®

Flyback Transformer Design For

TOPSwitch® Power Supplies
Application Note AN-17
When developing TOPSwitch flyback power supplies, total flyback component cost is lower when compared to other
transformer design is usually the biggest stumbling block. techniques. Between 75 and 100 Watts, increasing voltage and
Flyback transformers are not designed or used like normal current stresses cause flyback component cost to increase
transformers. Energy is stored in the core. The core must be significantly. At higher power levels, topologies with lower
gapped. Current effectively flows in either the primary or voltage and current stress levels (such as the forward converter)
secondary winding but never in both windings at the same may be more cost effective even with higher component counts.
time.
Flyback transformer design, which requires iteration through a
Why use the flyback topology? Flyback power supplies use the set of design equations, is not difficult. Simple spreadsheet
least number of components. At power levels below 75 watts, iteration reduces design time to under 10 minutes for a transformer

VDIODE
D2 L1
UG8BT 3.3 µH
7.5 V
ISEC
R1
39 Ω
R2
68 Ω
VR1 C2 C3
P6KE200 680 µF 120 µF
25 V 25 V
U2
D1 NEC2501-H
VR2
UF4005
1N5995B
BR1 6.2 V
C1
L2 33 µF IPRI
RTN
22 mH (CIN) D3
1N4148
VDRAIN

CIRCUIT PERFORMANCE:
Line Regulation - ±0.5%
C6 C5 (85-265 VAC)
0.1 µF 47µF Load Regulation - ±1%
DRAIN (10-100%)
SOURCE Ripple Voltage ± 50 mV
CONTROL C7 Meets VDE Class B
L 1000 nF
U1 C4
Y1
TOP202YAI 0.1 µF
N
J1
PI-1787-021296

Figure 1. ST202A Power Supply Operates from Universal Input Voltage and Delivers 15 Watts.

June 1996
 AN-17
that usually works the first time. This method, used for shape parameters include maximum duty cycle, average current,
continuous mode as well as discontinuous mode designs, has peak current, ripple current, and RMS current to completely
three distinct steps: define transformer primary current and determine operation in
either continuous or discontinuous mode. Transformer design
1) Identify and estimate a set of independent variables parameters include primary inductance, number of primary
(input) depending on application details, transformer turns, number of bias winding turns, gapped effective inductance,
core, and selected TOPSwitch. maximum flux density, AC flux density, ungapped core relative
permeability, estimated gap length, effective bobbin width,
2) Identify and calculate a set of dependent parameters insulated primary wire diameter, insulation thickness, bare
(output). conductor cross section, primary current capacity, and secondary
design parameters. Voltage stress parameters determine the
3) Iterate specified independent variables until selected maximum TOPSwitch off-state drain voltage and output rectifier
dependent parameters fall within defined limits for a peak inverse voltage.
practical flyback transformer.
Of all these dependent parameters, only three require examination
A simple PC spreadsheet (available from Power Integrations and comparison within limits during iteration. Maximum flux
for Excel or compatible spreadsheet programs) automates the density BM, gap length LG , and primary current capacity CMA
transformer design method presented in this application note. are checked with each iteration until all three parameters are
(Note: this improved version has been completely revised and within specified limits. The remaining dependent parameters
may give slightly different answers compared to earlier versions. are either intermediate calculations or parameters used by the
Refer to the last page of this application note for a complete manufacturer for construction or the designer for specifying
description of the changes.) components.

A new parameter, the ratio of primary ripple current to peak Understanding primary and secondary current waveform shape
current (KRP), is introduced to describe the TOPSwitch drain in both continuous and discontinuous mode operation is
current waveform shape and simplify subsequent calculations necessary before beginning transformer design.
such as RMS current and AC flux density.
Figure 1 shows a typical flyback power supply using the
Application specific independent variables include minimum TOP202 TOPSwitch from Power Integrations, Inc. TOPSwitch
and maximum AC input voltage, line frequency, TOPSwitch combines an integrated high voltage MOSFET switch with a
switching frequency, output and bias voltages, output power, complete switching power supply controller and protection
bridge rectifier conduction time, size of input energy storage circuitry in a single 3 pin TO220 package. The TOPSwitch
capacitor, power supply efficiency and power loss allocation power supply operates from 85 to 265 VAC and delivers 15
between primary and secondary circuitry. Variables depending Watts at 7.5 Volt output. AC power is rectified and filtered by
on the transformer core and construction include effective core BR1 and C1 (CIN) to create the high voltage DC bus applied to
cross sectional area and magnetic path length, ungapped the primary winding of T1. The other side of the transformer
effective inductance, bobbin physical winding width, margin primary is driven by TOPSwitch. D1 and VR1 clamp voltage
width (for creepage distance and safety isolation), number of spikes caused by transformer leakage inductance. D2, C2, L1,
primary layers, and number of secondary turns. Variables and C3 rectify and filter the power secondary. TOPSwitch bias
depending on TOPSwitch include switching frequency, reflected voltage is provided by D3 and C4 which rectify and filter the
output voltage, ripple to peak current ratio, and TOPSwitch bias winding. EMI filter components L2, C6, and C7 reduce
voltage drop. conducted emission currents. Bypass capacitor C5 filters
internal TOPSwitch gate charge current spikes and also
For a given application and transformer core, 20 of these 23 compensates the control loop. Regulation is achieved when the
independent variables will be calculated or estimated once and output voltage rises sufficiently above Zener diode voltage
then remain fixed during iteration. Only three variables, (VR2) to cause optocoupler photodiode current to flow.
number of secondary turns NS, ripple to peak current ratio KRP, Optocoupler phototransistor current flows into the TOPSwitch
and number of primary winding layers L will be changed during control pin to directly control the duty cycle and output voltage.
the iteration process. R1 together with series impedances of VR2 and TOPSwitch
determine the control loop DC gain. R2 and VR2 provide a
Dependent parameters are divided into four groups: DC input slight preload to improve regulation at light loads.
voltage, primary current waveform shape, transformer design,
and voltage stress. DC input voltage parameters are simply the Figures 2 and 3 show typical voltage and current waveforms
minimum and maximum DC input voltage after the AC mains taken from the same power supply delivering 15 Watts from
have been rectified and filtered. Primary current waveform 110 VAC input voltage but with different flyback transformer

2 C
6/96
 AN-17

PI-1526-091395

PI-1527-051695
200 V 200 V

VDRAIN
VDRAIN

0V IP 0V

IR IR
500 mA 500 mA IP 500 mA

IPRI
IPRI
0 mA 0 mA 0 mA
10 V 10 V

VDIODE
VDIODE

0V 0V
-10 V -10 V
-20 V -20 V

5A 5A 5A

ISEC
ISEC
0A 0A
0 10 20 0 10 20
Time (µs) Time (µs)

Figure 2. Voltage and Current Waveforms for Transformer Figure 3. Voltage and Current Waveforms for Transformer
Primary and Secondary in Discontinuous Mode. Primary and Secondary in Continuous Mode.

primary inductance. TOPSwitch turns on to effectively apply The stored energy is completely delivered to the load. TOPSwitch
the DC input voltage across the transformer winding with the drain voltage VDRAIN relaxes and rings back towards the DC bus
“dot” side at lower potential than the “no-dot side”. Primary voltage when no current is flowing in either primary or secondary.
current IPRI increases linearly with a rate of change (di/dt) that
varies directly with DC input voltage and inversely with primary Figure 3 shows trapezoidal current waveforms which define
inductance. Ripple current IR is defined as the incremental “continuous” mode of operation resulting from high primary
linear current rise (di) over the entire TOPSwitch on time (tON). inductance. Secondary current is still flowing when TOPSwitch
Peak primary current IP is the final value occurring as TOPSwitch turns on at the beginning of the next cycle. The stored energy
turns off. Energy, proportional to the square of peak current IP, is not completely delivered to the load. Energy (due to non-zero
is stored by magnetic field in the transformer core as if the magnetic field) remains in the core when TOPSwitch turns on
primary winding were a simple inductor. The secondary again which causes the initial step in TOPSwitch current. Note
winding carries a reflected voltage proportional to primary that TOPSwitch drain voltage VDRAIN stays at a high value equal
voltage by turns ratio with the same “dot” polarity. While to the sum of the DC input voltage and reflected output voltage
TOPSwitch is on, output diode D2 and bias diode D3 are reverse until TOPSwitch turns on again.
biased which prevents secondary current flow. When TOPSwitch
turns off, the decreasing magnetic field induces an abrupt Current never flows in the primary and secondary winding at
voltage reversal on all transformer windings such that the “dot” the same time. Neither primary or secondary current is actually
side is now higher potential than the “no-dot” side. Diode D2 continuous. In flyback power supplies, continuous/
and D3 become forward biased and secondary current rises discontinuous mode refers to magnetic field continuity in the
quickly to a peak value (proportional by the inverse turns ratio transformer core over one complete switching cycle. (The
to primary peak current IP). Primary current immediately drops flyback power supply is an isolated version of the simple buck-
to zero. TOPSwitch drain voltage quickly rises to a voltage boost converter where continuous and discontinuous modes are
equal to the sum of the DC input voltage and reflected output easily defined by inductor current continuity.)
voltage. Secondary winding current now linearly decreases at
a rate that varies directly with output voltage and inversely with Each primary current waveform has a peak value (IP), a ripple
secondary inductance. Duty cycle is defined as the ratio of current value (IR), an average or DC value (I AVG), and an RMS
TOPSwitch on time tON to switching period T. D can also be value (IRMS). IP determines the number of primary turns and the
calculated from tON and switching frequency f S as shown. core size necessary to limit peak flux density and must also be
below TOPSwitch peak current limit. IAVG is the average or DC
primary current (as well as the power stage DC input current)
t ON
D= = t ON × f S which is proportional to output power. IRMS causes power losses
T due to winding resistance and TOPSwitch RDSON. The ratio
(KRP) of ripple current IR to peak current IP defines the continuous
Figure 2 shows TOPSwitch and output diode triangular current or discontinuous waveform. KRP also simplifies subsequent
waveforms which define “discontinuous” mode of operation calculations. Transformers designed for discontinuous operation
resulting from low primary inductance. The secondary current have a higher peak current and a ripple current to peak current
linearly decreases to zero before TOPSwitch turns on again. ratio KRP of one. Practical continuous designs have lower peak

C
6/96 3
 AN-17
currents and a ripple to peak current ratio KRP of less than one Thomson. Applications requiring lower profile can benefit
but typically greater than 0.4. KRP is inversely proportional to from EFD cores available from the European vendors. EER
primary inductance so a continuous design with lower KRP will cores offer a large window area, require few turns, and have
have a higher inductance. Continuous transformer designs bobbins available with high pin counts for those applications
have a practical primary inductance upper limit approximately requiring multiple outputs. ETD cores are useful in the higher
four times that of a discontinuous design at the same input power designs when space is not a problem. PQ cores are more
voltage and output power due to the difference in peak currents expensive but take up slightly less PC board space and require
and value of KRP. less turns than E cores. Safety isolation requirements make pot
cores, RM cores, and toroids generally not suitable for flyback
The primary current waveforms shown in Figures 2 and 3 power supplies operating from the AC mains.
deliver the same output power and therefore (assuming equal
efficiency) must have equal IAVG. The discontinuous current Flyback transformers must provide isolation between primary
waveform has a higher peak value and therefore must have a and secondary in accordance with the regulatory agencies of
higher RMS current value. Discontinuous mode requires less the intended market. For example, information technology
inductance and reduces transformer size but operates with equipment must meet the requirements of IEC950 in Europe
higher losses and lower efficiency due to higher RMS currents. and UL1950 in the U.S. These documents specify creepage
Continuous mode requires higher inductance and larger and clearance distances as well as insulation systems used in
transformer size but offers improved efficiency and lower transformer construction. 5 to 6 mm creepage distance is
power losses. The trade-off between transformer size and usually sufficient between primary and secondary (check with
power supply efficiency depends on the packaging and thermal the appropriate agency and specification). Isolation is usually
environment in each application. specified by electric strength and is tested with a voltage of
typically 3000 VAC applied for 60 seconds. Two layers of
Some control loop comments regarding continuous mode are in insulation (Basic and Supplementary) can be used between
order here. Most designers tend to avoid the continuous mode primary and secondary if each layer exceeds the electric
whenever possible because the feedback control loop is more strength requirement. Three layers of insulation (reinforced)
difficult to analyze. Discontinuous mode power supplies are can also be used if all combinations of two layers (out of total
modeled with a single pole response and are simple to stabilize. three layers) meets the electric strength requirement.
Continuous mode offers improved efficiency, reduced losses,
lower component temperatures, or higher output power but Figure 4a shows the margin winding technique used in most
analysis is more difficult because a right half plane zero and flyback transformers. The margin is usually constructed with
complex pole pair all shift with duty cycle. However, stabilizing layers of tape slit to the width of the desired margin and
a continuous mode TOPSwitch power supply is quite wrapped in sufficient layers to match the winding height. The
straightforward. Adequate phase margins are achievable over margin is generally half the required primary to secondary
all line and load combinations because the 70% maximum creepage distance (2.5 mm in this example). Cores and
TOPSwitch duty cycle DCMAX (from the data sheet) limits right bobbins should be selected large enough that the actual winding
half plane zero and complex pole pair migration. Phase margin width is at least twice the total creepage distance to maintain
is generally higher than expected once the damping effect of transformer coupling and reduce leakage inductance. The
effective series power path resistance and output capacitor ESR primary is wound between the margins. To reduce the risk of
is taken into account. Crossover bandwidths of 1 KHz (or interlayer voltage breakdown due to insulation abrasion,
wider) are easily achievable with phase margins of at least 45 improve layer to layer insulation, and decrease capacitance,
degrees. Refer to AN-14 for circuit techniques to use in the primary layers should be separated by at least one layer of
continuous mode designs. UL listed polyester film tape (3M 1298) cut to fit between the
margins. Impregnation with varnish or epoxy can also improve
Transformer core, winding, and safety issues must also be the layer to layer insulation and electric strength but does not
discussed before beginning design. reduce capacitance. The bias winding may then be wound over
the primary. Supplementary or reinforced insulation consisting
Transformer core and construction parameters depend on the of two or three layers of UL listed polyester film tape cut to the
selected core and winding techniques used in assembly. Physical full width of the bobbin may then be wrapped over the primary
height and cost are usually most important when selecting and bias windings. Margins are again wound. The secondary
cores. This is especially true in AC mains adapter power winding is wound between the margins. Another two or three
supplies normally packaged in sealed plastic boxes. Applications layers of tape is added to secure the windings. Insulation
allowing at least 0.75 inches of component height can use low sleeving may be needed over the leads of one or all windings
cost EE or EI cores from Magnetics, Inc., Japanese vendors to meet creepage distance requirements at lead exits. Nylon or
TDK and Tokin, or European vendors Philips, Siemens, and

4 C
6/96
 AN-17

SECONDARY
SAFETY
INSULATION BIAS
TAPE BIAS (ALTERNATE
SECONDARY LOCATION)
PRIMARY
(Z WOUND) (INSULATED)

BIAS
M M
PRIMARY
(a) MARGIN WINDING

ALTERNATE
PRIMARY
WINDING

(b) C WINDING
PI-1521-091395 PI-1678-091395

Figure 4. Margin Wound Transformer. Figure 5. Triple Insulated Wire Wound Transformer.

Teflon sleeving with a minimum wall thickness of 0.41 mm insulated wire. The double or triple insulated wire is then
should be used to meet the safety agency requirements. Consider wound. Another layer of tape is added to secure insulated
the core as isolated dead metal (which means the core is winding.
conductive but not part of any circuit and safely insulated from
the consumer). The sum of distance from primary winding (or Figure 5 also shows an alternate position for the bias winding
lead exits) to the core added to the distance from the core to the wound directly over the secondary to improve coupling to the
secondary (or lead exits) must be equal to or greater than the secondary winding and reduce leakage inductance (to improve
required creepage distance. load regulation in bias winding feedback circuits). Note that
because the bias winding is a primary circuit, margin wound
Both Z winding (Figure 4a) and C winding (Figure 4b) techniques transformers must have another layer of supplementary or
for multiple primary layers are shown. Note that the “dot” side reinforced insulation between the secondary and alternate bias
which connects to TOPSwitch is buried under the second layer winding.
for self shielding to reduce EMI (common mode conducted
emission currents). Z winding decreases transformer Refer to AN-18 for more information regarding transformer
capacitance, decreases AC TOPSwitch losses, and improves construction guidelines.
efficiency but is more difficult and costly to wind. The
C winding is easier and lower cost to wind but at the expense of Flyback transformer design now begins by specifying the three
higher loss and lower efficiency. groups of independent variables shown in the spreadsheet
(Figure 6).
Figure 5 shows a new technique using double or triple insulated
wire on the secondary to eliminate the need for margins (insulated Application Variables:
wire sources can be found at the end of this application note).
In double insulated wire, each layer is usually capable of Output power PO, output voltage VO , AC mains frequency fL,
meeting the electric strength requirement of the safety agency. TOPSwitch switching frequency f S (100KHz), minimum
In triple insulated wire, all three combinations of two layers (VACMIN), and maximum (VACMAX) AC mains voltage come
taken together must usually meet the electric strength directly from the application.
requirement. Special care is necessary to prevent insulation
damage during winding and soldering. This technique reduces For efficiency (η), start with an estimate based on measurements
transformer size and eliminates the labor cost of adding margins in similar power supplies (or use a value of 0.8 if data is
but has higher material cost and may increase winding costs. unavailable).
The primary winding is wound over the full width of the bobbin
flange. The bias winding can be wound if desired over the Efficiency can be used to calculate total power loss PL in the
primary. One layer of tape is usually necessary between power supply as shown below. Some power losses occurring in
primary or bias and secondary to prevent abrasion of the series primary components such as the bridge rectifier, common

C
6/96 5
 AN-17
A B C D E F
1 INPUT OUTPUT
2 ENTER APPLICATION VARIABLES
3 VACMIN 85 Volts Minimum AC Input Voltage
4 VACMAX 265 Volts Maximum AC Input Voltage
5 fL 60 Hertz AC Mains Frequency
6 fS 100000 Hertz TOPSwitch Switching Frequency
7 VO 7.5 Volts Output Voltage
8 PO 15 Watts Output Power
9 n 0.8 Efficiency Estimate
10 Z 0.5 Loss Allocation Factor
11 VB 10.4 Volts Bias Voltage
12 tC 3.2 mSeconds Bridge Rectifier Conduction Time Estimate
13 CIN 33 uFarads Input Filter Capacitor
14
15 ENTER TOPSWITCH VARIABLES
16 VOR 85 Volts Reflected Output Voltage
17 VDS 10 Volts TOPSwitch on-state Drain to Source Voltage
18 VD 0.4 Volts Output Winding Diode Forward Voltage Drop
19 VDB 0.7 Volts Bias Winding Diode Forward Voltage Drop
20 KRP 0.92 Ripple to Peak Current Ratio (0.4 < KRP < 1.0)
21
22 ENTER TRANSFORMER CORE/CONSTRUCTION VARIABLES
23 EE22-Z Core Type
24 AE 0.41 cm^2 Core Effective Cross Sectional Area
25 LE 3.96 cm Core Effective Path Length
26 AL 2400 nH/T^2 Ungapped Core Effective Inductance
27 BW 8.43 mm Bobbin Physical Winding Width
28 M 0 mm Safety Margin Width (Half the Primary to Secondary Creepage Distance)
29 L 2 Number of Primary Layers
30 NS 5 Number of Secondary Turns
31
32 DC INPUT VOLTAGE PARAMETERS
33 VMIN 9 3 Volts Minimum DC Input Voltage
34 VMAX 3 7 5 Volts Maximum DC Input Voltage
35
36 CURRENT WAVEFORM SHAPE PARAMETERS
37 DMAX 0.51 Duty Cycle at Minimum DC Input Voltage (VMIN)
38 IAVG 0.20 Amps Average Primary Current
39 IP 0.74 Amps Peak Primary Current
40 IR 0.68 Amps Primary Ripple Current
41 IRMS 0.32 Amps Primary RMS Current
42
43 TRANSFORMER PRIMARY DESIGN PARAMETERS
44 LP 6 2 3 uHenries Primary Inductance
45 NP 54 Primary Winding Number of Turns
46 NB 7 Bias Winding Number of Turns
47 ALG 215 nH/T^2 Gapped Core Effective Inductance
48 BM 2 0 8 5 Gauss Maximum Flux Density (2000 < BM < 3000)
49 BAC 959 Gauss AC Flux Density for Core Loss Curves (0.5 X Peak to Peak)
50 ur 1845 Relative Permeability of Ungapped Core
51 LG 0 . 2 2 mm Gap Length (Lg >> 0.051 mm)
52 BWE 16.86 mm Effective Bobbin Width
53 OD 0.31 mm Maximum Primary Wire Diameter including insulation
54 INS 0.05 mm Estimated Total Insulation Thickness (= 2 * film thickness)
55 DIA 0.26 mm Bare conductor diameter
56 AWG 3 0 AWG Primary Wire Gauge (Rounded to next smaller standard AWG value)
57 CM 102 Cmils Bare conductor effective area in circular mils
58 CMA 3 2 1 Cmils/Amp Primary Winding Current Capacity (200 < CMA < 500)
59
60 TRANSFORMER SECONDARY DESIGN PARAMETERS
61 ISP 7.95 Amps Peak Secondary Current
62 ISRMS 3.36 Amps Secondary RMS Current
63 IO 2.00 Amps Power Supply Output Current
64 IRIPPLE 2.70 Amps Output Capacitor RMS Ripple Current
65
66 CMS 1079 Cmils Secondary Bare Conductor minimum circular mils
67 AWGS 1 9 AWG Secondary Wire Gauge (Rounded up to next larger standard AWG value)
68 DIAS 0.91 mm Secondary Minimum Bare Conductor Diameter
69 ODS 1.69 mm Secondary Maximum Insulated Wire Outside Diameter
70 INSS 0.39 mm Maximum Secondary Insulation Wall Thickness
71
72 VOLTAGE STRESS PARAMETERS
73 VDRAIN 5 7 3 Volts Maximum Drain Voltage Estimate (Includes Effect of Leakage Inductance)
74 PIVS 4 2 Volts Output Rectifier Maximum Peak Inverse Voltage
75 PIVB 5 9 Volts Bias Rectifier Maximum Peak Inverse Voltage
76
77 ADDITIONAL OUTPUTS
78 VX 12 Volts Auxiliary Output Voltage
79 VDX 0.7 Volts Auxiliary Diode Forward Voltage Drop
80 NX 8.04 Auxiliary Number of Turns
81 PIVX 6 8 Volts Auxiliary Rectifier Maximum Peak Inverse Voltage
82

Figure 6. Spreadsheet for ST202A Flyback Transformer Design.

6 C
6/96
 AN-17
mode choke, and TOPSwitch are not associated directly with
energy stored in the flyback transformer core. The remaining
power losses, occurring in the output rectifier and clamp Zener
diode when energy is released from the flyback transformer, are
now defined as secondary loss PLS. Loss Allocation Factor Z,
defined below as the ratio of secondary loss PLS to total loss PL,
is a scaling factor which distributes the losses between primary
and secondary. Loss allocation factor Z is typically between 0.4
and 0.6 which means that secondary loss PLS is usually 40% to
60% of total power supply loss PL .

1−η
PL = PO × ( )
η
Figure 7. Bridge AC Current, AC Voltage, and DC Voltage
P
Z = LS Waveforms.
PL
normally used and VD is typically 0.7 Volts.
Bias voltage VB is determined by the feedback control circuit
and is usually between 10 volts and 30 volts (see AN-16). Bias winding diode forward voltage drop (VDB) is also typically
0.7 Volts
For bridge rectifier conduction time tC, 3 milliSeconds is typical
(measure on a similar power supply or set equal to zero for a Ripple current to peak current ratio KRP determines how far into
conservative first design). the continuous mode a flyback transformer will operate.
Continuous mode transformers optimized for TOPSwitch
For input filter capacitor CIN , start with a standard value in applications operating from 100/115 VAC or universal input
microFarads between two and three times the output power in voltage should have a minimum KRP of 0.4. Applications
Watts (appropriate for universal or 115 VAC input). For operating from 230 VAC input voltage should have a minimum
example: 30µF to 45µF is a suitable capacitance range for a KRP of 0.6. Discontinuous mode transformers optimized for
15 Watt supply. 33µF is the lowest standard value within the TOPSwitch applications always have a KRP equal to 1.0.
range.
IR
TOPSwitch Variables: K RP =
IP
Reflected output voltage VOR appears across the transformer
primary when TOPSwitch is off and current is flowing through Transformer Core/Construction Variables:
the secondary and output rectifier diode. Transformers optimized
for TOPSwitch applications should be designed with a maximum The following effective parameters are specified by the core
reflected voltage VOR of 60V or less for the TOP1XX series and and bobbin manufacturer in data sheets: cross sectional area Ae
135V or less for the TOP2XX series. For more information, (cm2), path length Le (cm), ungapped inductance AL (specified
refer to AN-16. in either mH/(1000 turns)2 or nH/T2), and physical bobbin
winding width BW (mm).
VDS is the on-state TOPSwitch voltage from the data sheet
(typically 10 volts) at the specified value for peak TOPSwitch Margin width M, determined by insulation methods and
drain current IP. regulatory requirements discussed above, is usually between
2.5 to 3.0 mm for margin wound or set to zero for insulated wire
Output rectifier forward voltage drop VD depends on output voltage. wound transformers.
For lower output voltages (typically 8 Volts and below) a Schottky
diode is commonly used and VD is typically 0.4 Volts. In some cases, For number of layers L, one or two layers of primary winding
a Schottky diode can be used for output voltages as high as 12V are normally used. Higher number of layers increase cost,
depending on input voltage range and transformer turns ratio. For increase capacitance, reduce coupling, and increase leakage
higher output voltage, an ultrafast recovery PN junction diode is inductance.

C
6/96 7
 AN-17
Number of secondary turns NS is a key iteration variable. One voltage VDS:
turn per Volt of output voltage is a good value to begin with for
NS (for example: start with 5 turns for a +5V output). VOR
DMAX =
VOR + (VMIN − VDS )
The four groups of dependent parameters can now be calculated.
Average current IAVG is calculated from minimum DC input
DC Input Voltage Parameters: voltage VMIN, output power PO, and efficiency η:

Minimum DC input voltage VMIN depends on the AC input PO

voltage, bridge rectifier, and energy storage capacitor. I AVG =
Figure 7 shows how CIN charges to the peak of the AC input η × VMIN
voltage during a short conduction time tC. Because of full wave Peak primary current Ip is calculated from average current IAVG,
rectification, CIN has a ripple voltage at twice line frequency. ripple to peak current ratio KRP, and maximum duty cycle DMAX:
CIN must supply the entire average primary current during the
discharge time between the peaks of the AC input voltage.
2
Minimum DC voltage VMIN can be found from the following IP = I AVG ×
equation where PO is the power supply output power, η is an (2 − K RP ) × DMAX
estimate of efficiency, fL is line voltage frequency, VACMIN is the
minimum AC mains voltage, CIN is the value of the filter Ripple current IR is calculated from average current IAVG, peak
capacitor, and tC is an estimate for conduction time. As an primary current IP, and maximum duty cycle DMAX:
example, for 60 Hz, 85 VAC input voltage, efficiency of 0.8,
15 Watt output power, 33 uF input filter capacitance, and
estimated conduction time of 3.2 mS, VMIN is 93 Volts DC. I AVG
I R = 2 × ( IP − )
DMAX
1
2 × PO × ( − tC ) RMS current IRMS is calculated from maximum duty cycle DMAX,
2 × fL peak primary current IP, and ripple to peak ratio KRP. I RMS can
VMIN = (2 × V 2
)−( )
η × CIN
ACMIN also be calculated directly from DMAX, IP, and ripple current IR.

2
K RP
2 × 15 × (
1
− 3.2 mS ) I RMS = IP × DMAX × ( − K RP + 1)
2 × 60 3
= (2 × 85 ) − (
2
) = 93V
0.8 × 33µF
I R2
Maximum DC input voltageVMAX is simply the peak value of = DMAX × ( IP2 − ( IP × I R ) + )
the highest AC input voltage (VACMAX ) expected in the 3
application. Operation from 265 VAC input results in a
Transformer Design Parameters:
maximum DC bus voltage VMAX of 375 Volts DC.
Primary inductance LP (in µH) is determined by the flyback
VMAX = VACMAX × 2 = 265 × 2 = 375V transformer energy equation defined below. The flyback
transformer stores energy proportional to the square of primary
current. When TOPSwitch is on, primary current linearly ramps
Current Waveform Shape Parameters:
up over a current range, defined earlier as ripple current IR, and
increases the energy stored in the flyback transformer core.
DMAX is the actual duty cycle occurring when the TOPSwitch
When TOPSwitch turns off, the stored energy increment
power supply delivers maximum output power from minimum
associated with ripple current IR is delivered to the load and
input voltage. DMAX has an upper limit equal to the minimum
secondary losses (rectifier and clamp). Inductance LP can now
value of the TOPSwitch Data Sheet parameter DCMAX (64%).
by calculated from output power PO, efficiency η, loss allocation
DMAX is calculated from reflected voltage VOR, minimum DC
factor Z, peak current IP, switching frequency fS, and ripple
input voltage VMIN, and TOPSwitch on-state Drain to Source
current to peak current ratio KRP (which determines IR).

8 C
6/96
 AN-17
Maximum flux density BM is a dependent iteration variable to
( Z × (1 − η)) + η
PO × ( ) be manipulated between the limits of 2000 and 3000 Gauss by
η varying number of secondary turns NS which directly varies
LP = 10 ×
6
K number of primary turns NP as previously shown. BM is
fS × IP2 × K RP × (1 − RP ) calculated from peak current IP, number of primary turns NP,
2
effective gapped inductance ALG, and effective core cross
sectional area Ae. BM can also be calculated from effective
Primary inductance LP (in µH) can also be determined from a primary voltage (VMIN-VDS), output voltage VO , output diode
simple function of ripple current IR, effective primary voltage voltage VD , and maximum duty cycle DMAX:
(VMIN-VDS), maximum duty cycle DMAX, and switching frequency
fS as shown below but the resulting value for primary inductance
N P × IP × ALG
may be slightly different due to the selected value for loss BM =
allocation factor Z and TOPSwitch on-state Drain to Source 10 × Ae
voltage VDS. The energy equation given above is preferred for
selecting the value of inductance LP while the ripple current
IP × ALG VMIN − VDS DMAX
equation given below is best for verifying the LP value using = NS × × ×
in-circuit measurements. 10 × Ae VO + VD 1 − DMAX

(VMIN − VDS ) × DMAX BAC is the AC flux density component. The equation gives peak
LP ( MEASURED) = 10 6 ×
I R × fS AC flux density (rather than peak to peak) to use with core loss
curves provided by the core vendor. BAC can be calculated from
maximum flux density BM and ripple to peak current ratio KRP.
Number of primary turns NP depends on number of secondary BAC can also be calculated from effective primary voltage
turns NS, output voltage VO, diode forward voltage drop VD, (VMIN-VDS), duty cycle, frequency, effective core cross sectional
effective primary voltage (VMIN-VDS), and maximum duty cycle area, and number of primary turns NP:
DMAX:

V − VDS BM × K RP (VMIN − VDS ) × DMAX × 108

N P = NS × MIN ×
DMAX BAC = =
VO + VD 1 − DMAX 2 2 × fS × Ae × N P
Relative permeability µr of the ungapped core must be calculated
to estimate the gap length Lg. µr is found from core parameters
Ae (cm2), Le (cm), and ungapped effective inductance AL:
The number of bias winding turns NB is calculated from the
output voltage VO, output diode voltage V D, secondary number
of turns NS, target bias voltage VB, and bias diode voltage VBD: AL × LE
µr =
0.4 × π × Ae × 10
VB + VBD
NB = × NS
VO + VD Gap length Lg is the air gap ground into the center leg of the
transformer core. Grinding tolerances and ALG accuracy place
a minimum limit of 0.051 mm on Lg. L g (in mm) is calculated
ALG is the effective inductance for the gapped core in nH/T2.
from number of primary turns NP, core effective cross sectional
Some core vendors offer standard gapped core sets with specified
area Ae, primary inductance LP (in µH), core effective path
ALG. The transformer manufacturer either procures the gapped
length Le, and relative permeability µr:
core for the given ALG value or grinds the gap to meet the
inductance specification in the finished transformer. ALG is
 0.4 × π × N P2 × Ae Le 
also used to simplify subsequent calculations. ALG is calculated Lg =  −  × 10
from primary inductance LP (in µH) and number of primary  LP × 100 µr 
turns NP. Note that ALG is specified in nH/(turn)2.
Effective bobbin width BWE takes into account physical bobbin
L width BW, margins M, and number of layers L:
ALG = 1000 × P2
NP
BWE = L × ( BW − (2 × M ))

C
6/96 9
 AN-17
Primary insulated wire diameter OD in mm is found from CM
effective bobbin width BWE and number of primary turns NP: CMA =
I RMS
BWE This completes all calculations necessary for the primary
OD = winding. Secondary peak current, RMS current, average output
NP current, output capacitor ripple current, and secondary minimum
and maximum conductor diameter must also be calculated.
The bias winding is usually wound with the same wire diameter
as the primary to reduce the number of different wire gauges Peak secondary current ISP is a simple function of peak primary
necessary for production. current IP, primary turns NP, and secondary turns NS.

Actual magnet wire outside diameter OD is slightly larger than NP

the diameter DIA of the bare copper conductor. Insulation ISP = IP ×
thickness varies inversely with bare copper conductor American NS
Wire Gauge (AWG) size which means that smaller diameter
conductors have thinner insulation thickness. Data from several Secondary RMS current ISRMS is found from maximum duty
different manufacturers were tabulated to generate an empirical cycle DMAX, secondary peak current ISP, and ripple to peak
expression for total insulation thickness INS (in mm) as a current ratio KRP (KRP is identical for primary and secondary).
function of heavy insulated magnet wire outside diameter
(in mm). 2
K RP
ISRMS = ISP × (1 − DMAX ) × ( − K RP + 1)
INS = (0.0594 × LOG(OD)) + 0.0834 3

Output current IO is simply the ratio of output power PO to output

DIA = OD − INS Voltage VO:

PO
Another empirical equation determines the AWG for magnet IO =
wire with a given bare conductor diameter DIA. Integer AWG VO
values are the standard sizes of available wire so the calculated
AWG value should always be rounded up to the next integer or
Output capacitor ripple current IRIPPLE is not a true transformer
standard value (the next smaller standard conductor diameter)
parameter but is needed for capacitor selection and easy to
before proceeding with the current capacity or CMA calculation.
calculate from other transformer parameters. IRIPPLE is found
from secondary RMS current ISRMS and output current I O.
AWG = 9.97 × (1.8277 − (2 × LOG( DIA)))
I RIPPLE = ISRMS
2
− IO2
Magnet wire for transformer winding usually has the cross
sectional area specified in circular mils. A circular mil is the
Minimum secondary bare conductor diameter DIAS (in mm)
cross sectional area of a wire with a diameter of 1 mil (or
based on previously calculated current capacity CMA and
0.0254 mm). The effective cross sectional area in circular mils
(CM) of a standard AWG size bare conductor wire is found secondary RMS current must be determined.
from the following simple expression.
From the primary CMA and secondary RMS current ISRMS, the
 50 − AWG 
minimum secondary bare conductor CMS is calculated.
 
CM = 2  3 

CMS = CMA × ISRMS

“Circular mils per Amp” or CMA is a convenient way to specify
winding current capacity. CMA, which is the inverse of current
density, is simply the ratio of cross sectional area in circular Minimum secondary AWGS is then calculated from another
mils to the RMS value of primary current. CMA should be empirical equation. Secondary calculated wire gauge AWGS is
between 200 and 500 and is calculated from cross sectional always rounded down to the next integer value which selects the
wire area in CM and RMS primary current IRMS. next larger standard wire size.

10 C
6/96
 AN-17

AWGS = 9.97 × (5.017 − LOG(CMS )) Maximum peak inverse voltage PIVS for the output rectifier is
determined by transformer primary and secondary number of
turns NP and NS, maximum DC input voltage VMAX, and output
(Secondary conductors larger than 26 AWG should not be used voltage VO .
due to skin effects. Refer to AN-18 for suggestions on parallel
conductor techniques.)
NS
PIVS = VO + (VMAX × )
Bare conductor diameter (in mm) is now determined. NP

 50 − AWGS  Maximum peak inverse voltage PIVB for the bias rectifier is
4 × 2 3
 

25.4 determined from a similar equation using number of bias turns
DIAS = ×
1.27 × π 1000 NB.

The maximum wire outside diameter ODS (in mm) for a single NB
layer based on number of secondary turns and bobbin width
PIVB = VB + (VMAX × )
NP
must also be calculated:
Additional or auxiliary output winding number of turns NX and
BW − (2 × M ) rectifier diode peak inverse voltage PIVX can be determined
ODS = from the desired value for auxiliary output voltage VX, auxiliary
NS
rectifier diode forward voltage drop VDX, output voltage VO,
Secondary wire insulation thickness can now be calculated output rectifier diode forward voltage drop VD, and number of
from the bare conductor outside diameter (determined by secondary turns NS.
CMA) and the insulated wire outside diameter (determined by
number of turns and effective bobbin width). Note that secondary VX + VDX
NX = × NS
insulation thickness INSS (in mm) is the insulation wall thickness VO + VD
rather than the total insulation thickness used in the primary
winding calculation.
NX
ODS − DIAS PIVX = VX + (VMAX × )
INSS = NP
2
Obviously, if insulation thickness INSS is not a positive number, Iteration can now be used to reach a final and acceptable
another transformer design iteration is necessary with either solution for the flyback transformer design.
more secondary layers, a smaller number of secondary turns, or
a transformer core with a wider bobbin. Iterate number of secondary turns NS or primary ripple to peak
current ratio KRP until maximum flux density BM is between
For insulated wire secondaries, INSS must be equal to or greater indicated limits and check that gap length Lg is higher than
than insulation thickness of the selected wire. indicated minimum value. BM will decrease and L g will increase
as NS or KRP is increased.
Parallel combinations of wire with half the diameter may be
easier to wind and terminate but the effective secondary CMA Examine primary current capacity in Circular Mils per Amp
will be half the value of the single winding. (CMA). If CMA is below the specified lower limit of 200,
consider increasing number of primary layers from one to two
Voltage Stress Parameters: or use the next larger core size and perform new iteration. If
CMA is greater than 500, consider using the next smaller core
Maximum drain voltage is the sum of maximum DC input size. (CMA greater than 500 simply means that the wire
voltage VMAX, an estimated drain clamp voltage term based on diameter is oversized for the expected RMS current).
VOR , and an estimated voltage term related to typical blocking
diode forward recovery. Refer to AN-16 for more detail. The transformer design is now complete. The transformer

VDRAIN = VMAX + (1.4 × 1.5 × VOR ) + 20V

C
6/96 11
 AN-17
manufacturer needs the following information: References

Core part number and gapped effective inductance ALG Bisci, J., Part IV: Magnet Wire: Selection Determines
Bobbin part number Performance, PCIM, October 1994, pp. 37.
Wire gauge and insulation style on all windings
Safety or Electric strength and Creepage distance Leman, B., Finding the Keys to Flyback Power Supplies
specifications Produces Efficient Design, EDN, April 13, 1995, pp. 101-113.
Primary Inductance LP
Number of turns (NP, NS, NB, etc.) for each winding McLyman, C., Transformer and Inductor Design Handbook,
Bobbin pin connections Marcel Dekker, Inc. 1978
Winding layer placement and winding instructions
Temperature class of operation (class A is 105 °C, class B Insulated Wire Sources
is 130 °C, etc.)
Rubudue Wire Company
Spreadsheet Improvements 5150 E. LaPalma Ave, Suite 108
Anaheim Hills, CA 92807 USA
The order of the spreadsheet has been changed to simplify the iteration (714) 693-5512
process. Reflected voltage VOR and ripple to peak current ratio KRP are (714) 693-5515 FAX
now independent variables which make peak current Ip and duty cycle
DMAX dependent variables. Loss allocation factor Z is introduced to Furukawa Electric America, Inc.
distinguish between power losses occurring before energy is stored in 200 Westpark Dr., Suite 190
the transformer (primary losses) and power losses occurring after Peachtree City, GA 30269 USA
energy is released from the transformer (secondary losses). Primary (770) 487-1234
inductance LP is now calculated from output power PO, KRP, efficiency (770) 487-9910 FAX
η, and loss allocation factor Z. The spreadsheet now takes into account
primary magnet wire insulation thickness as well as the discrete steps The Furukawa Electric Co., Ltd.
of standard AWG wire sizes. Metric dimensions are used throughout 6-1, Marunouchi 2-chome,
(with the exception of Circular mils for wire cross sectional area). Chiyoda-ku, Tokyo 100, Japan
Drain Voltage VDRAIN now includes an estimate for the effect of leakage 81-3-3286-3226
inductance induced voltage spikes on typical primary clamp circuits. 81-3-3286-3747 FAX

Power Integrations reserves the right to make changes to its products at any time to improve reliability or manufacturability.
Power Integrations does not assume any liability arising from the use of any device or circuit described herein, nor does it
convey any license under its patent rights or the rights of others.

PI Logo and TOPSwitch are registered trademarks of Power Integrations, Inc.

©Copyright 1994, Power Integrations, Inc. 477 N. Mathilda Avenue, Sunnyvale, CA 94086

WORLD HEADQUARTERS AMERICAS EUROPE & AFRICA

Power Integrations, Inc. For Your Nearest Sales/Rep Office Power Integrations (Europe) Ltd.
477 N. Mathilda Avenue Please Contact Customer Service Mountbatten House
Sunnyvale, CA 94086 Phone: 408•523•9265 Fairacres
USA Fax: 408•523•9365 Windsor SL4 4LE
Main: 408•523•9200 United Kingdom
Customer Service: Phone: 44•(0)•1753•622•208
Phone: 408•523•9265 Fax: 44•(0)•1753•622•209
Fax: 408•523•9365

JAPAN ASIA & OCEANIA APPLICATIONS HOTLINE

Power Integrations, Inc. For Your Nearest Sales/Rep Office World Wide 408•523•9260
Keihin-Tatemono 1st Bldg. Please Contact Customer Service
12-20 Shin-Yokohama 2-Chome.Kohoku-ku Phone: 408•523•9265 APPLICATIONS FAX
Yokohama-shi, Kanagawa 222 Fax: 408•523•9365 Americas 408•523•9361
Japan Europe/Africa
Phone: 81•(0)•45•471•1021 44•(0)•1753•622•209
Fax: 81•(0)•45•471•3717 Japan 81•(0)•45•471•3717
Asia/Oceania 408•523•9364

12 C
6/96