winding yer own transformers, somewhat OT.

[D. Schouten]

Hi,

>Ok, I have been dabbling with winding hi voltage
transformers for projects like
>jacobs ladders and such and usually i do it exactly the way
it is the the books
>and it works, but this time...
>
> I recently bought a pre-made ferrite E-core complete with
a hollow spindle and
>clips..nifty... I thought I would treat myself instead of
going through all the bs
>of riveting a bunch of sheet steel Es and Is together to
make my own cores.

I haven't got any experience with 'Jacobs ladders' and
stuff, but as a SMPS designer perhaps I can say something
usefull here. At first are you sure you can use Ferrite for
your application, since Ferrite has totally different
characteristics than the Es and Is of sheet steel you
mentioned.
The biggest difference between these two materials is the
operating frequency range. The transformer cores made of
stacked silicon steel sheets are usually used in 50Hz/60Hz
mains applications but depending on material, these can be
used up to about 40KHz (with amorphous materials). In power
transformer applications, ferrites are used from about 10KHz
and up.

>Now, I know the number of turns is critical to the
tranformers behavior, but, is
>the number of layers critical also?
>Anyone have any answers? Maybe some of you power supply
ppl?

In high voltage apps, the way of winding the transformer is
very important since you desire very low interwinding
capacitance. Most high voltage output power supplies I've
seen, are using multi chamber bobbins on their transformers
where the number of windings are divided between the
different chambers. This is also important to avoid shoot
through between the windings.
You can make a multi chamber version of the bobbin (spindle)
you already have, by dividing the winding width of the
bobbin in smaller parts (lets say four) with 1mm carton
board. Then you have to split up the total number of
secondary turns you wish to place by four, and place them in
the chambers respectively.

I hope there's something usefull in the above. I personally
haven't made higher output voltages than 100VDC in my
designs yet, so these winding rules doesn't apply here.

Bye,

Daniel...