The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: Mike Poulton
Subject: Re: flyback transformer from scratch
Date: 4 Dec 2002 22:23:05 GMT
Organization: MTP Technologies
On 04 Dec 2002, "Jordan" said:
> I was planning on making a much more powerful unit, like the ones that
> are used in very old TVs. I need about 15 - 20kV AC (or more) output,
> and plan to drive it with a small coil on the side for the feedback to
> interrupt the transistor. That way it will run at its own resonant
> frequency. But I have one quick question. There are two circuits
> floating around for driving a flyback. One uses only one transistor,
> and the other uses two transistor in a push-pull configuration (not
> really sure what that means exactly). Again, I appreciate your help!
The single transistor system is closer to a real flyback converter.
Push-pull is infinitely better. A full H-bridge is probably best.
Whatever you decide to use, don't bother with 2N3055 transistors. They
are next to worthless for these circuits. They burn out all the time
and don't oscillate reliably. The best option is probably to use
mosfets and a separate oscillator (even a 555 timer) to drive it.
>> > During my endless search for a suitable flyback transformer for my
>> > HV
>> > project, the idea crossed my mind to wind my own transformer all by
>> > So far what I would need is a ferrite core (is there any specific
>> > type
>> > core I would need for a flyback or just a normal rectangular one?)
It must have the proper volume and magnetic permeability. If you borrow
one from a bad flyback, you can pretty much ignore those design
parameters. Just don't push more than a few hundred watts.
>> > thin (44 gauge?) magnet wire for secondary, 18 - 22 gauge wire for
>> > and some sort of machine to wind the secondary coils since it would
>> > take hours and days to wind it by hand. I would then coat maybe
>> > every 5
> layers of
>> > the secondary with varnish to keep it all together, then put a
>> > whole
>> > of coats all over the whole thing.
Not enough insulation, unless it will be highly underpowered.
Insulation betwen turns is not a big issue, but insulation between
layers is. Each layer of winding needs to be wrapped with mylar,
kapton, or some other good insulating material, and the coil should
probably be soaked with epoxy, silicone, or some other potting material
throughout the entire winding process. Don't forget an insulating
bobbin to wind it on -- the ferrite core is somewhat conductive.
>> I think this sounds like a hell
>> > of a project and would be very interesting to do. What do you all
>> > say? Am I
>> > plain crazy? Does it take like real designing to make one of these
I think you should understand why they call them "flyback" transformers,
and why you don't really want a "flyback" transformer. You just want a
plain ferrite-core transformer, driven in resonant mode. It's a bit
more than I want to explain here, but some googling on switching power
supply topologies will show you everything you need to know. Basically,
a "flyback" switching converter relies on suddenly and repeatedly
interrupting current flow in one direction through an inductor, then
using the resulting current flow through a magnetically coupled inductor
to power the load. Current does not flow in both windings at once, and
it flows only in one direction. Operation is completely non-resonant.
This is not really what you want, I don't think.
>> > How about the math involved to calculate the number of turns of the
>> > secondary and determining its resonant frequency, and all that good
>> > I'd like to know what you all think of this. Thanks for your input!
For non-resonant operation, input voltage is multiplied by the turns
ratio. For resonant operation, there's no good way to tell until you
try it. I'd say you should calculate the output voltage based on the
turns ratio, and any resonant improvement will be a bonus. Resonance is
essentially irrelevant when it's under load anyways. You will have to
determine the resonant frequency by experimentation, since the winding
capacitance will be nearly impossible to calculate. You can bet on it
being between 50 and 250kHz.
You can certainly do this. Making your own high voltage ferrite-core
transformer is not impossible. It's not even that hard, once you get
the hang of it. Just remember the insulation between layers, and read
everything you can on switching power supplies.
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup