Kzckxvkzooor: Service
Kzckxvkzooor: Service
Kzckxvkzooor: Service
KZCKXVKZOOOR
SERVICE MANUAL
KURZWEIL
Kurzweil is a product line of Young Chang America, Inc.
P/N 92002000
WARNING - When using electric products, basic precautions should always be followed, including the following:
1. Read all of the Safety and Installation Instructions and Explanation of Graphic Symbols before using the
product
2. This product must be grounded. If it should malfunction or breakdown, grounding provides a path of least
resistance for electric current to reduce the risk of electric shock. This product is equipped with a power
supply cord having an equipment-grounding conductor and a grounding plug. The plug must be plugged into an
appropriate outlet which is properly installed and grounded in accordance with all local codes and ordinances.
DANGER - Improper connection of the equipment-grounding conductor can result in a risk of electric shock. Do
not modify the plug provided with the product - if it will not fit the outlet, have a proper outlet installed
by a qualified electrician. Do not use an adaptor which defeats the function of the equipment-grounding
conductor. If you are in doubt as to whether the product is properly grounded, check with a qualified
serviceman or electrician.
3. WARNING - this product is equipped with an AC input voltage selector. The voltage selector has been
factory set for the main supply voltage in the country where this unit was sold. The setting of the voltage
selector can be seen just to the right of the power input receptacle on the back of the product. Changing the
voltage selector may require the use of a different power supply cord or attachment plug, or both. The reduce
the risk of fire or electric shock, refer servicing to qualified maintenance personnel,
4. Do not use this product near water - for example, near a bathtub, washbowl, kitchen sink, in a wet
basement, or near a swimming pool, or the like.
5. This product should only be used with a stand or cart that is recommended by the manufacturer.
6. This product, either alone or in combination with an amplifier and speakers or headphones, may be capable
of producing sound levels that could cause permanent hearing loss. Do not operate for a long period of time
at a high volume level or at a level that is uncomfortable. If you experience any hearing loss or ringing in
the ears, you should consult an audiologist.
7. The product should be oriented such that its location or position does not interfere with its proper
ventilation
8. The product should be located away from heat sources such as radiators, heat registers, or other products
that produce heat.
9. The product should be connected to a power supply only of the type described in the operating instructions
or as marked on the product.
10. This product may be equipped with a polarized line plug (one blade wider than the other). This is a
safety feature. If you are unable to insert the plug into the outlet, contact an electrician to replace your
obsolete outlet. Do not defeat the safety purpose of the plug.
11. The power supply cord of the product should be unplugged from the outlet when left unused for a long
period of time. When unplugging the power supply cord, do not pull on the cord, but grasp it by the plug.
12. Care should be taken so that objects do not fall and liquids are not spilled into the enclosure through
openings.
13. The product should be serviced by qualified service personnel when:
A. The power supply cord or the plug has been damaged; or
B- Objects have fallen, or liquid has been spilled into the product; or
C. The product has been exposed to rain; or
D. The product does not appear to be operating normally or exhibits a marked change in performance; or
E. The product has been dropped, or the enclosure damaged.
14. Do not attempt to service the product beyond that described in the user maintenance instructions. All
other servicing should be referred to qualified service personnel.
15. WARNING - Do not place objects on the product's power supply cord, or place the product in a position
where anyone could trip over, walk on, or roll anything over cords of any type. Do not allow the product to
rest on or be installed over cords of any type. Improper installations of this type create the possibility of
a fire hazard and/or personal injury.
Important: When connecting this product to accessories and/or other equipment use only
high quality shielded cables. The cable (s) supplied with this instrument must be used.
Note: This instrximent has been tested and found to comply with the limits for a Class B
digital device, pursuant to Part 15 of the FCC Rules. These limits are designed to
provide reasonable protection against harmful interference in a residential
installation. This instrument generates, uses, and can radiate radio frequency energy
and, if not installed and used in accordance with the instructions, may cause harmful
interference to radio communications. However, there is no guarantee that interference
will not occur in a particular installation. If this instrument does cause harmful
interference to radio or television reception, which can be determined by turning the
instrument off and on, the user is encouraged to try to correct the interference by one
or more of the following measures:
NOTICE
This apparatus does not exceed the Class B limits for radio noise emmissions from
digital apparatus set out in the Radio Interference Regulations of the Canadian
Department of Communications.
AVIS
Le present appareil numerique n'emet pas de bruits radioelectriques depassant les
limites applicables aux appareils numeriques de la class B prescrites dans le Reglement
sur le brouillage radioelectrique edicte par le ministere des Communications du Canada.
© 1992 All rights reserved. Kurzweil is a product line of Young Chang Akki Co., LTD.
Kurzweil, K2000, and MIDIScope are trademarks of Young Chang Akki Co., LTD. The names
Akai, SIOOO, Apple, Macintosh, DX7, IBM, DOS, MS-DOS, and Roland are registered
trademarks of the respective companies using these names
1
SETUP 1
K2000 OPERATIONAL CHECKOUT 5
K2000 POWER-UP CHECK 11
K2000R OPERATIONAL CHECKOUT 17
2
K2000 DISASSEMBLY/ACCESS 1
K2000R DISASSEMBLY/ACCESS 7
3
K2000 MAINTENANCE 1
K2000/K2000R TROUBLESHOOTING 7
DIAGNOSTICS 11
K2000R MAINTENANCE 47
4
OPTIONS A
TECHNICAL THEORY 1
5
K2000 PARTS LIST 1
K2000R PARTS LISTS 9
S CHEMAT I CS / D I AGRAMS
00
IV
Version 2.0 software, which is part of the SMP-R sampling option, allows for
graphic waveform editing via the built-in 240 x 64 display. Features include
powerful sample editing and DSP functions (time con^ression/expansion, sample
rate conversion, sample mixing and splicing, normalizing gain, audible sample
cueing, cross fade looping, auto looping, fade in/out and more)
Version 2.0 software will also be available separately from the Sampling
Option Kit.
1 . ENVIRONMENTAL SPECIFICATIONS
Temperature ranges
1 . PHYSICAL SPECIFICATIONS
ELECTRICAL SPECIFICATIONS
AC supply: selectable; lOOV, 120V, 220V, 240V; .75 amps at 120 volts nominal
1 . 6 GLOSSARY OF TERMS
Memory Banks The K2000*s memory is divided into ten spaces where
you can store any object you edit. These spaces are
called banks. Each bank can hold up to 100 objects of
each type, so we refer to them as the 100s bank, the
200s bank, and so on. The ID of an object determines
in which bank it is to be stored. An object with an
ID of 399, for example, would be stored in the 300s
bank. ROM objects are stored in the Zeros and 100s
banks. RAM objects can be stored in any bank.
NL DSP Function Non-linear DSP functions like SHAPER and WRAP add
waveforms to those already present in a sound, while
linear DSP functions act upon the existing waveforms
without adding new ones.
Note State Any K2000 is either on or off; this is its note state.
Normally^ any given note's Note State switches on when
you strike the key for that note. It switches off
when you release the key, and any sustain controls you
may have applied to the note (Sustain or Sostenuto
pedal, etc.).
Program Editor The set of parameters that lets you modify the sound
of ROM or RAM programs. Enter the Program Editor by
pressing the EDIT button while in Program mode, or any
time the currently selected parameter has program as
its value.
10
11
WARlsflNG: Do not attempt to operate the K2000 on any AC voltage other than the one selected - extensive
damage to the unit may result!
1 . Turn the unit off and remove the AC line cord from the AC power
connector.
2 . Open the cover of the fuse/voltage select compartment using a
small blade screwdriver or similar tool.
3. Pull the voltage selector card straight out of the housing,
using the indicator pin (See Figure 2.1 below).
4. Orient selector card so the desired voltage is readable at the
bottom.
5. Orient the indicator pin to
point up when the desired
voltage is readable at the
bottom (See Figure 2.2 below)
6 . Insert the voltage selector
card into the housing with the
printed side of the card facing
's
^
towards the AC power connector,
and the edge marked with the Fuse block
desired voltage first.
7. If changing from lOOV or 120V
to 220V or 240V, or vice-versa,
change fuse(s) and fusing
arrangement
^ as described under ?
Figure 2.1 - Voltage Selector ^ ^ Removal
Cardio i
..
"Fuses”
8 . Replace compartment cover.
Verify the indicator pin shows the desired voltage.
2 2 . FUSES
WARNING: FOR CONTINUED PROTECTION AGAINST THE RISK OF FIRE, REPLACE FUSES
ONLY WITH THE TYPE AND RATING SPECIFIED.
Fuse Replacement:
NOTE: When preparing the unit for use on European or other 220
- 240V/50HZ AC lines, the fusing arrangement must be changed in
order to fully protect the instrioment.
2.3.1 User RAM in the D2000 may be expanded with the installation of
SIMMs (Single In-line Memory Modules) in pairs. SIMMs for the
K2000 must be 30 pin, 8 bit, and 120ns or faster, with maximum
dimensions not exceeding 3.5” wide x 2.2” high x .25** thick -
SIMMs with ICs on both sides are not supported. The following
memory sizes are supported:
1 . Turn the unit off and remove the AC line cord from the AC power
2.6)
connector.
2 . Open the unit as described
in Section 3, "Disassembly". ^
1. Plug in the power line cord, and turn the unit on.
2. After the K2000 has initialized, press the ”DISK” button.
3. Confirm the ”Samples ####K” figure displayed in the center of
:
the status bar corresponds to the total amount of user RAM just
installed.
NOTE: The ICs containing the K2000's software are static sensitive
devices. USE ALL STANDARD ANTI -STATIC PRECAUTIONS WHEN
HANDLING THESE DEVICES!
1. turn the unit off and remove the AC line cord from the AC power
connector.
2. Open the unit as described in Section 3.
3. Locate the old software ICs U3 and U6, on the Engine Board
(Figure 2.6 above). Remove the old ICs using an IC puller or
similar tool. Be careful not to damage the traces on the
board.
4. Carefully insert the new IC labeled ”LO** in the socket at U3,
and the new IC labeled "HI” in the socket at U6. Make certain
the ICs are oriented with pin 1 toward the front of the unit.
Pin 1 is on the end of the IC with the indentation.
5. Inspect for proper installation - make sure the ICs are firmly
seated into the sockets and that none of the pins are bent.
6. Blose the unit as per the instructions in Section 3.
7. Always HARD RESET the instrument after changing the software.
Note that all user data in RAM will be erased. A hard reset
can be performed on power-up by pressing and holding the "1”,
”2” and ”3” buttons on the numeric keypad.
• Plug the included foot pedal into the rear panel jack marked "Foot
Switch 1".
• Scroll through the Program list with the Alpha Wheel (the large
knob to the right of the display)
• Press the Setup mode button, then use the Alpha Wheel to select
tri~zone setups with independent programs, MIDI channels and
controls in each zone.
• Press the Quick Access mode button and use the numeric keys to
select from programmable banks of ten programs or setups.
• Insert a program library disk into the floppy drive. Press the
Disk mode button, then scroll through the files with the Alpha
Wheel. Press "Load", then press "OK" twice. Then check out the
programs numbered in the 200s.
2.6.1 This section will walk you through the hookup of the K2000. We *11
take a look at the rear panel, then describe the power, audio, and
other cable connections.
2 . 7 BEFORE STARTING. . .
2.7.1 Before connecting anything, make certain the K2000 is properly and
safely situated. It will rest securely on most standard keyboard
and instrument stands. Otherwise, it should rest on a hard flat
surface. In this case, it mast rest on its rubber feet, and NOT
on the bottom panel. NEVER block the ventilation openings on the
rear panel; doing so can cause overheating which will seriously
damage the K2000. To provide adequate ventilation, the rear panel
should be at least four inches from any vertical surface.
2.8.1 The K2000 runs on lOOV, 120V, 220V, or 240 Volt AC power at 50 -
60 Hz. The voltage level is set with a selector on the rear
panel. Your dealer should have set the voltage switch to match
the voltage in your area, but it is a good idea to check it.
Excessive voltages can seriously damage any electric device. If
the voltage selector is not properly set, have your dealer or
other qualified service personnel reset it before you use your
K2000.
2.8.2 Facing the rear panel of the K2000, the power connector is on the
left-hand side. Connect the power cord to the K2000 first, then
to a properly grounded three-prong power outlet. If your power
source does not have the standard three-prong outlet, you should
take the time to install a proper grounding system. This will
assure you of avoiding problems with audio hum, and will reduce
the risk of a shock hazard.
NOTE; Refer to Figure 2.8 below for details on rear panel connections,
VOLTAGE SELECTOR
POWER AND FUSE
SWITCH MIX — FOOTSwrrcH
L(I^NO) R
AC POWER
CONNECTOR pHONpo
rHUNtb -A B- CONTROL PEDAL
L R L R MIDI MIDI SCSI
IN OUT I
2.9.1 The K2000 features six audio jacks, enabling you to route the
K2000*s sounds through its own internal effects processor, through
an external effects device, or through combinations of the two.
Of course the K2000*s effects processor can be bypassed as well.
On the rear panel are a pair of stereo mix jacks (MIX L and R)
and four additional jacks - A (L and R) and B (L and R) These .
settings you make for parameters on the OUTPUT page in the Program
Editor, and on the Master mode page.
2.9.3 The simplest configuration would involve connecting the left and
right MIX outputs directly to a mixing board or keyboard
amplifier. If you're using only one input to your sound system
from the K2000, use the left MIX output for a mono signal. We'll
assume for now that you're using only the MIX outputs.
2 . 10.2 The signal is routed using the OUTPUT page in the Program Editor.
While in Program mode, press the EDIT button, then press one of
the <more> soft buttons until you see "OUTPUT” on the display's
bottom line. Press the corresponding soft button, and the OUTPUT
page will appear. Set the value of the Pair parameter to A(FX) if
the current layer's sound is to be routed through the effects
processor before appearing at the MIX outputs.
2.10.3 A value of A (EX) mean two things: first, the layer's sound will
appear with effects at the MIX outputs, and second, the layer's
sound will appear without effects at the Group A outputs if
standard audio cables are inserted into the Group A output jacks.
Set the value of the Pair parameter to B(DRY) if the layer's sound
is to appear without effects at the MIX outputs. It will also
appear at the Group B outputs if standard audio cables are
inserted into the Group B output jacks. Connecting audio cables
to the A or B outputs will remove part or all of the signal from
the MIX outputs. Be sure to set the output pair for every layer
in the program; each layer can be routed independently (use the
CHAN/Bank buttons to scroll through the layers) . You'll also want
to set the pan position for each layer while you're on the OUTPUT
page. Depending on the program you're working with, you'll have
several different options for panning. Refer to "THE OUTPUT PAGE”
in chapter 6 of the Musician's Guide for details of these
options. Also see "PANNER” in Chapter 14 of the Musician's Guide
to learn how to gain even more control over a layer's pan
position.
2 . 11.1 The K2000's four separate outputs are arranged in two pairs, A and
B. These are mono output jacks, and you'll normally plug a
standard audio cable into one or more of them. Depending on the
OUTPUT page settings you use for a given program's layers, they
can function as two stereo output pairs, or as four separate
outputs
2.11.2 The diagram below illustrates the wiring of the K2000's audio
system. From the sound engine, the audio signal can be routed to
either the A or B output group. If the signal is routed to Output
2.11.3 If the signal is routed to Output Group and audio cables are
plugged into the A outputs, the signal will pass directly to the A
outputs. If no cables are plugged into the A outputs, the signal
will pass from the sound engine to the global effects processor.
The left and right sides are mixed before passing to the effects
processor. The effects processor splits the signal into left and
right sides again after the effects are applied. Effects will be
applied to both sides of the signal unless the Out A->FX parameter
is set L Only, in which case only the left side will have effects
applied. In either case, the signal will pass to the MIX outputs.
2.11.5 If you are using the K2000 for recording and multi-timbral
sequencing, you will need to use the separate outputs to increase
the mix down options, and for adding outboard effects to your
sounds. The separate outputs give a great deal of added
flexibility.
2 . 12.1 It is important that you understand the effect on the MIX outputs
when using the separate outputs. When a standard audio cable is
connected to any of the separate outputs, every sound (or part of
a sound) that is assigned to the output connected will be removed
from the MIX outputs. For example, imagine a cable is plugged
into the left output of Group A. The left side of every program
layer assigned to Output Group A will disappear from the left side
of the MIX outputs, and will appear only at the left Group A
output. If a layer is panned all the way left, it will disappear
from the MIX outputs entirely. If it is centered, the left side
will disappear, but the right side will remain at the right MIX
output. A layer that is panned hard right will not be affected in
this case. IF you use all four separate outputs, the MIX outputs
will not carry any signal.
2 . 12.2 The pan settings made on each program’s OUTPUT page apply to the
separate outputs just as they do for the MIX outputs. Keep in
mind, however, the signal at the separate outputs will always be
dry; only the MIX outputs make use of the global effects
processor.
2.12.3 Once the output groups are assigned on the OUTPUT page, go the
Master mode to specify how the signal will appear at the MIX
outputs. On the Master mode page are three parameters which
affect the routing of all signals that are sent to the MIX
outputs. The settings for these parameters will affect each layer
in every program.
2.12.4 The parameters Out A->Mix and Out B->Mix determine the panning of
the audio signal at the mixed outs. If they are set to Stereo,
whatever panning is applied to each program (as set on the OUTPUT
page in the Program Editor) will show up at the MIX outputs.
2 , 12.6 The parameter Out A->FX is intended for special cases when you
want an additional separate output that does not go through the
K2000*s effects processor. A value of L Only bypasses the effects
processor for all sounds assigned to the right side of Output
Group A.
2.14.1 Plug the stereo end of a stereo insert cable into one of the
K2000*s separate outputs, then plug the send side of the split end
into an input on an outboard effects box. If you connect the
effects box's output to the return side of the split end, you've
created a loop that will send the K2000's output into the effects
box, and return it wet to the K2000. Since the loop returns the
signal to the same output jack on the K2000, the K2000's MIX
outputs act as if there were no cable at all plugged into the
separate output, and the signal will appear at one or both of the
MIX outputs. If you are using one of the Group A outputs, you can
apply the effect of both the outboard box and the K2000's global
effects processor to the corresponding side of the K2000's MIX
outputs. IF you do not wan to double the effects in this way, go
the Effects mode, set the FX mode parameter to Master, and set the
Wet/Dry Mix parameter to 0%. Also, you can set the Pan parameter
of programs routed to Output Group A fully right, then go to
Master mode and set the Out A->Mix parameter to mono, and the Out
A->FX parameter to L Only. This will send programs routed to
Output Group A to the MIX outputs without going through the
K2000's effects processor, but will still send the signal to both
the left and right MIX outputs.
2.15.1 Plug the stereo end of an insert cable into one of the K2000's
separate outputs, then connect the return side of the split end to
the audio output of another instrument. This will route the other
instrument's output to the K2000. You can leave the send side
unconnected. The output from the other instrument will appear at
the K2000's MIX outputs, along with any sounds from the K2000 that
are routed to the MIX outputs. If the insert cable is plugged
into the A LEFT output, the K2000's effects will be applied to the
signal, unless the Out A->FX parameter to L Only, in which case
the K2000's effects will not be added to the signal. If the
insert cable is plugged into the A RIGHT output, the K2000's
effects will be added regardless of the setting of the Out A->FX
parameter.
10
2.16.1 VISUAL:
• Inspect unit for signs of damage or omissions.
• Verify the correct power entry module is configured
for proper AC voltage.
• USA units should have white nylon dot in the 120
volt positions.
• When converting to another voltage; fuse value may
require change.
• (Refer to appendix 4 of Musicians Guide)
11
12
13
default settings.
2.17.8 MIDI JACK TEST:
• Feed a MIDI signal into the MIDI IN jack. This MIDI
signal should appear at the THRU jack.
• Play the keyboard and test MIDI OUT jack by sending
MIDI signal into another instrument.
• Note: you can display MIDI data on the LCD. Press
MASTER, then UTILITY, then MIDI, and read data on
the LCD. (See chapter 10 in the Musicians Guide.)
14
15
16
2 18
. K2000R STARTUP
2.18.1 Startup - a Quick Checklist
If hooking up new gear is familiar to you, and you just want to get going,
here is a quick description of all the basic things you need to cover to get
started with your K2000R. If you need more information, thorough descriptions
of each step follow.
• If placing on a flat surface, affix the four pads (included with the
rack mount brackets) to the bottom of the K2000R, one in each corner.
• If rack mounting, attach the rack mount brackets, one to each side, at
the front edge of the unit using the screws already on each side.
• Connect a MIDI cable from the MIDI Out port of your MIDI controller to
the MIDI In port of the K2000R.
• If you hear distortion, reduce the gain on your mixing board, or use the
pad if it has one.
• Scroll through the Program list with the Alpha Wheel (the large knob to
the right of the display)
• If you MIDI controller can transmit on more than one MIDI channel, check
out Setup mode by pressing the Setup mode button. Then use the Alpha
Wheel to select tri-zone setups with independent programs, MIDI channels
and controller assignments in each zone.
• Press the Quick Access mode button and use the numeric keys to select
form programmable banks of ten programs or setups.
• Insert a program library disk into the floppy drive. Press the Disk
mode button, then scroll through the files with the Alpha Wheel. Press
"Load,” then press "OK" twice. Then check out the programs numbered in
the 200s.
• If you do not hear any sound, review these steps, then check the
Troubleshooting section in Section 4.
17
surface. In this case, it mast rest on its rubber feet (supplied in the
packing carton), and NOT on the bottom panel. NEVER block the ventilation
openings on the bottom or rear panels; doing so can cause overheating and
serious damage. To provide adequate ventilation, the rear panel should be at
least four inches from any vertical surface.
Power
The K2000R runs on 100, 120, 220, or 240 volt AC power at 50 - 60 Hz. The
voltage level is set with a selector on the rear panel. The voltage is
normally set to the country of destination. If the voltage selector is not
properly set, have a qualified technician, or your Kurzweil dealer, reset it.
18
As you face the rear panel of the K2000R^ the power connector is toward the
right hand side. When you have connected the cable at the K2000R end, plug it
into a grounded outlet (three hole) .If the power source does not have the
standard three hole outlet, take the time to install a proper grounding
system. This will assure of avoiding problems with audio hum, and will reduce
the risk of a shock hazard.
NOTE: A quick check of the K2000R and the audio system can be performed
without a MIDI controller connected to the K2000R. Hold down the CANCEL
button on the alphanumeric pad and press any other alphanumeric button. If
all is well, you will hear notes. Refer to Appendix 1 of the K2000R Musicians
Guide for more information about this feature.
IMPORTANT NOTE: A single SCSI device may be connected to either SCSI port.
When connecting more than one SCSI device to the K2000R (including an Apple
Macintosh™ computer) the SCSI chain MUST be terminated properly. Refer to
Chapter 13 of the K2000R Musicians Guide, specifically the section labeled
"SCSI Termination" for details. If the SCSI ports are not properly
terminated, data can be lost and/or the K2000R and SCSI devices may be
damaged
19
II >Lhanne>
Kfeyttapginfd:
1 Righteous: Piano
wSs^BSBb 0ctau4iiPanic
Set the volxime at a comfortable level. You will get the best signal-to-noise
ratio if you keep the K2000R at full volume (turn the volume knob fully
clockwise), and adjust the level from your mixing board. You may also want to
adjust the display contrast. This is done with the Contrast parameter in
Master mode. See Chapter 11 of the K2000R Musicians Guide. As you trigger
notes from your MIDI controller, you will see the MIDI LED flash. If it does
not flash, check you MIDI cables and connections.
20
WARNING: Do not attempt to operate the K2000R on any AC voltage other than
the one selected - extensive damage to the unit may result!
WARNING: CHANGING THE VOLTAGE SELECTOR MAY REQUIRE THE USE OF A DIFFERENT
LINE CORD OR ATTACHMENT PLUG, OR BOTH. TO REDUCE THE RISK OF FIRE
OR ELECTRIC SHOCK, REFER SERVICING TO QUALIFIED SERVICE PERSONNEL.
• Locate the Voltage Selector Card which is mounted to the AC entry module
directly beneath the AC connector.
• Slide the clear cover up and over the AC connector. NOTE: This will not
slide up unless the AC cord is removed from the connector.
• Locate the small Voltage Selection card (PC board, approximately 1 1/4" x
3/4") inserted edgewise next to the fuse remover lever. Now locate the
hole used to insert a removal tool (1/8” diameter)
• Using a scribe, small screwdriver or other suitable tool, insert same into
the hole and apply pressure outward (away from the rear of the K2000R) ; the
Voltage Selection will pull straight out. Take care not to snag the card
on the fuse removal lever.
• Reposition the Voltage Selector Card such that when reinserted the proper
voltage number appears, insert the card, and press fully home.
21
• Locate the Fuse Removal Lever labeled "Fuse Pull” and raise this lever
until the adjacent fuse is free of its clip mount. Remove the fuse.
• Insert the new fuse and press completely into the clip mount.
• Close the clear plastic cover and reconnect the AC power cord.
IMPORTANT INFORMATION
In North America and other locations where the AC mains supply voltage is 100
- 120V/60HZ, replace the fuses only with:
Type CMC, 5mm x 20mm, Time-Delay, 250 Volt, 1.25 Amp fuse
In Europe and other locations where the AC mains supply voltage is 220 -
240V/50HZ, replace the fuses only with:
22
• Turn the unit off and remove the AC line cord from the AC power connector.
• Locate the SIMM sockets on the Engine board (as per Figure 2.11 below).
There are two banks of SIMM sockets, the two inner sockets are bank A, and
the two outer sockets are bank B. When installing one bank (2 SIMMs),
install the SIMMs in the bank A sockets. When installing two banks (4
SIMMs) of the different sizes, make sure that the 2 SIMMs in the bank A
sockets are the same size, and the 2 in the bank B sockets are the same
size.
23
back of unit
u n> r> ,
• Insert a SIMM into the proper socket at a 60® angle (Figure 2.12).
• Press the SIMM firmly into the socket until it clicks into place.
• Pull the latches on each end of the socket outward^ rotate the SIMM to a
vertical position, and release the latches.
• Once all SIMMs have been inserted, inspect for proper installation - make
certain the top edges of all like SIMMs are at the same height and parallel
to the printed circuit board.
• Confirm the "Samples ####K" figure displayed in the center of the status
:
bar corresponds to the total amount of sample RAM installed in the unit.
24
Turn the unit off and remove the AC line cord from the AC power connector.
Locate the old software ICs - U3 and U6 on the Engine board (Figure 2.11).
Remove the old ICs using an IC puller or similar tool. Be careful not to
damage the traces on the Engine board.
Carefully insert the new IC labeled ”LO” in the socket at U3, and the new IC
labeled "HI" in the socket at U6. Make sure the ICs are oriented with pin 1
towards the front of the unit. Pin 1 is on the end of the IC which has the
indentation in the middle. Inspect for proper installation - make sure the
ICs are firmly seated into the sockets and that none of the pins of the ICs
are bent.
Always HARD RESET the instrument after changing the software. Note that all
user data in RAM will be erased A hard reset can be performed on power-up by
.
pressing and holding the ”1”, **2”/ and "3" buttons on the numeric keypad while
turning on the power switch. Continue to hold the buttons until a message
appears in the display.
25
3.1.2 Place K2000 upside down on clean^ smooth^ padded surface. CARE
MUST BE TAKEN TO POSITION THE UNIT SO AS TO PROTECT THE WHEELS
FROM DAMAGE! Using a #2 Phillips screwdriver, remove the six (6)
machine screws securing the bottom half of the shell to the top -
one in each corner and two near the center of the bottom.
Carefully lift the bottom half and place in a safe location until
reassembly.
NOTE: You may find a higher comfort level in placing the keyboard
away from your body thereby locating the major assemblies closer.
3.2.1 The Engine and Audio/Power boards are both attached to a metal
chassis and are removed as a unit using the following procedure.
To deviate from this procedure may place either board into danger
of damage from excess flexing; thus, use caution and be mindful of
any risks you choose to take. It is necessary to remove this unit
before accessing the LCD, Slider and Front Panel boards.
3.2.2 Disconnect all connectors from both the ENGINE and AUD/PWR boards.
Using a #2 Phillips screwdriver, remove twelve (12) self-tapping
screws securing both boards (six for each board) Carefully lift
.
3. 2. 4.1 Position metal chassis such that the boards are upside down on a
clean, flat surface. Locate and remove the two (2) machine screws
securing the Engine board connector panel to the chassis.
Carefully grip the chassis and Engine board while repositioning
right side up. Now the Engine board can be removed by gently
unplugging it from the AUD/PWR board.
3.2. 5.1 Removal of this board is the same as 3. 2. 4.1 above except there
are ten (10) machine screws securing the connector panel to the
metal chassis.
3.2. 6.4 Using figure 1 for reference, locate the slider board between the
Front panel and disk drive. Remove the single self-tapping screw
and the board is free.
3.2. 7. The LCD board is surrounded by the Front Panel board. Remove the
four (4) self-tapping screws securing the LCD and lift the board
free.
3. 2. 7. Reverse 3. 2. 7. to reinstall.
3.2. 8.1 Before removing the bottom half of the shell, remove the round
knob on the Front Panel. Also removing the retaining nut and
washer under this knob.
3.2. 8.4 Remove the nine (9) self-tapping screws securing the Front Panel
and carefully lift panel from instrument. NOTE: This board can
be easily damaged by careless handling. BE CERTAIN TO USE CAUTION
WHEN HANDLING THIS BOARD.
3.2. 8.5 Reverse 3.2. 8.4 above to reinstall the Front Panel board. Be
certain to remember to use the ninth screw to secure the bracket
(used for clamping the cables) in the center front of the panel.
3.2. 9.2 Note that the Power Transformer and Power Receptacle are hard-
wired as a unit. The harness from the secondaries of the
transformer are routed to J20 on the AUD/PWR board; thus, can be
easily removed from the instrument.
3.2. 9.3 Remove the four (4) machine screws securing the transformer to its
posts. Using a 7/32” nutdriver and a #2 Phillips screwdriver,
remove the two (2) nuts from the screws securing the Power
Receptacle to the back of the shell. Finally, remove the self-
tapping screw securing the green wire to the GND terminal on the
AUD/PWR board and unplug the harness connector from J20.
3.2. 9. 4 Reverse 3.2. 9.3 and 3.2. 9.2 above for reinstallation.
3.2.10.2 Remove ribbon connector from JIO on ENGINE board. Remove disk
drive power connector from Jll on ENGINE board. Remove two (2)
self-tapping screws securing aft end of disk drive mounting
bracket to shell. Disk drive will now lift out of shell.
3.2.11.1 Remove bottom of shell as per 3.1.2 above. Orient the K2000 such
that the keyboard is facing you and the unit is bottom side up.
3.2.11.5 Carefully remove the keyboard from the unit while guiding the
three ribbon cables/connectors through the foil slots. The
keyboard can now undergo maintenance/repair.
3.2.12.1 Remove the keyboard as per section 3.2.11 above and lay the
keyboard on a clean, soft surface to prevent damage.
3.2.12.3 Remove the spring from the back of the key taking note of the
positioning of the two ends. The wide end of the spring goes to
the the top of the key. This can be easily removed with a pair of
needle nose pliers by lifting one end of the spring out of its
socket
3.2.12.4 Insert a small/medium size flat blade screwdriver into the square
opening in the top of the key near the rear. Gently turn this
such that it releases the retaining clip while lifting the rear of
the key. Then tilt the key toward the front of the keyboard and
remove
• Using a #2 Phillips screwdriver, locate and remove the eight (8) machine
screws securing the cover; four (4) on each side of the unit.
• Disconnect ribbon connectors J411 and J412 on the K2000R Connector Board.
• Locate and remove the two -(2) hex head screws securing the connector
labeled "SCSI THRU” (J412) to the rear panel.
• Remove the ten (10) nuts securing the 1/4" jacks to the rear panel.
• Locate and remove the two (2) hex head screws securing the DB-25 connector
labeled "SCSI”.
• Locate and remove the six (6) screws securing the three MIDI connectors
using a #1 Phillips screwdriver.
• Locate and remove the six (6) screws mounting the Engine Board using a #2
Phillips screwdriver.
NOTE: It is NOT necessary to remove the Connector Board or the Engine Board in
order to install SIMMs or to replace the system software.
• Remove the cover. Connector Board, and Engine Board as per 3.3.1, 3.3.2,
and 3.3.3 above
• Remove the securing nut to the 1/4” jack at the bottom of the escutcheon
just to the right of the volume knob.
• Remove the six (6) flat head screw securing the escutcheon; two on top (one
at each end) and four on the bottom.
3 3.5.1
. Display. Removal
• Using a #2 Phillips screwdriver, remove the four (4) screws mounting the
Display to the frame.
• Trace the ribbon cable to J4 on the Engine Board and disconnect. The
Display can now be removed/ repaired/replaced.
3.3.6 Removal of any other items in the K2000R remain obvious and
requires no discussion.
4.0
4.0. MAINTENANCE, CARE AND PREVENTION, TROUBLESHOOTING
1 With a modicum of care^ the K2000 will provide years of use and
4.0. enjoyment. There are just a few important points to keep in
mind.
your cleaning cloth; you can then carefully wipe the surfaces
of the K2000.
4.2.1 Care should be always taken to keep dirt and dust out of the
floppy disk drive, thus few (if any) problems will result. If,
however, errors or failures in loading or saving are
experienced, dust or dirt in the drive may be the cause.
Consult your dealer or authorized service center for
information regarding products and techniques for floppy drive
cleaning. DO NOT ATTEMPT TO DISMANTLE THE UNIT OF THE FLOPPY
DRIVE!
4.3.1 The K2000 uses batteries to preserve its internal memory when
the power is turned off. The original batteries should last at
least one year before they need replacing. Replacement is
necessary when the LCD says "BATTERY VOLTAGE IS LOW" during
power-up, or when it is noticed the LEDs flash twice instead of
once during power-up. Once these warnings begin to occur, the
4.3.3 Most quality brand-name batteries now have "sell by" dates
printed on their packages. Carbon-zinc batteries will last for
at least a year after installation while alkaline batteries
should last for at least 2 years provided they are installed
before the date on the package. Rechargeable batteries should
not be used; the K2000 will not recharge them and their life
after charging on an external charger will be only a few months.
Battery life is not significantly increased by leaving your
K2000 on all the time; batteries may even suffer heat
degradation if the K2000 is left on continuously.
xxxxxxxxxxxxxxxxxxxxx
4.4.3 The battery voltage and wheel center values may be different on
your unit. The fourth line (represented by XXXX) is the one of
interest in diagnosing pedal problems (if it is flickering, move
the right wheel toward the center of its rotation; if it is
blank, all is OK)
4.4.4 To see how the display works, first plug the footswitch that
came with your K2000 into the FOOTSWITCH 1 jack on the rear
panel. When the footswitch pressed, the fourth line should
read:
PEDAL 1
4.4.5 When the footswitch is released, the fourth line should become
blank. When plugged into the FOOTSWITCH 2 jack, the action
should be similar except that "PEDAL 2” will be displayed.
4.4.6 Now try your third-party footswitch. If it acts the same way as
the Kurzweil unit, then all is well from a hardware standpoint.
IF the fourth line remains blank when the footswitch is pressed^
there is a problem with the footswitch; either its internal
switch, the connecting cable, or the plug. Likewise if "PEDAL
X" always shows whether the footswitch is pressed or not, there
is a short circuit somewhere along the line. Some footswitches
will act backward, that is, the display will show "PEDAL X"
immediately when the footswitch is plugged in and will go blank
when the footswitch is pressed. This type of footswitch can be
used with the K2000, but it must be plugged in Jbeifore the power
is turned on. Alternatively, performing a "soft-reset" (press
"+ -", "0", "CLR" together) after plugging in a backward
footswitch will result in proper operation.
4.4.8 Now try the control pedal. When connected, the fourth line will
change to show "CONTROL PEDAL - XXX”. Rocking the pedal forward
should make the numbers increase toward 255 while retiring it
should make them decrease. Moving the pedal to approximately
mid-position should yield a reading of around 130 (100 - 160 is
typical) Common control pedal problems are backward action
.
4.4.9 For any of these problems, first check the connection pattern
between the potentiometer in the pedal and the stereo plug. The
wiper must be connected to the tip, the lower end of the
potentiometer (the end the wiper moves toward when the pedal is
retired) connected to the sleeve, and the upper end to the ring.
If a volume pedal is being used and it will not go to zero.
4.4.10 As seen, the diagnostic program can also be used to check out
the built-in panel and performance controls of the K2000. Like
the Controller Slider, moving either wheel or pressing on the
keyboard will show a controller name and reading between 0 and
255. If you move the Alpha Wheel clockwise, the numbers will go
0-1-2-3-0-1-2 .. .while counterclockwise should produce 3-2-1-0-3-
2... If you press a button, its name will be shown and if it is
one of the mode buttons, its associated LED should flash.
4,4.11 The third line of the display shows the results of two
measurements that are made whenever your K2000 is turned on.
The battery voltage will be about 4.5 volts for new batteries
gradually declining over time to 3.2 volts, at which point you
will begin to receive warnings (see "BATTERY SELECTION AND
REPLACEMENT" above). The wheel referred to is the spring-
centering software will work perfectly. If it is outside that
range, the wheel will still be automatically centered but you
may not get the full control range in one direction. Please see
your Kurzweil/Young Chang dealer for any needed wheel
adjustments
4.5.1 The K2000 quite possibly has the lowest noise and widest dynamic
range of any instrument in the industry. The following tips
will enable you to make the most of this, and optimize the
K2000*s audio interface to your other equipment.
SECTION 4
4.5.4 Increasing the level too much can cause clipping distortion when
multiple notes are struck loudly. For dense sequences all
played through the same outputs, you will probably only be able
to increase the volume by 6dB or so without risk of distortion.
For monophonic instruments (lead guitar) or single instrument
tracks (such as drums), a substantially greater boost is
generally possible.
4.5.5 For the absolute maximum signal quality, the individual outputs
should be used. These are connected almost directly to the 18-
bit digital-to-analog converters with a minimum of noise-
inducing processing circuitry. A total dynamic range of over
lOOdB is available at these outputs. The MIX outputs are
naturally somewhat noisier because they represent the noise of
the individual outputs all mixed together, and the signal must
travel through more circuitry to reach them.
4.5.6 Programs which are routed through the K2000*s global effects
processor (Output Group A) will also be slightly noisier than
programs routed to Output Group B. As with an external effects
unit, maximizing the input signal level (using the methods
described above) will improve the signal-to-noise ratio of the
effects processor. When in Effects mode, you will see an
internal Wet/Dry mix parameter; in the Effects Editor you will
find numerous parameters for setting the level of the various
effects. Your best signal-to-noise ratio will be achieved by
setting the effects level parameters to maximum and adjusting
the Wet/Dry Mix parameter to set the overall effects mix. If
you are only using the effects unit for EQ functions, one of the
EQ or Tone Control functions accessible through the program
Editor will produce quieter results then the global effects
processor’s EQ functions (Parametric EQ, Para Bass, Para Mid^ or
Para Treble, for example)
4.6.1 A common problem with all electronic musical gear is the hum
that can occur in connection cables due to AC ground loops
Although ”3-prong to 2-prong” AC adapters are frequently used
break ground loops, they also break the safety ground that
protects you from electric shock. Using these adapters is
dangerous and SHOULD NOT be done I Furthermore, although using
these adapters may reduce low-frequency hum, high-frequency line
noise (such as motor switching noise) is likely to get worse in
this case, since the K2000*s AC noise filter will have no outlet
for the noise it filters if you disable the ground.
To reduce ground hum, you can increase your output signal levels
as described earlier in this section. Other safe procedures
include plugging your mixing board and amplifier into the same
outlet as your K2000, and making sure that all your gear is
properly grounded. If your* re using an external SCSI device,
plug it into the same outlet as well. AC isolation transformers
are extremely effective at eliminating ground loops, and are
recommended for critical installations. A 75 watt transformer
is sufficient for the K2000.
4.7.2 Very low line voltage or severe voltage dips are a problem for
any computer-based instrument. When the K2000 is set for 120
volt input (the normal North American setting), it should
function down to 90 volts. If the line voltage drops below 90
volts, a special circuit halts all activity to protect against
software crashes or damage. When the line voltage returns to
and stays at an acceptable level for at least one second, the
computer will automatically restart. The net effect is just as
if you had performed a soft reset. Continuous low line voltage
or transient dips will never produce symptoms other than
unexpected soft resets as just described. Any other problems
such as distortion, disk errors, or lost data are cause by
something other than line voltage fluctuations.
4.7.3 Soft resets from line voltage dips are most common. These are
easily identified because the reset occurs coincident with the
building lights dimming, stage lights or power amps being
switched on, or air-conditioning equipment starting up. The
solution in all cases is to get a more direct connection between
your K2000 (and any other computer-based equipment) and the
building’s power. Floodlights, large power amplifiers, and
motor-operated devices should use a separate extension cord;
preferably they should be plugged into a separate outlet.
4 . 8 TROUBLESHOOTING
4.8.1 Everything possible has been done to ensure the K2000/K2000R is
delivered free of defects. There is a good chance there is
nothing wrong with the unit, even if you are not seeing the proper
display or hearing any sounds. If so, carefully check the
following items:
5 . Reaffirm that the proper audio cables are fully connected to the
K2000/K2000R and to a sound system; also making certain the sound
system is properly connected to power and turned on. If sound is
still not heard, a good suggestion would be to test the audio
cables, replace them, or test for sound by connecting a high
quality pair of headphones to the headphone jack.
7. Lower the volume of the sound system, turn the K2000/K2000R off,
then on again. This is called a power cycle.
8 . Press the +/-, 0, and CLR (on the numeric pad to the right of the
front panel) simultaneously. This is called a soft reset
9. As a last resort, save to disk any RAM objects you've created, and
perform a hard reset as per paragraph 4.12.1 (page 11) for early
software versions or paragraph 4.19.1 (page 29) for software
versions VI. 30 and above. The K2000/K21000R will warn you about
deleting everything (only RAM objects will be deleted) Press
.
2. Disk is damaged.
Cannot write floppy disk. 1. Disk is not MS-DOS
formatted.
4. Disk is damaged.
Cannot format disk. 1. Disk is damaged.
10
This document is for use with K2000 Diagnostic software Version 0.4. Version
0.4 of the diagnostic software is included with K2000 Version 1.0 Engine
software. Version 1.2x of the K2000 Engine software include Version 1.0 of
the diagnostic software, which differs slightly from Version 0.4. If the unit
you're working on has Engine software Version 1.2x , please refer to the K2000
Diagnostics document for Version 1.0 of the diagnostic software.
4.11 INTRODUCTION
When you enter the K2000's diagnostic mode, you will see a menu of available
tests. At this point you can choose to run the standard Burn-in test, or you
can select individual tests to diagnose specific components. The Burn-in test
automatically initiates a series of tests to check the basic functionality of
the K2000's hardware.
There are three general types of tests: memory RAM, ROM, and I/O
(input/output The descriptions of each diagnostic test will identify their
) .
general types
IMPORTANT NOTE: Before you begin the diagnostics, be sure that the owner of
“th®K2000 has backed up any data stored in RAM, since the diagnostics will
completely clear the unit's RAM. Refer to section 13 of the Musician's Guide
for details on backing up RAM.
First, remove the battery access panel on the bottom of the unit and
disconnect the batteries. Wait 5 minutes for the battery capacitors to
discharge, then turn the unit over, plug it in, and turn it on. You'll see a
display presenting three options: Hard Reset, Cancel, and Diagnostics. See
Figure 4.2. Use the Alpha Wheel to select the desired option, highlighting
it in reverse. Then press the ENTER button to execute your choice.
Selecting DIAGNOSTICS will enter the K2000's diagnostic mode. Selecting HARD
RESET will reset the K2000 (hard reset) and return it to Program mode.
Selecting CANCEL will also execute a hard reset and return to Program mode.
11
irli]
.,S^NCi^y
Upon entering diagnostic mode, a menu of available tests appears. See Figure
4.3. The soft buttons are labeled by the bottom line of the display. The
options are "A Burn-in Test'', or "B Test Forever."
NOTE: The option labeled "Burning Test" on the bottom line of the display
refers to the Burn-in test, so called because it's used at the factory to
perform a general quality test of all the major components of the unit.
These tests are done multiple times to make sure the unit will perform
properly with extended use.
stib^lPUer SW
Rudia! Bbardll Opt i ottBoard;
MIDlf Uapt SinegiWawe:
SbUrid :ROM
SbUhdsRGM
1st NURRM
NOTE: Before rtinning the Bum-in test, insert a formatted floppy disk into
the K2000's floppy disk drive. The test will hang up without a disk in the
drive.
12
When the diagnostic test menu is displayed, press the A soft button (the
farthest to the left) at any time to begin the Burn-in test sequence. When
the Burn-in test is selected, the display changes to show the list of tests
included in the Burn-in sequence. As each test is completed, the display is
updated to show the result. See Figure 4.4. The Burn-in test is adequate
for testing the unit's major components; if all the Burn-in tests pass, the
unit should be in full working order.
NOTE: An SCSI device must be connected to the unit's SCSI port, otherwise the
SCSI test will fail. This does not necessarily indicate a malfunction of the
SCSI link.
LCD PudibBdard
Boot ROM Floppy; :
The Burn-in test can be aborted at any time by pressing the EXIT button, then
pressing any button on the front panel to return to the diagnostic test menu
(the display says "Press any key/' meaning a front panel button, not one of
the black-and-white keys) It may be necessary to press the EXIT button more
.
than once, because the K2000 can only recognize switch closures between tests.
Press any front panel button to interrupt the Burn-in test temporarily. The
test will resume when the button is released. Press any one of the keys, the
test will be interrupted and hold temporarily; it will resume when any front
panel button is pressed.
To run an individual test, use the Alpha Wheel to select the desired test from
the diagnostic test menu. The selected test is highlighted in reverse.
Press the ENTER button to run the selected test one time. When the test is
completed, the display will show the result of the test. Normally each test
need only be run once in the attempt to identify failures. Running a given
test continuously is recommended to monitor the K2000's output signal with an
oscilloscope, since many tests are completed too quickly to be monitored in a
single pass.
A selected test can be run indefinitely by pressing the B soft button (second
from the left) This will cause the selected test to run until it is
.
interrupted with the EXIT button. Remember, it may be necessary to press the
EXIT button more than once.
Press any front panel button to interrupt a continuous test temporarily. The
test will resiime when the button is released. Press one of the keys, the test
13
will be interrupted and hold temporarily and will resume when any front panel
button is pressed.
As the continuous tests are running, the bottom line of the display will
indicate the number of tests that have been completed, and the number of
failures that have occurred.
4 . 13 TEST TYPES
The tests described in this section are of three types, referred to as RAM
tests, ROM tests, and I/O tests. Memory tests check the physical memory
addresses of the data corresponding to various components. This is done by
writing an arbitrary test data pattern to memory and reading it back,
comparing the written data to the read data. ROM tests involve comparing the
data stored in the ROM components to a checksum. I/O tests also send a test
data pattern from the central processor to the component (s) being tested, and
check whether the data are being exchanged properly between the central
processor and the tested component s ( )
There are two stages to a memory test: the data test, and the address tfest.
In the data test, a test data pattern is written from the processor to the
memory component being tested. The diagnostic software expects the data
pattern that's written to memory to be read back exactly as it was written.
If this is the case, the test passes. If the observed data (the data read
from memory) differ from the expected data (the data written to memory) the ,
test fails.
Test Item 2
In the address test, the central processor (CPU-a Motorola 68301) checks every
memory address related to the component being tested. A test data pattern is
written to each address, and then read back. The diagnostic software expects
the data pattern to be read back from the same address to which it was sent.
If the observed address (the address read from) matches the expected address
(hhe address written to) , the test passes. If the observed address does not
match the expected address, the test fails. Address failures are displayed
14
in the following format: the expected address; the observed address. See
Figure 4.6 for an example.
NOTE: Many figures in this dociuoent show values of all zeros. You'll see
non- zero values for the address, escpeoted values, and observed values of all
tests. The e^q>ected and observed values themselves are not significant;
what' s important is that a difference between the expected and observed values
indicates a test failure.
Each of the tests described below will be either a memory test or an I/O test
(or, in some cases, a combination of the two) When performing single passes
.
of specific tests, the display will indicate the result of the test upon its
completion. In most cases, the result will be Success (Pass), Data Fail, or
Address Fail. Unless otherwise specified, failures will be displayed as
described in the previous paragraphs.
This test has both an I/O stage and a memory stage. It checks the connections
between the LCD and the engine board, as well as the memory of the LCD itself.
The processor writes a test data pattern to the LCD memory, then reads it back
expecting a certain value. If the observed value does not match the expected
15
NOTE: The sockets at U3 and U6 are labeled "SETUP" on the engine board.
Even so, the chips installed in these sockets are the Engine software EPROMs.
This discrepancy is due to a design change that was made after the engine
board was finalized and printed. For Engine software Version 1.0, the SETUP
sockets contain the Engine software, and locations U2 and U5 are unused.
Nevertheless, the Boot EPROM test checks the Engine software EPROMs.
16
SCSI Test
4.14.11
test is similar to the FDC test, and checks the functionality
SCSI port.
of the
If no SCSI device is connected to the unit's SCSI port,
will fail. this test
This does not necessarily indicate a malfunction of the
SCSI
operational, you should connect a
SCSI device before running this test.
17
The Swap Bit Test checks to make sure that the swap bit function has taken
place. The swap bit function also involves the GALs at UlO and Ull. If the
test fails, check the traces between UlO, Ull, and U35. If they appear
intact, one or more of the GALs may be defective.
The 13-step Calvin and Hobbes memory test checks the connections between the
CPU and the three VLSI sound engine chips: Calvin, Hobbes 0, and Hobbes 1.
Calvin handles sample fetching tasks, and Hobbes 0 and 1 primarily perform
wave form generation and DSP functions.
Calvin and Hobbes have undergone rigorous testing prior to mass manufacturing,
and are extremely unlikely to be defective. If one or more of the steps in
this test indicate a failure, the problem is probably to be found in the
connections between the chips and the CPU. The test results are displayed as
shown in Figure 6. The significance of failures in each step of the test is
explained below. Steps 2.1—2.3 test the connections to the internal RAM and
internal registers of Calvin and Hobbes.
When analyzing the results of these tests, the result should be a PASS message
for each test. The hexadecimal numbers that appear on the display are
irrelevant as long as they match for each test. The numbers shown in figure
5.6 are arbitrary, and may differ from the numbers seen in the unit under
test. This does not indicate a problem.
If any of the tests fail, check the connections between the components in
question, as described in the paragraphs below. If test #2 fails, check all
the connections between the CPU, Calvin, Hobbes 0, and Hobbes 1. Note that
expected observed values are not given on the line indicating the status of
Instead, a PASS or FAIL message will be seen for each of the three
steps in test #2.
NOTE: The Calvin and Hobbes tests are interrelated, and are sequenced from
the least significant test to the most significant. If any one test fails,
the subsequent tests are likely to fail as well. In this case, start by
checking the components related to the first test that failed. If a
malfunction occurs, repair the malfunction and run the Calvin and Hobbes test
again. It is likely this will result in the subsequent tests will passing
when retested.
18
In general, the data flow through these components is the following sequence;
Sample memory, Calvin, Hobbes 0, Hobbes 1. If test failures occur, check the
connections to Calvin before checking those to Hobbes 0, and Hobbes 0 before
Hobbes 1
2 1
. Calvin Memory Test Fail
Connection problem between CPU and Calvin
2 2
. Hobbes 0 Memory Test Fail
Connection problem between CPU and Hobbes 0
2 3
. Hobbes 1 Memory Test Fail
Connection problem between CPU and Hobbes 1
19
This test is intended for use with an oscilloscope connected to the K2000's
audio output, and enables analysis of the purity of the sine wave generated.
If the sine wave is impure, a problem with the sound engine is suggested. To
monitor the Sine wave test, connect the scope to any of the B outputs, the MIX
outputs, or the Headphone jack.
corresponds to two Sound ROM chips. In the unlikely event of a failure, the
engine board must be replaced, since the Sound ROM chips are soldered in
place.
purpose of this test is to determine whether the K2000 recognizes all of the
sample RAM that has been installed. The more RAM that has been installed,
the longer this test takes. The test takes approximately 80 seconds for each
RAM block containing a SIMM, regardless of the SIMM's size.
NOTE: for Version 0.4 of the diagnostic software, this test is significant
only if the unit contains SIMMs of 4 megabytes or smaller. If the unit
contains larger SIMMs, the test will indicate success even if the SIMMs are
20
not functioning properly. To test SIMMs larger than 4 megabytes^ Version 1.0
of the diagnostic software is required.
The K2000's sample RAM space is divided into two banks (Bank 0 and Bank 1),
representing the two pairs of SIMM sockets. Each bank is divided arbitrarily
into four blocks with a maximum address space of 8Mb, These blocks, A-D,
represent internal divisions of the address space; they don't correspond to
any physical division of the sample RAM.
When the Sound RAM test is run, the display will indicate which block in which
bank is currently being tested. This is shown by a series of dots that
appear as the test proceeds. As each block is finished, the dots are
replaced with either a success message or a failure message; see Figure 4.8.
If the block fails, the malfunctioning address is indicated, as well as the
expected checksum value and the observed value. See Figure 4.9.
„ <j;SQahd ?>
‘
Bank®
.
Bankl >
_ fiddress ExP 0bs^ Rddress ExK> Obs
R ;'Saccess'. M 'm'‘ m . i|;-v
B Success# m « a m m - . ^
C Success; V • •
*
,
' '
Sound 'RAM Test';- '
.. Banka
Address ExP :.0bs £ Address- ^ Obs ;
test; you will not receive a success or failure indication at the completion
of the 1st NVRAM test. Instead, you're pron^ted to power down the unit.
Wait 60 seconds after powering down, then power up again and select and
21
execute the 2nd NVRAM test. If a failure is indicated, power is not being
delivered from the batteries to the non-volatile RAM. Check the batteries
and the connection from the batteries to the engine board.
Address Fail Error in the address test stage of the memory test
Complete Test completed
Data Fail Error in the data test stage of the memory test
Even Fail Error in the high byte
Fail Test failure
Not Found SCSI device or sound RAM not found
Odd Fail Error in the low byte
Success Successful test
This section explains the use of the K2000's scanner diagnostics, which
enables confirmation of the proper function of numerous front panel buttons
and controls.
The K2000 uses a separate microprocessor to scan the keyboard, read the
performance controls, and operate the front panel buttons and LEDs. This
processor, which is called the Scanner, then communicates its findings to the
main 68301 processor (CPU) The scanner is on the Audio/Power board (at
.
There are two different diagnostic routines built into the scanner software.
The simpler routine can be entered only at power-up and can function
regardless of whether the engine board and/or LCD display is functioning
properly (an engine board must still be connected to provide clock signals,
however) This diagnostic routine reads out in binary format via the 8 LEDs,
.
The other diagnostic routine can be entered at any time, and uses the LCD to
display its results. Of course the engine board and LCD must be connected
and functioning properly to see the results. The LCD diagnostic routine is
easier to read than 8 binary-coded LEDs, so in most cases you'll want to use
it rather than the LED routine.
22
It is possible the battery voltage and wheel center values will differ from
unit to unit. Fresh batteries should read in the range of 4.3 - 4.8 volts.
If the battery reading is less than 3.5 volts, it is nearing the end of its
life (warnings will start to appear when it declines to 3.2 volts). The
wheel center should be between 120 and 136 (inclusive) for full range
operation.
A. When the Pitch Wheel is pushed back, the display should show "PITCH WHEEL"
and the niombers should smoothly increase toward 254 or 255. The numbers
will move in increments of 2 or more depending on how fast the wheel is
moved. If the wheel is pulled forward, the ntimbers should decrease toward
1 or 0, again in steps of 2 or more. When the wheel is released, the
number should return to a value between 120 and 136. If the centering or
range is off, there are two adjustment potentiometers on the wheel assembly
that can be moved to correct the problem. If the display flickers or
changes to BATTERY or CONTROL PEDAL when the Pitch Wheel is pulled full
forward or pushed full back, these pots should be adjusted. A perfectly
adjusted Pitch Wheel reads within 1 count of 128 in the center position,
reaches 254/255 just as it hits the back stop, and reaches 0/1 just as it
hits the front stop.
23
B. Moving the Mod Wheel should cause the display to show "MOD WHEEL" and the
numerical readout should react as with the Pitch Wheel. The range should
be from 0 or 1 to 254 or 255. If the range is insufficient or reaches the
0/1 and/or 254/255 limits too soon, there are 2 more adjustment
potentiometers on the Mod wheel assembly. If the display flickers or
changes to BATTERY or CONTROL PEDAL when the Mod Wheel is pulled full
forward or pushed full back, these pots should be adjusted.
E. If a switch pedal is plugged into the Switch Pedal 1 jack, the fourth line
will show "PEDAL 1" when the pedal is closed, and will go blank when it
opens. Some pedals (especially Roland) are normally closed, and open only
when pressed. Most others are normally open, and close when pressed.
The K2000 can use either type (as long as it is plugged in before powering
up) but the scanner diagnostics expect to see a normally open pedal.
After testing a normally closed switch pedal, the pedal must be unplugged
to clear the fourth line of the LCD. (This applies to Switch Pedal 2,
also.)
F. If a switch pedal is plugged into the Switch Pedal 2 jack, the fourth line
will show "PEDAL 2" when the pedal is closed, and will go blank when it
opens
G. If the Alpha Wheel is turned clockwise slowly, the display will show "SPIN
KNOB" followed by a nvimber that counts up (0, 1, 2, 3) and repeats.
Likewise if it is turned counter-clockwise, the numbers will count down.
Any other sequence indicates a problem-usually a bad connecting cable, but
possibly a bad switch contact.
H. If any button is pressed, the word "BUTTON" followed by its name will be
displayed. If it is one of the mode buttons, then its associated LED will
flash as well. It is normal for the LED associated with the PROGRAM
button to flash more brightly than the other buttons. If several buttons
have problems, the cause is usually the connecting cable. If only one
button is bad, then it needs replacing. Cable problems that seem to
disappear when the unit is disassembled are often due to a short circuit
caused by component leads on circuit boards puncturing the cable's
insulation. The ends of ribbon cable wires touching the conductive
coating inside the top cover can also cause improper operation.
I. Pressing any keyboard key down halfway will display the word "KEY" followed
by the key's note name (such as C#4 for the black key immediately to the
right of middle C) followed by the key's MIDI ID number (61 in this case).
This indicates that the "upper contact" of the key has closed. Pressing
24
lightly all the way down should display the same note name and MIDI number
a second time which indicates that the "lower contact" of the key has
closed. The two names/numbers should match. Most keyboard errors are
detected by the diagnostic which will display **ERROR ** in response. If
several keys seem to be bad, the problem is usually in the connecting
cables. Note: If none of the keys responds, yet the keyboard is known to
work, there may be interference from the pressure sensor. This will
usually go away if the unit is allowed to cool for awhile with the power
off. Units with serial numbers above 5000 should not be susceptible to
this
J. If a keyboard key is pressed down firmly, the pressure sensing circuit will
begin to respond and the display will change to "PRESSURE = XXX." "XXX"
will vary between 4 and 255 according to how hard the key is pressed.
When pressure drops below 4, the display switches back to the note
name/number display. There is a pressure sensitivity adjustment on the
audio/power board (at location R201, near the large filter capacitor) that
may be changed to suit an individual customer.
NOTE: There are two additional internal signals called AUXl and AUX 2 that
are normally unused. If the fourth line ever shows "AUX 1" or "AUX 2,"
then there is likely a short circuit (usually to a cable running under the
board) or current leakage in the audio/power board.
The LCD based diagnostic may be exited and normal K2000 operation restored by
pressing the EXIT and ENTER buttons together. The K2000 should restart just
as if a soft reset had occurred.
To enter the LED Scanner diagnostics, press and hold the 7, 8, and 9 buttons
while the power is off then turn the power on while continuing to hold them
down. The panel LEDs will flash in inverse sequence to indicate entry into
the LED diagnostic. After the light show, the software version number ($14)
will be shown briefly in hexadecimal, followed by the battery voltage in volts
(left hex digit) and tenths (right digit) .The battery voltage should rema in
on the display but early serial number units may display it only briefly.
The LED-based diagnostic uses the 8 mode LEDs for a two-digit hexadecimal
readout. The LEDs are sequenced from top left, down the left column, to the
top right, and down the right column. The values the LEDs represent are as
follows: Program mode = 128 decimal ($80 hexadecimal; Setup mode = 64 ($40);
Quick Access mode = 32 ($20); Effects mode = 16 ($10); MIDI mode = 8 ($08);
Master mode = 4 ($04); Song mode = 2 ($02); Disk mode = 1 ($01). See Fiqure
^
4.11.
25
When a key is pressed partially^ its MIDI note number is displayed dimly in
hexadecimal. The MIDI note numbers range from $24 (decimal 36) for the
lowest key to $60 (decimal 96) for the highest key. Pressing the key fully
should make the display brighten but otherwise not change. If the diagnostic
detects an error (lower contact made but not the upper contact for example )
some or all of the LEDs will blink. When the key is pressed hard enough to
activate the pressure sensor, the note number display is replaced by the
pressure value display.
To exit the LED-based diagnostic, power should be turned off then back on.
27
This document is for use with K2000 and K2000R Diagnostic software Version
1.0, which is included with K2000 and K2000R Engine software Version 1.2x. If
you're working on a K2000 with Engine software 1.0, please refer to the manual
for K2000 Diagnostic software Version 0.4.
Since the diagnostic software is the same for both the K2000 and K2000R, this
document will only refer to the K2000. However, all information also applies
to the K2000R.
4.18 INTRODUCTION
Upon entering the K2000's diagnostic mode, a menu of available tests will
display. The options are to run the standard Burn-in test, or to select
individual tests to diagnose specific components. The Burn-in test
automatically initiates a series of tests to check the basic functionality of
the K2000's hardware.
There are three general types of tests: RAM, ROM, and I/O (input/output)
The descriptions of each diagnostic test will identify their general types.
IMPORTANT NOTE Before beginning the diagnostics , be sure that the owner of
:
the K2000 has backed up any data stored in RAM, since the diagnostics will
completely clear the unit's RAM. Refer to section 13 of the Musician's Guide
for details on backing up RAM.
29
uiflGNdstiqSi
A .f .
When entering diagnostic mode, a menu of available tests appears. See Figure
4.13. The soft buttons are labeled by the bottom line of the display. The
options are "1 Burn-in Test", or "2 Test Forever."
NOTE: Before running the Burn-in test, insert a formatted floppy disk into
the K2000's floppy disk drive. The Burn-in test will hang up at the floppy
drive test if you begin it without a disk in the drive.
NOTE: An SCSI device must be connected to the unit's SCSI port, otherwise the
SCSI test will fail. This does not necessarily indicate a malfunction of the
SCSI link.
30
LCJJ -tflluclioBb^hd
Boot ROM '
FloPpy.:^:<<^-
Setup ROM 'SCSI-
PSRflM - Sound ROM
1^0 Port :;Sound RAM
Inter^ruPt
EXIT-StoP Dia?Js. Total-0000 Fail-000 &
Figure 4.14 - Bum-in Test Menu
The Burn-in test can be aborted at any time by pressing the EXIT button, then
pressing any button on the front panel to return to the diagnostic test menu
(the display says "Press any key/' meaning a front panel button, not one of
the black-and-white keys) It may be necessary to press the EXIT button more
.
than once, because the K2000 can only recognize switch closures between tests.
Press any front panel button to interrupt the Burn-in test temporarily. The
test will resume when the button is released. Press any one of the keys, the
test will be interrupted and hold temporarily; it will resume when any front
panel button is pressed.
To run an individual test, use the Alpha Wheel to select the desired test from
the diagnostic test menu. The selected test is highlighted in reverse.
Press the ENTER button to run the selected test one time. When the test is
completed, the display will show the result of the test. Normally each test
need only be run once in the attempt to identify failures. If you encounter
test failures, you may want to run the failed test continuously. Running a
given test continuously is recommended to monitor the K2000's output signal
with an oscilloscope, since many tests are completed too quickly to be
monitored in a single pass.
Press any front panel button to interrupt a continuous test temporarily. The
test will resume when the button is released. Press any one of the keys, the
test will be interrupted and hold temporarily, and will resume when any front
panel button is pressed.
As the continuous tests are running, the bottom line of the display will
indicate the number of tests that have been completed, and the number of
failures that have occurred.
The tests described in this section are of three types, referred to as RAM
tests, ROM tests, and I/O tests. Memory tests check the physical memory
addresses of the data corresponding to various components. This is done by
31
There are two stages to a memory test: the data test, and the
address test.
In the data test, a test data pattern is written from
the processor to the
memory component being tested. The diagnostic software expects the data
written to memory to be read back exactly as it was written.
If thxs xs the case, the test passes. If the observed data (the data read
from memory) differ from the expected data (the data written
to memory), the
Test :xtem.
Observed 5
Press anykey to continue.
Figure 4.15 - Data Failure
32
TeStvitem s maim
Please wait. Testin 9 i^i;flddr Err
Expected: 000000005 .
Since data cannot be written to ROM components, tests are done by reading the
data from the ROM components and summing the hexadecimal values of the data.
This sum is conpared to a checksum stored in the last two memory addresses of
each con^onent. The two sums must match for the test to pass.
In an I/O test, a test data pattern is sent from the CPU to the con^onent
being tested. The signal is processed by the component, and returned to the
central processor. If the signal returns to the central processor as
expected, the test is a success. If the observed pattern differs from the
expected pattern, the test fails.
Each of the tests described below will be either a memory test or an I/O test
in some cases, a combination of the two) When performing single passes
.
of specific tests, the display will indicate the result of the test upon its
completion. In most cases, the result will be Success (Pass), Data Fail, or
Address Fail. Unless otherwise specified, failures will be displayed as
described in the previous paragraphs
This test has both an I/O stage and a memory stage. It checks the
connections between the LCD and the engine board, as well as the memory of the
LCD itself.
The processor writes a test data pattern to the LCD memory, then reads it back
expecting a certain value. If the observed value does not match the expected
value, a data failure is indicated. This could indicate a processor error, a
faulty LCD, or a faulty connection from the LCD to the engine board.
This memory test checks the integrity of the Engine software data stored in
the ROM chips located on the engine board at U3 and U6. It reads the data
33
from each address and compares it to a checksum stored in the ROM chips. It
also checks for errors in the pin connections of each of these chips. If all
checksums match, the test is a success. If there are checksums that do not
match, the display will show either "Even Fail" or "Odd Fail." If an Even Fail
message appears, remove the chip located at U3 and test it in an EPROM
programmer. If the checksums match, the chip is good, and a connection
problem at U3 is indicated. If the checksums do not match the chip is
improperly programmed and should be reprogrammed or replaced. Likewise, if
an Odd Fail message appears, remove and test the chip from location U6.
NOTE: The sockets at U3 and U6 are labeled "SETUP" on the engine board.
Even so, the chips installed in these sockets are the Engine software EPROMs.
This discrepancy is due to a design change that was made after the engine
board was finalized and printed. For Engine software Version 1.1 and 1.2,
the SETUP sockets contain the Engine software, and locations U2 and U5 are
unused. Nevertheless, the Boot EPROM test checks the Engine software EPROMs.
Similar to the Boot EPROM test, this is intended to check the data and pin
connections of the ROM chips located at U2 and U5 on the engine board. For
Engine software Version 1.1 and 1.2, these locations are unused, and the test
will fail. This does not indicate a malfunction.
NOTE: This test checks only to ascertain whether the pseudo-static RAM chips
at U7 and U8 are good. It does not check to determine if the memory in the
P/RAM option (if installed) is good. To check whether the extra RAM is
recognized by the unit, it is necessary to enter Song mode or Disk mode when
finished with the diagnostics. The top line of the display will indicate the
amount of available RAM. This number should be approximately 120K if there
is no P/RAM extension, and approximately 760K with the P/RAM extension. If
the unit has the P/RAM option board (at connector Jl) , the display should show
the additional RAM space. If the display does not indicate approximately
760K of available RAM, the P/RAM option board may be defective, and should be
repaired or replaced.
In this test, the integrity of data transfer within the CPU (location U9) is
tested. This is done by writing a test data pattern from the low bit
addresses of the CPU to the high bit addresses. The high bits then read the
pattern and compare it to the original. If the patterns match, the test is a
success. If the test fails, the CPU may be bad, but it is more likely that
there is a problem with the I/O bus.
Also an I/O test, the Interrupt Test checks the connection between the IRQO
(interrupt request) pin and the floppy disk controller in the CPU. A
successful Interrupt test indicates that interrupt requests from the floppy
34
disk controller are being received by the CPU's IRQO pin. Failure of this
test indicates a defective connection between the CPU's IRQO pin and floppy
disk controller, and could identify the cause of malfunctions in the floppy
disk drive.
Another I/O test, this checks the connections between the engine and audio
boards. When you run this test, the LCD will always display "Success," but
this should be ignored, since it doesn't indicate the success or failure of
the test. In a successful test, the eight LEDs to the left of the LCD will
flash in sequence (from top left to bottom left to top right to bottom right)
at 300 ms intervals. A failure of any of the LEDs to light indicates a
defective LED or a faulty connection between the engine and audio boards.
This I/O test checks the operation of the MIDI In and Out ports. Before
executing this test, connect a MIDI cable from the K2000's MIDI Out port to
its MIDI In port. A failure of this test is indicated in the display as a
data failure as described in the overview. If this test fails, repeat the
test with a different cable. If it still fails, this indicates that one or
more pins in the MIDI Out and/or MIDI In port is defective.
A blank formatted floppy disk (DOS 1.4M) must be inserted in the floppy disk
drive before initiating this test. The test cannot proceed without a disk
inserted in the drive.
The FDC (floppy disk controller) test is an- I/O test, and checks the
functionality of the NEC PD72064 chip, the floppy disk drive, and their
connections. The test entails sending a test data pattern from the floppy
disk controller (FDC) to the floppy disk drive (FDD) and back. If the
display indicates a successful test, then the connections between the FDC and
FDD are functioning, and the test data pattern written by the FDC and
read
from the FDD are identical. If the display indicates "Read Fail," there is a
problem in the connection from the FDC to the FDD. If it indicates "Write
there is a problem in the connection from the FDD to the FDC. If the
display indicates "Data Fail," the read and write operations are normal,
but
the "read" signal and the "write" signal do not match. In this case, the
^isplsy will indicate a data failure as described in the overview.
This I/O test is similar to the FDC test, and checks the functionality of
the
SCSI port. If no SCSI device is connected to the unit's SCSI port, this test
f^il. This does not necessarily indicate a malfunction of the SCSI
link. To be sure that the SCSI link is operational, you should connect a
SCSI device before running this test.
The 13-step Calvin and Hobbes memory test checks the connections between
the
CPU and the three VLSI sound engine chips: Calvin, Hobbes
0, and Hobbes 1.
Calvin handles sample fetching tasks, and Hobbes 0 and 1 primarily perform
wave form generation and DSP functions.
35
Calvin and Hobbes have undergone rigorous testing prior to mass manufacturing,
and are extremely unlikely to be defective. If one or more of the steps in
this test indicate a failure, the problem is probably to be found in the
connections between the chips and the CPU. The test results are displayed as
shown in Figure 4.17. The significance of failures in each step of the test
is explained below. Steps 2.1—2.3 test the connections to the internal RAM
and internal registers of Calvin and Hobbes.
When analyzing the results of these tests, you're looking for a PASS message
for each test. The hexadecimal numbers that appear on the display are
irrelevant as long as they match for each test. The numbers shown in figure
5.16 are arbitrary, and may differ from the numbers you see in the unit you're
testing. This does not indicate a problem.
If any of the tests fails, check the connections between the components in
question, as described in the paragraphs below. If test #2 fails, check all
the connections between the CPU, Calvin, Hobbes 0, and Hobbes 1. Note that
expected observed values are not given on the line indicating the status of
test #2. Instead, a PASS or FAIL message will be indicated for each of the
three steps in test #2.
NOTE: The Calvin and Hobbes tests are interrelated, and are sequenced from
the least significant test to the most significant. If any one test fails,
the subsequent tests are likely to fail as well. In this case, start by
checking the components related to the first test that failed. If a
malfunction occurs, repair the malfunction and run the Calvin and Hobbes test
again. It is likely the subsequent tests will pass when retested.
In general, the data flow through these components is the following sequence:
Sample memory, Calvin, Hobbes 0, Hobbes 1. If test failures occur, check the
connections to Calvin before checking those to Hobbes 0, and Hobbes 0 before
Hobbes 1 . .
36
This test is relevant only if the sampling option has been installed in the
unit under test. If no sampling board is present, this test will fail.
This test is only necessary when the sampling option has been installed.
The "five- step sampling option test" checks the integrity of the sampling
option board and its connections to the CPU. The five steps check the
following items:
If there is a failure in any of the five steps, the LCD will show a failure
for the entire test. In this case you should check the connections of the
sampling option board. When done, run the test again. If it fails again,
replace the sampling option board.
37
This test is intended for use with an oscilloscope connected to the K2000's
audio output, and enables analysis of the purity of the sine wave generated.
If the sine wave is impure, a problem with the sound engine is suggested. To
monitor the Sine wave test, connect the scope to any of the B outputs, the MIX
outputs, or the Headphone jack.
This memory test takes about five seconds to run, and checks the data
integrity of the K2000's sound ROM (the factory sounds), and the connections
between the sound ROM and the CPU. The test is a simple checksum procedure
for each pair of Sound ROM chips. Failures are indicated as data or address
failures, as described in the overview. The display indicates the success or
failure of each of the four blocks of Sound ROM (A-D) . Each block
corresponds to two Sound ROM chips. In the unlikely event of a failure, the
engine board must be replaced, since the Sound ROM chips are soldered in
place.
This memory test checks the address space of the sample RAM (SIMMs) The .
purpose of this test is to determine whether the K2000 recognizes all of the
sample RAM that has been installed. The more RAM that has been installed,
the longer this test takes. The test takes approximately 20 aecondg for each
Megabyte of RAM.
The K2000's sample RAM space is divided into two banks (Bank 0 and Bank 1),
representing the two pairs of SIMM sockets. Each bank is divided arbitrarily
into four blocks with a maximum address space of 8Mb. These blocks, A-D,
represent internal divisions of the address space; and do not correspond to
any physical division of the sample RAM.
When the Sound RAM test is run, the display will indicate which block in which
bank is currently being tested. This is shown by a series of dots that
appear as the test proceeds. As each block is finished, the dots are
replaced with either a success message (OK) or a failure message. If the
block fails, the malfunctioning address is indicated, as well as the expected
checksxam value and the observed value. See Figure 4.18.
Each of the eight blocks corresponds to 8Mb of available RAM. The number and
size of the SIMMs installed in the unit determine which blocks are used. For
example, assume two pairs of 4Mb SIMMs are installed; thus 4Mb x 4, or 16Mb of
sample RAM. If the Sound RAM test is executed, and all RAM is recognized,
the display will show "8.0M Found" in Block A of each bank when the test
finished.
38
The sample screen shown in Figure 5 .17 indicates that there are 2 Mb of RAM
installed in Bank 0 (two 1Mb SIMMs) The test is currently checking Block C
.
of Bank !•
<lS6und RAM
H Nothing :> ^ , .X
^ c /: ^
;
This memory test checks the function of the backup battery power to the non-
volatile RAM (NVRAM) . This test works in conjunction with the 2nd NVRAM
test; no success or failure will be indicated at the completion of the 1st
NVRAM test. Instead, the tester will be prompted to power down the unit.
Wait 60 seconds after powering down, then press and hold the 1, 2, and 3
buttons while powering up again. Return to diagnostics mode and select and
execute the 2nd NVRAM test. If a failure is indicated, power is not being
39
Ssp!a!)®!tf Ssu!!!''
5®®^^ts. the LCD to
^ Of course the engine board and LCD must
<-• be connected
results. The LCD diagnostic routine is
easifrto^’read^thfn^rh^
it rather than the lL !ou!ine?
40
completed successfully, all 8 of the panel LEDs will flash ONCE about 1 second
after power-up. If the Pitch Wheel is too far off center for full range
operation, the LEDs will flash TWICE. If the battery voltage is too low for
reliable memory retention, the LEDs will flash THREE times. Four flashes
indicate both a low battery and an off-center Pitch Wheel.
The LCD scanner diagnostic may be started when the K2000 is displaying its
Program Page. Press the 4, 5, and 6 buttons simultaneously. The scanner
will respond by lighting each LED in sequence and then displaying the
following on the LCD:
It is possible the battery voltage and wheel center values will differ from
unit to unit. Fresh batteries should read in the range of 4.3 - 4.8 volts.
If the battery reading is less than 3.5 volts, it is nearing the end of its
life (warnings will start to appear when it declines to 3.2 volts) The .
wheel center should be between 120 and 136 (inclusive) for full range
operation.
A. When the Pitch Wheel is pushed back, the display should show "PITCH WHEEL"
and the numbers should smoothly increase toward 254 or 255. The numbers
will move in increments of 2 or more depending on how fast the wheel is
moved. If the wheel is pulled forward, the niimbers should decrease toward
1 or 0, again in steps of 2 or more. When the wheel is released, the
nximber should return to a value between 120 and 136. If the centering or
range is off, there are two adjustment potentiometers on the wheel assembly
that can be moved to correct the problem. If the display flickers or
changes to BATTERY or CONTROL PEDAL when the Pitch Wheel is pulled full
forward or pushed full back, these pots should be adjusted. A perfectly
adjusted Pitch Wheel reads within 1 count of 128 in the center position,
reaches 254/255 just as it hits the back stop, and reaches 0/1 just as it
hits the front stop.
B. Moving the Mod Wheel should cause the display to show "MOD WHEEL" and the
numerical readout should react as with the Pitch Wheel. The range should
be from 0 or 1 to 254 or 255. If the range is insufficient or reaches the
41
0/1 and/or 254/255 limits too soon, there are 2 more adjustment
potentiometers on the Mod wheel assembly. If the display flickers or
changes to BATTERY or CONTROL PEDAL when the Mod Wheel is pulled full
forward or pushed full back, these pots should be adjusted.
E. If a switch pedal is plugged into the Switch Pedal 1 jack, the fourth line
will show "PEDAL 1" when the pedal is closed, and will go blank when it
opens. Some pedals (especially Roland) are normally closed, and open only
when pressed. Most others are normally open, and close when pressed.
The K2000 can use either type (as long as it is plugged in before powering
up) but the scanner diagnostics expect to see a normally open pedal.
After testing a normally closed switch pedal, the pedal must be unplugged
to clear the fourth line of the LCD. (This applies to Switch Pedal 2,
also.
F. If a switch pedal is plugged into the Switch Pedal 2 jack, the fourth line
will show "PEDAL 2" when the pedal is closed, and will go blank when it
opens
G. If the Alpha Wheel is turned clockwise slowly, the display will show "SPIN
KNOB" followed by a number that counts up (0, 1, 2, 3) and repeats.
Likewise if it is turned counter-clockwise, the numbers will count down.
Any other sequence indicates a problem-usually a bad connecting cable, but
possibly a bad switch contact.
H. If any button is pressed, the word "BUTTON" followed by its name will be
displayed. If it is one of the mode buttons, then its associated LED will
flash as well. It is normal for the LED associated with the PROGRAM
^'^tton to flash more brightly than the other buttons. If several buttons
have problems, the cause is usually the connecting cable. If only one
button is bad, then it needs replacing. Cable problems that seem to
when the unit is disassembled are often due to a short circuit
caused by component leads on circuit boards puncturing the cable'
insulation. The ends of ribbon cable wires touching the conductive
coating inside the top cover can also cause inproper operation.
I. Pressing any keyboard key down halfway will display the word "KEY" followed
by the key's note name (such as C#4 for the black key immediately to the
right of middle C) followed by the key's MIDI ID number (61 in this case).
This indicates that the "upper contact" of the key has closed. Pressing
lightly all the way down should display the same note name and MIDI number
a second time which indicates that the "lower contact" of the key has
closed. The two names/numbers should match. Most keyboard errors are
42
J. If a keyboard key is pressed down firmly, the pressure sensing circuit will
begin to respond and the display will change to "PRESSURE = XXX." "XXX"
will vary between 4 and 255 according to how hard the key is pressed.
When pressure drops below 4, the display switches back to the note
name/number display. There is a pressure sensitivity adjustment on the
audio/power board (at location R201, near the large filter capacitor) that
may be changed to suit an individual customer.
NOTE; There are two additional internal signals called AUXl and AUX 2 that
are normally unused. If the fourth line ever shows "AUX 1" or "AUX 2,"
then there is likely a short circuit (usually to a cable running under the
board) or current leakage in the audio/power board.
The LCD based diagnostic may be exited and normal K2000 operation restored by
pressing the EXIT and ENTER buttons together. The K2000 should restart just
as if a soft reset had occurred.
In cases where the K2000 appears to be "dead," the scanner microprocessor and
user interface controls may still be functional. If the eight panel LEDs
flash at all shortly after power-up, this indicates the scanner processor is
running which in turns means that the power supply is probably functional.
To enter the LED Scanner diagnostics, press and hold the 7, 8, and 9 buttons
while the power is off then turn the power on while continuing to hold them
down. The panel LEDs will flash in inverse sequence to indicate entry into
the LED diagnostic. After the light show, the software version number ($14)
will be shown briefly in hexadecimal, followed by the battery voltage in volts
(left hex digit) and tenths (right digit). The battery voltage should remain
on the display but early serial number units may display it only briefly.
The LED-based diagnostic uses the 8 mode LEDs for a two-digit hexadecimal
readout. The LEDs are sequenced from top left, down the left column, to the
top right, and down the right column. The values the LEDs represent are as
follows: Program mode = 128 decimal ($80 hexadecimal; Setup mode = 64 ($40);
Quick Access mode = 32 ($20); Effects mode = 16 ($10); MIDI mode = 8 ($08);
Master mode = 4 ($04); Song mode = 2 ($02); Disk mode = 1 ($01). See Figure
4.21.
43
$21 9 $51 -
$81 Soft D
$22 8 $52 + $82 Soft E
$23 7 $53 ENTER $83 Soft F
$24 SETUP $54 MIDI $84 DISK
45
4.27.2 To replace the batteries, simply orient the unit such that the
rear panel is accessible, particularly the lower right corner of
the rear panel.
• Locate the round battery cover and, using a coin, rotate the cover counter-
clockwise 1/4 turn. It will naturally eject itself into the hand because
the batteries are spring loaded.
47
48
The M37450 is RESET by the same signal as the Engine Board and the Effects Processor which en-
ables simultaneous starting. The lOMHz clock for the M37450 is delivered from the Enqine via J7
^
(A-5).
The analog I/O section will address 8 analog inputs and two analog outputs. It contains only one
8-bit A/D converter and internal circuitry to select which of the 8 inputs is addressed at any
given time. There are two separate latched 8-bit D/A converters. There are separate reference
voltages for the A/D and D/A converters. These are connected to +5Vref. The M37450*s digital
circuitry is connected +5Vdig.
It is important to note that during power-up the +5Vref rises faster than +5Vdig which can cause
the M37450 to "latch up" and overheat. D2 (D-6) prevents this by keeping +5Vref from being more
than 0.6 volts more positive than +5Vdig during power-up.
The music keyboard has two PC boards with the keyswitches on them and a cable from each. Each
board has an 8 column by 8 row keyswitch matrix which covers 32 keys with two switches per key.
Connected together, the matrix is 8 columns by 16 rows. Because there are 61 keys on the key-
board, only 58 of the 64 positions on the treble circuit board are connected.
The matrix, part of which is diagrammed below, is connected such that the 8 columns are driven by
Port 2 and the 16 rows are sensed by Ports 0 and 1. During operation, the column being driven is
low and the remaining 7 columns are high. Pull-up resistors in RP7 and RP8 hold all of the rows
high. If a switch contact is closed, it will pull the row to which it is connected low.
Keyswitch isolation is accomplished by the diodes in series with each key contact. The M37450
drives each column in sequence, then reads the rows on Ports 0 and 1. This scan sequence is
repeated every 250uS and the resulting keyboard image (status) is stored in internal RAM. The
high scanning speed is necessary to resolve minute variations in keyboard velocity for expres-
sion.
Col COL
COL
45
HOW B
—
W-C7
^ u r'w iOK
: ‘"U u u
CZ-^2
ROW
$
Key* u L...^vwv—
u
Kcy
O-C-Z vf »
Figure
'M 5.
'Uj
1.1.1 - Portion of keyswitch matrix
{
Because the keyboard is large, the printed circuit wiring is extensive, and the boards do not
have a groundplane, there is potential for the keyboard to radiate RFI form the constant scan-
ning. R-C filters are used on all keyboard lines to "soften" the waveform edges and suppress RFI
from the clocks. Resistor networks RP1-RP6 are 100 ohm series resistors and 24 individual lOOpf
capacitors are used on the scanning lines.
J3 (1-5 to 1-6), which is the connector for the bass portion of the music keyboard, is a
26-pin
header, although only a 16-pin header is installed for normal operation. With a 26-pin header
installed, it and J2 (1-3 to 1-4) become a debug port. When JPl (E-7) is open-circuited (a thin
trace on the solder side must be cut), the M37450 begins operation in "microprocessor mode"
in
which case Ports 0 and 1 become an address bus and Port 2 becomes a data bus. The extra
10 pins
on J3 carry the remaining signals needed, including power, to interface to a "debug pod" which
contains EPROM (to hold a monitor program), RAM (to hold the program being debugged), serial port
logic to talk with the development host, and logic to recreate ports 0-2.
NOTE: This port is normally only used at depot level service. Do not attempt to use this port
for debugging unless specifically instructed. The purpose for this discussion was to define the
reason for the extra pins at J3 and the need for JPl to be intact for normal operation.
The K2000 front panel consists of 38 pushbuttons, 8 LEDs, and a continuous rotation "spin knob".
The buttons are wired into a 10 column by 4 row matrix. U31 (B-4 to B-5), receiving 4-bit binary
information from the M37450 drives one column at a time low. The 4 rows are held high by pull-up
resistors in RP9 (E-2). If a button is pressed, the appropriate row is pulled low and sensed by
the M37450. Because there are no isolation diodes for the front panel, the number of simultane-
ous button closures that can be sensed is limited to 2 unless they are all in the same row or
column.
The LEDs are wired in an 8 column by 1 row matrix, driven by a transistor current booster Q1 (B-
4). To turn an LED on, the M37450 will activate Q1 by driving Port 6, bit 7 high when the column
connected to the LED is selected. Q1 is normally off when the M37450 is actually sensing buttons
in order to minimize the saturation voltage of U31 and give a good logic low level to sense. Ql
is an emitter-follower which minimizes current drawn from the M37450. R81 suppresses possible
emitter-follower oscillations.
The spin knob is a rotary switch with 2 contacts, 36 positions, and continuous rotation. Each
contact is either open or connected to GND2. The switching sequence (Gray code) is such that the
M37450 program can detect which direction the switch is being rotated and how many positions.
One property of this code is that if the two contacts connections are reversed, the direction of
rotation will interpreted as reversed by the M37450.
U60 (C-5) is used to interrupt the Engine microprocessor by activating EXTIRQb at pin 6. This is
accomplished by taking column C high while column D is already high. The Engine acknowledges
this, and resets U60, by sending EXTIAKb to pin 1 of the flip-flop.
The wheels are standard rotary potentiometers actuated by large "thumbwheel" knobs. Only 1/3 of
the allowable rotation is used, thus the mechanical center must correspond fairly close to the
electrical center of the pots. Amplifiers U59-7 and U59-8 provide the necessary gain so that 1/3
rotation results in a full 0 to 5 volt swing for the M37450's A/D inputs. C194 and C191 filter
noise and smooth the output. R149, R150, D5 and D6 prevent overload of the M37450 A/D inputs in
the event of gross wheel mis-adj ustment
The Wheels circuit board itself contains a network for each wheel which allows electrical ad-
justment of the centering and exact amount of amplifier gain. Figure 5. 1.1, 2 below is the
schematic for one wheel; both circuits are identical. The circuit board’s pots are adjusted to
match the characteristics of each particular wheel’s pot.
The keyboard pressure sensor is a long strip between the keys and the keybed. When pressure
greater than that needed to hold a key down is applied, the strip is compressed and changes from
an open circuit to a resistance that varies with the amount of pressure.
R206 and D19 together make a 0.6 volt source with just a few ohms of dynamic impedance resulting
in a reasonably linear current/pressure curve. This voltage is applied to the pressure strip
through J5-6. (A-2) The resulting current flow ranging from 0 to about 1.0mA returns through J5-
7 and is converted to a voltage by U59-1 (C-2) R209 adjusts sensitivity and C159 filters out
.
noise. R147 and D4 prevent overvoltage, (from overpressure) from affecting the A/D inputs of the
M37450. D4 (D-2, Schott ky diode) clamps any overvoltage to no more than 0.3 volts above the A/D
supply voltage.
J19 (A-1) receives variable inputs from a '•control** pedal and can be used to control timbre,
pitch, etc. The circuit is optimized for a lOK linear taper potentiometer in the pedal, but will
work with higher resistances. Alternatively a control voltage of 0 to +5 volts can be input di-
rectly to J19 to produce the same effect. For use with a pot type pedal, R197 and R199 (B-1)
supply a current-limited reference voltage to the pedal. This voltage is equal to the +5ref when
the pot s resistance is lOK; thus there is no voltage drop across the limiting resistance.
Limiting resistance is necessary to prevent circuit damage or malfunction is a mono plug is
inserted into J19.
R195 pulls the pedal voltage input to zero when there is no plug in J19. R196 and D18 (C-l)
protect the A/D converter from overvoltage and C184 filters out noise.
J16 (I-l) and J17 (1-2) receive footswitch plugs and look for switch closures to ground. RP7-10
(G-2) and RP8-10 (F-2) are pull-up resistors and keep the M37450 inputs high when no
pedals are
or the pedal contacts are open. RFI suppression is accomplished in the circuits of
R188, R189, C180 and C181 (H-1 to H-2). At power-up, the M37450 looks at the status of J16 and
J17 and assumes that state to be "OFF**; the opposite state being **0N'*. This allows for use with
NO or NC pedal switches.
The LCD on the front panel requires an adjustable negative voltage source to control the
contrast. This voltage must not be applied until after the LCD has been initialized. The LCD
Vee circuit (F-7 to H-7) performs these functions.
The pulse-width modulated output from the M37450 is used as a third D/A output to control
the Vee
voltage. When filtered by R146 and C196, the duty cycle of the PWM waveform is converted
into a
voltage between 0 and +5 volts. U59-14 converts this to a -12V to +12V swing. Note that the
center voltage from the Wheels Interface (+2.5V) is needed to do this properly. If the Wheels
Interface is not plugged in, the LCD contrast will be wrong. D20 and R154 prevent the Vee
voltage reaching the LCD from exceeding +5 volts. Vee will be affected by changes in the +5
volt
digital supply. If it varies by more than 0.1 volt high or low, the LCD contrast adjustment
range may not suffice to compensate.
Q20 and R152 are connected so that the Vee voltage is forced to the positive rail when RESET
is
active, thus ensuring the display is inoperative when the processors are not running.
Approxi-
mately one second after RESET goes inactive, software will set a default level for Vee, about -
8.5 volts. This voltage may be changed by a message from the Engine Board to set to a user
stored value, if so programmed.
The audio path requires 6 digital signals to control switches and 2 analog signals to
control
variable gain amplifiers. The digital signals are noise-isolated from the analog circuitry
y bvy
R79-80 and R136-139 (F-6).
The 8-pin option connector (E-1) is included to utilize unused resources of the M37450
and to
provide for additional user control features in the future. Power (+12 and +5 volts), two analog
inputs, and one digital I/O line are provided.
The rear panel and bottom shield are used as a heatsink for the five TO-220 packaged power
semiconductors. For testing in unrestricted free air, just the rear panel is acceptable as a
"short-term" heatsink although it will get hot to the touch. NOTE: TBSTIHQ WITHOUT A HEATSINK IS
NOT ALLOWED AND WILL RESULT IN SHUTDOWN OR DESTRUCTION OF SOME OF THE POWER SEMICONDUCTORS IN A
MATTER OF SECONDS. The surfaces between the shield plate and the rear panel should be coated with
white thermal heatsink compound for improved heat dissipation.
The rear panel is at analog/chassis ground potential; thus all power semiconductors must use an
insulated mounting. The sketch below (figure 5. 1.2.1) details the mounting hardware necessary to
accomplish this. When installing this hardware it is critical that the semiconductor mounting
hole be aligned with the panel mounting hole. If they are not and the screw is forced in, the
screw threads will cut through the wall of the shoulder washer and short-circuit against the
semiconductor’s mounting tab.
^ JFLiT
flat dasher
WASHER
FOH/E/?
$£MI- FAO
comucroe
f H^5 jA/K \
A separate power transformer converts selected line voltages to 100, 115, or 230 volts. It
contains two center-tapped secondaries, 12 and 28 volts. Low voltage AC enters through J20 (A-
1). All voltages in this section will be given for the 115 volt configuration. Figure 5. 1.2.
below details the external transformer connections.
SL/KX/WHT BLUE
\15V A Btwe/WHT.
BLfiCK *-
BLUE
dRcvw/Yei, lyy -0 RED
^
fcV
ioo\/ REP/W«r.
6vf
BfKOViN -• RED
Figure 5. 1.2. 2 - Power transformer wiring
D23, D24 (B-1) and C192 (C-1) provide unregulated +20VDC to TP25 (C-1) from which +12Vana,
+5Vana, +5Vref and +8VDC are derived. D21, D22 (B-1) and C190 (C-1) deliver unregulated -20VDC
to TP22 (C-1) to provide -12Vana, -5Vana and -15Vbias. C192 is larger than C190 because of the
potentially greater load from the optional hard disk on the positive supply. Cl 97 and Cl 98
prevent low frequency RF rectifier switching noise from feeding back through the line cord.
NOTE: No secondary fuses are provided. Primary fuse In power entry module protects sufficiently
against current overloads.
VR5, C186 (F-1) provide +5Vana and VR2, CL88 (G-1) provide +5Vref; both circuits are sourced by
VR7, C207 (+12Vana 0 TP23, D-1 & E-1).
VR3, C193 (D-1) provide +12VDC for the optional hard disk at TP19 (H-1) and pin 1 of J18 (I-l).
R200 is required to lightly load VR3 when the hard disk is not installed; this prevents
oscillation.
VR6, C206 (D-2) provide -12VDC for the audio circuitry at TP24 (E-2) and becomes -12Vana.
VR5 and VR4 provide quiet +5 and -% volt power to the analog switches in the audio path. VR2
provides an isolated and quiet +5 volt reference to the A/D and D/A converters on the scanner
microprocessor and sever performance controls.
VRl, C182 (D-2) outputs -ISVbias for the LCD Vee generator and muting circuit. This regulator
may cease regulating at low line voltage, however the only effects are a possible reduction in
LCD contrast and premature clipping of large negative signal peaks.
The circuit of D25 (dual Shottky, B-2) and C202 (B-3) provide high current availability of +8VDC
(unregulated) 8 TP17 (C-2) This current source is used by the digital regulator circuit(s).
.
Due to the lower forward voltage drop of a Shottky rectifier, less heat is dissipated by the
rectifier and the transformer voltage can be lower; however because of the heavy current drain on
the +5 volt supply (up to 3.5 amps), it is necessary to heatsink D25 to the rear panel. C200 and
C201 suppress rectifier switching noise.
The circuit of U62 (C-3) and Q18 (E— 3, a VMOS power FET) makes an ideal pass transistor for a
linear regulator and provides +5VDC for all digital circuitry at the output end of L5 (1-2).
This circuit is wired as a voltage follower, thus the final output voltage at the source of Q18
is forced (by the gain of the op-amp, U62) to equal the reference voltage at its non-inverting
input. This voltage can be monitored at TP16 (H-2) The output is adjusted at R180 for exactly
.
5.00 volts (C-4) to optimize performance of the custom VLSI chips on the Engine Board.
It is extremely important for the Engine board to know when there is sufficient voltage for
proper operation so as to reset itself. The mute circuit (s) must also be asserted when there is
insufficient voltage. Simply put, when the output of U62 is within 1.2 volts of its supply
(close to its saturation point), Q16 and Q17 cut off saturating Q15 and discharging C176 quickly.
The reset circuit (described later) immediately asserts RESET to all microprocessors and mute
circuit (s )
R192 and C174 stabilize the feedback loop to prevent oscillation of the regulator. C178 and C180
improve regulation during sudden load changes. R193 provides a minimum load if the power supply
is ever operated with no other components on the board.
Three AA alkaline cells provide power to three RAM chips during power off. Standard alkaline
cells are used instead of the more expensive lithium cells because power drain is approximately
the same; thus reducing cost to the customer.
When power is off, current from the battery passes through R182 and D16 (1-3) to the BUVdd line
which is connected to the power pins of the 3 RAM chips (one on the Audio/ Power Board and the
remaining two on the Engine Board) D17 prevents battery power from backing into unwanted
.
circuitry.
With power up, Q19 provides power to the RAM chips. D17 ensures gradual application of full
voltage to the RAMs during power-up. C203 (E-7) provides 30 - 60 seconds of RAM power to allow
for battery change when needed.
The backlight for the LCD requires approximately 100 volts rms, at approximately 360Hz operation.
The circuit of Q25 and T1 (G— 4 to H— 5) fulfill this need. The backlight cable must be plugged
into J21 (1-5) for Q25 to oscillate. A reasonably good sine wave appears at J21-4. C208 and
R218 provide the positive feedback route.
Q24 and surrounding circuitry provide a controlled higher voltage from the unregulated +8 volts
so as to improve LCD brightness. The output voltage can be adjusted to balance between
brightness and the acoustic noise level from the panel by changing the value of R216 (F-5).
5 . 1 . 2 . 8 RESET CIRCUIT
The reset circuit provides a power-up delay between sufficient power for operation and release of
the RESET signal to the microprocessors. Also, at power-down (or severe line voltage dip), this
circuit re-asserts RESET when the +5VDCdig regulator drops out of regulation. The reset circuit
consists primarily of U61 (H-3). It is wired as a high impedance schmidt trigger and gives a
fast rise and fall time signal to the microprocessors. Feedback through R156 (H-3) latches the
circuit in either the high or low state. Current through R155 overrides the feedback current and
forces the latch to flip. The last stage (U61-8) enables both active-high RESET and active-low
RESETb.
The audio section, less effects, is on schematic page 3. This covers D/A conversion, filtering,
switching, mixing, and amplifying circuitry that handles the "dry" audio signals.
U37 and U39 (B-1 to B-4 are each dual 18-bit D/A converters with serial data inputs.
)
There is
thus one D/A converter for each audio channel. Although the Hobbes chips on the Engine Board are
capable of multiplexing one DAC among 2 or 4 channels, the one output per DAC mode was used on
the K2000 to minimize noise and distortion.
The AD1864 is a dual DAC with separate serial inputs, separate Latch Enable pins, and a common
clock pin. The Hobbes chip on the Engine Board was designed so that the sequence and timing of
the DAC signals would be directly compatible with the AD1864 and all other serial input DACs
claiming "PCM56 compatibility". The serial bit input rate is 9.968MHz and the word input rate is
actually 96.000KHz. This is not the same as "2X oversampling" because each pair of very slightly
different words is derived from the same sample but it does significantly reduce distortion of
very low level signals. Either current or voltage output connections can be used. Here the DACs
are wired for voltage output which is fixed by the chip design at +/-3 volts full-scale. Note
that only the Engine Fake board or specially programmed sounds ever approach this high amplitude
level; normal keyboard playing produces amplitudes in the 0.1 to .05 volt range.
For minimum digital noise feedthru, the power and ground wiring of the AD1864 is very critical.
The digital section of the chip requires + and -5 volts and has its own ground as does the analog
section. +5ana and -5ana power the analog section directly and the analog section ground is tied
to analog ground. +5ana and -Sana also go through RF chokes LI and L2 (B-5) and filter
capacitors C128, C169, C130, C172 (A-2 to A-4 and power the digital section. The chokes and
)
capacitors prevent digital activity in the chip from feeding back into the analog power supplies.
Ground for the digital section is connected to digital ground. In the layout, the digital
groundplane is under the chip*s digital pins while the analog groundplane is under the remainder
of the chip.
U38, U40 and surrounding components (E-1 to E-4 amplify the +/-3 volt full-scale signals from
)
the D/A converters to just over +/- 6 volts to optimize the signal-to-noise ratio through the
following filters. They also provide a small boost at very high frequencies between 15KHz and
20KHz to counteract a slight droop in the downstream filters. Sampling theory is valid only for
signals reconstructed as impulses. When reconstructed as stair-steps, the high frequencies near
1/2 the sampling rate are rolled off by as much as 4dB. At 20KHz in a 48KS/s system, the loss is
about 2.8dB. Although the U38 and U40 circuits are low-pass filters with a cutoff of about
25KHz, their "Q" is high which gives a boost to the frequencies just below cutoff. Note that
incorrect values for the resistors and capacitors could easily cause a very distorted frequency
response curve or even oscillation.
F11-F14 (F-1 to F-4) are passive, L-C type, low-pass filters that remove high alias frequencies
from the D/A converted signals. With the source and load properly matched, the loss through the
filters is slightly more than 1/2 (6dB) which brings the +/-6 volt boosted signal from the DACs
back to just under +/-3 volts.
Although the filters are efficient and have a very sharp cutoff, they are vulnerable to AC
magnetic fields because of the coils. In order to minimize audible hum, they must be kept away
from the power transformer and the low-voltage wiring between the transformer and J20 which also
produces a magnetic field because of the large currents flowing. In the K2000 case, the power
transformer must also be oriented such that the filters are in a "null point" of the
transformer's flux field. In addition, they could even pick up hum from another piece of
equipment, such as a power amplifier or another instrument, placed immediately below or above
the
K2000 case. TAKE CARE TO PREVENT THIS.
U34 and U35 (H-1 to H-4 amplify the +/-3 volt signals coming from the low-pass filters to +/-6
)
volts. They also provide sufficient power to drive the 4 separate output jacks and loads as low
as 600 ohms. R163 and R168-170 provide a source impedance to the jacks of 600 ohms, isolate the
amplifiers from capacitive loads (like long cables), and are also part of the RFI suppression
network along with C167, C170, C171, and C177.
During power-up and power-down, muting circuits on all audio outputs are asserted to prevent
transient noises and damage to speakers and human ears. The muting circuit description provided
here applies to all other outputs as well.
Using the Qll circuitry as an example (I-l), this J109 N-channel JFET is either ON or OFF. When
OFF, it is not conducting, thus any audio present passes through R163 and out Jll. When ON this
JFET shorts the output line to ground, thus muting the circuit. To control this function take
note -UNMUTE arriving in the upper left corner of the schematic (A-1), being delivered from the
collector of Q23 (H-7) on schematic page 2. This signal, when high, cuts off Qll, and all other
JFETs so connected and allows the passage of audio. When low, the JFETS are ON and the mutes are
asserted.
The scanner microprocessor has direct, individual control over the 6 analog switches and pairwise
control of the 4 variable gain amplifiers. Five of the analog switches (ASW 1, ASW 3 - ASW 6)
are in audio signal paths while the sixth is in a gain control signal path (MW_2).^ ^
Note: ASW_1 and ASW_^2 audio switching controls are drawn on schematic page 4.
Refer to figure 5. 1.3.1 below for a simplified detail of the Audio Switching Matrix.
In all of the audio path switching cases, the switch is connected to the summing junction of a n
inverting op— amp with the input resistor at the input of the switch. Signal current through the
input resistor either enters the summing junction when the switch is on or is diverted to ground
when the switch is off. This makes each switch a current-mode switch which allows it to block
large voltages when off and avoid large voltage swings when on. This minimizes distortion of the
signal and crosstalk. It also allows the switch to handle signal voltages greater than its own
power voltage. A simple detail of the Analog Switch follows in figure 5. 1,3. 2.
ES:n avTPur
coahrol
Note that the IN and NC terminals are connected when the control voltage is low {switch off) IN
and NO are connected when the control voltage is high (switch on).
The two sections of U1 (C-6 to C-7) do the final mixing of audio signals from the switching
matrix and Effects Unit. Each sums 5 signals from various sources into one. A low noise
amplifier is used here because the "noise gain" of the circuit is 5 even though the effective
gain for any one signal is unity. The lOOpF capacitors across the feed back resistors compensate
for stray capacitance at the large summing junction and also the finite load capacitance
presented by the cable to the volume slider. The signal level going into the volume slider is 6
volts peak if one of the mixed sources should reach peaks near 6 volts simultaneously.
The return signal from the volume slider varies from 0% to 100% of the signal driven to the
slider by Ul. The final mix output amplifier for the Right Channel is U33-1 (G-6) which simply
inverts the slider's signal at unity gain. If no plug is inserted into the Right mix output jack
(JIO, 1-6), the Right output signal is also sent to the Left output amplifier to be mixed with
the Left signal from the volume slider. This provides automatic mono output through the Left
output jac)c when no Right channel cable is used. Note the Right signal must be mixed without
inversion into the Left output amplifier to keep phasing correct in the mono signal. The 600 ohm
source impedance and muting arrangement for the mix outputs is the same as for the separate
outputs described elsewhere.
To drive the headphones adequately, a non-inverting gain of 2 amplifier with a power booster
stage is used. Using the Left channel as an example, op-amp U5-1 (F-6) provides the voltage gain
and Q2-Q3 provide current gain to drive low-impedance headphones. R91, between the bases and
emitters, allows the op-amp to drive the load directly for currents less than 6mA peaks and
eliminates crossover distortion. R160 {H-7) protects the transistors from user overload caused
by a short-circuited cable or driving a too great load.
It is possible for the headphone amplifier to clip at high volume slider settings. This occurs
at about 10 volts peak for typical headphone loads and would normally be extremely loud. This
clipping will not affect any other audio outputs.
Muting of the headphones is accomplished by Q4 and Q7 (F-6) by shorting the op-amp's (U5)
outputs. If the Audio/ Power Board is operated with a signal but the mute is activated for an
extended time (such as on the test bench), U5 will become quite warm, but will not be damaged.
Four voltage controlled amplifiers are also part of the audio switching matrix plus they allow
adjustment of the ratio of unchanged "dry" signal to altered "wet" signal from the Effect Unit.
The SSM— 2122 amplifiers used are duals but each half has independent inputs output, and control
terminals while the bias current terminal is common.
U9-14 (E-3), which controls the gain between SEPOUT_2 and OUT2_MOD, will be used for discussion.
The amplifier is a current— in/current— out device in which the current gain is varied by a control
voltage. For low frequencies, R108 ((B-2) converts the SEPOUT_2 signal into a current which is
fed into U9-10. Its value of 15K converts a maximum signal peak of 6 volts to a current peak of
400uA. When U9 is set for unity gain, its output current will be the same as its input current,
namely 400uA. This current, fed into the summing junction of op-amp U36 (F-3), will be converted
back to 6 volts by virtue of its 15K feedback resistor.
At frequencies greater than approximately 2KHz, R107 and Gill begin to boost the input current
for a given input voltage while R124 and C124 supply an exactly opposite cut in output voltage
for a given current. This boost/cut, which reaches a maximum of nearly 12dB at 20KHz, reduces
the apparent random background noise level of the SSM-2122 by about lOdB. The SSM-2122 will clip
at input/output currents greater than SOOuA peak, which is twice the bias current through F-
R68 (
4), so the maximum signal at high frequencies is limited. Under virtually all normal conditions
an audio signal carries substantially less energy at high frequencies than at low frequencies,
thus making this nearly impossible. R51 and C35 (E-3) prevent the SSM-2122 from oscillating.
Because the SSM-2122 *s input impedance is very low, they have minimal effect on the high
frequency response.
For gain control, the SSM-2122 has differential voltage inputs, +CTL and -CTL. When they are at
equal voltage levels, gain is unity. As the + input becomes more negative than the - input,
gain
reduces on a logarithmic scale at the rate of .005V/dB. The 6.5:1 voltage divider formed by R106
and R103 make this .039V/dB which for a 5 volt control range translates into 128dB of
control
range. The control voltage ultimately comes from an 8-bit D/A converter in the scanner
microprocessor, the control sensitivity becomes l/2dB per bit. R57 (E-3) is needed to balance
the control voltage inputs.
For the UO-14 example, the control voltage is switched by an analog switch. This allows that
particular amplifier to be set to unity gain while the other amplifier (U9-2, E-4), driven from
the same control voltage, continues to be controlled. Unlike the audio path switches, this
control path switch is operated in the voltage switching mode which is OK because any non-
linearity won't cause waveform distortion.
Control voltage buffers, U32-1 (B-6) and U32-7 (B-4) allow the relatively weak D/A converter
outputs from the M37450 to drive the qpproximately 600 ohm load presented by two SSM-2122
control
voltage inputs. R86, R92, C106, and C107 (B-4) smooth sudden changes in the control voltage
level to provide smooth, clickless changes in gain.
The Effects Mixer is shown on the top section of schematic page 4. U6-7 (E-1) is the actual
mixer stage which combines SEPOUT_l and optionally SEPOUT_2 into a mono signal for the
effects
processor. Analog switch U2 (C-1) determines whether current from the SEPOUT_2 signal
enters the
mixer stage or is diverted to ground. Because the reverb circuit is noisy and also because the
A/ D/A conversion process does not have a very flat frequency response, two poles
of high
frequency boost are added across the input resistors R3 (B-1) and R43 (E-1). The gain
of this
stage is unity from wither input signal at low frequencies. The effects A/D converter
clips at
+/-8 volts leaving some headroom for both inputs to approach their 6 volt
maximum.
5. 1.4. 3 EFFECTS INPUT LOW-PASS FILTER
The input anti-alias lo-pass filter is split between schematic pages 4 (F-1 to G-2) and
5 (A-5 to
B-6). The complete filter consists of two cascaded 3-pole Butterworth sections with component
values again scaled from the DSP-256 values to account for the 32MHz clock. Although
quite
different from the L-C filter described elsewhere, this circuit was selected to avoid
alterina
the desirable "DSP-256 sound".
10
The Effects Section is mono-in, stereo-out and can perform reverberation, echo, delay, frequency
equalization, chorusing, and flanging. Although the processing happens digitally, the input and
output data format is not compatible with the KISS format produced by the Engine Board; thus it
is necessary to convert the analog audio signal to be processed into digital form, have the
effects section process it, and convert it back into analog audio. The Effects Section could be
removed with little or no effect on the remaining circuitry.
Virtually all of the effects circuitry is licensed from DOD Digitech and is basically contained
in the DSP-256 stand-alone effects processor. One significant difference from the DSP-256
standard is the clock frequency has been reduced from 36MHz to 32MHz which is available from the
Engine Board.
An official circuit and software description from DOD is not available, thus the following
discussion represents Kurzweil engineering's analysis.
The Effects Mixer and Effects Input Low-Pass Filter circuits are described in 9,4.2 and 9.4.3
connected with schematic page 4.
The single channel of A/D conversion and two channels of D/A conversion needed would normally re-
quire one A/D converter and two D/A converter chips. However in the DOD circuit, one 16-bit
parallel input D/A converter, U16 (G-3 to G-4), is time-shared among the three functions. Analog
switches in UlO and U14 (C-5 to C-7) do the switching among the three functions of U16. The S&HL
and S&HR (Sample and Hold left and right) signals from the Effects LSI chip (U50, C-1 to D-4)
processed through U49, control these switches. Although each switch IC has 3 sections and only 3
switches are needed in all, the A/D function switch (UlO) was separated from the D/A function
switches (U14) to reduce crosstalk and the resulting distortion. Refer to figure 6 in 9.3.6 (for
schematic page 3) for simplified details on the analog switch operation.
The "A" switch section of UlO along with op-amp U12-7 (D-5) and C43 (D-6) perform as a sample-
hold circuit which holds the signal voltage stable while the A/D conversion is taking place.
Placing an oscilloscope probe at TPl will show a sampled version (stair-step approximation) of
the input signal. Voltages higher than 8.0 volts peak at this point will be clipped by the A/D
converter.
U50 performs A/D conversion by sending trial values to U16 which converts them into analog
currents which are converted into voltages by op-amp U13. comparator Ull then tells U50 whether
the trial value is less than or more than the analog input signal at TPl. Based on that result,
U50 sends out another trial value. Sixteen trials are performed to determine the signal voltage
to an accuracy of one part in 65,000 and the whole process requires 24 microseconds
(1 5uS/trial
. )
There are three adjustments around U16 which are required for quiet, distortion-free performance.
R63 (H-4) adjusts the linearity of U16 by slightly changing the most sifnificant bit value. It
is typically adjusted to minimize visible distortion of a 20mv sine wave. R122 (G-3) adjusts the
DC offset and is normally set for +50mv at TP6 with no signal to minimize residual distortion of
small signal amplitudes. R69 (G-3) adjusts the full scale signal level and is normally set for 8
volts full scale at TP6.
Throughout the Audio/ Power Board, Digital Ground (Dgnd) and Analog Ground (Agnd) are kept
separate. Since U16 has a single ground pin for both its analog and digital section, the two
ground systems are connected together under U16. This connection is called GJl in the schematic
and is actually two pads under U16 with a trace connected across them on the solder side of the
board. Plugging in the optional hard disk drive may also connect the two ground systems together
at J18 because most hard drives tie the +12 Return and +5 Return pins together at the drive's
power connector Typical DC power cables however should have sufficient impedance to prevent
.
serious groundloop problems. If not, it might be necessary to insert a choke in series with the
+12 return lead from the hard disk drive.
Two analog switch sections of U14 are used for the two channels of D/A output. One at a time is
turned on by signals from U50 through U49 and the corresponding voltage from U16 and U13 is held
on capacitors C45 and C48. The held voltage then appears in low impedance form at U15-1 (D-5 to
D— 7 As with the sampled input voltage, these two points should show stair-stepped
) .
U50 cannot provide settling time for smooth transition between new data being presented to U16
and the output analog switch closures to begin passing the converted voltage to the output. The
11
two flip-flops in U4 9 (A— 7) serve to dalay the control signals from U50 to the analog switches
sufficiently to allow for smoother transition from previous value to settling transient to new
value in a more linear fashion and eliminate distortion at low signal levels.
After conversion to sampled analog voltages at U15-1 and U15-7 (D-6 to D-7), the effected audio
signals are filtered by circuitry around U7 (F-6 to G-7). Each filter is two cascaded 3-pole
Butterworth sections identical to the input filter described elsewhere.
After filtering, the audio signals go through voltage-controlled amplifiers at the bottom of
schematic page 4. In order to get the phasing correct with respect to the "dry" signal that
bypasses the Effects Unit, two inverting amplifiers are required in each channel. The first
performs current-to-voltage conversion while the second reinverts the signal's phase and applies
one pole of the high frequency cut to cancel the lower boost pole added at the input mixer. The
high-frequency droop in the A/D/A process.
The Effects LSI chip (U50) is the largest IC on the Audio/Power Board. It is an 84-pin surface-
mount chip under the RFI shield along with the other digital circuitry. It has three basic
interface ports.
The MCA (Microcode Address) and MCD (Microcode Data) busses form the Microcode Port which is how
the chip receives instructions for performing any given effects algorithm. Every 125nS a new
microcode address is sent out on the MCA bus and somewhat later the corresponding microcode data
is received from the MCD bus.
The Analog I/O Port consists of the DAC bus and control signals MSREG, LSREG, S&HL, and S&HR.
Sixteen bit values destined for the D/A converter are output 8 bits at a time on the DAC bus and
then latched into U17 (E-3) and U18 (E-4) under control of MSREG and LSREG. S&HL and S&HR
control the analog routing switches as described earlier.
The Delay RAM port connects directly to 5 DRAM chips for delay line storage. U41 - U45 are each
64K by 4 bit DRAMs and are interconnected to make a 64K word by 20-bit memory. U50 sends out a
16-bit address multiplexed on 8 address lines (RMA bus) and reads or writes 20-bit data on the
RMD bus. Direct connections for the other DRAM signals (RAS, CAS, and WE) are also provided on
the Delay RAM Port. The DRAMs are always being cycled so their CE pins may be simply grounded.
The sample rate for the effects system is 33.3KS/s (Kilo-Samples per second), which is a sample
period of 30 microseconds. During each sample period it does one A/D conversion, two D/A
conversions, and cycles through the algorithm microprogram once. Figure 7 below shows a timing
diagram for one sample period. During the first 3uS the LSI chip closes the input switch (UlO
section A) and samples the input signal. The switch is opened and an additional 3uS elapses
waiting for the U12-7 op-amp to settle. The last 24uS is spent outputting 16 trial values and
reading the result from the comparator as described above. Simultaneously the output of the left
channel data is performed for the 3uS while the input is in the sample mode followed by the right
channel for 3uS wile the input op-amp is settling. Then the cycle repeats. The circuit is
always cycling like this even if the Engine Board selects no effects.
12
U50 reads its microcode instructions from a pair of 2KX8 high-speed static RAM chips, U24 (G-7)
and U25 (G-5). Actually, U50 only outputs 8 bits of microcode address; the other 3 are used to
identify 1 of 8 stored microcode routines. The SRAM is effectively dual-ported so that the
Effects Microprocessor can write into it while the LSI chip reads from it. This is necessary to
get the microcode routines loaded initially. Also for chorus and flange effects certain
locations with coefficients need to be periodically updated.
Address multiplexors U19, U20 and U22 select between addresses generated by the LSI chip when pin
1 is high and addresses generated by the Effects Microprocessor and latched in U29 and U47
when
^ is low. The LSI chip indicates via RVB^ENEX when it is not using the microdode RAM. Data
into and out of the microcode RAM is multiplexed on the MCD bus. For reading by the LSI chip,
the RAM's OE (Output Enable) pin is activated and its WE (Write Enable) pin deactivated. For
writing new data, the outputs of data latches U26 and U46 are enabled and the RAM's OE is
disabled and WE enabled.
The Effects Microprocessor is responsible for loading the Microcode RAM in response to program
change messages from the Engine Board and also periodically updating coefficients for the chorus
and flanging effects. It is a 6803 which has an enhanced 6800 core with 128 bytes of RAM, one
serial port, and two parallel ports built-in. Program memory is external. Its 4MHz input clock
is derived from the 32MHz clock by divider U56 (G-6 of page 2). It is internally divided by 4
again for a luS memory cycle and instruction times of 2 - lOuS.
The upper 8 bits of the external address bus are latched on the 6803 and thus can be connected
directly to the external address decoder and memories. The lower 8 bits are multiplexed with the
data on the DA bus. Address latch U55 (E-1) latches these which are then sent to the address
decoder and RAMs. An Address Strobe signal with the correct timeing is available directly from
the 6803.
Address decoding is performed by the 3 gates (U58 & U57, F-1) and two l-of-4 decoders (U54, F-2).
Gate U58-12 splits the 64K address space into an EPROM area of 56K (addresses $2000 - $FFFF) and
an ''other" area (addresses $0000 - $1FFF) .Gate U58-6 splits the "other" area into 7K of
external RAM from $0400 - $1FFF and IK of I/O and internal RAM between $0000 and $03FF. Then the
upper decoder (U54-10) splits, out the range between $0200 and $03FF and the lower decoder splits
that 4 ways for addressing 4 output latches (U26, U29, U46, and U47; F-3 to 1-3) which interact
with the microcode RAM as described earlier.
Program ROM for the Effects Microprocessor is a single 64Kx8 EPROM, U48 (G-1 to H-2). Much of
this space is taken by the original DSP-256 factory setups which are not used by the K2000. The
program code loaded into the EPROM is essentially what DOD supplied with a few minor patches to
skip over front panel related functions which would otherwise hang since there is not front panel
The CE pin (20) is used to activate the EPROM when addresses int he range of $2000 —
$FFFF are generated and the OE pin is used to deactivate it during all write cycles.
U51 (I-l to 1-2) is a low-power static RAM chip and serves both as a scratch RAM for the
microprocessor program and as a non-volatile RAM for effects setup storage. Non-volatility is
implemented by powering U51 from BUVdd and the use of Q8 (1-2) as a fail-safe, zero power
consumption Write Enable gate. During normal operation, RESETb, which is connected to the base
of Q8 through R184, is high. Thus when the Write signal from U57-H goes low to write, Q8
saturates and pulls its collector, which is connected to the WE pin on the RAM low. When Write
goes high or RESETb goes low, Q8 is cut off and R185 pulls the WE pin back high thus inhibiting
writing. When power off, the emitter is floating and there is no base current so R185 continues
to kee WE high with no power consumption (except leakage current).
The Engine Board software depends on the effect RAM being non-volatile. Whenever the Engine
software performs a "hard reset", such as after a battery power failure, it will reload the K2000
factory effects setups into this RAM which is a lengthy process.
In addition to communication through dual-port microcode RAM, the Effects Mocriprocessor communi-
cates with the Effects LSI Chip through a few dedicated control signals.
RVB_GATE is a signal generated by the microprocessor that will smoothly silence any sound that
may be reverberating in the chip and also clear the delay RAM. This is exercised whenever the
effects algorithm is changed.
13
RVB_OVFLO is produced by the LSI chip to indicate whenm the internal digital signal levels are
too great. Flip-flop U60-9 (B-4) latches this pulsed signal until the mocriprocessor can
recognize it. After recognition, the microprocessor resets the flip-flop.
Proper timing writes into the microcode RAM are critical and are synchronized with microprocessor
operation by several signals and circuits. The LSI chip actually controls the exact timing of
the write via the RVB_MCA and RVB^^MCD busses while the former is the actual Write Enable signal
to the microcode RAM. Counters U21 and U23 (B-1 to B-3) are also involved in sequential address
generation/
RVB_MUTE is produced by the Effects Microprocessor from power-up until the Microcode RAM is
initialized. The muting function is implemented at B-6 on schematic page 4 by overriding the
effects unit output amplifier control voltage via R18 and D1 (B-6 of page 4). When RVB_MUTE is
high, the U32-3 (B-6 on page 4) node is pulled high regardless of DACl_OUT and passed to U32-1
which is the control voltage to U8. Since the voltage— controlled amplifiers attenuate for
increasing control voltage, U8 is effectively cut off until the microcode RAM can be initialized.
14
M$9I BATA
T9/F^M CAftMe
15
3 115V red ti It
4 OV orange ft tt
5 115V yellow »• It
8 7.5V red ft
pin 6 It
11 16V blue It
pin 3 If
3 orange ft
Jl-3 tt
4 black ft
Jl-4 ft
"
5 black (large) Jl-5 It
6 blue ft
Jl-6 It
7 white It
Jl-7 II
3 orange tt ft
” J5-3
4 black tt It
” J5-4
5 black (large) ” ft
” J5-5
6 blue ft ft
" J5-6
7 white ft tt
" J5-7
connections
Tl red Pitch Wheel
T2 black ft ft
T3 orange tt tt
T6 orange It It
Mcru
/
K2000R
L-NGlNls
PCB
rERiniHRALS
/
K2000R
LNGtNE
INTERFACE
4— 4-Hhr-^s 4 4^^p-^S 4 »
- o 1, o - o
5,8 5,8 %s
4 4 » ll » >1 4 -HI~p— 4 ii l U »| 4 u W vr^Z
- o •- o
£2 5,8 5.8
4 u } r n-»| 4 4-Hh»->5
- o
%? 5,8
4 4-Hh=r->5 '
‘ =^' 4 4 !! » >8
1 11 1
I iliif!
JC4 >>>>>>>>
—
& —
-
OOOOOOOO c S
^ §
rri o^M ^ ^
ci> g>
ft
o»
j£%2SS28sd8 i-J^g225288^
/
K2000R
ENGINE
SIMMs
/
K2000R
SAMPLE
RAM ENGINE
PITCH
CENTER
R1
100K
K2000
SLIDER
PCB
BACKLIGHT
INVERTER
EXTERNAL
K2000
/
K2000R
PCD
AUDIO/POWER
PCB