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 #146
Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055
The device capacitance got me too when i was looking at my CM300 DRSSTC bridge. I noted that my voltage overshoot was dramatic at low DC input, but fell off as i actually approached a few hundred volts across the IGBTs. At the time i wasnt sure if it was real or if it was my probes, but since my IGBTs dont seem to blow up anymore, i figured the overshoot really was dropping off. Diode recovery was still a big issue, and slowing down the switching only helped moderately.
Registered Member #205
Joined: Sat Feb 18 2006, 11:59AM
Location: Skørping, Denmark
Posts: 741
Steve Conner wrote ...
Finn,
*edit* I got this from Richie by e-mail:
richie wrote ... Hi Steve,
snip
The only thing that seems strange is that he is trying to turn on the opposing switch immediately after the first complete current cycle ends. You have all the time in the world after this point, the only thing you sacrifice is a small amount of power by introducing deadtime here. You remember the "pulse thinning out" method of average-current control we talked about?
snip.
Steve, Richie, all
The reason for turning the swiches on at exactly 1/2 the resonant frequency was simply because everybody that I had read about at that time, did so.
The idea of introducing dead time seems compelling, and at first point I fell for it, but it seems like only 1/2 the truth that I can apply any amount of dead time btwn. complete current cycles.
Today, I hooked one set of IGBT`s up to the pulse generator, to better see what happens when only one single pulse is allowed to draw current. The situation is the one where the supply is starting to charge an empty capacitor. It will look different at the end of the charging cycle.
Yellow is gate, Cyan is current into resonant load, and Magenta is voltage acros the opposing swich in the leg.
As the gate is taken positive the IGBT starts to conduct, as shown in the negative going current cycle. At the same time, full buss voltage is applied to the opposing IGBT as seen on Magenta trace. Shortly after the current has transferred to the freewheling diode, the gate goes negative, closing the IGBT. When the current passes trough zero, it slams against the closed IGBT`s, the charge is trapped, and starts to oscillate. When the oscillation dies down, a bit more than half the bus voltage is deposited on the resonant cap. I guess the rest has been burned off in heat. I am not too sure about what really happens.....
Adding dead time robs me of my ZCS, unles I throw the swich at exactly 1.7uS deadtime:
Zooming in on the previous scopeshot, shows this, yellow cursors on Cyan trace:
Since the native frequency is 47.2kHz, the 1/2 period time is 10.6uS. Add 1.7uS and the half period time is 12.3uS corresponding to a drive frequency of 40.6kHz, the frequency I found experimentally yesterday. And which looks like this:
Perhaps it is time to build that super low inductance buss that will prevent current measurements, for lack of a rogowsky based system, slap the IGBT`s on a tunnel type heat sink, coz they do get warm, and try to ramp the voltage towards the 560 volts.
I would like to hear more about the "Pulse thinning method", and adding dead time in general, since it looks like it is not totally troublefree.
Registered Member #30
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706
Hi Finn,
In the experiment you did with the single pulse, the waveforms are quite OK. The ringing at the end looks awful, but it doesn't do any harm.
The only thing to do is make the pulse a little longer so you can be sure the current has commutated to the diode before you turn the IGBT off. Richie suggested turning off halfway through the second half cycle to buy some tolerance. This entire half-cycle flows in diodes, not IGBTs, so any switching instant during it will result in both ZCS and ZVS. When you widen the pulse the other waveforms shouldn't change at all! I'm still pestering him to join 4hv :P
Registered Member #89
Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145
Hey, you guys are awesome. I never actually realized that there's nothing preventing you from turning the IGBT off while the diode is in conduction anyway. TUrning it off to early may actually cause problems.
I see the point in it. You can actually switch halfway into the freewheeling cycle and have both ZCS and ZVS as far as I see. So even changing the duty cycle or frequency doesn't matter much within limits.
The only 'odd' thing I notice is apparent overshoot of humpy current waveform over zero. (I tought it needs to just touch zero). And it seems to have no effect at all..?
Registered Member #205
Joined: Sat Feb 18 2006, 11:59AM
Location: Skørping, Denmark
Posts: 741
Steve Conner wrote ...
Hi Finn,
In the experiment you did with the single pulse, the waveforms are quite OK. The ringing at the end looks awful, but it doesn't do any harm.
Awfull indeed. I was worried about the apparent DC bias it placed on the transformer....perhaps....
Steve Conner wrote ...
The only thing to do is make the pulse a little longer so you can be sure the current has commutated to the diode before you turn the IGBT off. Richie suggested turning off halfway through the second half cycle to buy some tolerance. This entire half-cycle flows in diodes, not IGBTs, so any switching instant during it will result in both ZCS and ZVS. When you widen the pulse the other waveforms shouldn't change at all!
That`s correct. A bit longer pulse is better.
Steve Conner wrote ...
I'm still pestering him to join 4hv :P
Don`t stop untill he does.
Marko wrote ...
The only 'odd' thing I notice is apparent overshoot of humpy current waveform over zero. (I tought it needs to just touch zero). And it seems to have no effect at all..?
I call it the "buttoc" waveform
It should just touch zero, but driving below 1/2 res makes it ring a bit past zero. this is in order to let the freewheling diode recover, to avoid much worse shoot trough. I don`t know if it is good or bad, but I cannot seem to avoid it..... This becomes a problem due to the big relatively slow bricks we are using. It would probably not be visible in an array of smaller, faster, IGBT`s, also known as a MUMISSS (MUlti MIni Solid State Switch). =:-o
Cheers, Finn Hammer
I guess spirit is high when acronymuslike abbreviations come to mind.
Registered Member #205
Joined: Sat Feb 18 2006, 11:59AM
Location: Skørping, Denmark
Posts: 741
All, I prefer to think that Richies website is a textbook example in website excellence, not likely to become the target of dis-satisfaction.
Anyway, today we took the the supply totally apart, in order to manifest "Iteration no. 3". I had previously decided to settle on an operating point with one deadtime ripple, so that the turn on of the IGBT`s became ZCS, and as we already know now, the turn offs are both ZVS and ZCS, so the supply could run in idle, with 25A on both cycles, without getting even luke warm. A lesson in swiching loses. To make sure we could handle the heat associated with 200 peak amps in this iteration, we installed the best heatsink I had in stock: This should be able to handle gracefully, the 400-600W we expect to create in heat.
This process involved cutting a lot of threads. Many times have I witnessed people on these boards complain about the troubles of cutting threads, so I thought this might be the right time to share with you, how i do it.
I`l be back with an update as soon as the internal wiring has come back in.
Registered Member #146
Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055
Just a side note: letting the IGBTs conduct for longer than a quarter cycle will have impact on the primary current during open circuit conditions. Test to be sure you dont excite some other resonance when the transformer is open circuited. I have found that in order to minimize bad "open circuit" performance you must 1) minimize stray inductance in the transformer itself and 2) carefully choose the operating frequency of the inverter. Under some cases, you might find the open circuit voltage rises far too high for safety. You guys might think im crazy for adding external inductance when you could just integrate it into the transformer itself, but i swear its "different" than you would expect... well at least thats how it worked out with the 3 or 4 of these supplies ive built so far.
Registered Member #205
Joined: Sat Feb 18 2006, 11:59AM
Location: Skørping, Denmark
Posts: 741
Steve Ward wrote ...
Just a side note: letting the IGBTs conduct for longer than a quarter cycle will have impact on the primary current during open circuit conditions. Test to be sure you dont excite some other resonance when the transformer is open circuited. I have found that in order to minimize bad "open circuit" performance you must 1) minimize stray inductance in the transformer itself and 2) carefully choose the operating frequency of the inverter. Under some cases, you might find the open circuit voltage rises far too high for safety. You guys might think im crazy for adding external inductance when you could just integrate it into the transformer itself, but i swear its "different" than you would expect... well at least thats how it worked out with the 3 or 4 of these supplies ive built so far.
Steve,
Thanks for your input. We _are_ going to add external inductance, at least if we can get the leakage inductance of the transformer itself low enough to do so.... We also intend to build voltage monitoring, so that the supply stops before it reaches it`s native top voltage. We have seen effects in simulations, at the end of the charging cycle, which might cover what you describe. It`s not a pretty sight.
You write that we should choose the operating frequency of the inverter with care. We beleive we do so, and relative to the LC resonant frequency of the load, we place it lower, to insert some dead time during which the freewheling diode comes out of conduction. This has already been determined to be important, because otherwise we get shoot through. During this dead time, the energy in the resonant circuit oscillates at a higher frequency, and we let the dead time last for one full cycle of this oscillation, so that we get ZCS at turn on. However we have yet to see the supply operate at more than 1/4 -1/2 power. We did this with a jacobs ladder . But the jacobs ladder is too unstable for us to extract usefull clues. We tried to make a salt water resistor as a load, but it also became a short. I guess because we didn`t have demineralised water, but only ordinary tap water. Could you expand on which parameters you optimise for, when you choose the operating frequency?
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.
CCPS (Capacitor Charging Power Supply)
