 - 18, ru.42 6.2 - 7, 7.8 - 8, 7.1 - 11, 7.12 - 45, 7.12 - 48, 7.12 - 74,
7.3 - 34, 7.3 - 37, 7.13 - 20, 7.13 - 18, 7.1 - 3 7.12 - 75
7.13 - 25, 7.15 - 30, 7.1 - 3, ru.43 7.1 - 11, 7.3 - 34, 7.12 - 51, ru.61 7.1 - 13, 7.3 - 35, 10.1 - 13,
7.15 - 30, 7.15 - 33 7.1 - 3, 7.15 - 10, 7.15 - 11, 7.12 - 28, 7.12 - 29, 7.12 - 35,
ru.18 7.1 - 7, 7.2 - 12, 7.2 - 13, 7.15 - 12 7.12 - 37, 7.12 - 44, 7.12 - 45,
7.2 - 14, 7.13 - 23, 7.13 - 27, ru.44 7.1 - 11, 7.15 - 10, 7.1 - 3, 7.12 - 46, 7.12 - 48, 7.12 - 54,
7.15 - 30, 7.1 - 3, 7.15 - 30 7.15 - 11, 7.15 - 12 7.12 - 55, 7.12 - 56, 7.12 - 58,
ru.19 7.1 - 3, 7.1 - 7 ru.45 7.1 - 12, 7.1 - 3 7.12 - 59, 7.12 - 60, 7.12 - 61,
ru.20 7.1 - 7, 7.2 - 12, 7.1 - 3, ru.46 7.2 - 12, 7.1 - 12, 7.1 - 18, 7.12 - 62, 7.12 - 68, 7.12 - 69,
7.2 - 14 7.2 - 14, 7.3 - 34, 7.13 - 18, 7.12 - 71, 7.12 - 74, 7.12 - 75,
ru.21 7.1 - 7, 7.1 - 8, 7.1 - 3, 7.3 - 34, 7.1 - 3, 7.13 - 9 7.1 - 3
7.3 - 19, 7.3 - 21 ru.47 7.1 - 12, 7.8 - 13, 7.1 - 3, ru.63 5.1 - 4, 7.4 - 5, 7.1 - 13,
ru.22 7.1 - 8, 7.3 - 19, 7.3 - 21, 7.8 - 13 7.4 - 4, 7.1 - 3, 7.12 - 28,
7.3 - 35, 7.3 - 36, 7.1 - 3 ru.48 7.1 - 12, 7.8 - 13, 7.1 - 3, 7.12 - 47, 7.12 - 48, 7.12 - 74,
ru.23 7.1 - 8, 7.1 - 3, 7.3 - 31 7.8 - 13 7.12 - 75
ru.24 7.1 - 8, 7.1 - 3, 7.3 - 31 ru.49 7.1 - 3, 7.1 - 12 ru.65 7.12 - 56
ru.25 7.1 - 9, 7.3 - 31, 7.3 - 32, ru.50 7.1 - 12 ru.68 7.1 - 13, 7.13 - 34, 7.1 - 3,
7.3 - 41, 7.1 - 3 ru.52 7.2 - 11, 7.1 - 12, 7.2 - 11, 7.13 - 34, 7.13 - 39
ru.26 7.1 - 9, 7.13 - 30, 7.13 - 27, 7.2 - 12, 7.2 - 13, 7.2 - 14, ru.69 7.1 - 13, 7.1 - 3, 7.12 - 13,
7.1 - 3, 7.15 - 8 7.4 - 4, 7.15 - 26, 7.1 - 3, 7.12 - 59, 7.12 - 74
ru.27 7.1 - 9, 7.2 - 12, 7.2 - 14, 7.15 - 31, 7.15 - 32, 7.15 - 33 ru.71 7.1 - 14, 7.3 - 28, 7.12 - 72,
7.15 - 31, 7.1 - 3, 7.15 - 32, ru.53 7.1 - 13, 7.12 - 47, 7.12 - 46, 7.1 - 3, 7.12 - 72, 7.12 - 73,
7.15 - 33 7.1 - 3, 7.15 - 21, 7.15 - 29, 7.12 - 74, 7.12 - 75
ru.28 7.1 - 9, 7.2 - 3, 7.2 - 11, 7.15 - 30, 7.15 - 32 ru.73 7.8 - 13, 7.1 - 14, 7.1 - 3
7.2 - 12, 7.2 - 13, 7.2 - 14, ru.54 7.2 - 12, 7.1 - 13, 7.2 - 13, ru.74 7.8 - 13, 7.1 - 14, 7.1 - 3
7.1 - 3, 7.15 - 21, 7.15 - 29, 7.2 - 14, 7.3 - 28, 7.3 - 35, ru.78 7.1 - 3, 7.1 - 14
7.15 - 30, 7.15 - 31, 7.15 - 32, 7.3 - 36, 7.12 - 6, 7.12 - 7, ru.79 5.1 - 4, 7.1 - 14, 7.4 - 14,
7.15 - 33 7.13 - 8, 7.12 - 10, 7.12 - 11, 7.6 - 4, 7.8 - 6, 7.13 - 11,
ru.29 7.1 - 9, 7.2 - 12, 7.2 - 14, 7.12 - 12, 7.12 - 14, 7.12 - 17, 7.1 - 3, 7.13 - 11
7.15 - 31, 7.1 - 3, 7.15 - 32, 7.12 - 18, 7.12 - 20, 7.12 - 22, ru.80 7.1 - 14, 10.1 - 12, 7.3 - 31,
7.15 - 33 7.12 - 23, 7.12 - 26, 7.12 - 29, 7.3 - 32, 7.3 - 41, 7.1 - 3
ru.30 7.1 - 10, 7.2 - 3, 7.2 - 11, 7.12 - 34, 7.12 - 35, 7.12 - 45, ru.81 7.2 - 12, 7.1 - 14, 7.2 - 13,
7.2 - 12, 7.2 - 13, 7.2 - 14, 7.12 - 46, 7.12 - 48, 7.12 - 56, 7.2 - 14, 7.1 - 3, 7.3 - 36
7.1 - 3, 7.15 - 21, 7.15 - 29, 7.12 - 58, 7.12 - 59, 7.12 - 66, ru.82 7.1 - 15, 7.1 - 3
7.15 - 30, 7.15 - 31, 7.15 - 32, 7.12 - 67, 7.12 - 72, 7.12 - 73, ru.83 7.1 - 3, 7.1 - 15
7.15 - 33 7.12 - 74, 7.12 - 75, 7.12 - 76, ru.84 7.1 - 15, 7.12 - 66, 7.1 - 3,
ru.31 7.1 - 10, 7.1 - 3 7.12 - 79, 7.12 - 80, 7.12 - 81, 7.12 - 66, 7.12 - 67, 7.12 - 74
ru.32 7.1 - 10, 7.2 - 3, 7.2 - 10, 7.1 - 3 ru.85 5.1 - 4, 7.1 - 3, 7.1 - 15
7.2 - 11, 7.1 - 3, 7.15 - 21, ru.55 7.1 - 13 ru.86 5.1 - 4, 7.1 - 3, 7.1 - 15
7.15 - 29, 7.15 - 30 ru.56 7.2 - 12, 7.2 - 13, 7.2 - 14, ru.87 7.1 - 3, 7.1 - 15
ru.33 7.1 - 10, 7.2 - 13, 7.1 - 3, 7.3 - 28, 7.3 - 35, 7.1 - 3, ru.89 5.1 - 4, 7.1 - 3, 7.1 - 15
7.2 - 13 7.12 - 14, 7.12 - 17, 7.12 - 18, ru.90 7.2 - 12, 7.1 - 15, 7.2 - 13,
ru.34 7.1 - 10, 7.2 - 11, 7.2 - 11, 7.12 - 20, 7.12 - 22, 7.12 - 23, 7.2 - 14, 7.1 - 3, 7.8 - 3,
7.2 - 12, 7.2 - 13, 7.1 - 3, 7.12 - 25, 7.12 - 26, 7.12 - 28, 7.8 - 14, 7.8 - 15
7.3 - 36 7.12 - 29, 7.12 - 34, 7.12 - 35, ru.92 7.1 - 16
ru.35 7.1 - 10, 7.2 - 13, 7.1 - 3 7.12 - 37, 7.12 - 43, 7.12 - 44, ru.93 7.1 - 16
ru.36 7.1 - 10, 7.2 - 11, 7.2 - 10, 7.12 - 54, 7.12 - 55, 7.12 - 56,
12
7.2 - 11, 7.1 - 3, 7.3 - 36 7.12 - 59, 7.12 - 60, 7.12 - 61, S
ru.37 7.2 - 11, 7.1 - 11, 7.2 - 10, 7.12 - 62, 7.12 - 68, 7.12 - 74, search 7.5 - 30
7.2 - 11, 7.8 - 12, 7.8 - 13, 7.12 - 75, 7.12 - 76
V
V/F characteristic control........7.15 - 5
V/Hz-characteristic.................7.15 - 6
Voltage 7.5 - 2, 7.5 - 5, 7.5 - 6, 7.5 - 7,
7.5 - 13, 7.5 - 21, 7.5 - 31,
7.5 - 32, 7.11 - 2, 7.11 - 8
reduction................................7.15 - 6
W
winding product....................7.15 - 20
write protection.........................4.2 - 3
12
X
KEB worldwide…
© KEB
Mat.No. 00F5AEA-K440
Rev. 1A
Date 10/2012