Ferrite cores for audio transformers?

[CJ]

you could use a pot core with a tunable slug,

mount the pot core on the front panel and put a knob on the adj. slug,

then you could really dial in the treble band you wanted to boost,

in conjunction with the regular front panel rotary switch you would have infinite frequency selection,

 

[bruce0]

you can make a BH tester with a 100 k resistor and a .1 cap,

use your scope in X-Y mode,

Ok, finally got back to this after being stuck over in g9 transformer land for a while. 

I can't seem to get enough current to saturate these things.  even on the lowest tap of a Pultec inductor (21mH) there are too many turns.  The MPP toroids saturate at around 20dB when driven through a 120 (ish) ohm impedance.  If the impedance is any less than that, the inductor just shorts out the driving amp, andy higher than that and it never clips (I can only provide 24 db, even when I run my little amp at 21V+- (chip rating is 18V +-). 

I guess I need to test the cores with fewer turns.

But What I did find out so far is IF i compare saturation of cores, to the MPP toroid that is very similar to the pultec toroid, the following cores saturate within 1 dB or so of where the Pultec clone did (20dB through 120 ohm impedance).

These cores produce similar or higher db level saturation point ( but all saturated under 24db on the 21mH tap)
All AL values specified below are as published ( I wound with the actual measured values varied by 10% ).
All of the saturation was measured at 20Hz on the 21mH tap

RM7 core - AL 250 - N48 material
RM8 Core - AL 630 - N48 material
26 x 16 4 slot pot core - H6A3 material - AL400

I think the main thing is to have enough of a gap to raise the saturation point, or a big enough core. 

I have also decided that the saturation point doesn't matter for the pultec clones, anything reasonably high is fine though  because the levels inside the Pultec are really low, and NEVER see anywhere near 20db, and in the HF section where the inductor is used is behind a relatively high impedance input trafo, and the levels are so beat down that it probably doesnt matter.

I will say however that my tests with HI perm cores (AL 3000, 8000, reused computer motherboard toroids, etc) show that they are not good, and distort at very low levels due to saturation.

I do think however that I am done winding toroids.  I am going to use the RM7/RM8 cores, easy to wind.

 

[gyraf]

What frequency areyou running it at? Usually you just use a lower frequency if you can't saturate it.

Jakob E.

 

[bruce0]

20Hz, which is as low as I can go.


I am sort of confused by the results I am getting. 

I tried the schematic with the .1uf cap and the 100K resistor, and I guess current is too low to saturate it the 21mH tap.  I don't think I tried just a few turns on a toroid on this schematic yet.

I tried the other schematic too, and I do get a distorted waveform something like the illustration but the signal source is distorting as well, because it is pretty near clipping when the saturation occurs, and then it gets basically 100ohms slammed across the output (at least that is what I think is happening).


I think i need to try some much smaller inductors to test this out first, because these are too close to the limits of my test equipment maybe.

 

[CJ]

what kind of signal generator do you have?

 

[bruce0]

what kind of signal generator do you have?

Here is a schematic of what I am doing.

I run a computer sound card, or a little BK technologies battery powered SINE/Square wave generator through this, which with +-18V rails does 22 or 23db, there is a meter built into the box too with led "brackets" at 0dB, 4dB, 16dB for easy zeroing in on particular test levels.

It is in a little Vector mod rack, I think there is a picture below.

Note, some of the values may be off... I kind of knocked this up out of one of my balanced receiver/VU cards.  If anyone can tell me how to make it better, that would be great.  it seems to add about 1.5dB of noise to my existing interface (compared to a straight loopback) when I run through two of these interfaces for output and input (the input without the push pull component).

I suppose I might get higher levels or something by using diodes instead of the red led to make it push pull AB. 

b