[Prev][Next][Index][Thread]

winding your own transformer




----------
From:  David [SMTP:davmckin-at-ix-dot-netcom-dot-com]
Sent:  Wednesday, March 04, 1998 4:13 AM
To:  Tesla List
Subject:  Re: winding your own transformer

Tesla List wrote:
> 
> ----------
> From:  Harri Suomalainen [SMTP:haba-at-cc.hut.fi]
> Sent:  Thursday, February 05, 1998 1:32 PM
> To:  tesla-at-pupman-dot-com
> Subject:  Re: winding your own transformer
> 
> >Toroidal?? Hmmm.. the pic I saw looked like the typical square
> shaped
> >transformer.. are they build from multipule cores? And what material
> would
> >I want for this thing?
> 
> Toroidal cores are not good for PSU-type applications. The reason
> behind this is simple: core shape will make it totally protected by
> winding. This includes thermal insulation. Double-E cores are much
> better and much more common for power supplys.


I made a mistake listing the core as a toroidal core.   The ETD core
is a special version of the E-core.  It has a round center leg.  This
maximizes the core area per mean length turn (allowing for about a 11%
decrease in winding losses over that of a square center leg).  This
configuration is very good for use in power supply and high frequency
applications.

> 
> >> configuration or possibly a current driven inverter operating
> >> at approximately 300 kHz.  As far as the windings being Cu
> >> foil - yes they need to be foil based on the guesstimation that
> >> this transformer is operating at 300 kHz and using a fairly high
> >> current with the necessity of having a low impedance as possible.
> 
> I definately disagree. Winding a few thinner wires for a bundle will
> do equally well or better. Copper foil windings have no way of making
> the proximity effect less severe. Both do have lots of surface area
> which is important due to skin effect. Litz will also do naturally.
> Copper foil is also nasty thing to be wound. Using copper foil with
> toroids is definately not nice.


For this particular application, I would still recommend using the foil.
Being the round center leg, the foil insulated with 2 x 1.5 mil layer
insulation would be easy to wind.  Additionally, I would think
that by winding this transformer with foil, one would be able to keep
the leakage inductance down - which is a very big problem in high
frequency power transformers.  Also, given the shape of the core and
window (round center leg and "grooved" outer legs) wouldn't copper foil
maximize the amount of copper one would be able to fit into the window?

> 
> >I had a reply on here about winding the coils in two parts, each
> >would the opposite direction from the other with the center tap
> >being the inside connection of both sides and the two parts wound
> >in opposite directions. I assume since I'm dealing with bare copper
> >here I need some kind of turn-to-turn insulation. And something for
> >the primary-to-secondary windings.  Bifilar winding is a must for
> >push-pull. That means you wind two wires side by side. Then you
> >form the center-tap by connecting different ends of windings
> >togeather. That is needed for minimizing leakage inductance
> >between the two windings.
> 


For this one, would it not be possible to use the start of the winding
as one side, then at the center of the winding bring out a tab then use
the finish of the winding as the other end (instead of winding two
separate coils to create the center tapped winding)?


> --
> Harri Suomalainen     mailto:haba-at-cc.hut.fi
> 
> We have phone numbers, why would we need IP-numbers? - a person in a
> bus


Those were my thoughts - right or wrong - but that's why were here
to help each other out.

David L. McKinnon