Minimizing destructive HV spikes on square wave push-pull MOSFET drive to transformer

[P E Schoen]

I had posted about success with fast MOSFET gate drivers on my 1500 VA DC-DC
converter. The UCC27321 drivers provide up to 9 amps so the MOSFETs now
switch on and off very quickly compared to the previous drivers, which were
actually gates of an LM324 op-amp. So, although I got better efficiency and
I was able to apply 25 VDC, the 30 volt TVS diodes from each drain to the 25
volt supply overheated, and I removed them.

The output waveform of the transformer was a very clean square wave with a
little bit of ringing during the 1 uSec dead-time transition, but the drain
waveform showed very high spikes at the transition. I was able to reduce the
amplitude somewhat, by adding a capacitor across the drains, as well as a
snubber, but one of the MOSFETs failed shorted. It was probably damaged
before I added the snubber and capacitor. But even after I replaced it, and
I added capacitors from drain to common, there were still unacceptably large
spikes, especially as I increased the voltage from my power supply.

There is also certainly a large current spike due to the load capacitors
being charged with a square wave, but I do have a 100A 100mV shunt from the
sources to common, and an external interrupt that should shut down the PWM
if the current is greater than 100A, but it could easily be much higher
before the PWM removes the gate drive. And then there would be a much larger
dead time. The MOSFETs are HUF75645 which are rated at 100V and 75A
continuous, and at least 450A pulse current for 10uSec. Turn on and turn off
times about 200 nSec. I have two in parallel on each leg of the push-pull.

So, I'm looking at ways to minimize these high voltage transients without
reducing efficiency too much. Some ideas are:

1. Adding capacitance to maintain the current flow through the inductor
during transition. This may be most efficient, but it will cause a lot of
ringing. And the capacitors need to be able to carry the maximum current of
the transformer primary, which could be as high as 50 amps or so. Maybe a
lot more if the current is reflected from the output capacitor charging.

2. Using RC snubbers to limit the peak voltage as well as dissipate the
energy in the resistor. I have tried a single snubber of 0.047uF and 2 ohms
across the primaries, which helped a little. But from my simulation it
seemed like it might be better to add snubbers across the D-S of each
MOSFET. Not really sure of the component values. For 50A peak the 2 ohms
will limit the spike to 100V. That's 5000W but only for less than 1uSec out
of a 250uSec pulse (2 kHz), so average power is about 5000/250 = 20W.
Probably much less.

3.Using TVS diodes to limit the voltage spikes to, say, 80V. The TVS diodes
I used were two 15V in series, so they were trying to suppress the spike at
a lower voltage than necessary, so they got hot. And that was before the
capacitors and snubbers. Probably a good idea to add them, and then design
the snubbers to keep them from absorbing any more than occasional spikes.

4. Slowing down the transitions of the gate drives. These high spikes seemed
to appear after I added the fast gate drivers, and the previous slower
drivers seemed to work well enough, although not as efficiently. So I can
probably add resistance and capacitance to cause slower turn-on and
turn-off, which will essentially quench the inductive spike in the MOSFETs.
Also reduced efficiency, but easier to dissipate the heat in the large heat
sinks than much smaller TVS diodes or snubber resistors.

I'm going to research some of the application notes on MOSFETs and gate
drivers, but some discussion would be appreciated. At least this is actually
about some "cool" electronics and not OT squabbling.

Thanks,

Paul

 

[Joerg]

I had posted about success with fast MOSFET gate drivers on my 1500 VA
DC-DC converter. The UCC27321 drivers provide up to 9 amps so the
MOSFETs now switch on and off very quickly compared to the previous
drivers, which were actually gates of an LM324 op-amp. So, although I
got better efficiency and I was able to apply 25 VDC, the 30 volt TVS
diodes from each drain to the 25 volt supply overheated, and I removed
them.

I am not so fond of the wimpy pull-up FET in those drivers but that's
another topic.

The output waveform of the transformer was a very clean square wave with
a little bit of ringing during the 1 uSec dead-time transition, but the
drain waveform showed very high spikes at the transition.

Can you post a scope shot? And schematics.

... I was able to
reduce the amplitude somewhat, by adding a capacitor across the drains,
as well as a snubber, but one of the MOSFETs failed shorted. It was
probably damaged before I added the snubber and capacitor. But even
after I replaced it, and I added capacitors from drain to common, there
were still unacceptably large spikes, especially as I increased the
voltage from my power supply.

There is also certainly a large current spike due to the load capacitors
being charged with a square wave, but I do have a 100A 100mV shunt from
the sources to common, and an external interrupt that should shut down
the PWM if the current is greater than 100A, but it could easily be much
higher before the PWM removes the gate drive. And then there would be a
much larger dead time. The MOSFETs are HUF75645 which are rated at 100V
and 75A continuous, and at least 450A pulse current for 10uSec. Turn on
and turn off times about 200 nSec. I have two in parallel on each leg of
the push-pull.

So, I'm looking at ways to minimize these high voltage transients
without reducing efficiency too much. Some ideas are:

1. Adding capacitance to maintain the current flow through the inductor
during transition. This may be most efficient, but it will cause a lot
of ringing. And the capacitors need to be able to carry the maximum
current of the transformer primary, which could be as high as 50 amps or
so. Maybe a lot more if the current is reflected from the output
capacitor charging.

2. Using RC snubbers to limit the peak voltage as well as dissipate the
energy in the resistor. I have tried a single snubber of 0.047uF and 2
ohms across the primaries, which helped a little. But from my simulation
it seemed like it might be better to add snubbers across the D-S of each
MOSFET. Not really sure of the component values. For 50A peak the 2 ohms
will limit the spike to 100V. That's 5000W but only for less than 1uSec
out of a 250uSec pulse (2 kHz), so average power is about 5000/250 =
20W. Probably much less.

3.Using TVS diodes to limit the voltage spikes to, say, 80V. The TVS
diodes I used were two 15V in series, so they were trying to suppress
the spike at a lower voltage than necessary, so they got hot. And that
was before the capacitors and snubbers. Probably a good idea to add
them, and then design the snubbers to keep them from absorbing any more
than occasional spikes.

TVS'es to muffle spikes is almost like putting oil on squealing brake
pads

4. Slowing down the transitions of the gate drives. These high spikes
seemed to appear after I added the fast gate drivers, and the previous
slower drivers seemed to work well enough, although not as efficiently.
So I can probably add resistance and capacitance to cause slower turn-on
and turn-off, which will essentially quench the inductive spike in the
MOSFETs. Also reduced efficiency, but easier to dissipate the heat in
the large heat sinks than much smaller TVS diodes or snubber resistors.

I'm going to research some of the application notes on MOSFETs and gate
drivers, but some discussion would be appreciated. At least this is
actually about some "cool" electronics and not OT squabbling.

It would be best to post schematics, layout and maybe snap a picture.

A (very wild) guess would be that the transformer is not optimized for
low leakage inductance and low capacitance yet, or maybe the hook-up of
that or the FET bridge is a bit loose. But without something to look at,
hard to say.

 

[Tim Williams]

RCD snubber from drain to ground helps. If you use a dV/dt snub (small
capacitor, small R*C time constant), you can slow down the edge, saving
switching losses in the transistor and absorbing the overshoot.

If you use a peak snub (large C, enough R to maintain DC level), the
overshoot spike is clamped, dumped into the capacitor, and dissipated as
heat in a dumb resistor rather than sensitive transistor junctions.

Note that a dV/dt snub carries load current while the voltage swings by.
This doesn't allow the stray inductance to discharge, so you still get
overshoot. To minimize overshoot, you need a much larger dV/dt snub than
otherwise, or you need a peak snub as well.

Switching transient with dV/dt snub, circuit:
http://webpages.charter.net/dawill/Images/SwitchingTransientTest.png
Waveform:
http://webpages.charter.net/dawill/Images/SwitchingTransient.png
Top: voltage, bottom: current 1A/V (-4 offset).
- Switch on: current blip due to diode capacitance, represented by C2.
Small ringing is due to the inductance between transistor, diode and output
load (the 36V supply), represented by L2. This series resonant circuit is
damped by Q1's Rds(on), V3's internal resistance (typically capacitor ESR),
and any loss components in the circuit (R4 in this case).
- On state: current rises; a lot of nothing else happens.
- Switch off: drain voltage rises and current falls. D2 turns on and C3
charges. Load current is transferred to C3. Technically, this waveform is
described not by a triangular slope, but by a fraction of the resonance
between L1 and C3.
- Damped "bouncing": as voltage rises past V3, D1 turns on and L2 charges
up. Now, the waveform is described by the resonance of C3, L2 and R4.
Voltage continues to rise, humps over, and falls. As the voltage falls, L2
discharges and D1 switches off. Now it goes back to the L1-C3 resonance,
which makes a short spike before D1, L2 clamp it again. This repeats until
the oscillation amplitude is less than Vf, at which point D1 remains on and
D2 remains off; the remaining oscillation is damped by R5. This occurs
after about two negative-going spikes in this case.

Ringing is well damped by R4, for simulation stability and clarity. Since,
in practice, this inductance is due to stray distance between components, it
is difficult to dampen directly. An alternative is an R+C damper across the
capacitance, in this case D1 and Q1, to dampen the falling and rising edges,
respectively.

The equivalent, applied to your circuit:
- L1 + R1 represents the load current, which in this case is delivered by
transformer. During the switching transient, it can be assumed reasonably
constant.
- Everything past the MOSFET and snubber is behind a transformer, so making
certain connections becomes dubious (e.g., placing R4 across only the
leakage inductance of a transformer!)
- L2 represents the leakage inductance of your transformer.
- V3 is the voltage which the waveform transitions to after the switching
event; since you're driving nearly full duty cycle into a PP transformer,
this is the opposite side MOSFET (and its body diode), which in this
circuit, is simply twice the supply voltage.

The magnitude of all components can be estimated from circuit geometry and
component measurements (you'll likely have to measure your transformer).
- C3 is calculated based on either the load current, delta V and desired
rise time (usually 50ns to 5us depending on use; for under-100kHz stuff,
200ns or so is fine), or the desired damping characteristics.
- You want the snubber to discharge appreciably while the transistor is on,
so you want 3 * R5 * C3 = t_on(min). This simulation used t_on = 10us, so I
picked 5us, close enough. I picked R5 somewhat large to illustrate the type
of nonlinear ringing you get from an underdamped diode type snubber.

The waveforms with a peak voltage snubber look similar, except:
- Rising edge is fast (obviously, it's normal)
- A fast ringing overshoot spike, determined by MOSFET-diode-capacitor path
inductance as the snubber diode "grabs" on
- "Top" voltage is determined by the capacitor, looking like a 1/4 sine wave
(relatively small capacitance) or a flat clamped flyback pulse (high
capacitance limit)
- When the leakage is done discharging, voltage relaxes back to its steady
value (i.e., the opposing half of the square wave), with ringing determined
by stray (diode and transistor junction) capacitance.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

I had posted about success with fast MOSFET gate drivers on my 1500 VA DC-DC
converter. The UCC27321 drivers provide up to 9 amps so the MOSFETs now
switch on and off very quickly compared to the previous drivers, which were
actually gates of an LM324 op-amp. So, although I got better efficiency and
I was able to apply 25 VDC, the 30 volt TVS diodes from each drain to the 25
volt supply overheated, and I removed them.

The output waveform of the transformer was a very clean square wave with a
little bit of ringing during the 1 uSec dead-time transition, but the drain
waveform showed very high spikes at the transition. I was able to reduce the
amplitude somewhat, by adding a capacitor across the drains, as well as a
snubber, but one of the MOSFETs failed shorted. It was probably damaged
before I added the snubber and capacitor. But even after I replaced it, and
I added capacitors from drain to common, there were still unacceptably large
spikes, especially as I increased the voltage from my power supply.

There is also certainly a large current spike due to the load capacitors
being charged with a square wave, but I do have a 100A 100mV shunt from the
sources to common, and an external interrupt that should shut down the PWM
if the current is greater than 100A, but it could easily be much higher
before the PWM removes the gate drive. And then there would be a much larger
dead time. The MOSFETs are HUF75645 which are rated at 100V and 75A
continuous, and at least 450A pulse current for 10uSec. Turn on and turn off
times about 200 nSec. I have two in parallel on each leg of the push-pull.

So, I'm looking at ways to minimize these high voltage transients without
reducing efficiency too much. Some ideas are:

1. Adding capacitance to maintain the current flow through the inductor
during transition. This may be most efficient, but it will cause a lot of
ringing. And the capacitors need to be able to carry the maximum current of
the transformer primary, which could be as high as 50 amps or so. Maybe a
lot more if the current is reflected from the output capacitor charging.

2. Using RC snubbers to limit the peak voltage as well as dissipate the
energy in the resistor. I have tried a single snubber of 0.047uF and 2 ohms
across the primaries, which helped a little. But from my simulation it
seemed like it might be better to add snubbers across the D-S of each
MOSFET. Not really sure of the component values. For 50A peak the 2 ohms
will limit the spike to 100V. That's 5000W but only for less than 1uSec out
of a 250uSec pulse (2 kHz), so average power is about 5000/250 = 20W.
Probably much less.

3.Using TVS diodes to limit the voltage spikes to, say, 80V. The TVS diodes
I used were two 15V in series, so they were trying to suppress the spike at
a lower voltage than necessary, so they got hot. And that was before the
capacitors and snubbers. Probably a good idea to add them, and then design
the snubbers to keep them from absorbing any more than occasional spikes.

4. Slowing down the transitions of the gate drives. These high spikes seemed
to appear after I added the fast gate drivers, and the previous slower
drivers seemed to work well enough, although not as efficiently. So I can
probably add resistance and capacitance to cause slower turn-on and
turn-off, which will essentially quench the inductive spike in the MOSFETs.
Also reduced efficiency, but easier to dissipate the heat in the large heat
sinks than much smaller TVS diodes or snubber resistors.

I'm going to research some of the application notes on MOSFETs and gate
drivers, but some discussion would be appreciated. At least this is actually
about some "cool" electronics and not OT squabbling.

Thanks,

Paul

 

[Jamie]

I had posted about success with fast MOSFET gate drivers on my 1500 VA
DC-DC converter. The UCC27321 drivers provide up to 9 amps so the
MOSFETs now switch on and off very quickly compared to the previous
drivers, which were actually gates of an LM324 op-amp. So, although I
got better efficiency and I was able to apply 25 VDC, the 30 volt TVS
diodes from each drain to the 25 volt supply overheated, and I removed
them.

The output waveform of the transformer was a very clean square wave with
a little bit of ringing during the 1 uSec dead-time transition, but the
drain waveform showed very high spikes at the transition. I was able to
reduce the amplitude somewhat, by adding a capacitor across the drains,
as well as a snubber, but one of the MOSFETs failed shorted. It was
probably damaged before I added the snubber and capacitor. But even
after I replaced it, and I added capacitors from drain to common, there
were still unacceptably large spikes, especially as I increased the
voltage from my power supply.

There is also certainly a large current spike due to the load capacitors
being charged with a square wave, but I do have a 100A 100mV shunt from
the sources to common, and an external interrupt that should shut down
the PWM if the current is greater than 100A, but it could easily be much
higher before the PWM removes the gate drive. And then there would be a
much larger dead time. The MOSFETs are HUF75645 which are rated at 100V
and 75A continuous, and at least 450A pulse current for 10uSec. Turn on
and turn off times about 200 nSec. I have two in parallel on each leg of
the push-pull.

So, I'm looking at ways to minimize these high voltage transients
without reducing efficiency too much. Some ideas are:

1. Adding capacitance to maintain the current flow through the inductor
during transition. This may be most efficient, but it will cause a lot
of ringing. And the capacitors need to be able to carry the maximum
current of the transformer primary, which could be as high as 50 amps or
so. Maybe a lot more if the current is reflected from the output
capacitor charging.

2. Using RC snubbers to limit the peak voltage as well as dissipate the
energy in the resistor. I have tried a single snubber of 0.047uF and 2
ohms across the primaries, which helped a little. But from my simulation
it seemed like it might be better to add snubbers across the D-S of each
MOSFET. Not really sure of the component values. For 50A peak the 2 ohms
will limit the spike to 100V. That's 5000W but only for less than 1uSec
out of a 250uSec pulse (2 kHz), so average power is about 5000/250 =
20W. Probably much less.

3.Using TVS diodes to limit the voltage spikes to, say, 80V. The TVS
diodes I used were two 15V in series, so they were trying to suppress
the spike at a lower voltage than necessary, so they got hot. And that
was before the capacitors and snubbers. Probably a good idea to add
them, and then design the snubbers to keep them from absorbing any more
than occasional spikes.

4. Slowing down the transitions of the gate drives. These high spikes
seemed to appear after I added the fast gate drivers, and the previous
slower drivers seemed to work well enough, although not as efficiently.
So I can probably add resistance and capacitance to cause slower turn-on
and turn-off, which will essentially quench the inductive spike in the
MOSFETs. Also reduced efficiency, but easier to dissipate the heat in
the large heat sinks than much smaller TVS diodes or snubber resistors.

I'm going to research some of the application notes on MOSFETs and gate
drivers, but some discussion would be appreciated. At least this is
actually about some "cool" electronics and not OT squabbling.

Thanks,

Paul

Have you tried the Parallel RC gate driving network? THe shunt cap
should give you a quick response of gate charge change and the R should
suppress the ring or greatly reduce it.

Jamie

 

[P E Schoen]

Tim:

Thanks for the very detailed and helpful reply. I need some time to absorb
all of that and respond, but in the meantime I am attaching a simplified
LTSpice ASC file that fairly accurately duplicates the waveforms I see on my
scope. I changed the coupling K factor to 0.998. I'm not sure how to measure
the leakage inductance, but using an LCR meter I read 180uH for each primary
winding and 32mH for the secondary, and I used those values for the
simulation.

I added 100 ohms gate resistors, but it did not seem to help much. And
0.047uF capacitors from drain to ground also didn't help much. I had 80V
peaks at 20V in, so it will be dangerously close to 100V with 25-28VDC input
(two batteries). Also, the peaks seemed even much higher during start-up.

In the simulation, I tried various values of gate resistors and capacitors
from gate to ground. The actual gate capacitance is about 4000pF, but I had
to add 0.1uF and 100 ohm to reduce the spikes, especially during startup.
This is an RC TC of about 10uSec, but the linear region is probably about
1/10 that with 10V gate drive, so maybe 1uSec out of 250uSec. The simulation
shows only about 5 watts in each MOSFET, with 330W output, so it does not
seriously affect efficiency. I show 10 watts in each capacitor, and overall
efficiency of 91.6%. So maybe this is a good way to reduce the spikes.

And I can also add the TVS diodes, just to be safe. As long as they don't
get more than warm, it's OK. I'd rather lose a couple watts than the
MOSFETs. BTW, as I was testing I encountered another overcurrent fault,
and thought I had blown another MOSFET. But I found that the driver had
partially shut down so it was effectively driving only one side. It appeared
to be leakage on the perfboard. I cleaned it with alcohol, detergent, and
water, touched up some solder connections, and it works much better.

I did a test with a 300 ohm load and I got an actual efficiency of 75% with
8.8V in at 5.1A (supply current limit). Subtracting the core loss I get 92%.
Once I'm sure I have the spikes under control, I can apply the batteries and
a real load. I'm looking forward to actually taking a ride on my electric
tractor!

Thanks,

Paul

=================================================================
Version 4
SHEET 1 880 680
WIRE 256 96 224 96
WIRE 256 128 256 96
WIRE 320 128 256 128
WIRE 432 128 384 128
WIRE 496 128 432 128
WIRE 544 128 496 128
WIRE 560 128 544 128
WIRE 592 128 560 128
WIRE -32 144 -160 144
WIRE 32 144 -32 144
WIRE 96 144 32 144
WIRE 128 144 96 144
WIRE 496 144 496 128
WIRE 224 192 224 96
WIRE 320 208 288 208
WIRE 432 208 432 128
WIRE 432 208 384 208
WIRE 496 208 464 208
WIRE 592 208 592 128
WIRE -640 240 -704 240
WIRE -576 240 -640 240
WIRE -496 240 -576 240
WIRE -448 240 -496 240
WIRE 128 240 128 224
WIRE 128 240 -448 240
WIRE 32 256 32 208
WIRE 128 256 128 240
WIRE 544 256 544 128
WIRE 544 256 496 256
WIRE 288 272 288 208
WIRE 288 272 224 272
WIRE 320 272 288 272
WIRE 416 272 384 272
WIRE -320 288 -384 288
WIRE -80 288 -256 288
WIRE 496 288 496 256
WIRE -640 320 -640 240
WIRE -704 336 -704 240
WIRE -384 336 -384 288
WIRE -256 336 -256 288
WIRE 32 336 0 336
WIRE 80 336 32 336
WIRE 112 336 80 336
WIRE 128 336 112 336
WIRE 80 352 80 336
WIRE 160 352 80 352
WIRE -496 368 -496 240
WIRE -32 368 -32 144
WIRE 80 368 -32 368
WIRE 160 368 160 352
WIRE 256 368 256 128
WIRE 320 368 256 368
WIRE 416 368 416 272
WIRE 416 368 384 368
WIRE 464 368 464 208
WIRE 464 368 416 368
WIRE 496 368 496 352
WIRE 496 368 464 368
WIRE 592 368 592 288
WIRE 592 368 496 368
WIRE -160 384 -160 144
WIRE 0 384 0 336
WIRE 592 384 592 368
WIRE -576 416 -576 240
WIRE -320 464 -320 288
WIRE -208 464 -320 464
WIRE -80 464 -80 288
WIRE -48 464 -80 464
WIRE -704 528 -704 400
WIRE -640 528 -640 384
WIRE -640 528 -704 528
WIRE -576 528 -576 480
WIRE -576 528 -640 528
WIRE -544 528 -576 528
WIRE -496 528 -496 448
WIRE -496 528 -544 528
WIRE -384 528 -384 416
WIRE -384 528 -496 528
WIRE -320 528 -384 528
WIRE -256 528 -256 416
WIRE -256 528 -320 528
WIRE -240 528 -256 528
WIRE -80 528 -240 528
WIRE 80 528 80 512
WIRE 80 528 -80 528
WIRE 160 528 160 512
WIRE 160 528 80 528
WIRE -240 560 -240 528
WIRE -160 560 -160 480
WIRE 0 560 0 480
WIRE 0 560 -160 560
WIRE -544 608 -544 528
FLAG -544 608 0
FLAG 592 384 0
FLAG 560 128 Vout
FLAG -448 240 in
FLAG 96 144 m1
FLAG 112 336 m2
SYMBOL ind2 112 128 R0
SYMATTR InstName L1
SYMATTR Value 180µ
SYMATTR Type ind
SYMATTR SpiceLine Rser=100u
SYMBOL ind2 112 240 R0
WINDOW 0 45 35 Left 2
WINDOW 3 41 61 Left 2
SYMATTR InstName L2
SYMATTR Value 180µ
SYMATTR Type ind
SYMATTR SpiceLine Rser=100u
SYMBOL ind2 240 176 M0
WINDOW 0 21 -5 Left 2
WINDOW 3 -9 113 Left 2
SYMATTR InstName L3
SYMATTR Value 32m
SYMATTR Type ind
SYMATTR SpiceLine Rser=10m
SYMBOL nmos -208 384 R0
SYMATTR InstName M1
SYMATTR Value IRFZ44N
SYMBOL nmos -48 384 R0
SYMATTR InstName M2
SYMATTR Value IRFZ44N
SYMBOL voltage -496 352 R0
WINDOW 123 0 0 Left 2
WINDOW 39 24 132 Left 2
SYMATTR SpiceLine Rser=8m
SYMATTR InstName V1
SYMATTR Value 24
SYMBOL diode 384 288 M270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName D2
SYMATTR Value MUR460
SYMBOL diode 320 224 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName D3
SYMATTR Value MUR460
SYMBOL polcap 480 288 R0
WINDOW 3 24 64 Left 2
SYMATTR Value 1000µ
SYMATTR InstName C1
SYMATTR Description Capacitor
SYMATTR Type cap
SYMATTR SpiceLine V=400 Irms=30 Rser=0.02 Lser=0
SYMBOL res 576 192 R0
SYMATTR InstName R1
SYMATTR Value 300
SYMBOL voltage -384 320 R0
WINDOW 123 0 0 Left 2
WINDOW 39 -43 57 Left 2
WINDOW 3 228 271 Left 2
SYMATTR SpiceLine Rser=100
SYMATTR Value PULSE(0 10 0.5u 10n 10n 499u 1000u 100)
SYMATTR InstName V2
SYMBOL voltage -256 320 R0
WINDOW 123 0 0 Left 2
WINDOW 39 -43 57 Left 2
WINDOW 3 100 298 Left 2
SYMATTR SpiceLine Rser=100
SYMATTR Value PULSE(0 10 500.5u 10n 10n 499u 1000u 100)
SYMATTR InstName V3
SYMBOL diode 320 144 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName D1
SYMATTR Value MUR460
SYMBOL diode 384 384 M270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName D4
SYMATTR Value MUR460
SYMBOL polcap -592 416 R0
WINDOW 3 24 64 Left 2
SYMATTR Value 2200µ
SYMATTR InstName C2
SYMATTR Description Capacitor
SYMATTR Type cap
SYMATTR SpiceLine V=25 Irms=20 Rser=100m Lser=0
SYMBOL polcap 480 144 R0
WINDOW 3 24 64 Left 2
SYMATTR Value 1000µ
SYMATTR InstName C3
SYMATTR Description Capacitor
SYMATTR Type cap
SYMATTR SpiceLine V=400 Irms=30 Rser=0.02 Lser=0
SYMBOL cap -656 320 R0
SYMATTR InstName C5
SYMATTR Value .47µ
SYMATTR SpiceLine V=250 Rser=100u
SYMBOL schottky -688 400 R180
WINDOW 0 24 64 Left 2
WINDOW 3 24 0 Left 2
SYMATTR InstName D5
SYMATTR Value MBRB2545CT
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL res -144 544 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R3
SYMATTR Value .001
SYMBOL cap 16 144 R0
SYMATTR InstName C4
SYMATTR Value .047µ
SYMATTR SpiceLine V=250 Rser=100u
SYMBOL res 16 352 M180
WINDOW 0 36 76 Left 2
WINDOW 3 36 40 Left 2
SYMATTR InstName R2
SYMATTR Value 2
SYMBOL cap -336 464 R0
SYMATTR InstName C6
SYMATTR Value 0.1µ
SYMATTR SpiceLine V=250 Rser=100u
SYMBOL cap -96 464 R0
SYMATTR InstName C7
SYMATTR Value 0.1µ
SYMATTR SpiceLine V=250 Rser=100u
SYMBOL cap 144 368 R0
SYMATTR InstName C8
SYMATTR Value .047µ
SYMATTR SpiceLine V=250 Rser=100u
SYMBOL cap 64 368 R0
SYMATTR InstName C9
SYMATTR Value .047µ
SYMATTR SpiceLine V=250 Rser=100u
SYMBOL res 64 528 M180
WINDOW 0 36 76 Left 2
WINDOW 3 36 40 Left 2
SYMATTR InstName R4
SYMATTR Value 2
SYMBOL res 144 528 M180
WINDOW 0 36 76 Left 2
WINDOW 3 36 40 Left 2
SYMATTR InstName R5
SYMATTR Value 2
TEXT 32 88 Left 2 !K1 L1 L2 L3 0.998
TEXT -536 576 Left 2 !.tran 0 200m 0 1u startup
TEXT -352 88 Left 2 ;Primary 2x8 turns 2V/turn at 600 Hz

 

[P E Schoen]

"Bill Sloman" wrote in message

I see your point, but the amount of energy is fairly small, and it is
difficult to convert it to a usable form. The zener type TVS, at least, can
handle repetitive surges, unlike the ZnO type surge suppressors that degrade
on each spike and eventually fail.

The schematic is essentially as posted in the simulation. The gate drives
are from a PIC16F684 PWM at 50% duty cycle and 1 uSec dead time.

Is the transformer bifilar wound? This would minimises the leakage
inductance between the two halves of the centre-tapped coil - if
you were using a centre-tapped coil.

Bifilar winding means winding the centre-tapped winding with
twisted pair, which gives you horrible inter-winding capacitance
(twisted-pair usually runs at about 150pF per metre) but absolutely
minimal leakage inductance (provided that you take the leads off
the transformer as twisted pair and only separate them by the bare
minimum at the driving transistors (and the current loop represented
by the alternative paths through the driving transistors should be
minimised too).

Sorry if this is irrelevant, but I once educated Tony Williams on this
point (to my immense surprise) and I've acted as if not everybody
has absorbed this information ever since.

You can see the transformer and the circuit board and wiring in the
following video:

At that time I was running at 500 Hz and the gate drives were from an LM324.
When I tried to run it again, with the battery freshly charged, the TVS
diodes burned up and shorted. It seems that the spikes are higher during
start-up and also the higher battery voltage may have contributed.

The transformer has 100 turns of about #18 AWG wound as the first layer
(secondary) on the toroidal core, which was then wrapped with Mylar. Then
the two primary windings were added, using 8 turns each of #10 AWG. There is
not much room left through the core, so it's about maximum fill. It is not
wrapped very tightly, as you can see. I'm trying to determine the pros and
cons of using an iron core transformer at higher frequencies for more power
in the same size. So I expect to get about 1500 VA out of it, which is about
3 times its rating at 60 Hz. I was hoping to be able to get as much as 3-4
kVA, but I think the copper losses will be too large. Possibly a toroid with
a larger hole will allow larger wire, and at the higher frequency not as
much iron will be needed.

Of course the best design might be using ferrite and 100kHz or higher. But
then there are are other design problems. And you must define "better".

Thanks,

Paul

 

[Joerg]

"Bill Sloman" wrote in message


I see your point, but the amount of energy is fairly small, and it is
difficult to convert it to a usable form. The zener type TVS, at least,
can handle repetitive surges, unlike the ZnO type surge suppressors that
degrade on each spike and eventually fail.


The schematic is essentially as posted in the simulation.

Which simulation? If it was a post many moons ago that would have rolled
off here.

... The gate
drives are from a PIC16F684 PWM at 50% duty cycle and 1 uSec dead time.

Why so much? That's a ton of dead time during which things can do plenty
of ringading. Can't remember any switcher design of mine that had more
than 100nsec.

You can see the transformer and the circuit board and wiring in the
following video:

Muttley looked kind of disappointed that he didn't get a ride

One issue is visible in the video, the rest is hard to see: A few wires
from the transformer are flopping about, way too long. They need to be
close to the core in an orderly fashion. When they head towards the
electronics they must be twisted or at least parallel. Else you'll have
lots on unwanted leakage inductance and ... spikes.

At that time I was running at 500 Hz and the gate drives were from an
LM324. When I tried to run it again, with the battery freshly charged,
the TVS diodes burned up and shorted. It seems that the spikes are
higher during start-up and also the higher battery voltage may have
contributed.

The transformer has 100 turns of about #18 AWG wound as the first layer
(secondary) on the toroidal core, which was then wrapped with Mylar.
Then the two primary windings were added, using 8 turns each of #10 AWG.

Can't see the secondary but try to arrange that primary as mention
above. Same for the secondary. You should see a noticeable reduction in
spikes. Then it'll be on to the other wiring, that needs some close-ups
from the transistor area.

There is not much room left through the core, so it's about maximum
fill. It is not wrapped very tightly, as you can see. I'm trying to
determine the pros and cons of using an iron core transformer at higher
frequencies for more power in the same size. So I expect to get about
1500 VA out of it, which is about 3 times its rating at 60 Hz. I was
hoping to be able to get as much as 3-4 kVA, but I think the copper
losses will be too large. Possibly a toroid with a larger hole will
allow larger wire, and at the higher frequency not as much iron will be
needed.

Copper losses, and there'll also be increasing core losses at higher
frequencies. Depends on the quality of the laminations in there.

Of course the best design might be using ferrite and 100kHz or higher.
But then there are are other design problems. And you must define "better".

Better in that case would probably mean "smaller". The efficiency won't
be that much greater considering that you are propelling a mini-tractor
with another 200lbs or so sitting on it, over uneven terrain.

Off-topic but needs to be said: Kudos for taking in Muttley, for giving
him a home. That is great. We did the same with a Shepard who is now
almost 13. She had major behavioral issues like aggressiveness towards
other dogs, it took us years to un-train that because she was more or
less a street dog before. But it worked. Tonight there will be three
two-legged guests and there will be three Labradors, and not a worry in
the sky.

 

[Fred Abse]

I'm not sure how to measure
the leakage inductance

Short the secondary and measure the primary, A perfect transformer would
look like a short circuit. Since there is no such thing as perfect, you'll
see some inductance. That is leakage inductance.


Looking at your simulation, the startup current as C1 and C3 charge up is
well over the 50 amps specified for your FETs. You need to add some
sort of soft precharge arrangement.

I'd advise that you:

Measure the leakage inductance, as above.

Measure mutual inductance, using the usual aiding-opposing technique. You
can calculate K from L1 L2, and M.

Make a more comprehensive model of the transformer.

 

[P E Schoen]

"Joerg" wrote in message

Which simulation? If it was a post many moons ago that would
have rolled off here.

It was in my post on the "High power 2kHz DC-DC - Success with gate drivers"
and other information was in an older thread (early April) about running a 3
phase motor on an SLA battery. I have now also put the ASC file on my
website:
http://www.enginuitysystems.com/pix/12V-300V_CT_Doubler.asc

and some screen shots:
http://www.enginuitysystems.com/pix/24V-320V_Schematic-s.png
http://www.enginuitysystems.com/pix/24V-320V_Simulation-s.png

... The gate drives are from a PIC16F684 PWM at 50% duty cycle
and 1 uSec dead time.

Why so much? That's a ton of dead time during which things can do
plenty of ringading. Can't remember any switcher design of mine
that had more than 100nsec.

I was just "playing it safe". But during startup (and voltage change) the
PWM will be throttled back so I have to quench the spikes anyway. I can't
short the primary as I could with a full H-bridge.

Muttley looked kind of disappointed that he didn't get a ride

He has to earn his keep. I'm also building a dog cart so he can pull me
around. I plan to use PVC pipe and have it so that it can be disassembled to
fit in a car trunk.

He had a long ride in the car over the past weekend. We went camping in a
KOA Kabin in Williamsport, MD.

One issue is visible in the video, the rest is hard to see: A few
wires from the transformer are flopping about, way too long.
They need to be close to the core in an orderly fashion. When
they head towards the electronics they must be twisted or at
least parallel. Else you'll have lots on unwanted leakage
inductance and ... spikes.

Can't see the secondary but try to arrange that primary as mention
above. Same for the secondary. You should see a noticeable reduction in
spikes. Then it'll be on to the other wiring, that needs some close-ups
from the transistor area.

I'll try to clean them up and see if there's any improvement.

Copper losses, and there'll also be increasing core losses at higher
frequencies. Depends on the quality of the laminations in there.

It seemed like 2kHz was about the best, although 4kHz was not bad. I
actually did tests awhile back up to 16kHz, and it still seemed to be OK

Better in that case would probably mean "smaller". The efficiency
won't be that much greater considering that you are propelling a
mini-tractor with another 200lbs or so sitting on it, over uneven terrain.

Size and weight are not critical for a tractor, especially when it will take
a couple hundred pounds of batteries to make it really practical. For a
"real" tractor, with a mowing deck or plow, it will take more power and more
batteries. But there are some working designs "out there". I'm documenting
my progress, and helping and learning from others, on
http://www.mytractorforum.com/index.php. I'm PStechPaul.

Off-topic but needs to be said: Kudos for taking in Muttley, for
giving him a home. That is great. We did the same with a Shepard
who is now almost 13. She had major behavioral issues like
aggressiveness towards other dogs, it took us years to un-train
that because she was more or less a street dog before. But it
worked. Tonight there will be three two-legged guests and there
will be three Labradors, and not a worry in the sky.

I'm glad I was able to take him in. He terrorized my cat, and viciously
attacked a young dog in an obedience class, and I almost had him put down at
the recommendation of the instructor. But he's been a great dog and he's
helped me maybe more than I helped him. His early history and lots of
pictures are on his website: www.muttleydog.com.

I've been active for a long time on the Cesar Millan website and some of us
have a new website: http://dogsnharmony.com/community/

We are organizing a big pack meet in Boone, NC in September, and I think we
have about 10 people and 20 dogs!

Thanks,

Paul

 

[P E Schoen]

Just a quick update. I tried adding 0.047uF capacitors to the gates, and it
seemed to make the spikes worse, especially as the voltage increased. This
might have been because the gate drive voltage was still not fully regulated
below 12V, but it did not seem like a very promising method:

10.1V 1.12A 25V peak
12.0V 1.38A 46V peak

Compare to the results without these capacitors:

10.0V 1.10A
12.0V 1.30A 32V peak
20.0V 1.95A 53V peak
25.0V 2.31A 66V peak

I added a 0.22uF capacitor across the drains:

12.0V 1.33A 33V peak
15.0V 1.59A 40V peak
20.0V 2.00A 53V peak

The amplitude of the first spike was about the same, but it just rang
longer. So I added a 12 ohm resistor in series for a snubber (in addition to
the 0.047uF and 2 ohm already in place):

12.0V 1.35A 28V peak
15.0V 1.63A 35V peak
16.5V 1.72A 38V peak
20.0V 2.06A 45V peak
25.0V 2.47A 58V peak

That seemed to be some improvement. The 12 ohm 10W resistor was pretty hot.
There is an additional 4 watts at 25V and that seemed about right.

I added a 600 ohm load to the output:

10.0V 3.72A 29V peak
12.0V 4.31A 34V peak

So, that was actually worse than the unloaded test with no snubber. But the
odd thing is that now the spike appears as the MOSFET turns on and the
voltage drops. Probably the energy stored in the snubber is being released
as the drive changes state. This is kind of like Whack-a-Mole!

Thanks,

Paul

 

[Joerg]

"Joerg" wrote in message

It was in my post on the "High power 2kHz DC-DC - Success with gate
drivers" and other information was in an older thread (early April)
about running a 3 phase motor on an SLA battery. I have now also put the
ASC file on my website:
http://www.enginuitysystems.com/pix/12V-300V_CT_Doubler.asc

and some screen shots:
http://www.enginuitysystems.com/pix/24V-320V_Schematic-s.png
http://www.enginuitysystems.com/pix/24V-320V_Simulation-s.png

It really looks like your transformer windings are too loose. Having a
laminated core at such a high frequency probably won't help either.

I was just "playing it safe". But during startup (and voltage change)
the PWM will be throttled back so I have to quench the spikes anyway. I
can't short the primary as I could with a full H-bridge.

With a "loose" transformer an H-bridge can be better.

He has to earn his keep. I'm also building a dog cart so he can pull me
around. I plan to use PVC pipe and have it so that it can be
disassembled to fit in a car trunk.

He had a long ride in the car over the past weekend. We went camping in
a KOA Kabin in Williamsport, MD.



I'll try to clean them up and see if there's any improvement.

I think tidying up the transformer and also the wiring in the primary
power loops will help a lot. The large those loops are the worse it'll
become with the spiek energy. You can snubber much of that but you'll
burn lots of energy in the snubber resistors that could have been avoided.

It seemed like 2kHz was about the best, although 4kHz was not bad. I
actually did tests awhile back up to 16kHz, and it still seemed to be OK

If you could get a hold of a nice big ferrite core that would allow

Size and weight are not critical for a tractor, especially when it will
take a couple hundred pounds of batteries to make it really practical.
For a "real" tractor, with a mowing deck or plow, it will take more
power and more batteries. But there are some working designs "out
there". I'm documenting my progress, and helping and learning from
others, on http://www.mytractorforum.com/index.php. I'm PStechPaul.


I'm glad I was able to take him in. He terrorized my cat, and viciously
attacked a young dog in an obedience class, and I almost had him put
down at the recommendation of the instructor. But he's been a great dog
and he's helped me maybe more than I helped him. His early history and
lots of pictures are on his website: www.muttleydog.com.

He is amazingly self-controlled with that steak on the plate. Wish I had
that much self-control when it comes to chocolate.

I've been active for a long time on the Cesar Millan website and some of
us have a new website: http://dogsnharmony.com/community/

We are organizing a big pack meet in Boone, NC in September, and I think
we have about 10 people and 20 dogs!

Yesterday we had fun. The dogs did their own thing, picked a toy, licked
each other, rolled on their backs, whatever they wanted to do. The
highest number we had in the house together was six, all big ones except
a young puppy that will (hopefully) be a guide for a blind person some day.

 

[P E Schoen]

"Joerg" wrote in message

It really looks like your transformer windings are too loose. Having a
laminated core at such a high frequency probably won't help either.

With a "loose" transformer an H-bridge can be better.

I think tidying up the transformer and also the wiring in the primary
power loops will help a lot. The large those loops are the worse it'll
become with the spiek energy. You can snubber much of that but
you'll burn lots of energy in the snubber resistors that could have
been avoided.

If you could get a hold of a nice big ferrite core that would allow

Yes, that 500Hz whine seemed to startle him and he kept barking for awhile.
It was pretty loud for me, too, and I have significant high frequency
hearing loss.

He is amazingly self-controlled with that steak on the plate. Wish I
had that much self-control when it comes to chocolate.

I don't know how long he would have held back if I had gone into the house
for a minute! But I did "test" him once by leaving a box of pizza on the
chair and I went in for a little while. He was still sitting there,
drooling. Of course I handsomely rewarded him with a few bites!

Yesterday we had fun. The dogs did their own thing, picked a toy,
licked each other, rolled on their backs, whatever they wanted to
do. The highest number we had in the house together was six, all
big ones except a young puppy that will (hopefully) be a guide
for a blind person some day.

It's wonderful that you can help disadvantaged people in this way. I have
found that Muttley can be dangerous to small, young dogs, for no apparent
reason, and his worst "attacks" were when I was not really paying attention,
so I don't think it was my own "fear" he was reacting to. He seems fine with
most dogs, however:

My dream is to establish an intentional community which is not only
dog-friendly, but dog-centric. I want to set it up as a full-featured
campground, where people will have their own cabins or small apartments or
camping trailers, but most activities will be shared, such as cooking,
eating, recreation, and maintenance. It can operate as a campground as well,
for revenue, and I would also like it to be a dog psychology, training, and
rescue center, as well as perhaps offer veterinary care, boarding, and
training for service dogs. Some of my ideas are here:
www.newkoinonia.com

Back to the project. I just added snubbers of 0.22uF 600V capacitors and 12
ohms, from each drain to GND, and it seems to work beautifully:

With the large filter capacitors out of the circuit:
10.0V 1.01A 22.9V peak
12.0V 1.19A 27.7V peak
15.0V 1.42A 34.6V peak
16.5V 1.53A 38.1V peak
20.0V 1.77A 46.5V peak
25.0V 2.10A 58.8V peak

With the capacitors in parallel (6600 uF at 400V total), and 12.5k bleeder
resistance:
10.0V 1.13A 23.1V peak
12.0V 1.32A 26.9V peak
15.0V 1.60A 33.9V peak
16.5V 1.71A 36.6V peak
20.0V 1.99A 44.7V peak
25.0V 2.38A 55.5V peak

Adding the 600 ohm load:
10.0V 3.62A 24.5V peak 112.7V 21.2W/36.2W=58%
12.0V 4.30A 29.5V peak 135.7V 30.7W/51.6W=59%
14.0V 5.01A 33.1V peak 159.7V 42.5W/70.1W=61%
15.0V 5.38A 34.9V peak 170.9V 48.7W/80.7W=60%

The snubber resistors got barely warm, so I figure only 1-2 watts each. I
can live with that, and my MOSFETs are happy and cool!

Paul and Muttley
www.pstech-inc.com

 

[Joerg]

"Joerg" wrote in message




Yes, that 500Hz whine seemed to startle him and he kept barking for
awhile. It was pretty loud for me, too, and I have significant high
frequency hearing loss.

Actually I have to design something where I need to make sure that
service animals won't be bothered by such noise, so I'll have to scope
that out later today. To see if 30kHz is high enough.

I don't know how long he would have held back if I had gone into the
house for a minute! But I did "test" him once by leaving a box of pizza
on the chair and I went in for a little while. He was still sitting
there, drooling. Of course I handsomely rewarded him with a few bites!


It's wonderful that you can help disadvantaged people in this way. I
have found that Muttley can be dangerous to small, young dogs, for no
apparent reason, and his worst "attacks" were when I was not really
paying attention, so I don't think it was my own "fear" he was reacting
to. He seems fine with most dogs, however:

Dogs are mostly quite peaceful if they are all off leash. Probably like
armies, while in uniform they shoot at each other. If they'd all be in
T-Shirts and shorts they'd all just have a beer together.

So, that toilet bowl in the background, is that a "canine drinking
fountain"? Cats always accuse dogs of drinking out of toilet bowls

My dream is to establish an intentional community which is not only
dog-friendly, but dog-centric. I want to set it up as a full-featured
campground, where people will have their own cabins or small apartments
or camping trailers, but most activities will be shared, such as
cooking, eating, recreation, and maintenance. It can operate as a
campground as well, for revenue, and I would also like it to be a dog
psychology, training, and rescue center, as well as perhaps offer
veterinary care, boarding, and training for service dogs. Some of my
ideas are here:
www.newkoinonia.com

Yes, there is a lack of places to down-scale when one gets older. In our
area there are some older "micro homes" in some villages. I could
imagine living in one some day, but preferably not in some sort of condo.

Back to the project. I just added snubbers of 0.22uF 600V capacitors and
12 ohms, from each drain to GND, and it seems to work beautifully:

With the large filter capacitors out of the circuit:
10.0V 1.01A 22.9V peak
12.0V 1.19A 27.7V peak
15.0V 1.42A 34.6V peak
16.5V 1.53A 38.1V peak
20.0V 1.77A 46.5V peak
25.0V 2.10A 58.8V peak

With the capacitors in parallel (6600 uF at 400V total), and 12.5k
bleeder resistance:
10.0V 1.13A 23.1V peak
12.0V 1.32A 26.9V peak
15.0V 1.60A 33.9V peak
16.5V 1.71A 36.6V peak
20.0V 1.99A 44.7V peak
25.0V 2.38A 55.5V peak

Adding the 600 ohm load:
10.0V 3.62A 24.5V peak 112.7V 21.2W/36.2W=58%
12.0V 4.30A 29.5V peak 135.7V 30.7W/51.6W=59%
14.0V 5.01A 33.1V peak 159.7V 42.5W/70.1W=61%
15.0V 5.38A 34.9V peak 170.9V 48.7W/80.7W=60%

The snubber resistors got barely warm, so I figure only 1-2 watts each.
I can live with that, and my MOSFETs are happy and cool!

Those are fairly reasonable values. You can get it better if you clean
up that transformer

 

[Chieftain of the Carpet Crawlers]

On Tue, 12 Jun 2012 04:39:37 -0700 (PDT),

TEXT 216 -24 Left 2 ;Primary 2x8 turns 2V/turn at 600 Hz

2 volts per turn ain't too hot. err IS too hot, actually.

 

[P E Schoen]

"Chieftain of the Carpet Crawlers" wrote in message

2 volts per turn ain't too hot. err IS too hot, actually.

That was at 600 Hz. I'm now running at 2 kHz so it's more like 6 volts/turn,
which is about right for a 24 V supply. And my readings show that it is not
saturating.

I was able to get the contraption "running" on Monday, but after a "short"
ride there was a "short" in two of the MOSFETs. Here is a possibly
entertaining movie of my (mis)adventure:

Enjoy! Now I have to replace those MOSFETs and get some proper TVS diodes.
But I think I may also have to devise a better "soft start". I found that
you can't use the PIC16F684 PWM to drive a push-pull except at 50% duty
cycle. So I set up a 2.5 second start-up which used an ISR to provide a 12%
PWM at start-up. Only 256 uSec wide, but still enough to cause a large
current surge. I'll have to dig into this a bit deeper. Maybe an inductor in
series with the output capacitors? But it seems like anything big enough to
be effective (like 100mH at 5A) is way too big and expensive. So that's one
reason to go with a much higher frequency, like 50kHz or more, and use
ferrite.

But another possibility may be a current limiter on the DC input, even a
linear device. It would just have to limit at 30 amps and 24 volts for a
very short time to charge the capacitors. Let's see. 300V and 6600 uF is 297
W-Sec. So an MJ11029 (PNP darlington, 60V, 50A, 300W) would do the job. I
just happen to have a few of them. Maybe set the current limit to 60A or 80A
with two in parallel, so a short circuit will trip the breaker before they
burn up. Probably not the most efficient method, but at 15 amps each (for
normal operation below current limit), they will dissipate 18W each. Not too
bad for a 750W+ converter. But the sense resistor will also dissipate about
5 watts each, and the base drive will take some power as well, although less
than a watt.

At least I think this will protect the MOSFETs from high current
destruction, and the TVS diodes (80V) should take care of overvoltage
spikes.

Yeah!

Paul

 

[josephkk]

Just a quick update. I tried adding 0.047uF capacitors to the gates, andit
seemed to make the spikes worse, especially as the voltage increased. This
might have been because the gate drive voltage was still not fully regulated
below 12V, but it did not seem like a very promising method:

10.1V 1.12A 25V peak
12.0V 1.38A 46V peak

Compare to the results without these capacitors:

10.0V 1.10A
12.0V 1.30A 32V peak
20.0V 1.95A 53V peak
25.0V 2.31A 66V peak

I added a 0.22uF capacitor across the drains:

12.0V 1.33A 33V peak
15.0V 1.59A 40V peak
20.0V 2.00A 53V peak

The amplitude of the first spike was about the same, but it just rang
longer. So I added a 12 ohm resistor in series for a snubber (in addition to
the 0.047uF and 2 ohm already in place):

12.0V 1.35A 28V peak
15.0V 1.63A 35V peak
16.5V 1.72A 38V peak
20.0V 2.06A 45V peak
25.0V 2.47A 58V peak

That seemed to be some improvement. The 12 ohm 10W resistor was pretty hot.
There is an additional 4 watts at 25V and that seemed about right.

I added a 600 ohm load to the output:

10.0V 3.72A 29V peak
12.0V 4.31A 34V peak

So, that was actually worse than the unloaded test with no snubber. But the
odd thing is that now the spike appears as the MOSFET turns on and the
voltage drops. Probably the energy stored in the snubber is being released
as the drive changes state. This is kind of like Whack-a-Mole!

Thanks,

Paul

I fussed with the simulations a lot and the only thing that seems to keep
the turn off spikes down is zeners drain to source.

?-)

 

[Chieftain of the Carpet Crawlers]

"Chieftain of the Carpet Crawlers" wrote in message



That was at 600 Hz. I'm now running at 2 kHz so it's more like 6 volts/turn,
which is about right for a 24 V supply. And my readings show that it is not
saturating.

I was able to get the contraption "running" on Monday, but after a "short"
ride there was a "short" in two of the MOSFETs. Here is a possibly
entertaining movie of my (mis)adventure:

Enjoy! Now I have to replace those MOSFETs and get some proper TVS diodes.
But I think I may also have to devise a better "soft start". I found that
you can't use the PIC16F684 PWM to drive a push-pull except at 50% duty
cycle. So I set up a 2.5 second start-up which used an ISR to provide a 12%
PWM at start-up. Only 256 uSec wide, but still enough to cause a large
current surge. I'll have to dig into this a bit deeper. Maybe an inductor in
series with the output capacitors? But it seems like anything big enough to
be effective (like 100mH at 5A) is way too big and expensive. So that's one
reason to go with a much higher frequency, like 50kHz or more, and use
ferrite.

But another possibility may be a current limiter on the DC input, even a
linear device. It would just have to limit at 30 amps and 24 volts for a
very short time to charge the capacitors. Let's see. 300V and 6600 uF is 297
W-Sec. So an MJ11029 (PNP darlington, 60V, 50A, 300W) would do the job. I
just happen to have a few of them. Maybe set the current limit to 60A or 80A
with two in parallel, so a short circuit will trip the breaker before they
burn up. Probably not the most efficient method, but at 15 amps each (for
normal operation below current limit), they will dissipate 18W each. Not too
bad for a 750W+ converter. But the sense resistor will also dissipate about
5 watts each, and the base drive will take some power as well, although less
than a watt.

At least I think this will protect the MOSFETs from high current
destruction, and the TVS diodes (80V) should take care of overvoltage
spikes.

Yeah!

Paul

You know, a simple choke in series with the primary will smooth the
square waves greatly before entering the xformer.

I have gained efficiency this way. Transformers do not like sharp slew
rates.

 

[Chieftain of the Carpet Crawlers]

At least I think this will protect the MOSFETs from high current
destruction, and the TVS diodes (80V) should take care of overvoltage
spikes.

You can place a ferrite bead on a leg of the FETs too. Right at their
entry to the PCB. We also got results from these practices.

Ours were about a quarter inch long. Maybe slightly less.

 

[josephkk]

"Chieftain of the Carpet Crawlers" wrote in message



That was at 600 Hz. I'm now running at 2 kHz so it's more like 6 volts/turn,
which is about right for a 24 V supply. And my readings show that it is not
saturating.

I was able to get the contraption "running" on Monday, but after a "short"
ride there was a "short" in two of the MOSFETs. Here is a possibly
entertaining movie of my (mis)adventure:

Enjoy! Now I have to replace those MOSFETs and get some proper TVS diodes.
But I think I may also have to devise a better "soft start". I found that
you can't use the PIC16F684 PWM to drive a push-pull except at 50% duty
cycle. So I set up a 2.5 second start-up which used an ISR to provide a 12%
PWM at start-up. Only 256 uSec wide, but still enough to cause a large
current surge. I'll have to dig into this a bit deeper. Maybe an inductor in
series with the output capacitors? But it seems like anything big enoughto
be effective (like 100mH at 5A) is way too big and expensive. So that's one
reason to go with a much higher frequency, like 50kHz or more, and use
ferrite.

Actually the PWM output is not designed for push-pull but two pins a touch
of code and a pair of and gates can steer the PWM output to suit your
system. PIC uCs are kind of known for cycle by cycle PWM, add a couple of
lines for the alternating drive for push pull.

 

[P E Schoen]

wrote in message

Just winging it with a diode clamp and resistor bleed gets those
spikes under control with negligible efficiency reduction. The
start-up transients get close to the limits, but then again your
circuit allows the FETs to peak at 300A, the maximum obtainable
with 10V gate drive, and with considerable resulting VDS that
probably blows them out. If your current sense can turn the gates
off at a peak currrent well below the worst case level corresponding
to 10V VGS, you should be okay. It wouldn't hurt to clamp each
drain with a protective TVS at about 1.5x worst case steady state
to catch the off chance events.

Fred, thanks for the idea with the diode clamp. I had tried something like
that before, but it seems to work quite well. So I have added a linear
current limiter to the battery, set at about 80 amps, but I also changed the
output capacitors to what they actually are, 3300 uF each, with 0.016 ohms
ESR. I found that it will take about 300 mSec for the output voltage to come
up to 300V, and during that time the capacitors seem to be dissipating about
300 watts each!

Actually, it will work without those capacitors, since the square wave has
only about 1 uSec during which the transformer is not being driven. I needed
them previously when I had them connected in a doubler configuration to work
on 12 VDC. So I should be able to use something much smaller, and/or use the
capacitors in the VF drive. But I think I may need an inductor to ease the
peak current in them as well.

In some ways it's amazing that the components lasted as long as they did.
The capacitors have 0.5*300*300*6600= 297 W-Sec so at 24V that would be 12
amp-seconds, but since the original simulation showed the peak voltage being
reached at about 50 mSec it would be 240 amp-seconds. If I limit the current
to 80 amps it should reach the desired voltage in about 297/(24*80) = 154
mSec. But if the capacitors are also dissipating 300 watts each during
charge, that would be added, and there would be more like 900 watt-seconds,
so the peak might be reached at 450 mSec. Of course real components do not
always match the simulator (or vice-versa), and the more accurate the
simulation the longer it takes. I'm running it now and it seems to be
stabilizing at about 270 mSec, but it took probably 10-15 minutes. I'll see
what it looks like when it reaches 350 mSec at which point the PWM stops.

Wow, the RAW file is 1 GB! Looking at the last 20 mSec, the input power is
783W, output is 276W, the MOSFETs are 9.7W, the two series pass are 23W
each, and there is still 214W in each of the output capacitors! At this
point the input current has dropped to 33A.

"Where has all the power gone, long time passing..." Bob Dylan, PP&M

Actually, though, this is still "reactive" power, because the capacitors are
still charging. The actual ripple current is about 1 amp, so the real power
is minimal. I needed to continue the simulation until full charge was
reached. Looking at the first 20 mSec of startup, the capacitor current is
just 3.7A. The series pass current limiters are dissipating about 900W, so
that might be a problem. But they will only do so for a short time, and
should gradually drop to the 23 watts at 250 mSec or so.

At least the linear regulator reduces the transients during turn-on, since
the MOSFETs only see about 3 volts at first. It may be better to design a
switching current limiter, but I'm trying to reduce complexity and enhance
reliability, so a brute force linear circuit might be acceptable.

Thanks,

Paul

=============== Here's my modified ASCII file ==================

Version 4
SHEET 1 880 680
WIRE 0 -32 -352 -32
WIRE -752 0 -800 0
WIRE -624 0 -656 0
WIRE -352 0 -352 -32
WIRE -160 0 -272 0
WIRE -352 80 -352 64
WIRE -272 80 -272 64
WIRE -272 80 -352 80
WIRE -704 96 -704 64
WIRE -704 96 -768 96
WIRE -352 96 -352 80
WIRE -272 96 -272 80
WIRE 256 96 224 96
WIRE 256 128 256 96
WIRE 320 128 256 128
WIRE 432 128 384 128
WIRE 496 128 432 128
WIRE 544 128 496 128
WIRE 560 128 544 128
WIRE 592 128 560 128
WIRE -960 144 -976 144
WIRE -944 144 -960 144
WIRE -832 144 -864 144
WIRE -800 144 -800 0
WIRE -800 144 -832 144
WIRE -752 144 -800 144
WIRE -624 144 -624 0
WIRE -624 144 -656 144
WIRE -160 144 -160 0
WIRE 96 144 -160 144
WIRE 128 144 96 144
WIRE 496 144 496 128
WIRE -352 176 -352 160
WIRE -320 176 -352 176
WIRE -272 176 -320 176
WIRE 224 192 224 96
WIRE 320 208 288 208
WIRE 432 208 432 128
WIRE 432 208 384 208
WIRE 496 208 464 208
WIRE 592 208 592 128
WIRE -768 224 -768 96
WIRE -704 224 -704 208
WIRE -704 224 -768 224
WIRE -832 240 -832 144
WIRE -704 240 -704 224
WIRE -704 240 -736 240
WIRE -624 240 -624 144
WIRE -624 240 -640 240
WIRE -592 240 -624 240
WIRE -512 240 -592 240
WIRE -464 240 -512 240
WIRE -448 240 -464 240
WIRE -320 240 -320 176
WIRE -320 240 -448 240
WIRE 128 240 128 224
WIRE 128 240 -320 240
WIRE 128 256 128 240
WIRE 544 256 544 128
WIRE 544 256 496 256
WIRE 288 272 288 208
WIRE 288 272 224 272
WIRE 320 272 288 272
WIRE 416 272 384 272
WIRE -320 288 -384 288
WIRE -80 288 -256 288
WIRE 496 288 496 256
WIRE -688 304 -784 304
WIRE -512 304 -512 240
WIRE -976 336 -976 144
WIRE -688 336 -688 304
WIRE -688 336 -848 336
WIRE -384 336 -384 288
WIRE -256 336 -256 288
WIRE 0 336 0 -32
WIRE 112 336 0 336
WIRE 128 336 112 336
WIRE -592 352 -592 240
WIRE -464 368 -464 240
WIRE 256 368 256 128
WIRE 320 368 256 368
WIRE 416 368 416 272
WIRE 416 368 384 368
WIRE 464 368 464 208
WIRE 464 368 416 368
WIRE 496 368 496 352
WIRE 496 368 464 368
WIRE 592 368 592 288
WIRE 592 368 496 368
WIRE -160 384 -160 144
WIRE 0 384 0 336
WIRE 592 384 592 368
WIRE -976 416 -976 336
WIRE -688 416 -688 336
WIRE -80 416 -80 288
WIRE -320 464 -320 288
WIRE -240 464 -320 464
WIRE -208 464 -240 464
WIRE -80 464 -80 416
WIRE -48 464 -80 464
WIRE -976 528 -976 496
WIRE -688 528 -688 496
WIRE -688 528 -976 528
WIRE -592 528 -592 416
WIRE -592 528 -688 528
WIRE -544 528 -592 528
WIRE -512 528 -512 368
WIRE -512 528 -544 528
WIRE -464 528 -464 432
WIRE -464 528 -512 528
WIRE -384 528 -384 416
WIRE -384 528 -464 528
WIRE -320 528 -384 528
WIRE -256 528 -256 416
WIRE -256 528 -320 528
WIRE -240 528 -256 528
WIRE -80 528 -240 528
WIRE 160 528 160 512
WIRE -240 560 -240 528
WIRE -160 560 -160 480
WIRE -96 560 -160 560
WIRE 0 560 0 480
WIRE 0 560 -96 560
WIRE -544 608 -544 528
FLAG -544 608 0
FLAG 592 384 0
FLAG 560 128 Vout
FLAG -448 240 in
FLAG 96 144 m1
FLAG 112 336 m2
FLAG -960 144 batt
FLAG -240 464 g1
FLAG -80 416 g2
FLAG -96 560 src
SYMBOL ind2 112 128 R0
SYMATTR InstName L1
SYMATTR Value 180µ
SYMATTR Type ind
SYMATTR SpiceLine Rser=100u
SYMBOL ind2 112 240 R0
WINDOW 0 45 35 Left 2
WINDOW 3 41 61 Left 2
SYMATTR InstName L2
SYMATTR Value 180µ
SYMATTR Type ind
SYMATTR SpiceLine Rser=100u
SYMBOL ind2 240 176 M0
WINDOW 0 21 -5 Left 2
WINDOW 3 -9 113 Left 2
SYMATTR InstName L3
SYMATTR Value 32m
SYMATTR Type ind
SYMATTR SpiceLine Rser=10m
SYMBOL nmos -208 384 R0
SYMATTR InstName M1
SYMATTR Value IRFZ44N
SYMBOL nmos -48 384 R0
SYMATTR InstName M2
SYMATTR Value IRFZ44N
SYMBOL voltage -976 400 R0
WINDOW 123 0 0 Left 2
WINDOW 39 24 132 Left 2
SYMATTR SpiceLine Rser=8m
SYMATTR InstName V1
SYMATTR Value 24
SYMBOL diode 384 288 M270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName D2
SYMATTR Value MUR460
SYMBOL diode 320 224 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName D3
SYMATTR Value MUR460
SYMBOL polcap 480 288 R0
WINDOW 3 24 64 Left 2
SYMATTR Value 3300µ
SYMATTR InstName C1
SYMATTR Description Capacitor
SYMATTR Type cap
SYMATTR SpiceLine V=400 Irms=30 Rser=0.016 Lser=0
SYMBOL res 576 192 R0
SYMATTR InstName R1
SYMATTR Value 300
SYMBOL voltage -384 320 R0
WINDOW 123 0 0 Left 2
WINDOW 39 -43 57 Left 2
WINDOW 3 228 271 Left 2
SYMATTR SpiceLine Rser=100
SYMATTR Value PULSE(0 10 0.5u 10n 10n 499u 1000u 350)
SYMATTR InstName V2
SYMBOL voltage -256 320 R0
WINDOW 123 0 0 Left 2
WINDOW 39 -43 57 Left 2
WINDOW 3 100 298 Left 2
SYMATTR SpiceLine Rser=100
SYMATTR Value PULSE(0 10 500.5u 10n 10n 499u 1000u 350)
SYMATTR InstName V3
SYMBOL diode 320 144 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName D1
SYMATTR Value MUR460
SYMBOL diode 384 384 M270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName D4
SYMATTR Value MUR460
SYMBOL polcap -480 368 R0
WINDOW 3 24 64 Left 2
SYMATTR Value 2200µ
SYMATTR InstName C2
SYMATTR Description Capacitor
SYMATTR Type cap
SYMATTR SpiceLine V=25 Irms=20 Rser=100m Lser=0
SYMBOL polcap 480 144 R0
WINDOW 3 24 64 Left 2
SYMATTR Value 3300µ
SYMATTR InstName C3
SYMATTR Description Capacitor
SYMATTR Type cap
SYMATTR SpiceLine V=400 Irms=30 Rser=0.016 Lser=0
SYMBOL cap -528 304 R0
SYMATTR InstName C5
SYMATTR Value .47µ
SYMATTR SpiceLine V=250 Rser=100u
SYMBOL schottky -576 416 R180
WINDOW 0 24 64 Left 2
WINDOW 3 -17 -52 VRight 2
SYMATTR InstName D5
SYMATTR Value MBRB2545CT
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL res -144 544 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R3
SYMATTR Value .001
SYMBOL cap -336 464 R0
SYMATTR InstName C6
SYMATTR Value 0.1µ
SYMATTR SpiceLine V=250 Rser=100u
SYMBOL cap -96 464 R0
SYMATTR InstName C7
SYMATTR Value 0.1µ
SYMATTR SpiceLine V=250 Rser=100u
SYMBOL diode -288 0 R0
SYMATTR InstName D6
SYMATTR Value MUR460
SYMBOL diode -336 0 M0
SYMATTR InstName D7
SYMATTR Value MUR460
SYMBOL cap -368 96 R0
SYMATTR InstName C10
SYMATTR Value 2.7µ
SYMBOL res -288 80 R0
SYMATTR InstName R6
SYMATTR Value 1K
SYMBOL res -848 128 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R2
SYMATTR Value .003
SYMBOL diode -976 352 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName D8
SYMATTR Value 1N4148
SYMBOL res -704 400 R0
SYMATTR InstName R4
SYMATTR Value 220
SYMBOL pnp -736 304 M270
WINDOW 0 51 63 VLeft 2
WINDOW 3 78 85 VLeft 2
SYMATTR InstName Q2
SYMATTR Value 2N2907
SYMBOL diode -912 352 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 -21 32 VBottom 2
SYMATTR InstName D9
SYMATTR Value 1N4148
SYMBOL npn -752 208 R270
SYMATTR InstName Q1
SYMATTR Value ZTX849
SYMBOL npn -752 64 R270
SYMATTR InstName Q3
SYMATTR Value ZTX849
TEXT 32 88 Left 2 !K1 L1 L2 L3 0.998
TEXT -536 576 Left 2 !.tran 0 400m 0 2u startup
TEXT 216 -24 Left 2 ;Primary 2x8 turns 2V/turn at 600 Hz