I saw a post today about leakage and temperature, so I thought I’d do a quick test myself to see how close it came. The test distinctly lacked scientific conditions – for starters I wasn’t even wearing a white coat at the time, but the results seem to match the theory.
For germanium transistors, leakage current mainly depends on temperature and doubles for every 10°C rise in temperature.
Measurement #1: taken in my garage / workshop after it had been closed up all day – it’s not very well insulated, so it gets really hot. The temperature was probably about 40 degrees (ambient temperature outside was high thirties).
79hfe, 0.32mA leakage
Measurement #2: Same transistor about an hour later after sitting in very cool air conditioning, I’m guessing, but it’s probably about 20 degrees in the house, so I’d say close to a 20 degree difference to the first reading.
WHAT’S THE POINT?
If you’re working with germanium transistors, try and work in normal conditions (a reasonable room temperature). If you have a circuit sitting just on the edge of gating at 35 degrees, it may not work at 15 degrees. Likewise if noise is a concern at a low temperature due to leakage, it’s going to get way worse when it heats up.