Simulating Leakage w/ Resistors

[squirrels]

I was reading up about simulating germanium transistors leakage with resistors and after a little experimenting I’m surprised with the results. By putting a resistor between the collector and base of a transistor raises the leakage (or at least gives the illusion of it). Using a GT402B and a 3.3m resistor I was able to raise the leakage from .104 to .322. I didn’t think it would have actually made a difference in the effect but after dropping it into Q3 of my GB88 Greasebox it dropped the collector voltage from 3.9v to 1.9v and it sounded fantastic. The higher voltage didn’t sound bad until the tone knob was about half way up, muffled, splatty and weak.

Anyone else try this or is this a fluke? It looks like the Soul Bender uses the same method to bias Q3 too.

[Sphere80]

I tried this to get some Russian low leakage transistors working. Things got moving a little better than before.  How has the value of the resistor  affected things for you?

 I saw this advice in the AION Tone Bender mk2 instructions and have been trying it out. He recommends testing 1M-2M2 between collector and bases of Q1 & Q3 if they aren't leaky enough.  

 What other circuits have you tried it in?  Maybe a Tonebender Mk1?

[squirrels]

I use the Germanium Tester from the main site to test my Germs. What I was doing was socketing the transistor and holding the resistor between the base and collector and doing the math based on what my multimeter was showing. The smaller the resistor, the higher the leakage and the bigger the resistor, less added/simulated leakage. I think I started with a 470k and wound out with 3.3m. Q3 in the Greasebox wound up with 1.9v on the collector.

I haven’t tried it out on anything else yet but I was thinking of a Mk. I or Mk. II or even a Buzzotron. I’m a big fan on the Mk. III/IV circuit and the Mk. II I made a while back didn’t really sound that great through my OR15.

[Sphere80]

What are some other good circuits that would benefit from this and what positions?  Q1 & 3 in a TBmki/Zonk, Q1 in a TBmkii.

I've seen this change for Q1 on silicon versions of the TBmkii.   Can you post the Soul Bender schem?

[squirrels]

I haven’t really thought about it too much but it seems like it could make more builds work trying out, especially with the Russian Germs. I don’t have a big stock of anything but all I have are low leak transistors including a few 2N1304s that have the right gains for a Mk. II or similar.

[squirrels]

Maybe a Fuzzrite?

[Sphere80]

Sure, that one already uses large resistors from collector to base as part of the design.   The layout on the main page has people reporting using MP38A's in it.  

[nocentelli]

squirrels wrote

I haven’t really thought about it too much but it seems like it could make more builds work trying out,....

I have are low leak transistors including a few 2N1304s that have the right gains for a Mk. II or similar.

Not sure how relevant it is to this discussion (I have never used a Ge transistor), but the skreddy lunar module is a Si MKII-alike, and it has a weird 100k resistor across C-B of Q2 (i.e. Q1 of the "fuzzface" part of a MKIi circuit) which isn't usually there. I always wondered if it was some sort of "leakage simulation" and whether it contributed to the more refined sound of the LM compared to other Si MK2 type fuzzes

[Travis]

Sphere80 wrote

Sure, that one already uses large resistors from collector to base as part of the design.   The layout on the main page has people reporting using MP38A's in it.

I always assumed that was negative feedback or to put a bias voltage on the base

I have tried this simulated leakage trick with a 2m pot in a MKIII and while it worked I ultimately wasn’t liking the sound. I’d say whatever works for you is good but I have preferred to select a part that works without modifications