Flyback Type Transformers

[alan sailer]

I've been exploring the depths of my ignorance lately and have a question about HV transformers.

I've been using several transformers for plasma power supplies. The first and least useful is a transformers called a Chirk.
The data sheet is here;

https://chirk.com.tw/HTV/RU3222.pdf

It is rated at 20kV p-p-k. So as I understand it this means that it will supply 10kV maximum to a plasma tube.

The second transformer is a vintage FLY352 that was meant as a replacement for a color TV flyback. When I did some searching I found an old color TV schematic that used this flyback along with a rectifier tube to supply 25kV to the CRT.

So this would mean that the FLY352 would output 50kV p-p-k which would be half-wave rectified to create the 25kV?

[klugesmith]

Normal flyback operation has very unsymmetric positive and negative voltages.
Positive high voltage appears for less than 10% of each cycle, while horizontal deflection yoke gets "high" voltage for rapid current ramp.
The other 90% of each cycle, HV winding has negative voltage with 1/10 magnitude, while horizontal yoke gets opposite polarity "small" voltage for slow, linear current ramp.
At some point color CRT systems started using voltage multipliers.   AFAIK the AC source is still on same core as the horizontal deflection (flyback) "transformer".
[edit]By the way, real flybacks have something in common with traditional ignition coils for engines.  The coils-and-core component is not driven by an AC voltage source, nor an AC current source. It's driven by a switch, ideally alternating between switch voltage=0 and switch current=0, with non-50% duty cycle.  In one phase the primary winding voltage can be much higher than DC supply voltage.

[alan sailer]

klugesmith,

As far as I understand I am not operating either transformer in flyback mode. I am using an IR2153 half bridge driver with two MOSFETs.

The output from both transformers is a pretty clean +/- sine wave symmetrical about ground.

Cheers.

[klugesmith]

Got it: real transformers, sinusoidal voltages.   AFAIK the air gap within core, necessary in a flyback, is generally unuseful in this application. Reduces inductance so increases the magnetizing current.

As no expert, I think the parameters that do matter include:
Voltage limit due to insulation.
Voltage per hertz limit due to core saturation. Could be expressed as a volt-second product, e.g. area of a 25 kV x 6 us rectangular pulse. For sine waves, same limit would apply to area of the above-zero halfcycle, and to the below-zero halfcycle.
Internal capacitance and self-resonant frequencies, probably not useful in original application.

[alan sailer]

Yes. Since my plasma glass doesn't use a lot of power I have only had issues
with the voltage limitation due to insulation.

And to me, the fact that the voltage is swinging around ground means that p-p-k
voltage is not important rather it is the maximum +/- voltage. I would not think that the insulation
cares about polarity. But I have been wrong about other things that seem obvious.