What do you see in this distorted sine waves?

Discussion in 'Solid State' started by elnaldo, Aug 9, 2017.

  1. elnaldo

    elnaldo Addicted Member

    Messages:
    6,436
    Location:
    Buenos Aires GMT-3
    Hello. After some weeks of use, a Scott 460-A amplifier started to develop intermittent distortion in one channel. I suspect the differential pair transistors I've replaced (probably fake or just a bad one), but I'd like to know what you experienced techs see in this distorted waves:

    All the screen-shots taken from headphone out, with no load, and loaded with 8 ohm speakers. When loaded the distortion is very hard, but after a while, it reduces a lot. (that makes me suspect from some output transistor too)

    1st pict, 1KHz sine-wave, no load at the speaker output (speaker set OFF from the selector). Scope at the headphone out.

    pic_650_1.gif

    2nd pict, same signal, same level, just enabled the speaker set from the speaker selector. A real speaker connected, not a dummy load. Scope at the headphone out.
    pic_650_2.gif

    3rd pict, same as 2nd, just after some seconds. Scope at the headphone out.
    The distortion gets better or worst , not steady, perhaps you turn the volume up or down, and then it's gone, then it comes back, and so. Yesterday the unit played fine all the afternoon. Bias was OK 2 weeks ago, even if now it shows that notch at the zero point.
    pic_650_3.gif
     
    Last edited: Aug 9, 2017
  2. Binkman

    Binkman Addicted Member

    Messages:
    5,827
    Location:
    Raleigh, NC
    bump.
    :D
     
  3. petehall347

    petehall347 the brandy coffee man Subscriber

    Messages:
    23,137
    Location:
    uk.. the middle bit
    i would be looking for poor connections . maybe a fuse holder or dry solder joint .
     
  4. bradfro

    bradfro AK Subscriber Subscriber

    Messages:
    108
    Location:
    Eden Prairie, MN
    How about using freeze mist on components to see if you can get a solid failure
     
  5. ev13wt

    ev13wt Super Member

    Does the headphone jack have its own amp? Just a big resistor?
     
  6. dr*audio

    dr*audio Fish fingers and custard! Subscriber

    Messages:
    18,028
    Location:
    Cleveland, Ohio
    Loaded it's missing the top half of the waveform, so that means the positive half of the amp lacks current drive. Not the diff pair. Start with checking the resistors in series with emitters and collectors of drivers, outputs and pre-drivers. If all ok, test hfe of all transistors.
     
  7. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    31,853
    Location:
    Southern NJ
    My Sherwood amp will do something much like this if you lose a fuse. Its not a split supply, but it has a resistor across the fuse holder. You end up with a little bit of the negative part of the wave and no positive side. If its going open randomly, bad connections are a good possibility. If this is split rail I'd be looking especially at the + side. Could even be a cracked trace on the board that opens randomly.
     
    Retrovert likes this.
  8. elnaldo

    elnaldo Addicted Member

    Messages:
    6,436
    Location:
    Buenos Aires GMT-3
    I'll check soldering at those points first
     
  9. Retrovert

    Retrovert AK Subscriber Subscriber

    Messages:
    1,845
    Location:
    New York City, NY
    Yes, certainly sounds like a thermally induced failure that could be identified by cooling.

    Another option besides the output devices is current through a cracked or otherwise damaged resistor (my addition), bad trace (as above noted), or bad solder joint (as above noted) which heats it, causing expansion and then failure. As the device or contact (whichever one it turns out to be) cools it contracts and the value returns to something more normal.

    This can arise after thermal cycling some number of times. Hairline cracks worsen and eventually fail.
     
  10. Alan0354

    Alan0354 Super Member

    Messages:
    1,742
    Location:
    Bay Area
    Do you mean this happens only AFTER you replaced the differential pair transistors?

    The first picture looks like oscillation on the sine wave. I cannot explain the second and third picture from the differential transistors.

    Have you try tapping the components on the circuit board around the output section to see if there is any cold solder or crack components?

    Particularly take closer look at the bias spreader.

    I can't find the schematic.
     
  11. sregor

    sregor AK Subscriber Subscriber

    Messages:
    3,015
    Location:
    Mass
    Looks (to me) like intermittent output transistor (NPN). Freeze spray would probably verify. Could also be cold solder. My 2 cents.
     
  12. dr*audio

    dr*audio Fish fingers and custard! Subscriber

    Messages:
    18,028
    Location:
    Cleveland, Ohio
    This is more of a parasitic oscillation or the transistor going in and out of it's active region. This and the clipped waveform makes me think there is intermittent drive in the positive half of the amp.
     
    smurfer77 likes this.
  13. elnaldo

    elnaldo Addicted Member

    Messages:
    6,436
    Location:
    Buenos Aires GMT-3
    I'll verify mechanical problems and check the positive half driver and output transistors for a fail.

    I mentioned the diff. pair as it was the only semiconductor replaced. I had problems also to set bias at 1st, suspected the stv-4h diode, but it worked fine after resoldering. Perhaps it was this same intermittent issue.
     
    Last edited: Aug 13, 2017
  14. Alan0354

    Alan0354 Super Member

    Messages:
    1,742
    Location:
    Bay Area
    I just don't think the diff pair is the problem. Do the tapping, measure the bias spreader voltage and report back.
     
  15. dr*audio

    dr*audio Fish fingers and custard! Subscriber

    Messages:
    18,028
    Location:
    Cleveland, Ohio
    Usually this is caused by an open resistor in series with the emitter or collector of a driver transistor.
     
    Retrovert and elnaldo like this.
  16. elnaldo

    elnaldo Addicted Member

    Messages:
    6,436
    Location:
    Buenos Aires GMT-3
    Here is the schematic: https://www.dropbox.com/s/9q2jzhkhec1ro5t/hfe_hh_scott_460a_schematic.pdf?dl=0

    Is Q9 11 13 15 a current sensor circuit, or just a feedback? ... I understand those transistors can reduce the bias to the drivers.

    Are those also suspicious parts? I think fail in that circuit could cut the bias to the drivers.

    Anyway I'm checking this in a couple of days, starting with the suggestions already posted, I think I have enough information by now to identify the problem.
     
  17. Alan0354

    Alan0354 Super Member

    Messages:
    1,742
    Location:
    Bay Area
    Seems like the schematic has error. If you look at base of Q9 and Q11. Under normal operating condition, Q13 and Q15 is off and there is no path to provide DC base drive to Q9 and Q11.

    That said, looks like Q9 and Q11 are bias spreader. Q13 and Q15 are from current limiting. Q13 shut down the top half ( Q17 & Q21), Q15 shut down the bottom half (Q19 & Q23). If your wave form looks like the top half is cutting off, problem might be Q9 and Q13 related as Q13 shut down Q9. I cannot be 100% sure as the schematic on this part doesn't make sense.

    This is a strange bias spreader.
     
    Last edited: Aug 14, 2017
  18. elnaldo

    elnaldo Addicted Member

    Messages:
    6,436
    Location:
    Buenos Aires GMT-3
    SEEMS SOLVED BY NOW :banana:

    Just "deep resoldered" (solder wick + removing resistors + cleaning solder pads) the driver and output transistors and emitter resistors on the positive half.

    Now, tapping on the board and relay, freezing components, turning volume up and down, have no effect on the output signal.

    I'll try it for a couple of days before returning it to the owner .
     
    smurfer77 and redk9258 like this.
  19. Alan0354

    Alan0354 Super Member

    Messages:
    1,742
    Location:
    Bay Area
  20. dr*audio

    dr*audio Fish fingers and custard! Subscriber

    Messages:
    18,028
    Location:
    Cleveland, Ohio
    Hmmm, better to make measurements first, then solder. You don't know if you really fixed anything.
     
    smurfer77 likes this.

Share This Page