Connect with us

Testing a NPN Darlington Transistor

Discussion in 'Troubleshooting and Repair' started by ADRT, Sep 27, 2015.

Scroll to continue with content
  1. ADRT

    ADRT

    192
    32
    Nov 25, 2014
    I'm testing a NPN Darlington transistor to see if it is OK. Transistor is for an electronic ignition system on an ATV. Transistor marking ETD41-035. I'm using a Fluke 179 true RMS meter on the diode test setting. Here are the readings I'm getting.

    B-C: .543 vdc
    B-E: .639 vdc
    C-B: infinite
    E-B: 1.54 vdc

    So going by the test procedure I should be seeing infinite E-B correct?
     
  2. dorke

    dorke

    2,342
    665
    Jun 20, 2015
    Yes.
    The Darlington is defective.
     
  3. dorke

    dorke

    2,342
    665
    Jun 20, 2015
    Unless...
    It is not a "pure Darlington".i.e has internal B-E resistors ,TIP120 like.
     
    ADRT likes this.
  4. ModemHead

    ModemHead

    69
    20
    Dec 2, 2010
    The readings would be consistent with a functional Darlington that has bias resistors and a CE protection diode. Random example (2SD2014):
    Darlington.png
     
    ADRT likes this.
  5. ADRT

    ADRT

    192
    32
    Nov 25, 2014
    Unfortunately I can not find the datasheet or any information. Just found a thread on another forum in which the gentlemen stated the same, but was given an alternative transistor to replace with. The replacement worked in his case, and was also being used in an electronic ignition system.
     
  6. ADRT

    ADRT

    192
    32
    Nov 25, 2014
    Personally I think the ECM is fine. I helped my cousin troubleshoot a similar problem with his ATV and it turned out to be the stator. This one was removed by a colleague and handed to me to try and fix. I can't find anything wrong with it.
     
  7. Old Steve

    Old Steve

    734
    169
    Jul 23, 2015
    Just in case it does turn out to be bad, MJE5740 or MJE5742 are recommended replacements, according to this:-
    http://www.hondashadow.net/forum/908215-post12.html
     
    ADRT likes this.
  8. dorke

    dorke

    2,342
    665
    Jun 20, 2015
    Since your measurements revile there is no C-E reverse diode.
    A better equivalent replacement should be this one :
    ST901T
     
  9. Old Steve

    Old Steve

    734
    169
    Jul 23, 2015
    ModemHead said that there is a reverse diode.
     
  10. ADRT

    ADRT

    192
    32
    Nov 25, 2014
  11. dorke

    dorke

    2,342
    665
    Jun 20, 2015
    ADRT,
    good find,but...
    In this case your original darlington can actually be good.
    Best thing would be to build a simple test circuit for it .
    like this:
    closing the switch should light the led(a 20ma led) .

    Voltages on the darlington should be about:
    VCE=0.9V
    VBE=1.4V

    Darlington tester.jpg
     
    ADRT likes this.
  12. ModemHead

    ModemHead

    69
    20
    Dec 2, 2010
    My bad, I used weasel words (like a politician) "is consistent with".

    Strictly speaking, no E-C measurement was included, and even if there was, with bias resistors, a diode, and the B-C junction, there would be two current paths from E to C, so it would be hard to *absolutely* confirm the presence of the diode.
     
    Old Steve likes this.
  13. Old Steve

    Old Steve

    734
    169
    Jul 23, 2015
    I was going to add that a C-E test in both directions was a good idea.

    And note my weasel-words- I didn't say "there is a reverse diode". Instead I quoted you. :D
     
    ModemHead likes this.
  14. dorke

    dorke

    2,342
    665
    Jun 20, 2015
    ModemHead,
    You are correct ,there is no C-E measurement!
    I jumped to the conclusion(Freudian?) that the "infinity" was of E-C :oops:.

    ADRT,
    can you please test the E-C bidirectionally and post?
     
    ModemHead likes this.
  15. ModemHead

    ModemHead

    69
    20
    Dec 2, 2010
    Thinking further (a sometimes dangerous thing), if the diode is *not* present, an E-C diode-test reading should be above the reasonable range for a diode forward voltage, due to the bias resistor drop in addition to the B-C junction drop. (A Fluke 179, like many DMMs, uses an approximate 1mA test current.) So a "reasonable" diode forward voltage reading in the E-C direction should confirm the diode is there.
     
  16. ADRT

    ADRT

    192
    32
    Nov 25, 2014
    So, E-C: .525, C-E: infinite. So there is a diode.
     
  17. ADRT

    ADRT

    192
    32
    Nov 25, 2014
    I'm going to try this test circuit this evening just to cover all my bases. Personally I hate working on the components of a machine without actually testing it in place myself. I think he has other problems.
     
  18. ADRT

    ADRT

    192
    32
    Nov 25, 2014
    So tested this in circuit and it tested fine. Talked to the gentleman about it and sure enough he actually didn't check the stator coils as I suspected. Thanks for all the help guys. :)
     
  19. dorke

    dorke

    2,342
    665
    Jun 20, 2015
    Your welcome,it's our pleasure :)
     
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

-