Germanium transistors: I need some advice

[agiant]

I got some old Ge transistors and I want to start making my own pedals. I am a total noob.

I got these:

oc71:
leaks=77 Hfe=67.9
leaks=91 Hfe=32.5
leaks=141 Hfe=51.6
leaks=163 Hfe=49.5
leaks=209 Hfe=42.8
leaks=211 Hfe=62.7
leaks=212 Hfe=57.5
leaks=242 Hfe=76.4

oc72:
leaks=192 Hfe=39.4

ac128:
leaks=180 Hfe=75
leaks=183 Hfe=61.3
leaks=194 Hfe=60
leaks=137 Hfe=57.3
leaks=154 Hfe=55.6
leaks=129 Hfe=52
leaks=136 Hfe=50
leaks=130 Hfe=51.5
leaks=147 Hfe=48
leaks=140 Hfe=47
leaks=113 Hfe=37
leaks=130 Hfe=44
leaks=119 Hfe=44
leaks=134 Hfe=35
leaks=198 Hfe=33

I followed the Keen method to measure them (http://www.geofex.com/article_folders/f ... fftech.htm).

In some cases the gain is pretty low, right?

The question: what can I do with them? (especially with the low gain ones)

I'd like to start building some classics: a Rangemaster, a Tonebender MKII and a Fuzzrite.

Are they suitable for these pedals?

Thanks!

[Ben79]

Those are lowish gain for some of the transistors in those circuits but I think there is a way round by putting them in darlington pairs....this multiplies the gain so you'd end up with enormous gain but I think there might be a way round this with some resistors. Someone who knows what they're talking about will be able to shed some more light on this.

[cherler]

You can also use them as diodes for clipping if they're not usable for gain.

[crochambeau]

Cascade stages or plant your low gain Ge after a high gain stage of a different type (it's not like mixing matter and anti matter).

With single transistor circuits the value of the individual transistor are very important; as you add more transistors the individual defers to the group, sort of like people that way.

[culturejam]

Those are lowish gain for some of the transistors in those circuits but I think there is a way round by putting them in darlington pairs....this multiplies the gain so you'd end up with enormous gain

You also end up with multiplied leakage and noise with darlington pairs. Just something to consider.

[eatyourguitar]

the darlingtons can have a low threshold for saturation and also a lot more non-linearity. I would expect a germanium darlington to act more like a comparator than an amplifier. if you like fuzz on your fuzz then darlington is fine. the numbers from the handheld tester are sometimes higher than the geofex method. you can use any method to test as long as you post what method you used. I think you can probably add 20HFE to some of those numbers and relate it to what you see on the forums. I think few people are using the correct geofex method now. most people use special transistor testers.

[agiant]

thanks for the responses!

I followed the Keen method because I read that common multimeters (the ones with transistor tester) do not work well for germanium transistors ... I supposed that everyone were doing the same! It's pretty easy to do it well.

[eatyourguitar]

the RG Keen method is a good standardized method but it is not the only one. therefor I would not say it is "better" exactly. but it is acceptable. the tolerances for the power supply, the resistors, the multimeter, the probes etc... is whatever people have on hand. the Geofex method does not have one specific tolerance unfortunately. a nice piece of equipment with known specs may provide a more accurate standardized test for transistors. how do people obtain regulated 9.0v supplies for the test?

[Liquefactionist]

Remember, Hfe/Beta is a parameter that is only Current Gain of the device, not the transconductance. If you drive it with a lower input impedance, you can get the gain back if the leakage is low because you care about voltage gain not current gain for guitar pedals. So, if you put a silicon follower (or jFET or the like) at the front to drive the lower load, and then go from there...

[eatyourguitar]

This is good information about getting the maximum voltage gain from a transistor by controlling the input impedance. I also want to point out for anyone that is new to learning electronics that the Hfe and the almost interchangeable alpha/beta rating of a transistor does not change with inpedance. It changes with different testing equipment that is calibrated differently. Any machine that can consistently gets the same measurements from the same DUT (device under test) under the same temperature is said to make actual factual measurements of Hfe. The problem is that there are different standardized tests so therefore the data from one lab is not portable to be compared with other data taken from elsewhere. If you regularly verify the data you are given for the transistors that you are buying then you will notice a consistent shift across all the data points between the two labs. This can be used to convert and localize data more quickly without local testing after the chain of trust as been established with your part supplier.