Connect with us

Amplifier died after working for a while!!!!

Discussion in 'Electronic Design' started by Himlam8484, Jul 17, 2007.

Scroll to continue with content
  1. Himlam8484

    Himlam8484 Guest

    Hi people,

    I made a amplifier following the Project 3A ( link is I make some chnages such as
    replace current source to a mirror current source, and added some big
    capacitor at the power supply.

    When it worked, it is really nice, but last about one hour and it
    died. I replace some died power transistor, and test again. I try to
    increase the quiesent current by adjusting the pot VR1, and it broken

    Can someone explain for me? What did it do like that?

    I am looking forward hearing from people soon!!!

    Van Hau
  2. Eeyore

    Eeyore Guest

    Well there are many possibilities but the first thing I'd like to know is how
    you heatsinked the power transistors. They are easily killed by excessive
    temperature rise.

    " If the temperature continues to increase, the heatsink is too small. This
    condition will (not might - will) lead to the destruction of the amp. Remove
    power, and get a bigger heatsink before continuing. Note also that although the
    power transistors are mounted to the board, never operate the amp without a
    heatsink - even for testing, even for a short period. The output transistors
    will overheat and will be damaged. "

    A short circuit on the output would kill that design too.

    I would use MJE15035 rather than BD140 for that design too.

  3. Tam/WB2TT

    Tam/WB2TT Guest

    I would start by measuring the quiescent current in the output and driver
    transistors. Also, see if there is any DC current flowing to the speaker.

  4. nospam

    nospam Guest

    Sounds like thermal runaway. Did you thermally couple Q9 to Q5 and Q6?

    Q9 is supposed to do thermal compensation of the output quiescent current.

  5. Jamie

    Jamie Guest

    for one thing, the VR1 is there to reduce the current to it's minimum
    and most likely used to balance the output to 0 volts with no signal on
    the input while making adjustments.

    You don't show how you have mounted your power transistors?, I do
    hope you are using heatsinks?
    And I hope you are monitoring the heat level ..
  6. Eeyore

    Eeyore Guest

    It's there to *adjust it* to the recommended value.

    Absolutely not in any way at all.

    You do talk absolute crap.

  7. neon


    Oct 21, 2006
    One solution HEAT
  8. Fred Bloggs

    Fred Bloggs Guest

    It most likely burned up with excessive bias through the output
    transistors... This is not a good design, there is no necessity for
    bootstrapping the output stage in a modern circuit...
  9. Eeyore

    Eeyore Guest

    The bootstrapping amused me slightly.

    Without it, the performance would suffer (not that it's anything special
    anyway). I'd use (I DO use) an active load myself of course.

  10. Eeyore

    Eeyore Guest

    It's possible he doesn't have good thermal coupling to the bias setting circuit

    The thermal coupling should only be between Q5, 6 and 9 of course NOT Q7 and 8.
    I see the text makes no mention of this and suggests that no coupling is

    Admittedly there's only 3mA through Q5 and 6 at idle but if they're not
    heatsinked (as the text suggests) they'll get rather hotter with some signal.

    An 'un-heatsinked' TO-220 can only dissipate ~ 2 watts without exceeding Tj max.
    At 35 volts that's only 57mA ! I would definitely heatsink Q5 and 6.

    A poor design throughout.

  11. Phil Allison

    Phil Allison Guest

    "Fred Bloggs

    ** Bollocks.

    ** ROTFLMAO !!!!!!!!!!!!!!

    Slaves to audiophool, voodoo inspired fashion are asinine fuckwits indeed

    The Bloggs congenital, autistic, MORON regularly obliges, on cue - as
    usual .......

    Bowel cancer is too good.

    ....... Phil
  12. John B

    John B Guest

    Perhaps you would have been better to have built and tested the design
    as given before trying to improve it with your own modifications.

  13. Himlam8484

    Himlam8484 Guest

    when i did not connect a speaker, i can not measure the voltage drop
    across power resistor.
  14. Eeyore

    Eeyore Guest

    Why not ? What are you measuring ? I'm doubtful that you're measuring the right

    You need to measure the voltage across R13 or R14 *with no load*. The voltage
    should be about 25 mV.

    Alternatively measure the current in either the positive or negative supply. It
    should be about 80mA.

  15. Jamie

    Jamie Guest

    Look again.

    this isn't my first rodeo .

    That circuit design is so common and old it isn't funny.

    I don't know why I waste my time.
  16. Eeyore

    Eeyore Guest

    I've looked at it in some detail. There is for starters, no way whatever that
    the bias adjustment affects the DC offset as you suggested.

    I have no idea what you mean by that.

    It's certainly 'basic'.

    You certainly shouldn't waste it talking nonsense.

  17. Jamie

    Jamie Guest

    Like I said, waste of space that surrounds you.
    please do others a favor and give them your air, it's being
    wasted on you.
  18. Eeyore

    Eeyore Guest

    Whatever your personal opinion of me, I can totally wipe the floor with you when
    talking about audio for sure.

    I'm pretty damn bloody expert on high power and high performance audio
    amplification btw. It's a little speciality of mine. It might have something to do
    with taking a particular interest in the subject from age ~ 16. There are several
    hundred thousand professional amplifiers of 200 - 1000 W/ch rating out there in
    the big wide world of my design.

    I'm not impressed by a gesturing clown like yourself who mistakes a bias pot for an
    offset adjust.

  19. Jamie

    Jamie Guest

    yeah sure, what ever you say expert.

    You can stop digging any time now and save your energy.
  20. Meat Plow

    Meat Plow Guest

    He is an expert and it shows when ignorant twits like you make mistakes
    then don't own up to them. I too am very impatient with twits like you not
    so much in electronics but in computer hardware/networking/intra-internet.
    Take your own advice.
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