Ge transistor gain prediction for FF

[Steben]

The reason I mention this is to find a thumbrule in determing the real gain of a fairly leaky device such as a germanium transistor. In order to get a FF going I don't think a specialist's work is needed.
Not all are leaking 300µA...

We all know 70 for Q1 and 110 for Q2 are great gains.
I mean: can you select a bunch of trannies in the range of 80-100 for Q1 and 120-140 for Q2 and start plugging them in, knowing that the real gain is lower?

Second question as I write this: aren't these values especially chosen to work within the stock FF resistors? I mean: does a FF need these values to really sound smooth and sweet, even if you have trimmers to bias the lot? In other words: Is a fine well-biased FF only "fantastic" with the 70-110 trannies?

[smallbearelec]

can you select a bunch of trannies in the range of 80-100 for Q1 and 120-140 for Q2 and start plugging them in, knowing that the real gain is lower?

You should be screening them for leakage first, so that the gain figure is a "real," adjusted gain. As long as the leakage is under 300 microamps, they should then bias easily. See either GEOFEX's "Technology Of" article or the Fuzz Face FAQ in How Tos on my site, smallbearelec.com.

aren't these values especially chosen to work within the stock FF resistors? Is a fine well-biased FF only "fantastic" with the 70-110 trannies?

Yes to the first, no to the second. Though the character of a pair using high gain devices (e.g. 110/190) is different. Time to breadboard...

SD

[brett]

Hi
my "pair" measure 140 and 140.  They were picked by ear from about 30 devices with very low leakage (mostly 50 to 150mA).
Funny thing is that mine doesn't sound particularly high-gain to me (but I suppose that I mostly use it with the gain control at "noon" to "two o'clock", whereas other people report using them full bore).
I'd summarise things like this: Magic numbers?  What magic numbers?