Transformer construction question (about internal and external shielding)

[kagliostro]

In this last days I'm trying to learn all I can about practical transformer construction

(I'm building an adequate desk with drawers and motor support for my unused winding machine)

In the best constructions, transformers often have an internal shield between windings and/or sometime an external one in copper seen as an umbelical ring

I would like to have council about the realization of this shields

1) As far as I can know there are shield between primary and secondary windings, which is the better way to obtain it ?

a winding of the same copper wire used in primary or secondary winding ?
or
a sheet of thin copper or brass ?

2) if a wire is used, one of the extremities must be isolated ?
    if a sheet is used, the ring must result open or closed ? I mean no connection between start and end of the sheet (I think a     closed loop will have negative effect because of the developed heat)

if the winding of primary are positioned in an alternate way (Primary/Secondary/Primary), a pair of shields must be used ? Or only one ? See this picture to understand what I mean

P Primario = Primary
S Secondario = Secondary
Antiorario = Counterclockwise
Orario = Clockwise

---

In some transformers there is an external shied build as an umbelical ring

3) Is correct to say that this shield can be applied in both cases where the transformer has or don't has end bells ?

4) The correct position of this shield is around the windings and lamination and NOT inside the laminations only around windings, and the extremities of the copper sheet must be connected together by soldering, is this right ?



Thanks

K






 

[eleventeen]

I have almost nothing to offer you in terms of knowledge....


have you seen this link? http://ludens.cl/Electron/trafos/trafos.html


I *did* build a fairly interesting toroid transformer once, it was over 40 years ago. I did not understand it, my boss at the time told me how to build it and I did, and it worked. The primary was some number of hundreds of turns of maybe 20 ga wire, and it was quite irritating to wind all that wire around the toroid, which was maybe 3-3.5" in diameter. The secondary was 40 paralleled windings of 1-1/2 turns each. I collected all the secondaries in rings above and below the toroid. It looked like something out of Star Trek. But that sucker put out 5 volts at 50 amps all day long.


You said: As far as I can know there are shield between primary and secondary windings, [/size]which is the better way to obtain it ?
[/size]
[/size]Is that true? It would seem to me that a shield between pri and secondary windings would reduce coupling.

[kagliostro]

Thanks for the link Eleventeen


I forgot to specify that the image of the transformer with interleave windings refers to an Output Transformer


K

[HotBluePlates]

In the best constructions, transformers often have an internal shield between windings and/or sometime an external one in copper seen as an umbelical ring

Know that these are 2 different things.

Internal shield between windings = Electrostatic Shield. Externally-visible Copper Band = Flux Band

The electrostatic shield will be a copper band which completely encloses the inner winding (no interleaving). The copper band has one end soldered to the other to form a complete loop. The lack of interleaving is not an issue, because the electrostatic shield applies to a power transformer to keep noise on the primary line off the secondary. A power transformer doesn't need high frequency response, so no interleaving is okay.

The flux band (Figure 13.4 and text) goes around the outside of the transformer enclosing the open side of the windings and core, meaning the portion of the bobbin. It lays over top of the windings and in the opening of the bobbin, but also goes around the core. See the picture in the link. This is to minimize the radiation of flux from windings to surrounding objects. It too is a closed loop of copper, with the end soldered to the start of the band.

These should both be sheets of copper, and not formed from wires wound around the transformer or windings.

3) Is correct to say that this shield can be applied in both cases where the transformer has or don't has end bells ?

Presence of endbells does not matter, but most manufacturers have endbells on their power transformers. The endbell sits over one part of the flux band on a completed transformer.

4) The correct position of this shield is around the windings and lamination and NOT inside the laminations only around windings, and the extremities of the copper sheet must be connected together by soldering, is this right ?

You now know this depends on whether you're talking about a flux band or an electrostatic shield, as they are loacted in different places in the transformer.

[kagliostro]

Many Thanks for the explanation HotBluePlates

the answer to question #4 is clear, I put the question without specifying I referred to the external shield so you may have understand I was asking about the inner shield

--

The fact that the inner shield is a closed loop confuses me a bit because I was thinking to it as a huge winding that can develop a high temperature if it is shorted, in practice this doesn't happen ?


EDIT: Electrostatic shields apply to PT - Flux Band apply to PT and OT, is this correct ?


Franco

[HotBluePlates]

The fact that the inner shield is a closed loop confuses me a bit because I was thinking to it as a huge winding that can develop a high temperature if it is shorted, in practice this doesn't happen ?

In practice, it doesn't happen.

The reason the shield works is because it is a shorted-turn. In simple terms (and probably not 100% accurate), there exists a capacitance between the primary and secondary windings because they are conductors separated by an insulator (which is the definition of a capacitor). If the wall current to be applied to the primary contained not only the 50/60Hz fundamental sine wave that you want but also higher-frequency noise, the small capacitance between windings allows it to couple very effectively from the primary to secondary.

The electrostatic shield acts like a 3rd plate added between the existing capacitor "plates" made up of the primary and secondary. Now you might see the situation as 2 series capacitors: primary-to-shield and shield-to-secondary. Because the individual capacities are in series, the total capacitance from primary to secondary is reduced (recall how series capacitors give a total capacitance smaller than either individual capacitor).

The shield must be a "shorted-turn" because you're seeking to sap the strength of the affected flux/signal. It also needs to be an enclosed structure to be an effective shield. Now a shorted turn should have a high circulating current, but you'e not using a thin wire but a large, wide sheet of copper.  That means resistance is low, so Power = Current²*Resistance (Heat Dissipation) is also relatively low.

EDIT: Electrostatic shields apply to PT Flux Band apply to PT and OT, is this correct ?

The "electrostatic shield" is the copper band between primary and secondary. The "flux band" is the copper band around the windings and core of the finished transformer. The are physically similar but perform different functions in different places on the transformer.

You could simply use a potted transformer instead of a flux band, and accomplish the same end result. However, the potting can will do nothing to keep noise on the primary from coupling into the secondary, as an electrostatic shield does.

Question: What is the application of the transformer you diagrammed? If it is not a power transformer, neither of these things apply. You don't see flux bands on any output transformer. And output transformers often have sectioned and interleaved windings to increase high frequency response/bandwidth; that makes it physically impossible/impractical to place an electrostatic shield between primary and secondary.

[PRR]

Loudspeaker transformers do not need shields. Signal levels are high, secondary impedance is very low, and there should not be much crap on the plate winding.

Power transformers for stage-amps generally do not need shields. Again the general racket overwhelms any capacitive spikes through the transformer.

Power for hi-fi.... if your wall voltage is clean, no shield is needed. If your wall-voltage is nasty, you probably want a Whole System power filter to keep the hash out of turntable, tape, CD, preamp, and power amp.

[kagliostro]

Many Thanks HotBluePlates & PRR

So, the situation is this

For guitar amps

Electrostatic Shield in PT may be a choice but not strictly necessary (also to say a line filter can do the job)

Flux Band in PT may be a good choiche

Electrostatic Shield in OT is difficult to apply, is not necessary and may be detrimental

Flux Band in OT aren't really useful to shield the signal present from external noise

For HiFi amps

Electrostatic Shield in PT may be a good choice but must be used with other devices on the whole components of the HiFi

Flux Band in PT is a good choiche

Electrostatic Shield in OT  is not necessary, very difficult or impossible to apply and detrimental

Flux Band in OT may be useful to shield the signal present from external noise ????

Am I correct ??

Question:

a Flux Band in a Guitar Amp OT, may prevent signal outgoing from the OT becoming noise if catch from preamp ??

Is to consider also that having recovered transformers to reuse, very often they didn't has End Bells, so Flux Band may become a choice also for OT ???

Many Thanks Again

K

[HotBluePlates]

Question:

a Flux Band in a Guitar Amp OT, may prevent signal outgoing from the OT becoming noise if catch from preamp ??

Is to consider also that having recovered transformers to reuse, very often they didn't has End Bells, so Flux Band may become a choice also for OT ???

No, the chassis is the shield between the preamp and output transformer. A flux band is all bout containing hum radiated by the power transformer.

Toroidal power transformers don't use flux bands because the windings themselves wrap around the core and serve to restrain radiated flux.

[kagliostro]

Thanks Again HBP

K

[J Rindt]

Thanks for the link Eleventeen


I forgot to specify that the image of the transformer with interleave windings refers to an Output Transformer


K

Yeah...that is a great link...especially the tranny winding guide.
I have the E-I core from a Merc Mag 5E3 OT.
I have been trying to give it away fro the last 2 years.
Now I will just take a shot at winding it myself.
Thanks Again