Rough and ready mc34063 ht supply.

Maybe this is one for Nige.
I need to test some high voltage components for a induction welder.
So I need a rough supply around 2.5kv and a few watts, doesnt need to be clean.
I just threw togther something with a tv lopty, to avoid gigohm resistors and handling eht I decided to try and derive feedback from one of the other low voltage windings on the flyback (I'm driving the primary with 12v), only to find the coupling is pants, shorting the eht o/p hardly increases the current pulled from the supply.
I solved the issue deriving feedback direct from the eht winding which is the o/p of the supply.
The circuit does now work well enough, the '34063 will regulate from 300v up to a few kv but I really need a proper xmas tree volts divider mr25's just arent up to it.

Do lopty's have rubbish coupling, or have I got a baddun?

LOPTX's are designed for a specific purpose, and not for anything else - in particular you don't want them to be able to supply more than 1mA or so.

Yes of course.
However I do have a supply good enough to do what I need.
A ma at 3kv is 3w so thats realistic.
Once I've got this thing sorted I'll arrange something better.

You see people with 555's making silly arcs with these things, the rectifier and secondary must be able to take a lot of overload.

dr pepper said:

You see people with 555's making silly arcs with these things, the rectifier and secondary must be able to take a lot of overload.

Click to expand...

It amazing me, considering how many of them fail under normal use

I suspect those making silly arcs don't last very long.

Not my kind of thing, but your probably right if they fail on their own mashing the heck out of them cant be good.
Once I've tested some caps with it, job done, sorted.

I have an olde fashionde lopty, from an itt set, the one that converted pal to ntsc or something like that, anyway it doesnt have a rectifier, for a completed nicer version I'll do something with that and a stack of diodes.
I ha d wondered about not having a rect, flyback topolgy as you know can be arranged to produce lower forward volts at the secondary, and much higher reverse during flyback, this would average out on a cap, but methinks the amount of current to reverse the polarity due to reactance would probably blow something up.

I managed to test all these caps with my bodged supply, found 1 open and 1 leaky, when opened you can see tracking marks.

So I will make a more permanent version, I ended up with 60kc switching freq, much higher than the lopty would normally operate at however it doesnt warm up off load so seems ok.
Tv transformers in my experience often use 3c85 core material or similar, I'dve thought flybacks would be lower freq ferrite, but this ones fine at 4x the original freq.
5ma doesnt seem to heat up the high v diodes either.
I have a box of lopty's so when it fails no problem.

Seems the old machine is having a hard time, one of the elctrodes got damaged and shorted to ground and this happened:

Looking at those capacitors, they all seem to be "bulging" on the ends. Not a healthy sign.
JimB

2 of them have exploded and vaporized the copper tracks, this pcb was repaired by my electronics supplier a year ago, I wonder if they fitted caps with too high a esr.
I'm not going to tell the boss that.