Fender Vibro Champ Output Transformer

[CJ]

was not too thrilled with the sound of the Vibro Champ output, Model P-TF22905

so a torch job was mandated,

first off, they had the leads out of phase on the pri so i got pos feedback instead of negative,

Second, this is a Single Ended amp, meaning DC on the primary,

was the transformer gapped?

no, not even laced Lap 3, it was laced Lap 1, so we think a rebuild is in order,

the re-issue had 2683 turns of #38  (guess)  and 60 turns of #20 (guess)

on a EI core with a 5/8 tongue so call it 625EI, square stack so 0.39 inches sq core area

299 ohms pri  0.2 ohms sec

8K to 4 ohms,  about 44 to 1 turns ratio

wound pri-sec

do not know if the original was gapped, maybe somebody wanted some saturation to mellow out the odd harmonics,

so if you need an easy DIY output for 6V6 SE, just wind the above only put in an air gap of at least 1 mil.

 

[sahib]

Great stuff CJ.

It's in my list. I will definitely try that out in the new year.

 

[Gus]

CJ how much of a difference would there be gapped or having using 3 lam stacking as you mentioned in your post?
Inductance vs level?

 

[12afael]

just pri-sec this one is easy!

 

[CJ]

air gap versus inductance formula can reduce down to

L = 1/air gap,  abbrev.  l g in most books, l g = length of gap,

so f(x) = 1/x for your data plotter,

Y axis equals inductance, x axis equals air gap,

i would guess that you want gap around the knee of the graph of greater,
gapping close to the y axis will lend an unstable transformer as far as mechanical shock causing the inductance to change,

see that if the gap were zero, that the inductance would be infinite, however, even if th air gap were close to zero, other losses would set a limit for maximum inductance, such as fringing loss,core loss, copper loss,




graph of 1/x>

 

[CJ]

stacking 1 by 1 or 3 by 3?

Finke-Mag Matals:

http://www.magmet.com/pdf/TransformDesignConsiderat.pdf

"the hysteresis loop can be sheared to improve the incremental permeability with D.C by stacking E laminations in groups of 2x2, 3x3 or 4x4. Fig 10 shows the incremental perm of 0.014 thick, 2425 EE laminations made of 50% Ni-FE , stacked differently.

Such stacking methods allow to maximize the inductance for AC signals with superimposed DC at a very low cost."

main thing to see from graph is 3 by 3 stacking gives a somewhat flat perm curve with DC but 2 by 2 and 1 by 1 drop in a linear fashion with DC.

you want the perm to stay constant with different levels of DC so that the inductance stays the same depending on the application.

 

[CJ]

ok i was concerned about the lack of an air gap on the vibro champ output,

further research shows that every lamination has a natural air gap,

and the air gap ratio is this gap divided by the magnetic path length,

you can imagine that small lams are harder to squeeze the air out of, a little bend goes a long way, where as you can clamp a big core down prettiy tight, and small irregulaities will not make as much air gap as with the tiny cores, no wonder Beyer uses the pre welded cores with the spin on coils, the perm on small nickel cores can change radically depending on how tight they are, so you want the small lams at a constant tension for quality control,

Magnetic Path Length, MPL is  the average path around the core,

gap ratio = L-a/MPL, where L-a is the air gap, and MPL is the magnetic path length,

use the same units for L-a and MPL, like centimeters for europe,  or inches for those of us still suffering the Imperial system, ;D,


for small lams like the 625 EI as used in the champ output, the MPL is small so the ratio of the natural airgap of a 1 x 1 stacked 625 core to it;s MPL is high enuff to reduce dc flux without gapping.

so the larger the lam, the less natural airgap, as the MPL gets big quick..so the gap becomes more important as the xfmr gets bigger, if unbalanced dc is running thru the coil.



interesting note:

total design flux = ac flux + dc flux,

ac flux does not depend on the air gap,

dc flux does,

so when you insert an airgap, you are reducing dc flux but not ac flux.

if dc flux did not depend on the gap, then there would be no reason to use an air gap.
(except maybe to help stablalize (sp) permeability of the xfmr, but thats in chapter 6,)

formula for dc flux caused by current to feed the 6v6,
B-dc=0.4PiNI-dc/L-a+MPL/u-r,

example:  2000 turn primary coil with 40 ma dc current ,

B=dc=1.256*2000*0.04/gap,  B-dc = 100/gap +MPL/u-r

see that for hi perm cores the MPL/u-r will drop out, eg, hi perm 80 , 625 EI,

mpl = 3.75, u-r = 100,000, mpl/ur= a small number, so we can simply write:

B-dc=100/gap, if the gap is 0.005 cm, you have 100/.005 = 20,000 gauss of dc flux that we have to lower, (bigger gap)

0.01 cm would drop us to 10,000 gaus dc flux, which allows room for some ac flux from the guitar signal,

now if you use regular steel grain, 625 EI, which has MPL = 3.75,  you have to consider the MPL/u-r term  in the formula,  3.75/1500=0.0025,

so for a steel core with low perm, the MPL/u-r term should be included.

100/-005 +.0025=100/.0075 = 13,000 gauss dc flux, so for the same gap, the lower perm materials will get their dc bias lowered more than the super 80 stuff,

another thing to grasp is that dc flux goes up with turns, ac flux goes down with turns, so with an SE transformer, you are fighting 2 things at once, raise the turns to lower b max, and you ampere-turns go up, which increases dc flux, so you get your turns for ac flux first, then compute the dc flux, then the gap you need to reduce dc flux, which then lowers inductance, so you raise turns, see how this process requires a few tries to get into the ball park of happiness?

now the effective perm, a little air gap reduces 80 ni to nothing quicker than it will drop steel perm, but we are still working on this so 420 break needed before my head explodes,