Connect with us

BC857 Destroyed

Discussion in 'Electronic Design' started by [email protected], Sep 24, 2007.

Scroll to continue with content
  1. Guest

    Hello

    I have a circuit with a BC857B where the transister gets destroyed.

    I am a litle out of ideas. When mounted on the board i have seen
    arround 150 transistors out of 1800units where destroyed.

    The transistor in first place seems shorted between collector and
    emitter, the ressistance is between 5 to 250 Ohm.
    The funny part is both "diodes" in the transistor can be measured from
    base. And the one with 250Ohm i have tried to activete in the circuit
    and the transistor seems to react and make lower ressistance between
    collector emitter. (so in fact it looks like a normal working
    transistor with a build in ressitor of 5-250ohm between collector and
    emitter)

    The product has been produced for several years with out problems,
    they started in earlier this year..

    I have got several transistors decapped, so there seems not to be any
    burned inside the transistor and the boundings where ok. I have only
    seen the chip from the top, and collector emitter is in top of each
    other.

    The product is assambled by proofesinel production house and been both
    reflow and wave soldered
    (only wave soldered on the bottom side)

    Another funny things are there are 2 of the same transistor on the
    board, we are only seeing one of them with error so far.

    I have measred the currents in the circuit it seems ok.

    There has been tried to change transistor vendor, and there has been
    seen the same problem, but only 25 of 1800 with same error.

    Does anyone have a auggestion to the error, or seen similar ??

    Best Regards

    Rapzak
     
  2. You have not said anything about the circuit, or the conditions when
    it gets destroyed (on test, or in the field, etc).
     
  3. Phil Allison

    Phil Allison Guest

    "John Devereux"

    ** The OP has described a C-E punch-through failure.

    Invariably caused by C-E overvoltage.

    Wanna bet his dead BC857s are all relay drivers ?

    Whatever, no-one cannot reliably diagnose the issue remotely with a Kraut
    FOOL like him supplying the only info.





    ....... Phil
     
  4. Joerg

    Joerg Guest

    Best to post a schematic and a description of what is connected. Motors,
    relays, solenoids etc.? Did any of those change? Line lengths?
    Environment? Often this kind of stuff is caused by spikes that for some
    reason became larger than they used to be.
     
  5. Kasper

    Kasper Guest

    Hello

    I now the one with circuit, since this is not a private project i can't make
    the schematic public :(

    But yes, it has to do with the relay, but not directly, the driver for the
    relay is BC847 and is protected by diodes.
    (across each relay coil) and no one of the bc847 is destroyed...

    The transistor is sitting in a voltage doubler for making a higher relay
    voltage when the relay activates.
    You charge up a capacitor and add it to the 10V.

    The supply voltage is 10V, so it doubles it to 20V.
    I have measred the current in the C-E to have a peak of arround 40mA.
    And there is BAV99 across each relay.

    The transistor seems to be destryed in the assambly + test, but this is not
    for sure.

    But i want to point out the circuit has been runing in several years and in
    several thousands of the same units without problem, like also to say the
    product has passed EMC/Surge/Burst test...

    I just wondered if anyone else have seen a transistor with these problems,
    since the transistor is not comepletly shorted between C - E like a normal
    burned transistor.

    Rapzak
     
  6. Kasper

    Kasper Guest

    For the schematic i need approvals to do that, but read my other post for 2
    min ago

    Some of them atleast are destroyed in the production house under powered
    test or ESD what so ever...
    (since they are aprroved to ESD things..)

    so no wires length or anything has changed, and as i said it is less than 8%
    there have the error....

    Kasper
     

  7. Yes, I've seen this kind of thing. Probably spot damage which you'd be
    able to see if the microphotography was done properly.

    Show us the circuit, including all the details of the load and the
    driving circuitry. It's almost certainly a design flaw, barring
    something strange like the incorrect or counterfeit components. Are we
    talking about the BC587BS dual? That's not much of a transistor, and
    is probably quite easy to damage.

    Best regards,
    Spehro Pefhany
     
  8. Joerg

    Joerg Guest

    You don't need to post the whole thing, only the stuff around that
    transistor. Just cut it out of a copy, scan and post.

    Did they migrate to a RoHS process with higher reflow temperatures?

    8% is a whole lot.
     
  9. D from BC

    D from BC Guest

    Does the circuit function within safe limits in SPICE?
    The description seems like a circuit that can be accurately simulated
    for diagnosis in SPICE.


    D from BC
     
  10. Kasper

    Kasper Guest

    Hello

    I am just trying to get this running...

    This is not a design i have made, it is 4-5 years old, så i don't know so
    much about the details jet.

    Rapzak
     
  11. Kasper

    Kasper Guest

    Hello

    No, it is a single in sot23...

    I have also seen a report with these spot from dual transistor, but is is
    quite difficult to see between C and E, while they are laying in top of each
    other on the chip, i only have a top view

    Rapzak
     
  12. The relay coils are 12V, 576 ohms, right?

    So if the doubler is energized simultaneously with the 3 relays the
    peak collector current is over 100mA with only 2mA of base drive?

    It's also possible that there's something wrong with your diodes,
    since without a diode, the voltage will spike negative at the
    collector of TR8 (more than enough to break down the transistor) when
    the doubler is shut off. There are single diodes in the same package
    with one connection unconnected. If the BAV70s on failed units have
    other markings than "A4" you should check that out, and check the
    diodes electrically in any case.

    Best regards,
    Spehro Pefhany
     
  13. Eeyore

    Eeyore Guest

    If you expect some help you'll have to change that attitude.

    A 2 transistor circuit isn't going to be 'secret' enough to matter. The very
    idea is a joke !

    Graham
     
  14. Kasper

    Kasper Guest

    No 18V 1276 Ohm for 1 and the 2 other is 1620 Ohm.
    Also the software locks for not activating all 3 releays at once.
    Then we could discuss just when the uC starts up, all 3 + boost is
    activated... hmm...

    The allowed peak current is 200mA.
    The biggest problem with that are we have seen problems back to last year,
    so the problem is made over several production lots.
    But we have ofcause checked the diodes on several bad units, and they
    measured correctly.

    Rapzak
     
  15. Eeyore

    Eeyore Guest

    I've seen diodes like the BAV99 destroyed by flyback energy from a relay coil.

    Replace it with a 1N400x.

    Graham
     
  16. Eeyore

    Eeyore Guest

    What you need to do is to discover the fault condition.

    Comparing your blown transistor with other blown transistors will tell you
    NOTHING of much value. It's an absurd idea.

    Graham
     
  17. Kasper

    Kasper Guest

    This is small PCB mounted relays with high ressistance, so i doubt that is
    the problem.
    Any way they have not been broken on any of the verified devices.

    Rapzak
     
  18. Kasper

    Kasper Guest

    If they have found the source to the problem, this could be the same
    problem, so i don't see that as stupid.
    This is also because that i believe it is not a common way the transistoer
    is destroyed do to the functionality it has afterwards.

    Rapzak
     
  19. Eeyore

    Eeyore Guest

    You're 100% confident no diodes have gone open circuit ?

    Graham
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-