ampage
Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum. New: view Recent Searches.
New: visit Schematic Hell!
The sunn still shines online!

 
Listen to great tunes streaming live right now!

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  
previous: Tony T I guess it's not the material that ... -- 8/28/1997 6:10 PM View Thread

Re: Transformers/ Speakers

8/28/1997 9:07 PM
R..G.Re: Transformers/ Speakers
This is a common misconception/guitar myth, and one that Weber and some transformer makers have propagated.  
 
I laughed for quite a while when I read Weber's "raise your strings and see how much quieter they are" rationalization of this. In fact, I laughed at a number of things in his book, but that's not germane to this discussion.  
 
In winding a standard EI lamination, the differences in thickness from one bobbin to the next is of no consequence, for any normal and reasonable thickness of bobbins.  
 
It is possible to make bobbins so thick that you could DETECT that the bobbins were thick, with appropriate sensitive instruments, but that same thickness would so interfere with the normal function and design of the transformer that it would be impractical to make.  
 
It is true that a magnetic field in free space decreases by inverse square law. The thing that matters here is that the windings are not in free space. The windings occupy pretty thoroughly all of the space inside the shell that the EI laminations make around the winding window. If you go through the math for a coil of wire in free space, the fields of the two sections of wire of any two adjacent cancel between the wires, and add in the center of the coil(I had to do this in my first physics course. It still hurts.) That's the point in winding coils in the first place - the field is concentrated in the center.  
 
In the case of the EI transformer, the coil fields add to a maximum in the core, helped along by the fact that the core "conducts" magnetic fields several thousand (!) times better than air does, so the magnetic field prefers to stay in the iron.  
 
This means that to a first approximation, the entire magnetic field of any turn of wire anywhere inside the winding window stays in the transformer core. Of course, that isn't perfectly true, but we call the departure from perfect "leakage inductance" and it's separately calculable.  
 
It's also something that is defeated by intermingling sections of winding, not by twiddling bobbin thicknesses. The "unity coupling" of a McIntosh output transformer relies on winding the halfprimaries in bifilar, side by side, not on the thin bobbins.  
 
For coils with a height in the winding window of any significant fraction of the window, the thickness of the bobbin is insignificant compared to the winding build itself, so the bobbin thickness is insignificant to the result in two different ways - thickness of the bobbin compared to the winding build height and the several-thousand-to-one preference of the magnetic field to be in the iron than in the air around the wire.  
 
The difference between a bobbin of 0.030 and 0.045 will be imperceptable, probably even at the detail instrumentation level for a coil of any significant size wound on it. I don't think I could devise a test that would with an electronic mesurement of the current and voltage on a coil reliably detect bobbin thickness for normal transformer winding practices, much less hear the result if that was the only variable.  
 
And this completely ignores the fact that paper bobbins vary a lot in thickness themselves, as well as not always fitting very well around the tongue of the cores. The paper could be thinner, but have a bend that bows the lamination away from the iron.  
 
It's tempting to think that the old ways must be better, but it's important not to get blinded to the physical reality.  
 
Let me say it again: The bobbin material does not matter. It is possible to design equivalent sounding transformers with paper or plastic bobbins if you have the cores and windings equivalent.  
 
That does not say that any two transformers, one with nylon and one with a paper bobbin, HAVE been wound the same. If the designer of the nylon bobbin one was sloppy or greedy and cut corners and if the designer of the paper one expected to get solid gold prices for it, you might get two different results. That would be because the two transformers were wound differently, not because the bobbins were paper and/or nylon.

 
Replies:
John Martin Well, thats a relief! (re:RG) I ... -- 8/28/1997 9:28 PM
Tony T So what transformer manufacturer/su... -- 8/28/1997 11:15 PM
Rusty Bretz R.G. , Thanks for the t... -- 8/30/1997 5:11 AM