If you need assistance, please send an email to forum at 4hv dot org. To ensure your email is not marked as spam, please include the phrase "4hv help" in the subject line. You can also find assistance via IRC, at irc.shadowworld.net, room #hvcomm.
Support 4hv.org!
Donate:
4hv.org is hosted on a dedicated server. Unfortunately, this server costs and we rely on the help of site members to keep 4hv.org running. Please consider donating. We will place your name on the thanks list and you'll be helping to keep 4hv.org alive and free for everyone. Members whose names appear in red bold have donated recently. Green bold denotes those who have recently donated to keep the server carbon neutral.
Special Thanks To:
Aaron Holmes
Aaron Wheeler
Adam Horden
Alan Scrimgeour
Andre
Andrew Haynes
Anonymous000
asabase
Austin Weil
barney
Barry
Bert Hickman
Bill Kukowski
Blitzorn
Brandon Paradelas
Bruce Bowling
BubeeMike
Byong Park
Cesiumsponge
Chris F.
Chris Hooper
Corey Worthington
Derek Woodroffe
Dalus
Dan Strother
Daniel Davis
Daniel Uhrenholt
datasheetarchive
Dave Billington
Dave Marshall
David F.
Dennis Rogers
drelectrix
Dr. John Gudenas
Dr. Spark
E.TexasTesla
eastvoltresearch
Eirik Taylor
Erik Dyakov
Erlend^SE
Finn Hammer
Firebug24k
GalliumMan
Gary Peterson
George Slade
GhostNull
Gordon Mcknight
Graham Armitage
Grant
GreySoul
Henry H
IamSmooth
In memory of Leo Powning
Jacob Cash
James Howells
James Pawson
Jeff Greenfield
Jeff Thomas
Jesse Frost
Jim Mitchell
jlr134
Joe Mastroianni
John Forcina
John Oberg
John Willcutt
Jon Newcomb
klugesmith
Leslie Wright
Lutz Hoffman
Mads Barnkob
Martin King
Mats Karlsson
Matt Gibson
Matthew Guidry
mbd
Michael D'Angelo
Mikkel
mileswaldron
mister_rf
Neil Foster
Nick de Smith
Nick Soroka
nicklenorp
Nik
Norman Stanley
Patrick Coleman
Paul Brodie
Paul Jordan
Paul Montgomery
Ped
Peter Krogen
Peter Terren
PhilGood
Richard Feldman
Robert Bush
Royce Bailey
Scott Fusare
Scott Newman
smiffy
Stella
Steven Busic
Steve Conner
Steve Jones
Steve Ward
Sulaiman
Thomas Coyle
Thomas A. Wallace
Thomas W
Timo
Torch
Ulf Jonsson
vasil
Vaxian
vladi mazzilli
wastehl
Weston
William Kim
William N.
William Stehl
Wesley Venis
The aforementioned have contributed financially to the continuing triumph of 4hv.org. They are deserving of my most heartfelt thanks.
Registered Member #347
Joined: Sat Mar 25 2006, 08:26AM
Location: Vancouver, Canada
Posts: 106
Update - I've gotten several PMs asking for the gerbers. I've attached all the design info in a zip file, including schematcs and PCBs in pdf, the Altium Designer Summer 08 project, and gerbers. If you want the project in some other format, don't hesitate to ask, I'll see what I can export it to.
Do note that there are a couple of errors as there usually are on a first run of boards. The gate drive voltage pot on the control board is wired wrong, so the gate drive voltage regulator setting is fixed with a resistor. On the inverter board, the terminal strips that the IGBTs connect to are about 0.4mm too narrow, so the leads need to be bent a little bit to fit. No terminal strips are available that fit the IGBTs perfectly.
Registered Member #1232
Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881
Looking at the pictures I'd say you definitely need to keep some low ESR electrolytics right at the PFC converter. Stray inductance in the MOSFET/diode/output cap loop of a PFC circuit needs to be kept extremely low to prevent voltage spikes when the MOSFET turns off. I'm suprised you didn't blow the switching transistor from over voltage by putting the electrolytics at the end of flying leads!
The Inrush-bypass diode is definitely a good idea. If you make this a beefy conventional silicon diode it will also likely protect the PFC stage in the event of a DC bus short if the inverter fails.
> This PFC should be good to about 12A mains draw from 120V to 240V now.
It will likely power limit at slightly more than the original design power because the voltage loop will run out of headroom, that is what normally happens. Even though all components can stand the voltages associated with 240V operation, and the currents associated with 110V operation, the control loop often cannot make this combination of voltage and current available at the same time without being modified.
I successfully modified an active PFC stage from a 150W PSU to deliver 1500W in a SSTC application, but I think I had to change about 22 individual things to get there!
Registered Member #347
Joined: Sat Mar 25 2006, 08:26AM
Location: Vancouver, Canada
Posts: 106
GeordieBoy wrote ...
Looking at the pictures I'd say you definitely need to keep some low ESR electrolytics right at the PFC converter. Stray inductance in the MOSFET/diode/output cap loop of a PFC circuit needs to be kept extremely low to prevent voltage spikes when the MOSFET turns off. I'm suprised you didn't blow the switching transistor from over voltage by putting the electrolytics at the end of flying leads!
I did have a 100uF cap placed where the original caps were on the PCB to handle the fast spikes. I took a look at the current through the wires between that cap and the external caps (440uF), and it seems most of the ripple is being handled by the external caps, but the high frequency components that the 100uF cap had to handle were making it get quite hot.
Here's a shot of the current through the wires between the 100uF cap on the PCB and the 440uF external cap, 4A/div
GeordieBoy wrote ...
> This PFC should be good to about 12A mains draw from 120V to 240V now.
It will likely power limit at slightly more than the original design power because the voltage loop will run out of headroom, that is what normally happens. Even though all components can stand the voltages associated with 240V operation, and the currents associated with 110V operation, the control loop often cannot make this combination of voltage and current available at the same time without being modified.
I successfully modified an active PFC stage from a 150W PSU to deliver 1500W in a SSTC application, but I think I had to change about 22 individual things to get there!
I took a look at the control loop, and there seems to be lots of headroom in this design, I calculate about a 12A max draw at 240V based on what the voltage loop can demand.
Registered Member #1232
Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881
For PFC design I normally work on the basis of 1uF per watt of power when the DC bus voltage is 400V. This capacitance can be split equally between the PFC output and the inverter input with a common-mode choke in between. For 12A draw off 240VAC that works out at about 1400uF at the PFC output, and another 1400uF across the DC bus at the inverter. A single 100uF electrolytic right at the PFC is almost certainly unable to support the ripple if you've loaded it up to nearly 3kW.
Registered Member #347
Joined: Sat Mar 25 2006, 08:26AM
Location: Vancouver, Canada
Posts: 106
Well, I've had my first IGBT failure, while trying the coil out at higher bus voltages. I had no problems running up to about 350V, the problem occurred the first firing at 395V. I was running about a 75-100uS burst with a peak current of maybe 600-700A. I'm not sure if the IGBTs started the failure, one of the bus caps had its leads vaporized and was sent flying.
There appears to be arcing between bus + (lower pad on capacitor) and the half bridge output (top layer):
I'm not quite sure if the IGBTs failed first, or the cap. Only two IGBTs failed, both on the negative leg of the H bridge, the same side as the capacitor failure. The remaining bus caps recharged after the loud bang, no fuses blew, and no IGBT cases cracked. I was also [luckily not] shocked that the failed cap still had voltage on it after cutting it open! I don't see any damage inside the cap, it appears only the leads were blown off.
I've got two theories how the failure occurred.
1: High voltage and high dv/dt on the bridge output caused tracking and an arc between the bridge output and bus +. The arc continued and vaporized the leads on one of the capacitors, and the high current through the IGBTs caused the low side ones to fail.
2: A weakened lead on one capacitor breaks under high current, causing an arc. This arc continues and makes its way between bus + and bridge output, and the high current causes the low side IGBTs to fail short circuit.
In either case, once the IGBT's gates have short circuited to ground, the gate driver shouldn't be able to drive the gates high on opposite leg's IGBTs, so shoot through doesn't cause a massive failure of all the IGBTs in that side of the H bridge. If this is the case, an initial IGBT failure shouldn't be able to cause massive shoot through current on the bus caps, so I can't see that causing a cap to blow up.
Any other ideas/insight?
GeordieBoy wrote ...
For PFC design I normally work on the basis of 1uF per watt of power when the DC bus voltage is 400V. This capacitance can be split equally between the PFC output and the inverter input with a common-mode choke in between. For 12A draw off 240VAC that works out at about 1400uF at the PFC output, and another 1400uF across the DC bus at the inverter. A single 100uF electrolytic right at the PFC is almost certainly unable to support the ripple if you've loaded it up to nearly 3kW.
-Richie,
These chargers run much smaller bus caps, about 0.4uF/W. I think that's because there's no hold up time requirement, the only limits are ripple rating and heating. I've got the original 440uF worth of caps back in the PFC stage now.
Registered Member #1232
Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881
The voltage and dv/dt produced at the output of a 400V IGBT bridge is not sufficient to initiate breakdown of FR4 laminate. Flyback SMPSUs with reflected voltages in the high hundreds of volts typically only use a few millimetres of creapage distance over similar material without any problems.
If there are traces of carbon tracking then it is probably a common-mode noise problem between the DC bus and mains earth. Do you have common-mode capacitors from DC bus - to earth and DC bus + to earth?
My theory for the dramatic failure would be this:
One of the 400V electrolytics developed an internal short, (395V is too close to the 400V rated voltage, especially when you consider that the capacitors are likely being thrashed in terms of their RMS current ripple rating too. Are they new devices or pulls from old equipment?) The remaining bank of paralleled capacitors discharged into the shorted component fusing its connecting leads and sending it flying across the room.
Were you running the coil from the active PFC pre-converter when the failure occured? The output voltage of an active PFC pre-converters is only semi-regulated, and the voltage loop has a sluggish response time. When an active PFC stage is presented with a pulsed load on the DC bus it wouldn't be uncommon for the bus voltage to sag and then overshoot considerably after each "bang" of the DRSSTC. The bus caps are usually rated at 450VDC for a 380VDC boost voltage and the over-voltage trip should be set around 425.
Registered Member #347
Joined: Sat Mar 25 2006, 08:26AM
Location: Vancouver, Canada
Posts: 106
GeordieBoy wrote ...
If there are traces of carbon tracking then it is probably a common-mode noise problem between the DC bus and mains earth. Do you have common-mode capacitors from DC bus - to earth and DC bus + to earth?
I do have a 22nF cap from bus - to ground. Early on I had the ground fall off and cause some arcing between the heatsink and somewhere under the board, as the secondary base current found the easiest path to ground through the mains. I added the cap at that point, and didn't have any problem until now.
GeordieBoy wrote ...
My theory for the dramatic failure would be this:
One of the 400V electrolytics developed an internal short, (395V is too close to the 400V rated voltage, especially when you consider that the capacitors are likely being thrashed in terms of their RMS current ripple rating too. Are they new devices or pulls from old equipment?) The remaining bank of paralleled capacitors discharged into the shorted component fusing its connecting leads and sending it flying across the room.
These are new caps. I thought about an internal short too, but discounted it when I found that the failed cap still held a charge. Could a short circuit blow itself clear and let the cap still hold a charge? I'll unroll the failed cap later and see if there's evidence of an internal short.
GeordieBoy wrote ...
Were you running the coil from the active PFC pre-converter when the failure occured? The output voltage of an active PFC pre-converters is only semi-regulated, and the voltage loop has a sluggish response time. When an active PFC stage is presented with a pulsed load on the DC bus it wouldn't be uncommon for the bus voltage to sag and then overshoot considerably after each "bang" of the DRSSTC. The bus caps are usually rated at 450VDC for a 380VDC boost voltage and the over-voltage trip should be set around 425.
It was running off a variac at the time.
This battery charger has 400V rated bus caps, so they're pushing the ratings too.
Registered Member #347
Joined: Sat Mar 25 2006, 08:26AM
Location: Vancouver, Canada
Posts: 106
I think I may have a lead on the cause of the failure. I notice that at high primary currents, the hi to lo transition of H bridge output '1' is being delayed significantly This delay doesn't occur when the primary current is low. The delay is shown in the scope shots below:
Setup: 200VDC bus, no secondary, 90uS burst length Channels: 1: H bridge side 1 output WRT bus (-) 2: H bridge side 1 low side IGBT gate vs. ground WRT bus (-) 3: H bridge side 2 output vs. ground WRT bus (-) 4: Primary current, 200A/div
Overview of burst, ~650A peak:
Side 1 hi to low transition, delayed 150ns from side 2. The slew rate of the transition also looks slower on side 1 compared to side 2. Also, the miller plateau lasts for less than 100nS, while the plateau on the other side (not shown) lasts about 200nS:
Side 1 lo to hi transition. Much better matched with side 2:
I've checked that every gate resistor and diode is fine, and all the gate drive transformer outputs have the same drive capability of 4A peak driving a short circuit. This delay doesn't appear on the other side of the H bridge, which is for the most part a mirror image of the problem side. I'm doubtful the problem is due to layout given the other side is fine.
The only thing I can think of that would cause this delay is a damaged TVS. I'm using bi-directional TVSs to stop the TVS diode from taking any current that should be handled by the IGBT anti-parallel diodes. If one of the two anti-series diodes in the TVS went short circuit, then perhaps the freewheeling current would be handled by the slow TVS, and the slow reverse recovery is holding the side 1 voltage high.
Does this theory sound reasonable, or am I completely barking up the wrong tree?
Banned on 3/17/2009. Registered Member #487
Joined: Sun Jul 09 2006, 01:22AM
Location:
Posts: 617
I think it sounds reasonable to me. Ive actually damaged an IGBT's gate that caused it to act weird very much like what you're describing. It wouldn't blow the 5 amp fuse and the coil still worked but just not very well.
This site is powered by e107, which is released under the GNU GPL License. All work on this site, except where otherwise noted, is licensed under a Creative Commons Attribution-ShareAlike 2.5 License. By submitting any information to this site, you agree that anything submitted will be so licensed. Please read our Disclaimer and Policies page for information on your rights and responsibilities regarding this site.
"MMI 1" - New DRSSTC
