Testing AC187 AC188

[stuart_morgan_64]

Towers state gain as 100, when testing the transistors I am reading anywhere from 14 to 120. Can someone explain the reason, and would i need to change the ones reading low gain.

[G6Tanuki]

Measured gain varies with emitter-collector current - bearing in mind that these are 'power' transistors, is your tester pushing enough current (up to 100mA or so) through the device-under-test to make a sensible guess at the gain??

Also remember - these are germanium devices - is your tester competent to test these, or does it assume everything's silicon??

[mhennessy]

If one half of an output pair has insufficient current gain, then you can see that on a 'scope, or measure it with a distortion analyser. If some observations like that lead you to suspect a lack of current gain, then it's time to replace it. Numbers from a transistor tester are an indication at best, and current gain is a decidedly "shift" parameter at best - designers are good at producing circuits that don't rely on a particular value where possible. Usually, a certain minimum value might be required, but after that, it should be reasonably non-critical.

14 sounds suspicious to me. Transistors can fail with low current gain, and germanium transistors particularly can leak heavily, which might confuse the tester. But the acid test is simple: is the circuit doing what it needs to be doing? Simple transistor testers are of limited utility.

[paulsherwin]

Simple DMM transistor test functions aren't good at dealing with Ge types.

[stuart_morgan_64]

Hya, thanks for the response, what I use is a Atlas DCA - Semiconductor Analyser - Model DCA55. I am going through some Philips plastic record players, all seem to use these devices.

[mhennessy]

That is one of the better testers - I have one myself - and it's OK with germanium, but does get caught out occasionally.

Despite knowing that you have a decent tester, I still stick to my earlier advice

[stuart_morgan_64]

Ok thanks, one thing I have read in the manual of the tester, the gain can be very heat sensitive, even body heat, I had just removed two of the leads with a desoldering gun. may account for some strange readings.

[mhennessy]

Yes, current gain varies with temperature (along with sample, collector current, planetary alignment and just about anything else). But it's unlikely that desoldering it would change it permanently...

[jim_jobe]

And where the gain is measured by just having a single resistor between the power source and the base you can get very misleading results caused by self heating of the junction. This is often apparent when trying to measure hFE, using a DVM, of transistors of the OC170/AF117 family. In such cases thermal runaway kicks in and you can't get a stable reading. These transistors are usually quite ok in a stabilised circuit. (This has nothing to do with the tin whisker problem).
Jim