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 #152
Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384
The problem is that big reactive currents always exists, actually the highest W:VA ratio that can be gotten is 1:4 (assuming linear V-A characteristic). The switching transistors have to switch this reactive current and they get hot.
The idea is not only to make the transistors switch at zero current (this is actually not the biggest problem), but mainly to take the reactive currents away from them. This is usually realised by a parallel resonant LC tank which is only "poked" at the resonant frequency by the switching transistors, which ideally switch only the real power.
The only circuit I am aware of that can do this, is the Mazilli inverter. However there are parasitic oscillation problems with it which I haven't been able to solve so far.
Anyone aware of an inverter topology which is able to take the reactive current away from the switches?
Or advise a different place where can I ask about this, as I'm not sure if this isn't too much of a specialised question for these forums
Registered Member #30
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706
The Mazzilli is just a current-fed inverter. There are lots of other types of current-fed inverters, and all of them are short-circuit proof. Magna-Power use the technique to build big industrial SMPS, and have a technical paper here:
The SLR converter is a kind of compromise between voltage-fed and current-fed. It has a DC bus capacitor, but it's short-circuit proof.
(The problem is that it does not take the reactive current away from the switches, the idea is to have it oscillate in a LC tank (like in the mazilli inverter). In the SLR, the transistors do switch ZCS , however they conduct all of the reactive current. This is what I want to avoid.) Nevermind, I was talking about the SLR converter, not LCLR
The current fed inverters are nice but 1) they require additional switching stage for the current supply, and 2) transistors have to switch all of the reactive current - the same problem all over again.
Hmm, maybe I didn't express me clearly, what I want is that in case of output short circuit, the current in the switches should be very low (as in the mazilli inverter), not at maximum.
Registered Member #1232
Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881
The 3rd order LCLR circuit does keep the circulating VARs out of the semiconductors - that is why it is commonly employed for induction heating (load PF is generally poor.) Inverter current also reduces when the parallel resonant tank is short-circuited, although ZCS cannot be maintained under this fault condition! (...current tends towards low level triangle waveform with 90 degrees phase-lag.)
Also check out the parallel resonant converter for a more detailed (and mathematical!) description of the LLC circuit for DC to DC conversion:
Registered Member #89
Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145
Hi Jan,
I asked similar questions several times now only to get ignored in pretty much all cases.
actually the highest W:VA ratio that can be gotten is 1:4 (assuming linear V-A characteristic).
What do you mean by this 1:4? You can have power factor as low as you want, just what would it be if you removed the transformer and shorted the inverter with a piece of wire?
Or you meant the maximum that you could achieve with your devices?
The idea is not only to make the transistors switch at zero current (this is actually not the biggest problem), but mainly to take the reactive currents away from them.
Absolutely agree - I've been trying to point it out many times but did not receive much attention.
Hard switching would not be big problem alone if you didn't have to switch 10 times your real output current at the moment of it's maximum! It is very easy to pass over your device SOA even though input DC current into the inverter looks low.
I'm pretty sure jan knows all this well but I'd like if others could review my thinking:
An arc is an unstable, negative-resistance load which is close to short on inverter output, and if we use a voltage fed inverter it's almost only the transformer's leakage inductance what is keeping the magnetizing current within SOA. So the inverter may actually blow up if we make too good work designing the transformer that does not store enough energy in it's leakage inductance!
Freewheeling diodes are taking all the reactive current and that is why we are telling people building 'bridge flyback drivers' to have them rated same as switching devices are!
Putting a rectifier and a capacitor (like a TC tank cap) directly across the secondary, is even worse thing to do, since discharged capacitor is indeed like a pretty good short. A good transformer stores very little energy, doing nothing but transforming the inverter's output impedance.
It's then the matter of basic physics - when a charged capacitor is connected to other empty one, in order for conservation of charge to be obeyed, by each charging of the output cap as much energy as it received will have to be turned into heat in the process!
The forward converter tricks this by placing a large inductive reactance in series with rectifier output, which stores this energy and releases it during off time through freewheeling diodes.
Since inductor is after the rectifier no energy can be returned to the primary side of the transformer and thus inverter sees nearly resistive load. Small diodes across the devices are required only to freewheel the small primary magnetizing current.
****
With high voltage transformers, the output buck inductor would need to have very high inductance and stand high voltages which would make it very impractical.
The only circuit I am aware of that can do this, is the Mazilli inverter. However there are parasitic oscillation problems with it which I haven't been able to solve so far.
I can think of 3 methods that may possibly be used:
1. hard-switched current fed inverter 2. parallel resonant current fed inverter 3. LCLR 4. series resonant inverter
1. It is possible to build a hard-switched current fed inverter which would have no resonant circuits at all, instead of it it simply moved the buck inductor to primary side where it is much more practical to design.
I don't know much about this topology, most information I have is from here:
It appears to need a buck converter to simulate the current source, and must be a full bridge.
I don't know whether it is possible to build such an inverter as a halfbridge or/and without buck converter. It is apparently possible for resonant current-fed inverter.
2. Parallel resonant current-fed inverter can be a bridge, not necessarily push pull like a royer converter. Switch peak voltage should be halved in that case which should give a good efficiency boost over a push-pull.
I understand this topology poorly too, actually only information about it I ever got was from st app-note about CFL drivers. DC link choke has two windings wound for high coupling on a same powdered iron core, with one winding being between +rail and high side switch, while other between low side switch and GND. Used this way the inductor does exactly the same thing as in royer oscillator.
Bridge output drives a parallel LC and voltage feedback is taken from the L to excite the transistors.
3. LCLR: I don't know whether it is usable for driving ferrite transformers, but it is definitely capable of producing lots of reactive power.
tank current goes up with no load, so I fear I would not be able to avoid saturating the transformer. Need to gather more information as I'm new to topology.
*EDIT* several replies have appeared before I completed mine, richie came up with links and topologies I see for first time in my life.
So, my question is, what prevents LCLR from saturating the ferriite transformer? How much can tank voltage ever rise in no load condition? Will I be alright if I simply use the primary which can take the V/turn it gets from inverter output voltage, or more?
4. series resonant: without load and no current control series resonant inverter would self destruct.
Steve Conner used DRSSTC current limiter in his series resonant induction heater to limit the resonant rise when no workpiece is present - I believe something similar could be done for arc-drawing inverters.
If I wanted steady current control I'd try to use some sort of hysteresis controller: turn off when the current reaches the max limit, and turn back on when it decays under the min limit.
I really think there is enough information here now to get the thread going, and hopefully not get ignored like lots of my posts do.
Registered Member #152
Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384
Marko wrote ...
What do you mean by this 1:4? You can have power factor as low as you want, just what would it be if you removed the transformer and shorted the inverter with a piece of wire?
If you shorted it by a wire it would be 1:<insert some really fat number here>. Assume a power supply with constant impedance (such as inductive reactance). The most power this supply can put out happens at 1/2 maximum output voltage and 1/2 maximum output current, this is where my 1:4 came from. By the way this is also the point where an electric arc becomes unstable and goes out.
[EDIT: what about controlling the frequency of the inverter based on load? With short circuit you would bring the frequency really high so only little reactive current would flow, and you would decrease it as the output voltage rises. Surely this would not get rid of all the reactive current but it would decrease it by a good deal.
I think I will mention this video: It seems like some type of self-tuning inverter but the guy doesn't want to tell anything. ]
Actually I have tried the LCLR to drive a flyback, but the problems were: 1) the matching inductor would sometimes resonate with the tank cap, 2) if you get too big matching inductor, output power is poor, if you get it too small- (1) happens much harder and more often. I had a crappy scope and didn't understand some resonances, but I'm afraid the "too small" and "too large" regions for the matching inductor actually overlap on a quite big range
Registered Member #152
Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384
Here is a full-bridge extension of the Mazilli parallel resonant current-fed converter. It does not require additional chopper/current source transistor.
However I have no idea where I would get the feedback for the inverter from. The problems are 1) you need to switch at zero voltage not zero current so CT is not an option, and 2) you need to ensure some cross-conduction so the supply voltage does not skyrocket during dead time.
Registered Member #89
Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145
If you shorted it by a wire it would be 1:<insert some really fat number here>. Assume a power supply with constant impedance (such as inductive reactance). The most power this supply can put out happens at 1/2 maximum output voltage and 1/2 maximum output current, this is where my 1:4 came from. By the way this is also the point where an electric arc becomes unstable and goes out.
Sorry but I don't understand what you mean by this...
When load is matched to the inverter's output impedance (which is inductive in this case) your real power will be sqrt2 apparent inverter power.
The inverter will suffer the most when it's output is shorted, because in tihs case it will be drivng the lowest impedance.
Any inverter I would want to draw arcs from would have to stand that condition.
Here is a full-bridge extension of the Mazilli parallel resonant current-fed converter. It does not require additional chopper/current source transistor.
Jan, have you looked at the ST app note I posted? The second circuit in it (figure 5)
I do not think this circuit you drawn would work since current fed converter *needs* shoot through to work, because link choke current must never be interrupted.
In your schematic transistors would short out the tank cap if they go on simultaneously.
You need to use the double-winding choke as it is shown in the app note on both halfbridges.
1) you need to switch at zero voltage not zero current so CT is not an option, and 2) you need to ensure some cross-conduction so the supply voltage does not skyrocket during dead time.
I would take the feedback by a voltage transformer across the tank - a ferrite toroid wound with couple of tens of turns of primary and secondary winding, divided, clamped and fed into 74HC14...
You would then run into a problem of not being able to transmit shoot-through over GDT - you would need to use some sorts of isolated gate drivers with electronics which would generate required delays in site.
After summing up all that this gets very very complex for the job of just driving a flyback -
Royer isn't used for no reason, but because it's far simpler for closely the same job.
Also, I don't understand problems of the royer circuit well enough to be able to say at all whether a bridge topology would actually solve them!
If you are only into driving ferrite transformers, 50..100kHz, it may all be a bit of overkill. Do you really need perfect ZVS and ZCS so badly that you want to go through all of that complexity, to drive a flyback?
I'd probably try to research LCLR or series resonance more.
Now I'm starting to think that your series resonant inverter idea isn't bad at all as long as you find a way to control the current.
I was wrong in my last post about the current control - you would not be well with use of CT's to keep the current constant because you would saturate the transformer without load.
Proper way would be to control tank voltage according to max allowed primary V/turn of your transformer. Having done that you could run even without load, but you would have to take care that your surge impedance is high enough not to saturate in first drive cycle before the control circuit has a chance to shut off.
Actually I have tried the LCLR to drive a flyback, but the problems were: 1) the matching inductor would sometimes resonate with the tank cap, 2) if you get too big matching inductor, output power is poor, if you get it too small- (1) happens much harder and more often. I had a crappy scope and didn't understand some resonances, but I'm afraid the "too small" and "too large" regions for the matching inductor actually overlap on a quite big range
Did you use feedback or a fixed oscillator? IH guys had success with simple direct feedback from the tank current. Matching inductor should never resonate with tank capacitor, only with DC blocking cap and tat is only if you take current feedback from inverter output. I'd expect inverter to blow up instantly in that condition...
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.
The probem of short-circuitable inverters
