Transistor testing

[Neilriser]

Hi all,please can someone have a look at these figures and give me there advice.Yes ive been through the section on testing/transistors but my figures dont add up? i have 8 x pnp germanium diamond shape transistors [4 from another machine] they all read pretty similar maybe bar one,but not as the test charts states?am i to assume they are all either good or faulty? they i gather are difficult im told [unless some members know ware?]to get hold of as they are matching pairs,and i dont want to discard them if there ok.
[digital used set to 2k]
TEST POL TRAN1 TRAN2 TRAN3 TRAN4 TRAN5 TRAN6
B to E NEG B 591 520 462 534 509 440
B to C NEG B 491 435 380 437 407 339
B to E POS B O O O O O O
B to C POS B O O O O O 871
C to E NEG E 922 435 352 519 644 523
C TO E POS E 315 221 199 236 292 262
C to E and B POS E 480 261 221 296 409 266

http://www.vintage-radio.com/repair-...or-faults.html

[Station X]

No units given, so I assume they're ohms?

The 871 figure sticks out as indicating a fault.

I've found the best way to test transistors out of circuit is to use a DMM's diode test function:-

https://www.vintage-radio.net/forum/...ad.php?t=42194

[vidjoman]

Possibly the best way to test them is to use one of the cheap LCR testers mentioned in recent posts. It will tell you whether good, bad, polarity, gain etc. Doesn't matter which pins you connect to - it'll tell you which is which. Very cheap and they work.

[Neilriser]

Thanks for all replies,still confussed please con you clarifie.Ive been pointed to the post on transistors testing chart on more than one occasion it is a guide ive been following as a struggling to learn beginner,it clearly states that a pnp germaniun transistor can be easily tested on a 2k-ohms setting giving certain results and any different readings make the transistor faulty,however my readings are different from that table and yet they might be good? yes ive got a diode symbol on my digital meter,but doesent tell you to use this? can we please clarifie for new beginners like myself please.At the moment trying to source transistors when i might not have to.Many thanks.Neil

[Station X]

yes ive got a diode symbol on my digital meter,but doesent tell you to use this?

There's a fully illustrated guide here:-

https://www.vintage-radio.net/forum/...ad.php?t=42194

[Station X]

Back in post #1 you're confusing things by quoting readings of 'O' ie leter O not digit 0. If this represents an open circuit then THESE READINGS are good. If on the other hand it represents a short circuit then THESE READINGS are bad.

Carry out this simple experiment. Set your meter to the ohms range, but do not connect any leads. What you see on the display denotes an open circuit. Now plug in the leads and touch them together. The display should change. What you now see denotes a short circuit.

[jonnybear]

I find the easiest way to check, go no go transistors is with a analogue meter set to ohms x 10 range, digital meters I have found for testing transistors usually give inconclusive readings, as you will very rarely get two transistors that read alike.
John

One way to test is put an ohm meter on the collector and emitter and then a slightly dampened finger twixt base and collector, the meter should then shoot up. You can get an indication of gain by trying the same with said finger across emitter and collector. Also this method can discriminate collector and emitter, much more 'shoot up' the right way round.

[paulsherwin]

Most basic DMMs have a transistor hfe test function which provides a reasonable go / no go indication.

[David Simpson]

Folks,
Surely there is someone closer to Bristol with a CT446, than myself(600 miles away) ? For all that Neil requires testing - a wee padded envelope & a couple of stamps would suffice.
Also, some oscilloscopes(Hameg, Wavetek, etc) will visually test semi-conductors.

Regards, David

[dave cox]

The figures for C-E look suspect (ALL), especially the ones with B-E shorted.
I would expect them to read open circuit on a 'typical' 2K ohm range but it will depend on the meter. What type of transistor is this and what meter are you using ?

Matching transistors is not so difficult, especially if you know the electrical environment that they need to be matched for. For small signal transistors it could be done with a battery and a sprinkling of resistors! Of course you need a good number of candidates to start from

I am a little (!) closer to you than David Simpson and could run them through a curve tracer but I would need to know the typical operating parameters ...

dc

[Neilriser]

Many thanks to everyone .Dave happy to drop them to you anytime and iI have service data. How do we tie up? Neil.

[Herald1360]

Matched pairs should have their current gains within about 10% of each other across the likely range of operating currents. Badly mismatched pairs will work, but probably sound fairly awful.

Best you can do with four random devices is pick two pairs as close to each other as possible- hopefully they won't be too mismatched.

If AD149s are cheaper and their voltage/current ratings are at least as high as the NKTs I would go for those and maybe buy a few extra for selection purposes. You should be able to sell on the leftovers at not too much of a loss.