Testing AC128 Transistors For Leakage and Gain

Now that I have a test rig set up, I can start to measure some transistors.  I went through my bag of Tungsram AC128s and started to number them so I can keep track of the good ones.  If there are any.

Here are the first 25, lined up like little soldiers.

Searching the interwebs, I found a schematic for a germanium buffer to build using leaky transistors.  I suspect I might be building some of those too.

The process is pretty simple.  Put a transistor in the socket on the tester.  Note the voltage across the collector resistor.  I called this the 'static' voltage since there isn't any current flowing through the circuit.  If you take that figure and divide it by the value of the collector resistor (the 2472 ohms one I cooked up), you get the leakage in microamps.  Basic ohm's law.

Then we switch the tester on and measure the voltage again (across the resistor).   Note that voltage and then subtract the static voltage from that.  Multiply by 100 and we have the transistor's gain.

One of the transistors under test here.  You can see the DMM leads connected across the collector resistor. 

I started keeping a chart of the results.  I realized I could use Excel for this, with some formulas to calculate the leakage and the gain, which I had started to do manually. 

The lowest leakage I've seen so far is about 155 microamps.   R.G. Keen says 100 is good, that 200 is seen "quite frequently" and that 300 is "suspicious."  Above 500 is a bad transistor.

I'm seeing a lot right around 200, a couple over 300, and a few under 150.  Not too bad.  I figure I'll just try some in the Red Pepper and see how it sounds.

What concerns me more than the leakage is the fact that all of my gains are less than about 72.  The prime gains for a Fuzz Face are about 70 for the first transistor (Q1) and about 100-120 for Q2.  I'm wondering if I will get any that are close to 100.  We shall see.