Connect with us

PNP Beta

Discussion in 'Electronic Design' started by Paul Burridge, Aug 18, 2004.

Scroll to continue with content
  1. Maybe it's just coincidence, but on testing a few dozen spare trannies
    I have yet to sort out properly; all the PNP ones seem to exhibit
    *far* higher gain (tested on a DVM with that facility) than the NPNs.
    Do PNPs typically have higher gains and if so, why?

  2. No. It usually is the opposite. I haven't seen a PNP with hfe more than
    1000, but there are a number of NPNs that go well beyond.
  3. CFoley1064

    CFoley1064 Guest

    Subject: PNP Beta
    Some of the DMMs with hfe testing have a bit of an offset problem with the
    current sources. That would account for what you're observing, I'd guess. If
    you've got a means of measuring base current using another meter, you might
    want to check. Also, you might want to breadboard an NPN and PNP to derive
    hfe, and compare with the meter results.

    Good luck
  4. Active8

    Active8 Guest

    Why is this troll still designing, I mean hacking, with beta?
  5. Fred Bartoli

    Fred Bartoli Guest

    Hey, he's not hacking but "freaking".
    BTW, beta is a french noun meaning something like silly billy.
  6. I wonder why you lightweights would assume from my post that I was
    designing around beta? I bought a bunch of mixed NPNs and PNPs from
    some guy on ebay. I had reason to believe they might have been zapped
    in a microwave so sticking a few of them in the DVM was the speediest
    way to ascertain their functionality. Whist doing so, the pattern
    between betas for the two types became apparent. HTH.
  7. Paul Burke

    Paul Burke Guest

    So beta testing is a way of making your code idiot proof?

    Paul Burke
  8. Active8

    Active8 Guest

    I thought the term came from the fact that it was the next step
    after alpha (in-house) testing. Thanks for clearing things up ;)
  9. Active8

    Active8 Guest

    Assume my ass. We *know* you're a lamer.
    You mean your troll.
    and still *are* a hack.
    But you *can't* design period, let alone the right way.
    I just add catsup and taste test them. Usually a wide sigma clues me
    Beta patterns for an alpha moron.
  10. I thought the "in" thing was concurrent alpha and beta testing.

    Best regards,
    Spehro Pefhany
  11. I thought the latest was "revenue alpha test". VHS killed off beta a
    long time ago.
  12. Tim Shoppa

    Tim Shoppa Guest

    What are the part numbers? What betas are you seeing? Are you sure
    you don't have some Darlingtons mixed in? Beta is a very loose
    quantity (varying by up to a factor of 10 for a single 2N part number)!

    If what you have are factory rejects (common on the surplus market even
    today - I went through that phase in the early 70's and even then it
    was a waste of my time and money - Poly-Paks!) you might have
    gotten the too-low-beta rejects from the NPN bin and too-high-beta
    rejects from the PNP bin...

  13. These don't appear to be rejects, Jim. Their betas are all within
    surprisingly close tolerances. None of the PNPs vary by more than +-12
    at 800, for example. The NPNs are similarly quite close to each other,
    though we're only talking around an average figure of 120 for those.
  14. Tim Wescott

    Tim Wescott Guest

    Are you sure your PNP's aren't darlingtons? Have you measured the
    base-emitter drop? 800 is a bit low for a darlington, but it's way high
    for an NPN.
  15. Tim Wescott wrote...
    What a conversation, trying to generalize on semiconductor
    properties from two sets of no-name eBay auction winnings.
  16. I read in that Winfield Hill
    Sometimes it's interesting to see how much you can 'Sherlock' out of an
    apparently 'insufficient data' problem. In this case, a report of the
    Vbe and Ic values during the beta tests would be helpful.

    My original guess was that the tester was faulty. Maybe the NPNs pin-
    outs are backwards (C and E swapped), so that the reverse beta is being
    indicated, but magnified 10 times by the fault.
  17. Tim Wescott

    Tim Wescott Guest

    Hey, I'm just trying to cut through the poor guy's befuddlement, god
    knows why.

    If I wanted to know the answer to the "PNP Beta" question I'd just look
    in the Digi-Key catalog and find out that among the jelly-bean
    transistors "general purpose" units have betas of around 100, "power"
    units have significantly lower beta, etc.

    Then I'd look at some modern transistor manufacturers (Zetex comes to
    mind) and see what they sell for "high beta" units in both NPN and PNP.
    I'd find betas approaching or exceeding 800 for the NPN's (I've looked
    before), but I have no idea what I'd find for PNPs.
  18. Tim Shoppa

    Tim Shoppa Guest

    Do the PNP's have a part number of any kind on them? They seem likely
    to be Darlingtons (you can check their Vbe drop.) Or they're
    leaky or have some kind of built-in bias resistors which is screwing up
    your DMM's measurement :)

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