A little spin-off on the subject, toroid transformers as output transformers in tube amps are total garbage. Apparently the electrical effectiveness is inversely proportional to it's frequency range.

 

A little spin-off on the subject, toroid transformers as output transformers in tube amps are total garbage. Apparently the electrical effectiveness is inversely proportional to it's frequency range.

Not so. Toroids can make excellent valve amplifier output transformer when designed for the purpose. The frequency range of a toroid is better than an EI transformer because of better coupling between the windings, the same reason that makes it a good output transformer. If you hoped to use an off-the-shelf mains toroid as a valve output transformer, think again - you will be two octaves short of bass response, and your leakage inductance will still be too high. You also need very tight matching between the anode currents of the output valves, as a toroid will easily saturated, and you are unlikely to find a gapped toroid for use with a single-ended amplifier.

 

Thank you for your reply! While working with tube amps, we had one custom made toroidal transformer made specifically to be an output transformer, and the result was quite lame!
Until this day from that single experience I rendered the toroidal transformers useless for output audio. It is always nice to share knowledge and learn from other people's experience.

 

Thank you for your reply! While working with tube amps, we had one custom made toroidal transformer made specifically to be an output transformer, and the result was quite lame!
Until this day from that single experience I rendered the toroidal transformers useless for output audio. It is always nice to share knowledge and learn from other people's experience.

I have an idea that R-cores would be very good, (GOSS single strip core, but with concentric bobbins) but there’s no-one locally with the winding machinery.

 

Thank You for the insights about toroid transformers for audio signal applications. I know that there must be a reason why they have not been used at all in sound system equipment, and I had previously guessed that it was only cost. I do have an audio mixer panel that appears to use small toroid transformers with powdered iron cores for the microphone inputs. But at those very low power levels probably any non-linear effects are not present, and the self-shielding properties are very useful.

 

Thank You for the insights about toroid transformers for audio signal applications. I know that there must be a reason why they have not been used at all in sound system equipment, and I had previously guessed that it was only cost. I do have an audio mixer panel that appears to use small toroid transformers with powdered iron cores for the microphone inputs. But at those very low power levels probably any non-linear effects are not present, and the self-shielding properties are very useful.

I’d be very surprised if iron powder were used for a signal transformer.
What you need is lots of permeability, to get a high value of magnetising inductance. it needs at least 4.8H to get 20Hz at 600Ω, and iron powder has a relative permeability of about 75, compared to 60,000 for silicon steel. Could they be ferrite?

 

Quite possibly they could be ferrite. And after a closer examination in much better light the core material is an unknown.
The tiny transformers are less than one cm in any dimension, including the molded plastic base that supports the toroid and the thru-hole mounting pins. The only obvious property is that they are certainly toroid transformers. I may probe one with a magnet to see if there is much attraction or not.

 

There’s also o-cores which are much closer to the mathematical toroid, which has a circular cross section.
A square cross sectional toroid has a CSA of D and a turn length of 4D.
A “real” torioid would require a diameter of 2D/√π or 1.13D for the same CSA, but would have a turn length 2√πD =3.54D, a saving of 12%

Hi,

What I do is calculate the area of a circle (circular cross section):
Ac=pi*r^2=pi*(d/2)^2=pi*d^2/4 with 'd' the diameter,

then area of a square with 's' one side:
As=s^2

then equate the two:
pi*d^2/4=s^2

then solve for s:
s=d*sqrt(pi)/2

So the square would have to have that side to have the same area as the circle.

The circumference of the circle is:
Cc=pi*d

and of the square is:
Cs=s*4=2*d*sqrt(pi)

then find the ratio: Cs/Cc
Cs/Cc=1.128

So yes it's about 12 to 13 percent more for the square. I had to correct my original post somehow it came out to twice that value.