Suggestion to test: Distortion using transformer saturation

[geertjacobs]

A former colleague of mine had an interesting idea for a distortion pedal. I found it really original at the time (a few years ago) and since he hasn't marketed it since then... 

He wanted to try and use the transformer saturation as a method of clipping:
- Find a small transformer (1:1 isolation or something as in octavia maybe?)
- find out when it goes into saturation
- make input buffer to boost signal into saturation area of transformer
- add output buffer
and that could be about it.

Do you think this would be feasible?

[R.G.]

Unfortunately, no.

It's a great idea if you are doing a fixed frequency. That's the basis of constant-voltage transformers. They saturate portions of the iron path, then pick off the (relatively) constant level of voltage from the saturation drive.

The core inductance saturates at a given level of volt-time product, not volts. So it saturates more and more easily as frequency goes down. That's why 50Hz transformers are bigger than 60Hz transformers - they need more core to not saturate at the lower frequency.

Hysteresis loop distortion is similar, but is also frequency dependent.

[geertjacobs]

Now you mention it: the hysteresis curve was what he was drawing.
It would only work as a very limited frequency booster/attenuator then?

[R.G.]

It would only work below a certain frequency, and all frequencies below that would be more and more distorted as the frequency got lower.

[Elektrojänis]

How about some kind of pre-emphasis and post filtering?

It might need a very crappy transformer or higher voltage power supply than 9v typical in stompboxes. Maybe a self wound transformer on some very crappy iron core. Nail or something... Use a rusty one... Rust probably makes no audible difference... just for style. 

[davebungo]

It must be worth experimenting with - what about mic-pre transformers?  Anyway, to me it might be quite good to have some frequency dependent distortion at the low end and cleaner highs.  If no-one has tried it, no-ne can say what it will sound like - give it a go and let us know what the experience is like.

[R.G.]

By all means, do experiment. Just because things don't work the way you first thought they did is no indication that you won't get something good.

[Jaicen_solo]

It might be easier trying to overdrive an inductor into saturation.
There was talk about winding a low impedance coil around an inductor (in a wah), and passing an amplified signal through the coil.
The theory being that the signal passing through the coil would alter the magnetic bias of the inductor and perhaps pushing it into saturation.
I'm not sure whatever came of it, but it certainly sounds feasible.

[Paul Perry (Frostwave)]

I think there was a suggestion somewhere (?geofex?) about modifying the characteristics of a transformer (or for that matter an inductor) by running a variable DC current thru another winding, varying the DC curent to saturate the core. I guess usign caps to isolate the DC from the audio signals then you could use the same winding.

[Ian]

A former colleague of mine had an interesting idea for a distortion pedal. I found it really original at the time (a few years ago) and since he hasn't marketed it since then... 

He wanted to try and use the transformer saturation as a method of clipping:
- Find a small transformer (1:1 isolation or something as in octavia maybe?)
- find out when it goes into saturation
- make input buffer to boost signal into saturation area of transformer
- add output buffer
and that could be about it.

Do you think this would be feasible?

I was under the impression, from a description in this thread: http://www.diystompboxes.com/smfforum/index.php?topic=38033.0, that the method described above was how the unique sounding distortion of the "tube" output on the Intersound IVP pre-amp was achieved. I'm still a novice at this, so I may have misinterpreted that post, but that was the only example I've found thus far of a guitar effect based specifically on transformer distortion. It's existence would seem to suggest there's something fruitful in this idea.

[Skreddy]

This idea is discussed in the Hertog Simulator? thread.
Here is my graphical idea of how it might work (still untested...) http://skreddypedals.com/circuits/SkreddZog.gif

[Ian]

Thanks Skreddy, that's very helpful. I hadn't recognized the content of that thread from its title during my earlier searches.