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
I have set up an FM singing arc like this: audio in > op-amp (gain ~20) > 4046 VCO 25-100kHz > gate drivers > GDT > halfbridge > rectified TV transformer. Everything checked OK (gate waveforms etc.) so I turned it on and got an arc. I started increasing the volume when suddenly the bridge popped. I have ABSOLUTELY NO idea how that could have happened. The VCO has ultimately limited its frequency above 25kHz so the bridge cannot pop because of overcurrent. The supply voltage for the halfbridge was 330V DC and 50 turns on the transformer primary, which is more than enough even for 25kHz.
Does anyone have an idea what happened? I can't see anything wrong with the circuit, don't want to put another set of FETs in and pop them again.
Registered Member #152
Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384
Uzzors wrote ...
The rectified flyback sticks out as a problem to me. I suspect it saturated because of the single rectifying diode.
I have driven these xformers without problems from a bridge. Like I said, I turned it on and got an arc, turned it off after a while to check the FETs and they were cold. So I turned it back on and started bringing the volume up from zero and then it popped.
Registered Member #162
Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141
I would probably repair the bridge and use a mains filament lamp as a light load, 'scope the output whilst increasing the audio modulation. You may see the problem on the 'scope, at least you'll have eliminated the bridge/driver as the source of the problem.
Registered Member #152
Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384
Sulaiman wrote ...
I would probably repair the bridge and use a mains filament lamp as a light load, 'scope the output whilst increasing the audio modulation. You may see the problem on the 'scope, at least you'll have eliminated the bridge/driver as the source of the problem.
I might try that, but I'm afraid I'd see just blurry trails on the scope because of the frequency modulation... (I have scoped the output of the gate drivers while inputting the audio, and you can't really see a thing on the scope).
Registered Member #1232
Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881
If you increased the FM depth too far the VCO may have actually stopped oscillating at the troughs of the audio waveform.
Keeping one MOSFET on for too long in the half-bridge is goind to cause the magnetising current on the LOPT to ramp up to dangerous levels if you haven't got a DC blocking cap in there. Did you use a DC blocking capacitor, or just tie the load between the bridge leg and the mid-point of a bank of electrolytics?
That would be my guess at what is happening. A similar thing can happen in TV horizontal deflection circuits. If you keep the HOT on for a fraction too long, the LOPT core can saturate and the transistor dies instantly from overcurrent.
-Richie,
PS. Also bare in mind that applying rapid and deep frequency modulation can upset the nice 50% duty ratio you get out of the VCO. This in turn upsets the volt-second ballance across your GDT, which can affect how hard each of the two MOSFETs are getting turned on. Instantaneous deviations of the half-bridge's duty ratio from 50% could also cause the mid-point of the capacitive divider to shift in voltage too. If the capacitive divider voltage doesn't shift quickly enough, then a net DC bias across the LOPT primary is likely to saturate it fairly quickly.
Registered Member #152
Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384
GeordieBoy wrote ...
If you increased the FM depth too far the VCO may have actually stopped oscillating at the troughs of the audio waveform.
Prior to firing the circuit up, I have tested the input of the VCO with a potentiometer ("outer" pins to supply rails, wiper to VCO input), and it seemed to work well from 0V all the way up to Vdd (frequency tested on scope).
GeordieBoy wrote ...
Keeping one MOSFET on for too long in the half-bridge is goind to cause the magnetising current on the LOPT to ramp up to dangerous levels if you haven't got a DC blocking cap in there. Did you use a DC blocking capacitor, or just tie the load between the bridge leg and the mid-point of a bank of electrolytics?
You are right, I haven't used that cap. But that was because I thought the frequency can never drop below the VCO minimum.
GeordieBoy wrote ...
PS. Also bare in mind that applying rapid and deep frequency modulation can upset the nice 50% duty ratio you get out of the VCO. This in turn upsets the volt-second ballance across your GDT, which can affect how hard each of the two MOSFETs are getting turned on. Instantaneous deviations of the half-bridge's duty ratio from 50% could also cause the mid-point of the capacitive divider to shift in voltage too. If the capacitive divider voltage doesn't shift quickly enough, then a net DC bias across the LOPT primary is likely to saturate it fairly quickly.
I thought about this too, after all it seems like the only possible explanation.
Registered Member #1232
Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881
> Thanks for the help
No worries. Problems like this can be quite frustrating to diagnose.
SMPSU designers have to be careful in things like full-bridges and push-pull converters. Any rapid modulations of the duty-ratio can cause unequal conduction times for the two pairs of switches. This causes a temporary imbalance in the volt-seconds across the transformer primary and a flux-imbalance in the core. If sufficient headroom is not allowed then transient core saturation can occur, causing excess current flow through the switches.
There is something called sub-harmonic oscillation that can damage SMPS's in this way if the control loop is badly designed!
In general problems like this can be prevented by:
1. Preventing rapid deep modulation of both the switching frequency or duty ratio. (This is essentially a control problem.) 2. Designing the transformer with some transient saturation headroom on top of that for normal steady-state operation. (This often takes care of itself if the transformer is conservatively designed: HF ferrite power transformer designs are usually constrained by core-loss not saturation.) 3. Gap the core very slightly to increase the saturation flux density. 4. Use small PP capacitors in the capacitive divider leg of the half-bridge. This means that imbalances in duty-ratio can be quickly absorbed by shifts in the mid-point voltage. (That's why the load isn't usually tied to the middle of the voltage doubler electrolytics.) 5. Use current-mode control to prevent flux-walking. Or at least peak switch-current limiting to prevent disasters.
If it were me trying to diagnose your problem, i'd pop a CT over the load wire leaving the half-bridge MOSFETs. Rectify the output of the CT, feed to a burden resistor and display the voltage on a storage scope. Run the circuit with no modulation, and turn up the scope trigger to a level where the current being measured is just too low to trigger the scope. Then gradually introduce audio modulation and see if it starts to trigger the scope. Any triggering shows that larger current pulses are flowing. You can determine how much larger, by how far you can increase the trigger threshold before the scope stops triggering.
If you have a decent scope with deep storage memory and a few channels you can scope the VCO and GDT drive too. Once you trigger on a overcurrent event, you can roll back the display to see what the gate drive signals were doing at that instant. This should reveal what is causing the problem.
If the waveforms are "jumping all over the place" with music as the modulator, then use a sig gen to produce a steady periodic modulating signal. Then the modulating frequency and the depth are controlable and repeatable.
Registered Member #152
Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384
Thanks, but no storage scope here
As far as I understand, introducing some deadtime might fix the problem, but I'm not exactly sure how to simply implement this, the circuit needs to be compact and I already have 4 ICs in it...
Another way might rebuild the circuit as a single-switch forward converter, that should fix the problem right? But the thing is that with 330V on the dc bus and a 800V FET, I can't quite use the 50% duty cycle from the VCO...
Registered Member #1232
Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881
To be honest the idea of audio modulating a flyback spark via FM sounds a little bit dodgy to me! How is it that you think the spark intensity get's modulated by deviating the switching frequency?
I may be partly responsible for the proliferation of this FM singing arc idea though. I showed that you can use FM of the switching frequency of a CW SSTC to audio modulate the discharge. But, this only works because the Tesla Resonator used as the load is a high-Q tuned circuit. The TC resonator actually converts the FM into incidental AM as the carrier frequency rides up and down one side of the bandpass response. It is these amplitude variations that result which modulate the spark temperature and produce the audio. The inverter effectively outputs full AC voltage amplitude all of the time, and relies on "load modulation" caused by the FM to vary the current drawn by the load. (It just so happens that Frequency Modulating an inverter running at full power is easier than implementing true AM via high-level modulation, and it circumvents some of the problems with implementing it via PWM.)
The reason that FM sounds strange for modulating flyback arc intensity is that this method relies on a high-Q resonant response from the load to produce this "slope detection" effect converting FM to AM.
So, how do you think FM causes audio to be emitted from a flyback transformer arc? Are you relying on some resonance of the transformer windings? Or do you have some other resonant matching network in there between the inverter and the flyback transformer?
It just seems to me that PWM might work better for a flyback transformer. Or more specifically modulating the switch's on-time and using a fixed off-time for resonant flyback. The variable on-time allowing you to precisely control the energy stored in the core at the end of each on time.
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.
Singing arc failure
