Diode Questions

[selfdestroyer]

I recently got a huge bulk of glass diodes and some have color bands and some do not, some have text on them and others don't. One of the first things I wanted to do was sort these by SI & GE and wondering if the following is good to follow:

For silicon diodes, the typical forward voltage is 0.7 volts, nominal. For germanium diodes, the forward voltage is only 0.3 volts.

I do have a Peak DCA Pro that I will use to read the forward voltage to help sort.

For clipping, what other factors are important other than forward voltage? I am reading up on diodes tonight but most articles I find are not necessarily geared towards audio clipping.

Last one, I am a still little confused on the Unicorn Tears syndrome. If a pedal uses typical symmetrical clipping  and uses some super rare Russian GE diodes with a Vf of 0.321 and I find a run of the mill GE diode with a Vf 0.321 will it clip the same? Do each diode "type" have its own characteristics?

I am sure there is a lot of "opinions" and best practices and I would love to hear them.

Note: I am not done absorbing these..
What I have been reading through while I type this:
http://www.diystompboxes.com/pedals/diodes.html
http://www3.eng.cam.ac.uk/DesignOffice/mdp/electric_web/Semi/SEMI_3.html
http://www.geofex.com/effxfaq/distn101.htm

[mgwhit]

Jon's diode thread is the best one around here.

[midwayfair]

My diode thread was not the most scientific thing ever (in fact, it was very unscientific!), so I'll add some thoughts here:

Several things MIGHT matter, but the four that will probably make a difference are:

1) Fv - this is by far the thing that matters most. You will be hard pressed to tell the difference between two diodes of the same Fv in the same circuit. Most of the time, this is all you care about.

2) Resistance - this is the next important thing. Measure the resistance across the diode. Silicon diodes will typically be very high -- several, or even dozens of megaOhms. I have some germanium diodes that measure very high, over a meg (OA126 measure ~7M, for instance), but many more that measure very LOW (I have some 1N192 that measure ~120K, more common is about half a meg). There were a couple circuits, like my Fallstaff boost, that had a change in the bass content when going between diodes in situations that I thought really shouldn't matter, and it took me a while to figure out that this was probably what was going on. The resistance could also explain why sometimes two germanium in series sound different from a single silicon diode.

3) Knee - this is how quickly the diode conducts. It's another one of those things that is legitimately different about diodes when Fv is otherwise the same. Silicon should conduct faster than germanium. I can't remember if LEDs are slower than germanium or just slower than silicon.

4) Miscellaneous unpredictable behavior -- Germanium has leakage, so it will sound different in hot weather than cold. The Fv rises with heat (this is how you test if a diode is actually germanium, BTW). LEDs are photosensitive on their own, which means that if you have multiple LEDs in a circuit, you could hypothetically affect the Fv of one with the light from another. I've actually seen a (hilarious) FET biasing circuit that used LEDs' photosensitivity to (poorly) auto-bias a FET. I'd also toss in reverse conduction here, but that's predictable (and useful, in the case of Zener diodes).

[selfdestroyer]

Jon's diode thread is the best one around here.

Thanks, I actually quoted Jon on some of his findings on my blog months ago since I found it very helpful on "sounds" of different clipping diodes. So needless to say I have seen this post and it has helped give me an idea what different diodes sound like.

Jon, thanks for the breakdown and food for thought. I will continue to read up and breadboard this week. I figured there were other "properties" that needed to be considered, I just had no idea what they were.

Thanks again guys

Cody

[nieradka]

Also, sometimes my peak dca gives weird readings for fv on germanium diodes, giving much higher fv than expected (like some models of ge diodes with fvs of .56). When reading with a multimeter, i get expected values (such as .26 for the same diode). Id use a multimeter for sorting.

[selfdestroyer]

Also, sometimes my peak dca gives weird readings for fv on germanium diodes, giving much higher fv than expected (like some models of ge diodes with fvs of .56). When reading with a multimeter, i get expected values (such as .26 for the same diode). Id use a multimeter for sorting.

Thanks, I will do some comparison tonight.

Cody