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 #205
Joined: Sat Feb 18 2006, 11:59AM
Location: Skørping, Denmark
Posts: 741
Marko wrote ...
I simply wanted to get an idea of difference in leakage inductance with primary-under-secondary and separate style transformer.
The difference is great. With the windings so close you get almost no leakage inductance. So you would need an external choke. Why do that when you can get it for free?
Marko wrote ...
What really limits current (with ghetto SG3525-style controllers) when you draw arcs or short the transformer? There must be a lot of secondary leakage inductance doing this.
The combination of L and C sets the impedance of the primary, and this determines the current. The fact that you deny the circuit to build up resonant rise is the real current limiting factor.
Marko wrote ...
If you aren't using a (insanely complex) resonant controller like Marco larger inductor is also needed to stabilize the inductance to a point so it can be tuned by a fixed controller like SG3525. (at least from what I 'figured out').
Apparent inductance drops proportionally with load, so I imagine fixed controller to have trouble without larger inductor.
Marco Denicolai`s controller does the same thing as your SG3525 and my TL494. I`s just a question of putting out a 50% duty cycle pulse train, at half the transformers resonant frequency. There is nothing more to it.
It is very clever and very simple. That`s the beauty of it.
The leakage inductance is a fraction of the total inductance based on the geometry of the winding layout relative to the core. It doesn`t change unless the total inductance changes.
Marko wrote ...
Finn: I see you copied the transformer much from Denicolai's design. I'm lost with his resonant controller, but how do you keep the thing in tune with only leakage inductance of primary?
Marco, from what you write, I start to get the feeling that you don`t have enough windings on your primary, so that when you increase the voltage across it, you run into saturation. It may look like a power problem but it is a volt/turn problem.
From my experience, it is very easy to get this topology basically right, (apart from details like gate waveforms) and we have made prototypes on smaller cores. Just make sure there are enough windings on the core, or else, you get problems when you "turn up the wick".
Do you have some documentation of your transformer, and the rest of your drive circuitry?
Registered Member #89
Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145
The difference is great. With the windings so close you get almost no leakage inductance. So you would need an external choke. Why do that when you can get it for free?
Finn you are awesome. Thanks a lot.
I`s just a question of putting out a 50% duty cycle pulse train, at half the transformers resonant frequency.
My point was on this (could be one of FAQ's); The transformer does not have stable primary inductance. The cap will be resonating with leakage inductance alone only when secondary is shorted, and with full primary inductance if secondary is open or missing. (as if it was an inductor on ferrite core.) The inductance will wary between those two points depending on load, so I imagined it very hard/impossible to tune for ZCS without feedback.
And as far as I see, denicolai's controller keeps ZCS for the full range of charging.
(sorry, I didn't invent Lenz's rule )
Finn, how are you going to regulate the output voltage of your CCPS? Are you just going to stop the oscillator when you reach the voltage you want?
Great project BTW :D
Steve: CCPS appears to be more constant current than a voltage source, so I don't think there can be real voltage regulation. Denicolai used a HV probe to stop the charging at certain point as you said, and I guess fin will do something similar.
Apart from that I can imagine some kind of complexity which would 'skip cycles' and regulate current that way, but sorry, I'm not completely confident how this works anyway..
Marco, from what you write, I start to get the feeling that you don`t have enough windings on your primary, so that when you increase the voltage across it, you run into saturation. It may look like a power problem but it is a volt/turn problem.
It is a small hard switched design with really nothing special. I calculated the core throughly and got quite close to saturation, but stayed just under it. I could clearly observe the core going into saturation just as I lowered the frequency. From everything else it appeared just as I had a big ballast on it, which I believed to be due to winding style. I can post more into another thread if you are really interested...
Registered Member #205
Joined: Sat Feb 18 2006, 11:59AM
Location: Skørping, Denmark
Posts: 741
Steve Conner wrote ...
Finn, how are you going to regulate the output voltage of your CCPS? Are you just going to stop the oscillator when you reach the voltage you want?
Great project BTW :D
Thanks, Steve.
You are right. The oscillator is running all the time. The gate driver chips have their enable pins controlled by a 74HC74 D-flipflop, so that they are in sync with the oscillator. The FF is controlled by a CA3031 opamp configured as a comparator with adjustable hysteresis. The voltage probe is going into this comparator to stop driver chip output, when desired voltage level is reached.
Registered Member #30
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706
Marko, this power supply is designed such that the secondary is always shorted from an AC point of view. It uses a bridge rectifier with a large capacitor on the output, and no filter choke as in usual SMPSs. So the resonating inductance is assumed to be always the leakage inductance.
With non-zero output voltage, the waveforms are affected in a strange way as the bridge rectifier commutates. The effect is to lower the Q of the oscillation, as energy is extracted to the load, but the ringing frequency doesn't really change, IIRC. However, if the Q falls too low, hard switching might set in.
Incidentally, it does no harm to run slightly below half the resonant frequency, and it might even give more leeway to let the waveforms change as the output voltage ramps up.
FWIW, the current-fed converter is related to the SLR, but you can regulate it just like a normal SMPS, by PWM of a buck converter in the DC link.
Registered Member #205
Joined: Sat Feb 18 2006, 11:59AM
Location: Skørping, Denmark
Posts: 741
This has been a rewarding weekend.
When Daniel arrived, and I proudly showed him the CCPS, I wanted to show him how long it took to get hot when the variac was turned up full. I had done this several times, but right this time it went *crack*, and blew a fuse. Enter rebuild mode. Thanks god those IGBT`s were cheap.
During the tuning procedure we noticed that the transformers were sensitive like a tesla coil. We took them out and moved the one around next to the one we had sig. gen. and scope on. Found out they influence each other a lot, depending on wether they are in phase or not. These transformers resonate at 80kHz, but could be brought to resonate at from 70 o 100kHz depending on phasing and distance. This sensitivity would be an argument for using close coupling in the transformer and external choke, which in turn could be adjusted easily. Rebuilt, the same bridge blew almost instantly. Decided to use the one good bridge to test how parallel transformers would share current, and they did well.
Today we went searching in the stock and found a couple of FUJI 2MBI300N-120 With internal current limiting, not suitable for DRSSTC duty but perfect for this job. We had given up driving the transformers with IGBT`s barely able to handle the current load, and decided to parallell the primarys and use just one heavy bridge. How to drive those gates? The UCC37322 chips didn`t handle the task well. We tried to use the one good bridge, only to realise that it will not deliver a 50% duty cycle. So what to do?
I had in my mind the picture from data sheets, of the TTL output from chips, and took a closer look, well right! it is a PN pair with the gates tied together. Woila! the beefy driver!
We lashed one together, and got some gate waveforms that resembled square waves. Tried to power up the beast, and, well it didn`t get hardly hot at all, so we made this video of a jacops ladder.
We didn`t get any of the fancy voltage probe stuff done, we built 5 bridges and blew 4 but we got there: We recorded something like 60 primary amps during this shot which is below 2.4kW because we were in a 10A fuse doing it. Sorry, we were too tired to make better measurements, after 2 days of 14 hours each, full time of learning and doing.
So this has been a fun day: Using the tools from the very first SSTC work, the plasmasonic and the beefy driver, we came out to the bleeding edge with a CCPS.
At present gate drive is per beefy driver and gate transformer, we will change that to dedicated drivers before next update.
Picture shows present status as per this writing moment.
Registered Member #89
Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145
Aw, the number of dead IGBT's completely sucks finn.. I'm glad you've got it working now.
Don't you fear that it may just have been overheating what killed them? Really lot of power and small heatsinks. Bricks don't look happy neither! (How did you fix them to heatsinks smaller than their footprint after all? )
Regarding transformers interfering, maybe you could put some shielding between/around them and just leave-them-alone?
I'm just thinking how could a self-resonant controller be built for such a thing, and how beneficial would it be. (Waveform shape is weird so, some kind of zero cross detector followed by logic)
The choke of few tens of uH is not a too big problem, few turns on #26 toroid would do for lots of current. Much harder is, as I think to construct a transformer with lower leakage inductance.
Registered Member #205
Joined: Sat Feb 18 2006, 11:59AM
Location: Skørping, Denmark
Posts: 741
Marko wrote ...
Don't you fear that it may just have been overheating what killed them? Really lot of power and small heatsinks. Bricks don't look happy neither! (How did you fix them to heatsinks smaller than their footprint after all? )
Current handling ability drops with temperature, so..... Oh, we were getting mean, so the bricks got fixed with tye wraps
Marko wrote ...
Regarding transformers interfering, maybe you could put some shielding between/around them and just leave-them-alone?
It`s ok as it is now. I just thought I`d mention it. When we had 2 separate bridges and one controller, putting a Pearson near one transformer and altering it`s freq. was a disaster.
Marko wrote ...
Much harder is, as I think to construct a transformer with lower leakage inductance.
This would be a matter of putting the windings on the same leg, one on top of the other. No problem at all. With the amount of turns per volt, there is no need for more than one layer anyway, so no extra insulation problems. And there would be room for twice the amount of copper.
Registered Member #205
Joined: Sat Feb 18 2006, 11:59AM
Location: Skørping, Denmark
Posts: 741
During the week I built this gate driver:
It is so strange that the controll circuitry is getting more complex than the circuit it is controlling.. another view:
And here is a shot of it driving the gate of a CM600, without gate resistor. It is 50% duty cycle, so it is really a 97.6 kHz pulse...
The gate driver consists of a IRF530 and a IRF 9540 on top. The gates of these MosFets are driven by an Agilent HCPL-J312 Optocoupler gate driver, which I used only because I could not get the Fairchild FOD 3180. The latter is a fast optocoupler worthy of use in feedback systems, due to a super low 200nS propagation delay. Well, this does not matter here, I`m oscillator driven. The opto gets it`s timing impulses from the TL494, buffered by UCC37321 because I had to invert the signals.
There are +24 and -5 volts regulators on board. I wonder how the high side drivers are going to respond to being swung 560 volts, but we`l see. I feel comfort by the fact that there is 50% dead time.
The gates of the 530/9530 pair had to be fed by a 500 ohm resistor, and discharged trough a 1n5819 diode and a 15ohm resistor. the former to avoid the shoot trough current spikes (100nS pulses of 4A) the latter to keep discharge current lower than the gate drivers 2.5A limit.
The IGBT`s have been changed to SEMITRANS SKM 400GB124D because they are easy to drive.
It may seem overkill, to use these bricks, but I have been reading up on this topology, and this paper gave me a jolt, because it states that with the resonant capacitor size I have available, I`l only be able to transfer around 2kVA. I`l have to look into that, but In the end, maby we are into a transformer rewind.
I won`t have time to do more work this weekend, but comments are welcome.
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.