Electrostatic shield on torroid

[2muchstuff]

Since I will be rewinding a torroid transformer to get the correct voltage out of it, why not add an electrostatic shield. Has anyone seen one before. I'm assuming that it a metal foil in-between the primary and secondary windings. Is it wrapped like the wire or is it wrapped around the outside and gets connected to earth ground. Anybody have any ideas.

[trilect]

Originally Posted by 2muchstuff Since I will be rewinding a torroid transformer to get the correct voltage out of it, why not add an electrostatic shield. Has anyone seen one before. I'm assuming that it a metal foil in-between the primary and secondary windings. Is it wrapped like the wire or is it wrapped around the outside and gets connected to earth ground. Anybody have any ideas.

Might sound silly, but I usually use stainless steel mixing bowls the smallest required to enclose the transformer. On the open side of the bowl I use a piece of gasket material between the bowl mouth and the thin stainless steel plate that I cover it with. Then I use a nylon bolt thru the center of the whole thing to hold it together and to whatever chassis it needs to be held too. Hope this helps.

[Al_The_Man]

Electrostatic shields are not used as much on Toroidal transformers, compared to square lamination type due to the fact that Toroids do not radiate fields out as far, also care has to be taken that any metal cover or screening does not cause a shorted turn effect by completing a path through the center and around the outside of the toroid.
If you insert it between primary and secondary, it may detract from the magnetic coupling to the secondary.
Al.

[gar]

050521-1631 EST USA

2muchstuff:

An electrostatic shield is often referred to as a Faraday shield.

Its usual use is between the primary and secondary and is connected to ground (chassis) to reduce the capacitive coupling from primary to secondary to reduce capacitively coupled noise.

Effectively this shield provides a capacitor from primary to ground and separately from secondary to ground.

It can be any good conductive material like gold, silver, copper, and aluminum. With these materials which are non-magnetic there is no effect on the magnetic coupling from primary to secondary.

However, the winding of the Faraday shield must be such as to prevent a shorted turn effect.

To do this on a toroid wind the secondary ( or if desired primary ) on the core first. Apply an insulating layer. Next starting at the outside circumference of the the toroid start a layer of copper foil with one edge that is along said circumference going all the way along that circumference and coming back to the start point. The copper foil can touch and should overlap at this point.

Next run insulating tape over this circumferencial edge so that when you push the copper foil thru the center hole of the toroid and back around to slightly overlap the circumferencial edge that it will not short to the starting
edge.

Attach a wire to the foil to bring out to ground the shield.

Next apply an insulating layer and over that wind the primary ( or if desired secondary ).

There are other ways to install the shield.

Why do you want to use a torroid instead of an E core? E cores are far easier to wind. Or just buy a custom transformer for you needs.

.

[2muchstuff]

Thanks for the reply to my question. I have some copper foil around here that I can use for the shield.

The reason for the use of the torroid transformers is that I got a good deal on four of them. Actually I got four new 650 watt torroids for $36.00. They were 120 volt @ 5 amp output and I will be unwinding and rewinding them to get the correct voltage that I need.

[gar]

050522-2039 EST USA

2muchstuff:

Some more clarification. A Faraday shield between primary and secondary is primarily to minimize common mode high frequency noise on either side from coupling to the other.

Any transformer will look like a low pass filter to normal mode signals because of the leakage lnductance between primary and secondary and that there will always be parallel capacitance across both primary and secondary.

A normal mode signal is the signal applied between the ends of the primary or secondary. Thus, the power signal and all its harmonics coming into the transformer are common mode signals.

The higher the leakage inductance for a given capacitance load the lower is the filter bandwidth. A toroid will have the lowest leakage inductance because it has the best magnetic path ( no air gap ). Therefore an E-core or larger air gap will provide a lower cutoff frequency. A Sola constant voltage transformer is a very good filter because of its high leakage inductance and resonant circuit.

A comon mode signal is a signal of the same amplitude and phase applied to both input leads of the transformer. This signal couples nothing thru the magnetic circuit because no current flows thru the transformer coil from this signal.

If you really need a Faraday shield, then the description in my previous post is not the easiest to make.

If you can get or make a copper foil strip with insulation on one side, then wind this as a coil in an overlapping fashion but never connect the ends together. The goal is to get a non-magnetic conductive shield around the outside of the inner winding and ground this but not create a shorted turn.

.

[2muchstuff]

I have done some reading and found somwhere that by adding an electrostatic shield to a transformer will reduce it's effiency. Is this reduction in effiency an amount to be concerened with or do you think that I'm trying to make a mountain out of a molehill with trying to reduce the amount of noise coming out of a power supply.

[jon]

is the primary winding always the inner one, and the secondary winding always the outer one on the Toroidal core transformer?

[gar]

050523-1041 EST USA

jon:

Primary and secondary are only relative terms. The primary is normally the input side. But a transformer is bidirectional so for example in a telephone application signals may flow in both directions.

2muchstuff:

You might tell us if your transformer is a 50-60 Hz power transformer, or a moderate frequency switching transformer.

Generally in a 50-60 Hz application transformer efficiency is about 90 to 95%.

The losses are primarily resistive, hystersis, and eddy current. Magnetizing current is an approximately constant resistive loss due to the magnetizing current. Magnetizing current grows very rapidly as you go into saturation. For greatest overall lifetime economy most magnetic circuits are driven slightly into saturation. This minimizes iron and copper but increases residual resistive losses.

Connect a Variac to an unloaded transformer, and a scope to measure the current to the transformer. Start at maybe 50 % of nominal voltage and gradually increase the voltage and go to maybe 110 %. As you go into satuation you will see a peaking and distortion of the current waveform.

A proper Faraday shield in a low frequency application is going to contribute essentially zero loss and therefore has no effect on efficiency, other than it will reduce the window area for the copper winding.

.

[Al_The_Man]

The practically universal custom is to wind the primary first, there is a difference in the efficiency of toriod over EI type, with Toroid being quite a bit more efficient due to less losses etc.
Some transformer manufacturers will mention that their products can be primary/secondary reversed but many times add a slight derating figure when used this way.
Al.

[gar]

050523-1541 EST USA

al:

The following is not to disagree with you, but rather to provide some numeric information on an E core transformer. Yes the toroid will be more efficient but does it make any major difference for a line frequency application and it is far more difficult to wind.

I ran an experiment on a Stancor P-8668 117 VAC 60 HZ input to 28 VAC at 2 A output. Input power was measured with a Simpson 75 W full scale wattmeter. The ouput when loaded was with a 13.3 ohm resistor. The output was measured with a Fluke 27 multimeter. The load resistance was measured with the Fluke 27 prior to loading the secondary. Load power was calculated from measured voltage across the resistor when connected to the secondary and the premeasured resistance.

Input voltage to the transformer was 122 volts. Input power unloaded was 4 watts.

With the 13.3 ohm load the input power was 72 watts and the load voltage was 29.5 volts. Thus, output power was 65.43 watts. The efficiency is 90.9%. The power loss in the transformer is 6.57 watts of which about 4 watts is transformer exicitation loss. Had I run at a lower input voltage the efficiency would have been somewhat better.

Also a somewhat higher VA transformer would have slightly better efficiency.

I would rather wind a coil for an E core transformer than a toroid. Also an electrostatic shield is easier to make.

As frequency goes up a tape wound toroid is better because of the thin tape used compared to E laminations. However, powdered metal or ferrite cores whether E or toroid are required as frequency goes high to reduce eddy current losses.

2muchstuff:

You really need to decide if an electrostatic shield is needed. It is a lot of work on a toroid and eats up space for copper primary and secondary.


Another point: Transformer efficiency will not be as good when the load is a capacitive input filter following a rectifier. This is not relative to the rectifier loss, but due to current peaking and the higher RMS current for a given average dc output current.

By the way a bridge rectifier is a particular configuration of four diodes, and has nothing to do with whether or not the four diodes are separate or in one package.

Also a bridge rectifier ( this is a full wave rectifier by default ) in comparsion with a full wave center tapped rectifier will have twice the power dissipation because there are two diodes in series with the load instead of one. But the diodes in the bridge rectifier only see 1/2 the peak reverse voltage of those in the full wave center tapped circuit.

.

[Al_The_Man]

Originally Posted by gar I would rather wind a coil for an E core transformer than a toroid. .

I would agree, if it is a new construction. But if it is a modification to the secondary, usually the Toroid is easier, although still awkward, I have dismantled EI type to modify the winding, but it is very labour intensive and if the laminations have been tack welded, or dipped in shellac, forget it.
Al.