Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

[SOLVED] Multiple parallel windings

Status
Not open for further replies.
R

red_alert

Guest
Hello,

I want to build a flyback transformer (ETD core) with a single output (12V/10A rated).

As I only have some 0.7mm magnetic wire, I intend to make few identical (rather having the same number of turns) interleaved edge-to-edge windings for both primary and secondary (the designed turns ratio is 1:1).

Obviously, every layer will have a slighty larger diameter (although the same number of turns) thus its resitance will be slighty different. May I also consider that the flux density will decrease, too?

The question is: may I consider those windings "identical" enough to put them in parallel or do I have to add some small series resistors to balance those voltages?

Of course, I could try to make some sort of litz wire using 4-5 strings in parallel but it's much easier to wind a single wire (beside having slighty more available space and stronger magnetic coupling because of interleaving layers).
 

The addition of multiple turns will not affect the flux density in the core, however it will have the effect of reducing the leakage inductance and therefore it will be better coupled. This will not have much of an effect in a coupled inductor but it does make quite a difference in a traditional transformer.
 

How do you plan to tie these thin wires, at the reel pinouts ?
 

The addition of multiple turns will not affect the flux density in the core

Actually, I was trying to say that the magnetic flux for the outer windings is somehow attenuated.

How do you plan to tie these thin wires, at the reel pinouts ?

Yes, I might use a single pair of pins for all the coresponding layers.

I'm only concerning about voltage variation of every single layer due to magnetic field intensity variation and different intrinsic resistance.
 

The easiest way is wind them bifilar fashion. Don't wire one then the other, tie several strands together and wind them simulaneously so they are the same length and exposed to the same flux density.

Brian.
 

    V

    Points: 2
    Helpful Answer Positive Rating
The easiest way is wind them bifilar fashion

Doing that, take care to do not tie the end of each winding to the beginning of the next winding; you have to connect to the extremity of the opposite one, otherwise would cancel the whole inductance of the bunch.
 

    V

    Points: 2
    Helpful Answer Positive Rating
The easiest way is wind them bifilar fashion. Don't wire one then the other, tie several strands together and wind them simulaneously so they are the same length and exposed to the same flux density.

If that's the best option, I'll try to twist some strands and go the classical way. But I'll still need to wind multiple layers as there will be not enough turns in a single one (due to increased stranded wire diameter).
 

The question is: may I consider those windings "identical" enough to put them in parallel or do I have to add some small series resistors to balance those voltages?

you don't have to add resistors like that no.

I know what you are getting at, for ourselves at home, its kind of the easiest and seeming neatest way to do it, the way you describe, instead of fiddling with winding them litz fashion....as long as you have enough pins to terminate on..

Of course you know that you have to have say the secondary "sandwiched" between primaries to get the interleaving effect.
in my book its fine doing what you describe, of course you will have looked into skin effect etc.
 

    V

    Points: 2
    Helpful Answer Positive Rating
I have 2 x 7 pins available and I could tie two layers on the same pins (or even all the coresponding layers to the same two pins).

I need 5 layers (separate windings) for primary and 5 for the secondary. Of course, I would interleave them one by one.

The problem is the inner layers would have the smallest resistance and the greatest voltage hence they will carry most of the current, the downside being they have the less cooling capabilities.
 

The problem is the inner layers would have the smallest resistance and the greatest voltage hence they will carry most of the current, the downside being they have the less cooling capabilities.
Smallest resistance yes, greatest voltage, how?

Current will be primarly balanced by working of the leakage inductance, current sharing is not purely detremined by winding resistance.
 

Smallest resistance yes, greatest voltage, how?

Theoretically, the magnetic flux is greater near the core (inner layers) thus the induced voltage per turn is supposed to be greater, too.

Actually, I've just built a similar flyback transformer with 5 identical (but separate) outputs. The outer windings had a slighty lower voltages though.

But maybe that was because the primary winding was wound first (single layer, no interleaving) thus the magnetic coupling was weak for the outer windings.

If I build these parallel connected windings completely interleaved (one secondary layer, one primary layer) chances are the induced voltages will be the same?
 

Theoretically, the magnetic flux is greater near the core (inner layers) thus the induced voltage per turn is supposed to be greater, too.

According to which theory? If the flux in the transformer winding differs (due to stray fields), the variation in range of a few percent.
 

Well, we're in the "few percent" realm anyway.. I'll put them in parallel and I'll check for any abnormal heating.
 

Yes you can have pri, sec, pri, sec ... etc and parallel the sec's and all will be well, as FvM says the leakage determines sharing for high freq switch mode power supplies, not so much the resistance, the increasing resistance with outer layers will not be a problem... be sure to get EXACTLY the same number of turns on your sec's...!
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top