No announcement yet.

Power transistor: Short or bias failure?

  • Filter
  • Time
  • Show
Clear All
new posts

  • Power transistor: Short or bias failure?

    My beloved Harmon Kardon 430 receiver just developed a loud hum in one channel. The speaker output on that channel is at -27.6V …not good. Should be 0V.

    The output transistors are 2SC1030 npn in “totem-pole” arrangment. B+ of 29V on the “upper” collector, base at 0.5VDC, emitter at 0 VDC (out to speaker) hooked to the collector of the “bottom” transistor that is tied by emitter to the B- (-29.2VDC) That base sits at -28.6 VDC. This is the GOOD channel.

    The BAD channel has similar B+ and B- rails, but the base of the top transistor is sitting at -27.1 VDC (instead of 0.5V) with the emitter at -27.6VDC. The bottom transistor base is -28.6VDC an the emitter is tied to the -29.2 VDC rail.

    I was thinking there is a short in the bottom transistor, but I would have expected the emitter and collector to be at the same voltage for a dead short. No way to test for a short until I remove the transistor from the circuit. No burning smell or smoke, but I did eventually blow the 3 Amp fuse in the B+ rail for the involved channel.

    Before I replace the power transistors, would a bias failure be an alternative explaination for these voltage readings? It would be easier to replace the bias components before unsoldering TO-3's from the board. I already replaced the driver transistors about a year ago, which successfully fixed a gradual loss of volume in both channels.

    Any advice would be appreciated.



  • #2
    I doubt it is a shorted output, but power off, set your meter to diode test, and check them for shorts. Shorted outputs usually blow fuses. But check anyway.

    I don't care if they are only a year old, check the drivers and predrivers for opens.

    It sounds like the bias string is open at the top - at the + rail. So the whole thing has snapped down to the negative rail. Look for an open transistor or resistor up at the positive end of the bias string.
    Education is what you're left with after you have forgotten what you have learned.


    • #3


      I've got the schematic and service manual in hand and I should be able to check those components in a day or two.




      • #4
        Some progress

        Regarding the HK430 problem I had. It's been awhile due to a house move, but replacing the differential amp and current source transistors got the voltages off of the negative supply rail. Prior to that, changing the drivers and predrivers made no difference.

        Now I'm looking at a DC offset of + 1.4 V on the problem channel output; which suggests additional problems. The diode string in the bias circuit is ok, so I'm going to change the coupling and power supply caps on the output board. Maybe a leak in there somewhere.

        If that fails, I may need to bite the bullet and change out the power transistors, which I believe are the originals on this unit.

        If I get that far, do I need to worry about counterfeit 2SC1030's?
        MJ15003 NPN power transistors are cheap and available and should be suitable replacements with better tolerances for power and voltage. Would the MJ's realistically sound any different than stock 2SC1030's? The latter question is probably opening a can of worms.




        • #5
          One other thought

          I suppose that a poorly matched pair of transistors at the differential amp would produce some DC offset. I don't know about 1.4VDC. I replaced the input and current transistors with BC547C NPN units. The "C" semi-selects for higher hfe, but they are not actually matched.



          • #6
            The outputs were not likely matched when new anyway. Solid state amps are so heavily fed back that yuo can used almost anything in them and it will sound fine. MJ15003 would be OK, best to replacce all of them in teh channel then.

            In my experience, outputs are eitherBAD or OK. Not going to cause a volt or so of offset.

            Compare DC voltages channel to channel. Buas problems will affect current draw more than offset.
            Education is what you're left with after you have forgotten what you have learned.


            • #7
              Finally, some resolution

              I recapped both power output boards, but no change.
              Finally, I went back for the nth time and mapped out current through various components. Turns out the current source transistor was intermittently conducting, even after I resoldered the leads. Swapped in a fresh transistor and the whole thing came alive!

              DC offset in the left channel is 20 mV now and the right channel runs around 15 mV.

              I'd forgotten how nice this unit sounds through my 70's Dahlquist speakers.

              thanks for the support.