Bipolar Junction Transistor Testing Basics

Discussion in 'DIY' started by EchoWars, Jul 15, 2005.

  1. marloubow

    marloubow Best Day Ever Subscriber

    Messages:
    903
    Location:
    Enid, Oklahoma
    Thanks EW
    The TO-3's in a KLH model 52a I'm working on are A'ok.
    Hoping a recap is all it needs.
    Nice to learn something practical today.
    martin

    :thmbsp::thmbsp::thmbsp:Yes...this should be a sticky:thmbsp::thmbsp::thmbsp:
     
  2. marloubow

    marloubow Best Day Ever Subscriber

    Messages:
    903
    Location:
    Enid, Oklahoma
    Oh! It is a sticky.
    I found this thread through a roundabout search.
    Well...good!
    Martin
     
  3. roger2

    roger2 . Subscriber

    Messages:
    2,893
    Location:
    indiana
    2sc3860 testing

    Will a standard diode test work for this transistor?

    I am getting "OL" on all six connection combinations. Is this because the transistor is bad or because my diode tester is unable to work through the 10k resistance?

    My first time to encounter this type of transistor. For those, like myself, not familiar with this it has a 10k resistor in series with the base. The datasheet refers to this as "on chip bias resistance".
     
  4. EchoWars

    EchoWars Hiding in Honduras

    Messages:
    28,783
    Location:
    Kansas City
    Good question. Try a known good transistor and after confirming that you can verify its status, add a 10K resistor to the base and see how it tests.
     
  5. roger2

    roger2 . Subscriber

    Messages:
    2,893
    Location:
    indiana
    Good answer, thanks. Should have thought of that myself :bash:

    Will report back later tonight.
     
  6. OverLoad

    OverLoad High On Solder Fumes

    Messages:
    1,377
    Location:
    Ontario, Canada
    I predict that with that high of a base resistor the diode checker can't cut the mustard.
     
  7. roger2

    roger2 . Subscriber

    Messages:
    2,893
    Location:
    indiana
    OverLoad was right. Apparently my diode tester will not work with 10k ohms in series with the base as it is in the 2sc3860.

    I have some extra 3.3k resistors so I twisted three of them together in series and the end of one of them to the base of a ksc1845. This rig gave me 4 convenient test points - 0 ohms, 3.3k, 6.6k, 9,9k.

    With no resistance in series with the base of the ksc1845 my tester showed .68v B-C. With 3.3k in series the tester worked but the B-C voltage was 1.96v. No go with either 6.6k or 9.9k in series with the base, got the "OL" reading just like on the 2sc3860.
     

    Attached Files:

    Last edited: Mar 5, 2013
  8. roger2

    roger2 . Subscriber

    Messages:
    2,893
    Location:
    indiana
    EchoWars ...I did some further testing on that 2sc3860 and would be interested to hear your thoughts.

    Since my battery operated DMM based diode checker was unable to give any results, I thought why not try a different instrument, i.e. my BK 878 LCR meter. Maybe the LCR meter outputs a stronger test signal than the DMM? In any case I did get some results testing resistance on the 1M ohm scale of the LRC meter:

    R B kOhm
    B C 580
    B E 615
    C E 55

    B R kOhm
    B C 200
    B E 620
    C E 65

    R and B in the heading indicates the meter lead. I was surprised to get readings in both directions. The kOhms numbers are approximate, the readings would not settle in on an exact number, and the alligator clips on my leads are kind of sloppy so taking the readings was not easy. But I did double check most of them.

    Would it be correct to imply from these readings that the on-chip base resistor of the 2sc3860, as well as all three of the transistor sections, are at least functional, i.e. neither open nor shorted? Or is that crazy talk?
     
    Last edited: Mar 5, 2013
  9. Bethlehemtom

    Bethlehemtom New Member

    Messages:
    40
    Location:
    Bethlehem, Pa.
    Thank you EW for for all the effort put into this topic.
    There is a book that I purchased from Radio Shack when I was maybe 18 that some may find useful. The title; "Getting Started In Electronics" by Forrest M. Mims. First published in 1983 then 2000. The book is in paperback/magazine format, black and white and gives a good basic education of how all the different electronic components work and influence the world that they live in. Plus it teaches the reader to read a simple schematic along the way. At the end of the book there is a section of 100 simple electronic projects that you can DIY, all via schematic.

    tom
     
  10. honestsoul99

    honestsoul99 New Member

    Messages:
    1
    Location:
    Gary, Indiana
    "Getting Started in Electronics" book can be found in PDF form in "alt.binaries.e-book.technical" newsgroup
     
  11. VintageWacko

    VintageWacko Member

    Messages:
    88
    Can someone please tell me if I need to remove the component from the PCB in order to do the tests? I've been asking on other threads but not getting a clear answer. My problem right now is quite a few of my readings check out except.

    For NPN D880, I'm getting a 1.2V from Collector to Base instead of reverse bias
    For NPN D880, I'm getting a 0.6V from Collector to Emitter instead of reverse bias

    For PNP B834, I'm getting a 1.2V from Base to Collector instead of reverse bias
    For PNP B834, I'm getting a 0.6V from Emitter to Collector instead of reverse bias

    Thanks.
     
  12. mbates14

    mbates14 Well-Known Member

    Messages:
    723
    I would remove the transistors and test them again. Sounds like they are leaky or partially open.
     
  13. hirscwi

    hirscwi AK Subscriber Subscriber

    Messages:
    1,526
    Location:
    Indianapolis
    Yes, definitely need to remove from the PCB for this test to work properly, otherwise the rest of the circuit distorts the result.
     
  14. VintageWacko

    VintageWacko Member

    Messages:
    88
    Testing transistors

    Thanks guys. Response much appreciated. I saw some "Kits" on the web that claimed to be able to test transistors in circuit, but I'm going to drop that idea.
     
  15. kenwood61

    kenwood61 KR-4070 started it all Subscriber

    Messages:
    4,852
    Location:
    Bend, OR
    If I understand this, no matter what NPN or PNP transistor you are measuring, you want to see a value in the 0.45 to 0.65V range?

    I just pulled a pair of 2SD381 and 2SD382 NPN transistors and followed your test (above). Mine are about 0.57V with each pair almost dead on. I was worried about these since they are on the power supply board of my Sansui BA-3000 amp. During testing, the cement resistors and one of these 382 transistors (closest in circuit to the big resistors) all got hot. This was BEFORE replacing caps, and the VD1212 diodes, which I'm in the middle of. So no telling what else I'm replacing contributed to this. My main concern now is just making sure I can define any faulty parts I missed on my first part order.

    Other question I hope someone can answer. These two pairs of TO220 transistors are the only ones in my amp that do NOT have thermal paste or insulators between the metal backs and the heat sinks. Is this normal??
     
  16. EchoWars

    EchoWars Hiding in Honduras

    Messages:
    28,783
    Location:
    Kansas City
    A diode test is not a definitive test...only an indicator that the transistor has a 'better than average chance' of being fully functional. I'd say the diode test is good to catch about 65% of failures.
     
  17. kenwood61

    kenwood61 KR-4070 started it all Subscriber

    Messages:
    4,852
    Location:
    Bend, OR
    The amp was working on the good channel. Having a bad left channel may have caused the overheating. The parts the blew up were on the bad channel driver board.
     
  18. kenwood61

    kenwood61 KR-4070 started it all Subscriber

    Messages:
    4,852
    Location:
    Bend, OR
    Measuring diodes in series

    Thanks for the thread Glenn!

    I have a question about measuring diodes in series. Specifically, 1N4148 in series to replace VD1212 diodes. When I measure a single 1N4148, no problem. But when in series, I can only get a measurement on the leads for each single diode. When I try to measure across both, I get "OL". Since the measurement for each diode is good, I know something else is up. Can you explain?

    Thanks!
     
  19. petehall347

    petehall347 the brandy coffee man Subscriber

    Messages:
    23,138
    Location:
    uk.. the middle bit
    its because your meter hasn't got the voltage / power to do it ..you can make a simple circuit with a 9v battery and the correct value resistor to limit the current then measure voltage drop over the diodes .
     
  20. kenwood61

    kenwood61 KR-4070 started it all Subscriber

    Messages:
    4,852
    Location:
    Bend, OR
    Thanks. Short of doing that to be 100% sure, is it a safe assumption that if the mechanical/solder connection is solid, and each diode is measuring correctly . . . that the diodes in series are good to install?
     

Share This Page