Still having trouble with "real gain"

[FenderTyler]

Seems like i just cant figure this out.  Can someone tell me if i am doing this right.  I hooked up the positive lead of the dmm to the collector on the trans., the negative lead to the - battery terminal, and the emitter to the + battery terminal.  I have a radio shack auto ranging 42 range dmm and set it on mA/A for the test. After a few minutes the display reads 124.1 mA.  Is this number the amount of leakage.  I then hooked up a 1 meg resistor from the base to the - battery terminal along with all the previous connections and after a few minutes the display read 204.5 mA.  After subtracting 124 from 204 i get 80.  Is this number the "real gain" of the transistor?  Am i doing all of this right?

[hellwood]

this is how i do it...
http://www.geofex.com/Article_Folders/fuzzface/fffram.htm

[R.G.]

Read the article at GEO, but yes, you have the idea.

Think about it - current gain is the current that flows in the collector for a given current in the base. If base current =0 (i.e. the base is open) then any current flowing through the collector cannot be actual gain. It is leakage, caused by whatever the transistor does without any base current. If there was a "gain" involved here, it would be infinite - something for nothing. And since there is something out for nothing in, what you get out is not controllable.

If you then add base current, you start into the area of real gain. You're putting in a base current, and collector current flows. DMMs are dumb. They assume that leakage will be zero, so they report 100% of any collector current as gain. They put in a fixed base current and so whatever collector current flows must be from that base current. This is ...correct... for silicon transistors.

But for germanium, there is leakage. So you have to measure the leakage, then subtract it from whatever you get on a real gain test with base current.