Design Considerations for Clamping Networks for Very High Voltage Monolithic Off-line PWM Controllers

AN1680/D
Design Considerations for
Clamping Networks for Very
High Voltage Monolithic
Off-line PWM Controllers
http://onsemi.com
Prepared by: Christophe BASSO
MOTOROLA SPS
BP-1029, Le Mirail
31023 Toulouse France
email: [email protected];
Tel.: 33 5 61 19 90 12
APPLICATION NOTE
INTRODUCTION
these devices and how to predict its efficiency in the
application.
In the large family of Switch-Mode Power Supply
(SMPS) components, the recently introduced high-voltage
monolithic switchers start to play an important role. First of
all because they provide an easy mean to instantaneously
build an efficient off-line supply but also because their
internal structure offers everything a designer needs:
internal clock, pulse-by-pulse limitation, Leading Edge
Blanking (LEB) etc. However, the internal MOSFET
exhibits a low-energy capability body-diode which no
longer protects the device against accidental avalanche.
This element thus needs an adequate protection network
against the electromagnetic leakage energy. This paper
details what network is best adapted to the protection of
The Leakage Inductance
Figure 1. shows a transformer wound across a standard
magnetic material. The primary side made of Np turns
creates the necessary force F which gives birth to two
components: fm who links both windings, but also fl1
which does not couple to the secondary and corresponds to
a leakage path through the air. Thanks to fm, a current Io
circulates in the secondary, but this current also gives birth
to another leakage flux fl2 whose polarity is opposite of
that of fm. It is important to note that fm produces Io
while fl2 is a consequence of it.
Ip
Io
fl1
fl2
Np
Ns
Vexcitation
Vo
fm
Magnetic material with a
linear reluctance ℜm
Figure 1. A Two-winding Transformer Showing the Leakage Paths
As one can see from the picture, φl1 and φl2 close
through the air. As any (magnetic) medium, the air is
affected by a Reluctance ℜ, or its inverse, the permeance P.
These permeances create in the primary and secondary two
leakage inductance with a value of: Lleak = N2 ⋅ Pair , with N
© Semiconductor Components Industries, LLC, 2000
September, 2000 − Rev. 0
being the primary or secondary turns. As an effect, these
parasitic leakage elements degrade the energy transfer
between the primary and secondary (ies). In a FLYBACK
converter, the presence of the leakage element will a)
generate a voltage spike at turn-off and b) divert a portion
1
Publication Order Number:
AN1680/D
AN1680/D
primary and leakage coils. When the ON periods stops, the
MOSFET opens and interrupts its current. Since no current
discontinuities can take place in an inductor, both magnetic
fields collapse and the voltage across the inductances
reverses in an attempt to keep the amps-turn constant: Lp
energy is thus coupled to the secondary and gives birth to
the output current charge. Since Lleakage cannot find a
circulating path, it pulls-up 2 the drain voltage until Dclamp
starts to conduct and protects the switcher at a maximum
theoretical level of: Vmains + Vclamp = 650 V. Figure 2. (b)
shows the results of an INTUSOFT’s IsSpice4 (San Pedro,
CA) simulations. When all the leakage energy is released, a
short parasitic oscillation takes place involving Lleakage and
all the parasitic capacitive elements present in the circuit
(transformer’s primary capacitance, MOSFET’s Coss etc.)
of the primary current into the clamping network. Point a)
implies the use of protection network to prevent a lethal
drain voltage excursion while point b) explains the root of
a degraded open-loop gain: the peak current needs to be
higher than theoretically calculated to deliver the full rated
output power.
The Principle of a Protection Network
The goal of the clamping network is to prevent the drain
voltage to exceed a given limit. For instance, in the new ON
Semiconductor MC3337X family, the maximum voltage
shall stay within 700 V. A worse case arises when the mains
is at its highest level, e.g., 285 VAC in a universal mains
application. To prevent the drain from reaching this value,
Figure 2. (a) shows how a perfect network would work:
when the MOSFET closes, the current builds-up in both
RATIO_POW = −0.183
RATIO_AUX = −0.142
MC33374
Vclamp
250 V
R6
3.6
C1
50 mF
C2
1 mF
C4
100 n
R7
75 k
13
1
Rp
100 m
17
Aux
16
D1
1N4146
R8
15 k
C5
10 mF
Lleak
4 mH
MOS
GND
VDCmains
330 V
CTL
VCC
+
FB
2
L3
5 mF
UP TI
Lp
250 mH
+
3
Dclamp
Int
4
7
MC
3337X
X2
MBR20100
10
C3
150 mF
Int
6
R12
270
Aux
12
C6
10 nF
14
MOC8101
15
UP TI
16
Figure 2. (a) A Simple FLYBACK Configuration Implementing a Clamping Network
http://onsemi.com
2
Rload
5
Vout
AN1680/D
700
500
VDS
CLIPPING EFFECT
X = 212.29U
Y = 650.84
Fosc +
300
1
[email protected]@
ǸLleak @ Clump
100
−100
210.20U
211.20U
212.20U
213.20U
214.20U
Figure 2. (b) The Drain is Safely Clipped Below 700 V at High Mains
Diverting the Primary Current
where the secondary diode catches-up with the primary
current. The slope of the decreasing primary current is
simply
Figure 2. (a) is interesting because it helps understanding
how the reflected secondary voltage resets the leakage
energy and how much of primary current this leakage
inductance “steals away” by diverting it into the clamp.
Everything is detailed on Figure 2. (c) graph. When the
MOSFET turns-off, a reset voltage is applied to the leakage
inductance. This reset voltage depends on the clipping
voltage but also on the FLYBACK’s. The higher this level,
the faster the leakage energy drops to zero and authorizes
the secondary current to take place. The time Δt needed to
complete the energy transfer is easily defined by:
Dt +
N @ (Vout ) Vf sec)
,
Lp
but this equation can also be written as:
N @ (Vout ) Vfsec)
Ip * Ipx
.
+
Dt
Lp
Replacing Δt and solving for Ipx gives:
Ipx
+1*
Ip
Lp @
L
@ Ip
leak
,
V
* (Vout ) Vf sec) @ N
clamp
ǒ
L
leak
.
V
clamp
*1
(V )Vf)@N
out
This last equation gives you the effective percentage of
primary current stolen by the leakage inductance. Applying
Figure 2. numerical values gives: Ipx = 98.4% of Ip. Since
Ip grows up to 2.73 A, then the theoretical peak secondary
current establishes at: 0.984 ⋅ 2.73 ⋅ 12.5 = 33.58 A.
Figure 2. (d) validates the calculation.
where Ip is the final primary current, N the transformer
ratio secondary to primary, Vfsec the secondary diode
forward drop and Lleak the primary leakage inductance.
Estimating the percentage of diverted current tells you
the real peak current you will actually put in the primary to
deliver the rated power. Figure 2. (c)’s Ipx point shows
Iprimary
Ip
Ipx
Vin/Lp
Vo @ N
Lp
ILleak
Iprimary
ILleak
Isecondary
ton
Ǔ
Vclamp * Vo @ N
L
leak
Dt
Figure 2. (c) Waveforms at Turn-off: the Leakage Coil Prevents an Immediate Transfer
http://onsemi.com
3
AN1680/D
30
x = 202.37U
y = 33.600
VCC + Vz
Isec. peak
VO.N
Ip = 2.73 A
20
10
Vdrain-source
Ileakage
Iprimary
0
Vout @ turn-off = 10 V
200.60U
201.79U
Ip
202.98U
204.17U
205.37U
Izener diode
Figure 2. (d) IsSpice4 Simulation of Figure 2’s Circuit
Dt
Protecting the MOSFET with an Active Element
The easiest way to clamp at a known level is to replace
the null-impedance Vclamp source by a zener diode or a
transient suppressor. Figures 4. (a) and 4. (b) detail the
connections and their associated waveforms. Since we have
two diodes in series, we have to care about both dissipated
powers. Diodes can be modeled by a voltage source V
(which equals Vzener or Vforward) in series with a dynamic
resistance Rd. The total average conducted power
dissipated is therefore:
Figure 3. (b) Clipping Waveforms with a Zener Diode
To account for the Rd term, MOTOROLA specifies a
clamping factor FC which gives the real peak zener voltage
at a given peak current: VZ(Peak) = VZ(Nom) . FC. From this
formula, we can write: Vz(Nom) + Rd ⋅ Iz(peak) = Fc ⋅
Vz(Nom). Solving for Rd gives:
2
(Fc * 1) @ V
z(Nom) ,
Rd +
P
PK(Nom)
Pavg = V ⋅ Iavg + Rd ⋅ I2rms.
For the zener diode, the first part is easily deducted from
Figure 3. (b):
with PPK(Nom) being the maximum peak power accepted by
the zener diode or the transient suppressor. From Figure
3(b), let’s now calculate the RMS and average values of the
zener current:
V @ Ip Dt
Pavg1 + z
@
2
T
or, once solving with Δt:
I zener(t) + Ip @ Dt * t,
Dt
Vz @ Ip 2 @ L
@F
leak
,
P avg1 +
2 @ (Vz * (Vout ) Vfsec) @ N)
to obtain the RMS value simply solve:
Ǹŕ
with F the switching frequency and Vz the nominal zener
level.
8
5
Ǹ3Dt@ T, T
is the switching period. Pavg2 is thus:
Vout
P avg2 +
Cout
Lp
4
I 2zener(t).dt or I zenerRMS + Ip @
0
1
Dclamp
1
T
D1
N:1
DCrail
Dt
Rd @ Ip 2 @ F @ Dt.
3
The IzenerAVG value, which affects the conduction losses
of the diode in series with the zener is evaluated by:
2
R1
I zenerAVG +
Ip @ Dt @ F .
2
As we previously wrote, the diode conduction losses are
expressed the same way as the zener’s, except that
Lleak
Rd + dVf
dId
extracted from the Vf versus Id curve in the diode
data-sheet at Id = Ip. To summarize we have:
7
Ip 2 @ L
@ F @ (V z ) 0.66 @ Rd @ Ip)
leak
2 @ (V z * N @ (V out ) Vf sec))
Ip 2 @ L
@ F @ (V ) 0.66 @ Rd @ Ip)
leak
Ip
f
P
+
cond_diode
2 @ (V z * N @ (V out ) Vf sec))
P
Drv
Figure 3. (a) Clipping with a Zener is Possible Solution
cond_zener
http://onsemi.com
4
+
AN1680/D
a 180 V zener diode, then the series diode starts to clip at
605 V. However, because the injection time into the
low-doped N-region takes about 50 ns, the dynamic
resistance of the MUR160 gradually drops to its nominal
value, accordingly generating an overshoot upon the drain.
Measurements have to be carried upon the final board to
confirm the safety of the final drain level.
The final clipping level will be affected by two
components: the zener clamping factor FC but also the time
the series diode takes to react. If we select a fast
MOTOROLA MUR160, the data-sheet specifies a turn-on
time of 50 ns. In presence of drain voltage rising with a 1.5
kV/μs slope, the diode will start to conduct at VmainsDC +
Vz . FC. If we take the highest mains level of 275VAC and
RATIO_AUX = −0.156
RATIO_AUX = −0.156
Rclamp
100 k
RC CLAMPING
NETWORK
14
R7
15 k
13
9
D4
1N4148
LI1
58 m
16
10
MC33363
Vaux
R5
110 k
13
6
11
7
10
8
9
Vclamp + Vin (Passive Network)
BVDSS (MOSFET Alone)
Aux
Vdrain
12
Rload
9.8
6
1
3
5
C10
470 mF
8
Lp
3.56 M
1
4
Vout
17
2
12
X1
MC33363
VDCmains
320
Aux
D1
1N5821
R1
3.5
Dclamp
1N4937
+
4
10
Cclamp
1 nF
test
Overvoltage Protect.
Vout.N
R8
4.7 k
Ip
11
20
C2
10 mF
C4
7
100 nF
R6
24 k
16
R10
921
C6
C5
820 p 100 nF
Dt
Figure 4. (a) An RC Network Clamps the Drain
Voltage at Turn-off
0
Figure 4. (b) Turn-off Waveforms with a
MOSFET Clipping the Peak
When to Use a Zener or a Transient Suppressor?
A Passive RC Network to Clamp the Drain
If the above solution provides a stable clamping level
rather independent from peak current variations, the cost of
those zener elements can degrade the overall price of your
SMPS. The alternative lies in implementing a passive RC
network as the one depicted in Figure 4. .
If we assume we do not have any external clipping
network, we can calculate the amount of energy ET
dissipated in the transistor every time it opens, assuming it
would safely avalanche the voltage (Figure 4. (b)). As we
said, the leakage tries to keep the current circulating at its
level (Ip, when the transistor opens) during Δt and pushes
the drain voltage up to BVDSS. Ip(t) can be expressed by:
There are little technology differences behind a standard
zener diode and a transient. However, the die area is far
bigger for a transient suppressor than that of zener. A 5 W
zener diode like the 1N5388B will accept 180 W peak
power if it lasts less than 8.3 ms. If the peak current in the
worse case (e.g., when the PWM circuit maximum current
limit works) multiplied by the nominal zener voltage
exceeds these 180 W, then the diode will be destroyed when
the supply experiences overloads. A transient suppressor
like the P6KE200 still dissipates 5 W of continuous power
but is able to accept surges up to 600 W @ 1 ms. If the peak
power is really high, then turn to a 1.5KE200 which accepts
up to 1.5 kW @ 1 ms.
Ip(t) + Ip @ Dt * t.
Dt
http://onsemi.com
5
AN1680/D
You calculate the energy by integrating over time the
cross-over area between current and voltage:
It is also sometimes interesting to know the clamp
voltage set by known RC elements. In that case, simply
apply:
Dt
E +
T
ŕ Id(t) @ VDS(t) @ dt + 12 @ Ip @ BVDSS @ Dt
0
V
+ 1 @ Vout @ N ) 1 @
clamp
2
2
Ǹ Vout
.
If we now introduce the term Δt previously calculated, we
get:
P
+1@ V
@ Ip @ Dt @ F
clamp
clamp
2
.
Δt has already been defined, but this time, BVDSS is
replaced by Vclamp. Once introduced in the previous
equation, we obtain the power dissipated in the clipping
network:
V
clamp
+ 1 @ Ip 2 @ L
@F@
clamp
leak
2
V
* (Vout ) Vfsec)N
clamp
Since this power will mainly be dissipated by the resistor
Rclamp at steady-state, we can write the following equality:
2
V
V
clamp
clamp
+ 1 @ Ip 2 @ L
@F
leak
2
R
V
* (Vout ) Vf sec)N
clamp
clamp
.
By solving for Rclamp we calculate its value for a given level of
clamping voltage:
clamp
+
[email protected]
@ (V
* (Vout ) Vf sec) @ N)
clamp
clamp
L
@ Ip 2 @ F
leak
It is important to minimize the ripple level Vripple
superimposed on Vclamp. A capacitor Cclamp will fulfill this
function. If we agree that the amount of charge Q will
equally split between Rclamp and Cclamp at turn-off we can
write: Vripple . Cclamp = IRclamp. T. Knowing that IRclamp =
Vclamp / Rclamp and then solving for Cclamp it comes:
C
clamp
+
V
ripple
V
clamp
@[email protected]
clamp
In steady-state operation, by neglecting the RMS current
circulating in Rclamp, the RMS current which flows in the
capacitor is the same as in the fast series diode:
I
C
+ Ip @
RMS clamp
clamp
@L
leak
@ Ip 2 @ F
If RC networks are more attractive than zeners when
talking about price, they suffer from a poor behavior when
the peak current varies. As a matter of fact, the clamping
level will always be calculated at the highest primary
current. The highest primary current depends, of course, on
the internal current limit (at the highest TJ) but also from
the turn-off response time due to the over-current
comparator propagation delay. For the MC3337X family,
this value is typically 280 ns, while the maximum current
limit increases by roughly 3.5% at the maximum operating
temperature. As an example, let us take a primary
inductance of 290 μH. With this value in mind, a high
mains of 285 VAC imposes a slope of 1.38 A/μs when the
MOSFET closes. The maximum current limit of the
MC33374 is set at 3.7 + 3.5% = 3.83 A. With the previous
slope and a total delay of 280 ns, the switch will close when
the current finally reaches: 3.83 A + 0.28 μs ⋅ 1.38 = 4.21 A.
To have an idea of the peak current dependency of the RC
network’s clamp level, we built a 13 W power supply
supposed to operate on wide mains. Despite a 400 mA
nominal peak current, we used an MC33373 to operate
without an heatsink. MC33373 authorizes peak currents up
to 3 A. The RC network was specified to clamp at 240 V @
Ip nominal and gave values of 11 kW and 100 nF. With a
118 μH leakage inductance and a 13.8 turn ratio, the normal
operating clamping voltage was measured at 238 V.
However, as one can see from Figure 5. , the clamping
level grows-up by adding successive voltage steps,
corresponding to every switch turn-off. At power-on,
Vclamp and Vout are both at zero. The internal error
amplifier pushes the duty-cycle toward 70% but each cycle
is fortunately truncated when Idrain exceeds the internal
limitation. Vclamp starts to grow and keeps rising until Vout
nominal is reached and forces the PWM modulator to
brake: Vclamp then diminishes to establish at the calculated
240 V. If for some reasons maximum Ipeak conditions
would stay longer than expected (e.g., Vout could not reach
it nominal value), Vclamp would continue its grow, no
longer protecting the internal MOSFET. To avoid this
condition, Figure 5. (b)’s plot depicting Vclamp = f (Ip) will
help the designer to track worse case conditions and react
by either lowering Rclamp or simply selecting a member of
the MC3337X family exhibiting a maximum peak current
of 400−500 mA (e.g., MC33369). Another solution is to use
an adequate zener diode whose clamping level will be less
sensitive to peak current variations.
This result depicts the average power the transistor would
be the seat of, if no mean were implemented to re-route the
energy spike elsewhere.
By wiring an RC network from drain to VDCmains via a
fast diode, we will prevent the drain voltage to rise above
VDCmains + Vclamp, the clipping voltage we want to impose
(Figure 4. (b)). Let us first consider that the voltage across
the RC network is constant (ripple is low compared to
average voltage) and equals Vclamp. The average power
dissipated at turn-off in the clamp is:
R
@ N2 ) 2 @ R
Watch-out for the Current Dependency!
BV
DSS
.
P + 1 @ Ip 2 @ L
@F@
T
leak
2
* (Vout ) Vf sec) @ N
BV
DSS
P
2
Ǹ3Dt@ T
http://onsemi.com
6
AN1680/D
V clamp (Ip)
1000
500
0
Figure 5. (a) Turn-on Sequence Showing Vclamp Running
Away to Large Values
Selecting the Right Active Components
0
1
2
3
Figure 5. (b) This Plot Shows How Vclamp Moves with Ip
To illustrate this point, Figure 6. shows a little spike
superimposed upon the drain voltage at turn-off due to this
diode turn-on time. This shot is directly taken from
Figure 4. circuit. Once again, this spike should never
exceeds BVDSS.
The series diode should be fast enough to clamp as soon
as the drain voltage exceeds Vclamp + VDCrail. As you can
understand, the switching time should be selected
accordingly with the drain voltage rising slope, dVDS / dt.
Figure 6. The Diode Turn-on Time Allows VDS to
Continue its Rise Before it Actually Clamps
The zener clamping level must be selected to be between
40 to 80 volts above the reflected output voltage when the
supply is heavily loaded. The given formulae will help you
determining the average losses of the zener but also its
maximum peak power (e.g., during power-on).
http://onsemi.com
7
AN1680/D
The following ON Semiconductor references can be used as active clipping elements:
Reference
Nominal Voltage (V)
Average Power (W)
Maximum Peak Power
1N5953B
150
1.5
98 W @ 1 ms
1N5955B
180
1.5
98 W @ 1 ms
1N5383B
150
5
180 W @ 8.3 ms
1N5386B
180
5
180 W @ 8.3 ms
1N5388B
200
5
180 W @ 8.3 ms
P6KE150A
150
5
600 W @ 1 ms
P6KE180A
180
5
600 W @ 1 ms
P6KE200A
200
5
600 W @ 1 ms
1.5KE150A
150
5
1.5 kW @ 1 ms
1.5KE180A
180
5
1.5 kW @ 1 ms
1.5KE200A
200
5
1.5 kW @ 1 ms
X = Full Scale Span
Another benefit of using a zener diode is to limit the
inverse voltage applied upon the fast series diode during
turn-on at VRRM = VDCmains. In the presence of an RC
clamping network, the clamping level unfortunately adds to
the input voltage and forces the adoption of a diode with
better VRRM parameter: VRRM = VDCmains + Vclamp.
Depending on the dVDS/dt slope and maximum reverse voltage conditions, the following MOTOROLA references can be used:
Reference
VRRM
Ton (typical)
IF max
MUR160
600 V
50 ns
3A
MUR100E
1000 V
25 ns
3A
1N4937
600 V
200 ns
1A
MSR860*
600 V
100 ns
8A
MSRB860-1*
600 V
100 ns
8A
* soft recovery diodes
Parameter Evolutions with Temperature
variations with the ambient temperature, they must not
jeopardize the SMPS operation. The below lines give an
idea on the way these parameters move with temperature:
Once protected, the equipment must survive in its
operating environment. That is to say, despite the
component’s key specs (e.g., Tfr for the series diode)
Conclusion
MC3337X’s MOSFET BVDSS: positive temperature
coefficient
MC3337X’s maximum current limit: +1.03% at TJ = 125°C
MC3337X’s oscillator frequency: 115 kHz at max TJ
Zener breakdown voltage: positive temperature coefficient
Diode reverse recovery time: positive temperature coefficient
Data-sheets
can
be
downloaded
http://mot-sps.com/books/current.html
In FLYBACK converters, leakage inductance always
stresses the switching elements. An efficient active or
passive protection network is, therefore, mandatory to
ensure the SMPS will survive in any operating conditions.
This paper details some possibilities on how to implement
the adequate level of protection.
at:
Spice models of diodes and MC3337X:
http://mot-sps.com/models/html/models.html
http://onsemi.com
8
AN1680/D
Notes
http://onsemi.com
9
AN1680/D
Notes
http://onsemi.com
10
AN1680/D
Notes
http://onsemi.com
11
AN1680/D
ON Semiconductor and
are trademarks of Semiconductor Components Industries, LLC (SCILLC). SCILLC reserves the right to make changes
without further notice to any products herein. SCILLC makes no warranty, representation or guarantee regarding the suitability of its products for any particular
purpose, nor does SCILLC assume any liability arising out of the application or use of any product or circuit, and specifically disclaims any and all liability,
including without limitation special, consequential or incidental damages. “Typical” parameters which may be provided in SCILLC data sheets and/or
specifications can and do vary in different applications and actual performance may vary over time. All operating parameters, including “Typicals” must be
validated for each customer application by customer’s technical experts. SCILLC does not convey any license under its patent rights nor the rights of others.
SCILLC products are not designed, intended, or authorized for use as components in systems intended for surgical implant into the body, or other applications
intended to support or sustain life, or for any other application in which the failure of the SCILLC product could create a situation where personal injury or
death may occur. Should Buyer purchase or use SCILLC products for any such unintended or unauthorized application, Buyer shall indemnify and hold
SCILLC and its officers, employees, subsidiaries, affiliates, and distributors harmless against all claims, costs, damages, and expenses, and reasonable
attorney fees arising out of, directly or indirectly, any claim of personal injury or death associated with such unintended or unauthorized use, even if such claim
alleges that SCILLC was negligent regarding the design or manufacture of the part. SCILLC is an Equal Opportunity/Affirmative Action Employer.
PUBLICATION ORDERING INFORMATION
NORTH AMERICA Literature Fulfillment:
Literature Distribution Center for ON Semiconductor
P.O. Box 5163, Denver, Colorado 80217 USA
Phone: 303−675−2175 or 800−344−3860 Toll Free USA/Canada
Fax: 303−675−2176 or 800−344−3867 Toll Free USA/Canada
Email: [email protected]
Fax Response Line: 303−675−2167 or 800−344−3810 Toll Free USA/Canada
N. American Technical Support: 800−282−9855 Toll Free USA/Canada
EUROPE: LDC for ON Semiconductor − European Support
German Phone: (+1) 303−308−7140 (Mon−Fri 2:30pm to 7:00pm CET)
Email: [email protected]
French Phone: (+1) 303−308−7141 (Mon−Fri 2:00pm to 7:00pm CET)
Email: [email protected]
English Phone: (+1) 303−308−7142 (Mon−Fri 12:00pm to 5:00pm GMT)
Email: [email protected]
CENTRAL/SOUTH AMERICA:
Spanish Phone: 303−308−7143 (Mon−Fri 8:00am to 5:00pm MST)
Email: [email protected]
Toll−Free from Mexico: Dial 01−800−288−2872 for Access −
then Dial 866−297−9322
ASIA/PACIFIC: LDC for ON Semiconductor − Asia Support
Phone: 303−675−2121 (Tue−Fri 9:00am to 1:00pm, Hong Kong Time)
Toll Free from Hong Kong & Singapore:
001−800−4422−3781
Email: [email protected]
JAPAN: ON Semiconductor, Japan Customer Focus Center
4−32−1 Nishi−Gotanda, Shinagawa−ku, Tokyo, Japan 141−0031
Phone: 81−3−5740−2700
Email: [email protected]
ON Semiconductor Website: http://onsemi.com
EUROPEAN TOLL−FREE ACCESS*: 00−800−4422−3781
*Available from Germany, France, Italy, UK, Ireland
For additional information, please contact your local
Sales Representative.
http://onsemi.com
12
AN1680/D
Similar pages