Transformer Question. EI Laminations

[Bruce / Mission Amps]

Got a couple new output trannys to try out in a 20-25 watt amp build project....
Two different vendors trying to get my business...

The first sample came with the lamination stack being 1" thick but used a lot of very thin EI laminations... the other one has the exact same 1" stack dims but the actual laminations are exactly twice as thick "each" and now there are almost exactly half the numbers of laminations in the stack as the previous one.
OK... I didn't ask if it is the same mu metal but can somebody comment on the net effect of this (other then cheaper to make)...

I'm puzzled and concerned about this and whether or not I proceed with an order.

[bbsailor]

Got a couple new output trannys to try out in a 20-25 watt amp build project....
Two different vendors trying to get my business...

The first sample came with the lamination stack being 1" thick but used a lot of very thin EI laminations... the other one has the exact same 1" stack dims but the actual laminations are exactly twice as thick "each" and now there are almost exactly half the numbers of laminations in the stack as the previous one.
OK... I didn't ask if it is the same mu metal but can somebody comment on the net effect of this (other then cheaper to make)...

I'm puzzled and concerned about this and whether or not I proceed with an order.

Have each vendor supply the AL or inductance factor of the lamination core material in micro henries (uH) per coil turn. Typically it will be between 2uH and 3.5uH per coil turn or possibly a little higher for larger core sizes. To calculate the inductance of a coil on each sample, square the number of coil turns and multiply by the AL value of a single turn to obtain the calculated value of the coil inductance on that lamination sample.

Joseph Rogowski

[Chuck H]

Right. Just like Joseph said. Simple

IMHE that sort of info would be, at best, guestimated by a winder. Possibly with missing pieces of info left out of the considerations. Magnetics shops always seemed more like garages than laboratories to me. My guess would be that the thinner lams are of a lesser material and are made thinner to minimize the effect of greater permeability. The thicker lams are probably made of some industry standard with a low enough permeability for that thickness. But this is just a guess. For guitar amp purposes the net result of either method is probably similar enough. One of my favorite OT's ever was a pull from a "The Fisher" 500B. The lams are a little thicker than average. So I guess what I'm saying is that you should listen to them and decide what sounds best. Though I'm also interested in the net effect of thickness alone. I've seen EI's stacked two or three at a time too. As in two E's and then two I's at a time. I'm sure that thicker, thinner, stacking more than one at a time and materials choice is all about saving money and or labor just as one would expect. It's probably also tried and proven to work suitably well before it gets to us as amp builders. I might have reservations about any import tranny though as such things are often made from rote physical designs with lesser materials to an uncertain result.

[Tom Phillips]

... I didn't ask if it is the same mu metal but can somebody comment on the net effect of this (other then cheaper to make)...

I agree with the others that, for your guitar amp product, you should choose by auditioning the sample. Then, before you order, I’d get verification that the production run will be made exactly the same as the sample that you approved.

I offer the following additional comments:
1. Mu metal is a special alloy that is used for shielding rather than for transformer cores.
2. Transformer laminations are made of low carbon and silicon alloy steels. There are several types. Most are common industry standards but there have been some alloy formulas that were proprietary and marketed under a trade name. I don’t know an easy test to determine the steel alloy used in a completed transformer.
3. Lamination material catalogs don’t offer all the various alloys in all thicknesses. I’m not sure why. Maybe because some combinations don’t sell well. Maybe because some alloys are a problem to manufacture in certain thicknesses.
4. In general there will be lower eddy current and therefore less core loss in a transformer made with thinner laminations. This assumes the same size core and the same steel alloy. For a guitar amp application there may very well be no discernible difference in the overall performance of your two transformers because of the different lamination thicknesses.

[J M Fahey]

Sure no mu metal there, which by the way saturates easily, just different grades of silicon steel laminations.

In general thinner is better, lower loss, but we don't know about steel composition or heat treatment.
The thinner one might be made out of recycled tuna cans.

No kidding, a supplier once tried to sell me EI laminations cut out of surplus (still flat) 55 gallon oil drum blanks.
Worst is they they worked, sort of , both sides were insulated after all, one with internal varnish intended to keep it away from oil and avoid its rusting and contaminating it, the outer one painted nice green and white (the oil company's colors, go figure), but it overheated on its own, even at idle.

You might make a basic measurement which will tell you all you need in a jiffy:

1) wire an 8 (or 10) ohms resistor in series with the 8 ohms tap, feed it 18V RMS out of an SS amplifier and keep lowering frequency (start at 200Hz) until you have same voltage across the resistor as you have across the winding (wire a spdt switch to move your voltmeter probes back and forth) , read frequency, that's your Fl -3dB point *under realistic conditions*
I forgot, you do this with all other taps taped, **specially** the plate and +B ones, there will be some very dangerous voltages there.

2) short plate to plate wires, now *raise* frequency, and measure back and forth until again voltage across resistor matches voltage across winding, this is your Fh -3 db point.

From 1 and 2 you might calculate transformer inductance (reflecting those 8 ohms into the plate to plate side) and loss/parasitic (choose one depending on what side of the pond you are) inductance, but only academic need for that, you already determined *practical* usage limits.

No need to say that the transformer with the best specs is the one to choose.

A very poor scatter wound winding will very probably show a nasty unbearable peak well within the audio band, even as low as 4 or 5 KHz .
Anything above 12 KHz is acceptable.