Most exotic 3.5kOhms Output Transformers

[petavgeris]

I have seen this said before but I have never been able to see it in practice except when the applied DC shrinks the physical gap.

dave

Dave,
bass performance is a result of magnetics. Trebble performance is a result of coil winding. It is so simple! DC bias excites the core, raising the effective permeability, so bass is better. Logical, isn't it?
Here there is no chance of gap shrinking, it is made of very stiff carton paper. Gap must be absolutely stable in thickness.

Greetings,
Peter

 

[dave slagle]

Dave,
bass performance is a result of magnetics. Trebble performance is a result of coil winding. It is so simple! DC bias excites the core, raising the effective permeability, so bass is better. Logical, isn't it
Greetings,
Peter

Hey,

Can you point me to one measured or documented result where this happens? Every inductance vs current curve that I have measured (or seen) always starts with inductance being the greatest at 0 DC current and the addition of DC current reduces the inductance.

dave

 

[JZatopa]

Thanks for sharing petavgeris. When you post more pics, could you post one with the transformer next to a can of coke, it is hard to visualize the size of the transformers in your pictures.

Considering the work and documentation you have created, are you planing on selling these on a custom order basis? If so, what price range are you looking at?

 

[petavgeris]

Hey,

Can you point me to one measured or documented result where this happens? Every inductance vs current curve that I have measured (or seen) always starts with inductance being the greatest at 0 DC current and the addition of DC current reduces the inductance.

dave

Exactly the opposite!
Inductance (all else being equal or unchanged) is related to the square of primary turns and the magnetic permeability. Permeability mu=B/H. The gap reduces permeability, so we have the effective permeability of the whole structure including the gap.
Study all B-H charts, they are more or less the same.
We first have initial permeability, then incremental permeability, then maximum permeability and beyond this level permeability starts to get smaller till saturation where it is zero.
I am really impressed that you have never seen any chart like this... Whatever I have measured simply justifies one of the most principal laws of magnetism. With little excitation you get sybstantially increased inductance!

Regards,
Peter

 

[dave slagle]

Exactly the opposite!
Inductance (all else being equal or unchanged) is related to the square of primary turns and the magnetic permeability. Permeability mu=B/H. The gap reduces permeability, so we have the effective permeability of the whole structure including the gap.

agreed

Study all B-H charts, they are more or less the same.

yes.

We first have initial permeability, then incremental permeability, then maximum permeability and beyond this level permeability starts to get smaller till saturation where it is zero.

here is where you go off track.
Initial and maximum permeability are AC functions. Incremental permeability is an ac function with an imposed DC field.

At no point does imposing a DC field increase permeability.

I am really impressed that you have never seen any chart like this...

I have... in many different forms.

Whatever I have measured simply justifies one of the most principal laws of magnetism. With little excitation you get sybstantially increased inductance!

that is true for AC excitation. in the case of DC bias just the opposite happens. (inductance decreases)

dave

 

[petavgeris]

I have a graph somewhere.
With a gap and DC bias there is an initial substantial increase in mu, then mu is almost dead flat and then it starts to go down due to saturation. That's one of the main virtues of SET amps. Push Pull gives terrible sound compared to Single Ended because there is no DC bias and initial effective permeability of iron is quite low, so low level detail gets lost. It is this low level detail where magic of music lies! This has nothing to do with the amplitude of the signal, it is purely a function of DC bias. Gap and DC is all the magic of SET amps!

 

[petavgeris]

Thanks for sharing petavgeris. When you post more pics, could you post one with the transformer next to a can of coke, it is hard to visualize the size of the transformers in your pictures.

Considering the work and documentation you have created, are you planing on selling these on a custom order basis? If so, what price range are you looking at?

Hello!
Yes, by all means. The amp chassis are now at the workshop and they're almost done! I hope that by next week I will post the pictures of this rather insane project. You will see one of the most exotic power amps ever made, if not the most exotic one!

I hope information mentioned in the first text is quite descriptive, including the dimensions.

Regarding any custom orders, for your information I can wind any transformer anybody might ever think of. I have already done it many times for local audiophiles and friends. I am mostly involved in custom order projects that are beyond levels of mainstream performance. This pair consists of four o rings in form of cut cores, ideal for a double c core stereo set. As soon as they were not needed, I decided to wind one pair of the best 3.5kOhms trannies that I could dream of, just for the one and only person that would like to acquire them.

I plan to sell them sometime in the near future for EUR 2.5k, with complete documentation, secondary connection instructions, full set of gapping material and lifelong support.

This is my last set of 0.1 and 0.23 millimeters cores of 10 sq. cm cross sectional area(32x32mm). I have several 0.1mm and 0.23mm cores in 6.25 sq cm cross sectional area (25x25mm), ideal for power amps of up to 5W SET OPTs. I plan to wind a pair of OPTs for 2A3 and a pair for type 45 tubes.

Thanks for your feedback.
Regards,
Peter

 

[tinitus]

sorry, Im not gonna mess with you pros here, but this sounds like its got something to do with one of Joe Rasmussens's transformer 'experiments'

I think he attached some kind of DC curcuit to keep the core saturation constant

 

[Tyimo]

Hi Petavgeris!
Thanks. I am also very "satisfied" 😀 with the quality of women here, but not with the quality of iron cores....

greets:
Tyimo

 

[cab]

what is a cnc laser jet cutter?

 

[dave slagle]

I have a graph somewhere.
With a gap and DC bias there is an initial substantial increase in mu, then mu is almost dead flat and then it starts to go down due to saturation.

I would like to see that graph.

That's one of the main virtues of SET amps. Push Pull gives terrible sound compared to Single Ended because there is no DC bias and initial effective permeability of iron is quite low, so low level detail gets lost.

or it could just be the airgap required for DC linearizes the inductance.

It is this low level detail where magic of music lies! This has nothing to do with the amplitude of the signal, it is purely a function of DC bias. Gap and DC is all the magic of SET amps!

I agree with the concept that music lies in the low level details but have yet to witness DC current increase the perm. As an experiment I measured a choke I had on hand. 49% nickel core with 0 DC current I measured 144hy's with a 0.5V 50hz signal applied. (this represents an AC flux of 3.5 gauss) Applying 1ma of DC current (Bdc of 236 gauss) did not alter the inductance at all. Granted this device had a substantial airgap which could hide a slight change of perm so I placed a minimal gap (0.0005") and remeasured.

This time I measured 810hy's and interestingly enough 1ma of DC did bump the inductance a bit. This was simply the DC current pulling the core together which has the effect of shrinking the gap. A few large clamps secured everything in place and the inductance now measured 950hy's. I then applied DC in .1ma increments and it wans't until 1.8ma that a change was noticed and that was the slight reduction of Q that predicts the onset of saturation. 2.5ma of current dropped the inductance to 890hy.

Finally I changed the excitation frequency to 120hz to reduce the net flux to 1.4 gauss and the inductance dropped to 900hy but I was unable to get an increase in inductance by applying DC. To close up this experiment I did the calculations based on the selected airgap and a 25% increase in perm translates to a 1.5% increase in inductance which should be observable if it existed.

dave

 

[jbercovici]

Hello,
this is a long time after...
What was the winding technique meaning how many P and S layers ? What was the insulation material between P and S layers ?
Did you experiment Finemet Nanocrystal cores ?

Julien