mosfet current rating question

[Rudolf]

Hi,
when designing primary for SSTC, should I take into account mosfet pulsed drain current or continuous drain current parameter? When having 5ms on-time, f0=204kHz and running in 50hz ramped mode using stacato interupter.

Thanks

[davekni]

Peak current needs to be under the peak rating, and RMS current needs to be under the continuous rating.  FET switching losses add to power dissipation, so you RMS current will need to have some margin below the FETs continuous current rating.

[Rudolf]

So I tried my newly built ramped SSTC, with half-bridge it worked like charm making up to 35cm long discharges with 20cm secondary.

But I wanted bigger sparks, so I made H-bridge of 4 irfp460s (using your awesome low inductance bridge design - which worked perfectly, no inductive spiking) .
When testing H-bridge at very low BPS(around 2 BPS) I noticed that discharge sometimes (every 5-10 sec or so) makes different sound. Also I noticed that at lower voltages around 90VDC it happens much more often. It sounds like when QCW discharge hits some object, but mine doesnt hit anything.

Then I tried measuring secondary voltage with scope by putting probe next to the coil, and I got this:



(secondary voltage with H-bridge, input for bridge:100hz two way rectified mains, unfiltered)
(this waveform is when it was normal sounding, sadly I couldnt capture waveform when it maked different sound, cause it happens pretty irregularly)

But i saw it few times on the scope and it looks almost same but the first, short spike is around 30% higher than everything else. Also Im not sure if the secondary voltage should look like that, cause with half-bridge it didnt look like that, secondary voltage didnt drop during the on-time.



socondary voltage with halfbridge (not ramped cause this photo is 2 weeks old and back then it was running in normal mode)(DC-filtered rectified mains) (in this photo im powering the bridge with filtered DC, but unfortunately it is only one I have of secondary voltage using half-bridge, I dont have any photo of it running in ramped mode)

The driver did not change, so I think that it isnt causing the problem.
Should I be concerned ?, or is it normal and doesnt matter ? (this is my first attempt of building ramped SSTC and I couldnt find anything like this on the web)

Thanks

[davekni]

I may not be understanding everything correctly, but will attempt to answer.  (Glad my H-Bridge layout post was helpful.)

How is your ramp implemented?  Are you ramping drive phase or ramping bus voltage?  The first two scope images look like ramped bus voltage that happens to have some initial charge on the bus (initial charge on the output of the ramped-voltage generator).  The initial VBus charge makes the initial spike in output voltage, which is followed by the intended ramp.

The final (third) scope image looks like bus voltage that sags at the start of a burst.  That could happen if there is some low-ESR bulk capacitance close to the H-Bridge, and some larger bulk capacitance farther away with higher resistance.  Another possibility is that the driver phase is better at the start and shifts later, but I can't think off-hand what would cause that.  If it is a driver issue, the upper scope traces with the slow Vbus ramp would mask any such initial phase change.

Concerning why sound is occasionally different, I have no guess.  If you capture the difference in scope traces eventually, that should provide a clue.

BTW, my QCW is running like a ramped SSTC.  Frequency is low (115kHz) for a normal QCW.  Primary capacitance is large, so not a significant factor in resonant frequency.  The only differences are higher coupling factor of 0.909 and primary current feedback rather than secondary current or voltage.
https://highvoltageforum.net/index.php?topic=1268.0
All the initial testing of this coil is using the example H-Bridge I'd posted.  Plan to upgrade to large IGBT bricks in a couple months.

[Mads Barnkob]

Concerning why sound is occasionally different, I have no guess.  If you capture the difference in scope traces eventually, that should provide a clue.

The description of change in sound only happening occasionally and with a thump-like flavor to it, takes me back to my initial days of coiling with SSTCs. I had the same "issues" and back then I tracked it down to unstable feedback, it appeared to me like a misfiring that was off the resonant frequency of the secondary circuit and sounded differently from sparks that was generated at the resonant frequency.

[Rudolf]

Hi!

First I want to thank both of u for helping me out with my  "issues".

So after your responses I decided to push my coil to it's limits, which surprisingly went out well. Now I'm getting at 260VAC around 50-55 cm open air discharges    I think that's pretty good for an SSTC combined with fact that I'm just beginner "coiler".

Now I'm building an audio interrupter using Kaizer SSTC/DRSSTC audio interrupter circuit. (I'm not sure about the rules - so I rather don't attach the schematic)
At first it didn't want to work correctly, but after bit of probing I found out that when the C11 (10uF cap at the input to the op-amp) is removed, the circuit work correctly.
(I already saw thread on this topic with exactly same problem, but there was no solution to the problem)

So here's my question: Is this C11 cap required for the circuit to work correctly or it can be omitted?


Thanks

[davekni]

Yes, that circuit unlikely to work as drawn.  Eliminating C11 works, but only if the audio source is DC-coupled or has some DC impedance across its output and the amplitude is high enough to reach the LM741's common-mode input range.  A better option might be to leave C11 in place and add two resistors after (to the right of) C11, one to +9V and one to ground.  That will set the DC operating point of the opamp independently of the music source.