Methods of Winding Large Ferrite Cores / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Methods of Winding Large Ferrite Cores

first previous 2 3 4 5 next

Move Thread

LAN_403

jpsmith123

Sat May 03 2008, 01:31AM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

Here's what the U100/57/25 core looks like with a bobbin:

The next step will be to wind a multilayer secondary able to produce 10 kv peak with no potting.

MRacerxdl

Sat May 03 2008, 02:41AM

Registered Member #989 Joined: Sat Sept 08 2007, 02:15AM
Location: SÃ£o Paulo, Brazil
Posts: 476

I think you will need to insulate layer by layer with something. I buyed a insulation plastic for transformer on a transformer store. It have 0,25mm and can insulate 6-8kV. I have proved with a NST in the coil. And after sometime on my homemade flyback.

The only thing I didnt like its the plastic is very thick if you compare with the wire :( (I used AWG 28) I think in your case its a thicker wire, and the thickness of the plastic its not a very large problem

jpsmith123

Sat May 03 2008, 02:15PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

As I see it there are two issues: (1) keeping the wire roughly in place after winding a layer, and; (2) keeping the electric field at the outer surface of the wire below the "corona threshold", say 30 kv/cm (reduced by some safety factor...so make it 20kv/cm).

I'm thinking that if I use 28 gauge teflon insulated wire, with a diameter of about 0.035", I can get about 28 turns per inch...or about 54 turns per layer on the 1.9" long bobbin. If I drive the core at, say, 20 volts (peak)/turn, then for a "conventional" winding, I'll have 2160 volts between the first turn of the first layer and the last turn of the second layer.

If instead of a "conventional" winding I run the wire back across the winding after completing a layer, so as to always wind each layer left-to-right, then I think I can reduce the voltage gradient by 50% and stay well below the point where I might have a problem.

Maybe I can get away using just teflon tape as "layer insulation" mainly to hold the wire in place?

MRacerxdl

Sat May 03 2008, 02:37PM

Registered Member #989 Joined: Sat Sept 08 2007, 02:15AM
Location: SÃ£o Paulo, Brazil
Posts: 476

Hmm, good idea.
I think its not bad idea using teflon tape as insulation for the layers, I only think the most teflon tapes are not too thick and you need to get much layers to get a good insulation. For 3kV inter-layer I think it can be done.

jpsmith123

Mon May 12 2008, 03:27AM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

I wanted to model the transformer with "CircuitMaker 2000", to see how it would work as a part of a system, with varying loads and drive circuits.

In order to get the best results, I wanted to know as accurately as possible what the coupling coefficient (k) would be, or at least, what the lowest practical value might be, so I first simulated the transformer with an EM simulator (magnetostatically), assuming that any AC vs DC current distribution would not affect k very much.

I felt that the lowest practical coupling would be with the windings physically arranged as shown (only one secondary winding shown):

I was surprised that, assuming a permeability (u) of 2200, the simulator predicts a k of 0.979; for u = 1800, k = 0.974...so in this range, k is a slowly varying function of u.

I'll post any interesting spice results here subsequently.

Dr. Dark Current

Mon May 12 2008, 06:07AM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

jpsmith123 wrote ...

I felt that the lowest practical coupling would be with the windings physically arranged as shown (only one secondary winding shown):
...

The lowest coupling will be if the windings are opposite each other on the core. The highest coupling will be if the 2 coils are wound on the same axis (i.e. one is under the other).

jpsmith123

Mon May 12 2008, 11:29AM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

Well in this case the lowest coupling will be with the arrangement shown, since there will be two secondary coils (opposite each other) on the core...only one is shown.

jpsmith123

Sat May 31 2008, 10:55PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

I'm wondering, has anyone here ever successfully used a multi-layer winding on a big ferrite core to develop a high voltage?

Of the several internet-published examples of HV transformers using large ferrite cores, most seem to use a single-layer secondary and drive the core fairly hard.

In the following linked example, the author explains that he at one point made a multi-layer HV secondary winding,

, and it seemed to work "fairly well", but despite this he ends up using single layer windings, and doesn't explain why he abandoned the multilayer winding.

And then, in the next case,

, the designer uses a single layer secondary, apparently taking-up only a small fraction of the available winding area of the ferrite core, and then he combines four such transformers (and their associated driver modules) together to get 20 kv at 5 kw power throughput.

This makes me wonder: Did these experimenters use single layer windings primarily to avoid the extra stray capacitance associated with multilayer windings?

GeordieBoy

Sat May 31 2008, 11:30PM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

I'm wondering, has anyone here ever successfully used a multi-layer winding on a big ferrite core to develop a high voltage?

Yes, flyback transformers in TV sets do this. But they use a diode-split secondary wound in pie sections to minimise the effects of stray capacitance.

end-to-end capacitance, or layer-to-layer capacitance can be a big problem in the secondary winding of HV transformers designed to operate at a high frequency. The high turns ratio makes even a relatively small capacitance on the HV side appear quite large when transformed to the LV side that is being driven.

In many designs like the SLR inverter, parallel resonance between the transformer secondary winding and it's parasitic capacitance is highly undesirable. That is why a single layer secondary is used, and it is kept well clear of other conductive objects.

Some other parallel resonant designs like the Royer oscillator tolerate secondary side capacitance better because it simply adds to that already connected across the primary winding.

-Richie,

Marko

Sat May 31 2008, 11:58PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

The most obvious reason to use a single layer winding, is that it's just so much easier to insulate.
You will have hard time building a long-living multilayer secondary without vacuum potting, as any small amounts of trapped gas provokes corona which over time destroys the insulation between layers. It has to be done on small cores where window area is scarce, like TV flyback transformers, but I would otherwise avoid it whenever I can.

Single layer winding will work even in free air, just insulation to core needs to be taken care off.

For large cores, there's usually just no need to use multi layer winding, as single layer can accommodate more than enough output voltage and wire thickness on large cores.
It's better to just leave the remaining space for insulation to core (which may be just air, but entire winding may also be potted into resin for greater protection).

Just my opinion about this.

first previous 2 3 4 5 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob