Announcement

Collapse
No announcement yet.

Hartke HA-3500 farty noise

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Hartke HA-3500 farty noise

    Ok have picked this up to start working on it again. I have an HA-3500 that when you play on it it sounds farty.

    Have smacked it a few times, no change. Did effect send to power amp in on another amp and it sounds clean. So that says output section.
    Pulled board and looked like lots of cold solder joints fixed those no change.
    Looking at the schematics I printed out long ago on the output board there is a heavy black line that goes from con501 to an101 and a couple of other spots, what does this black line denote?

    There are a few things I'd like to do can I check the AC ripple at cn502? (why ? I'm not sure)

    How would you bias this amp as I've seen sever trimpots on the output but have not touched them.

    Then I would like to work backwards from output to find what is causing the farty noise, but the PCBmounts on top of a heatsink with power transistors mounted on the side. What would be a feasible way to audio trace backwards on PCB when the best side is the bottom?

    Thanks,
    Nosaj
    soldering stuff that's broken, breaking stuff that works, Yeah!

  • #2
    That won't work. The solid state amp is massively fed back on itself, it will have any artifacts throughout.


    Follow the schematic, that heavy line is the ground through the power amp. Note it is in two sections joined by a 10 ohm resistor. The stuff on the right is grounded through the power supply ground, while the stuff on the left is grounded via the input conector on the PA board. Don't try to service this amp without the little input connector plugged in. or at least ground the An101 ground pins.


    Farty noises? Got a scope? verify both power supplies are up to the same voltage and clean. Scope them and see if they fail under the load at high levels. Scope the output. What does the farty look like? CLipping? One side only or both sides clipping? Run a sine through it, does that come through OK?


    I see only two trimmer controls. One is between the drivers - SVR302 - and it controls bias transistor Q308. That is the bias adjustment. The other is SVR301, and it is the offset adjust. Start with a working amp. With no signal and no load, monitor the output. Adjust 301 for zero volts DC on the output. Then with signal through the amp, adjust the bias as you would for any solid state amp. I'd adjust for the crossover notch to just disappear.
    Education is what you're left with after you have forgotten what you have learned.

    Comment


    • #3
      Ok Took the output board and started testing diodes d302 tests 1.17 one way and .386 the other way IT is a in60.
      d310 test 1.006 one way then .650 the other way it is a in4006.
      These diodes should be replaced I 'm sure, because the prevent backfeeding of voltages, correct?

      Took power connector that goes to output board(disconnected) powered on amp one test lead to chassis other to connector.
      AC 63.2 v B+ 84.2 B- 84.1..

      When you say both power supplies you mean the rails correct? 15-+

      Will go ahead and setup the scope and get some info for you.

      Thanks,
      nosaj
      Last edited by nosaj; 06-11-2013, 04:15 AM.
      soldering stuff that's broken, breaking stuff that works, Yeah!

      Comment


      • #4
        Run a sine through it, does that come through OK?

        This doesn't make sense to me, since I can hear the noise wouldn't the sine at this point not be a clean one? Will still do the test, just verifing what I think I understand.

        Thanks,
        Nosaj
        soldering stuff that's broken, breaking stuff that works, Yeah!

        Comment


        • #5
          Well I shouldve put things away since it it late. But I hooked a sine 1k to the input jack. Put a 8 ohm dummy resistor on output with channel 1 scope lead. Sine was definately clipped on top and bottom. Then I started hearing this high pitch noise( thinking it might be a cap venting I kept looking) then poof some output transistors smoked and blew the t8 fuse. Heatsink was extremely hot, Everything was mounted properly without the lid on.
          So tomorrow guess I will source out transistors and diodes. Anyone recommend a source for the output transistors?
          Thanks,
          nosaj
          soldering stuff that's broken, breaking stuff that works, Yeah!

          Comment


          • #6
            D302 and D310. Did you disconnect them from the circuit to test? D310 is in parallel with the relay coil, so you have that relatively low resistance confusing your readings. I doubt that has anything to do with the issue.

            The only rails coming into the power amp are the high voltage, apparently 84v. I don't care if they are 84v or 79v, what I want to see is both more or less the same voltage. Opposite polarity of course. The AC is used to make a V- for the speaker relay. They could have used the -84v rail, but by doing it from AC like that, it goes away instantly when power is turned off, so the relay drops out.


            Always check output with NO load first, just to see what comes out. In fact if a waveform looks clean unloaded but crumbles under load, that is a solid clue.

            Music is peaky, sine waves are steady. Playing music might be farty because peaks freak it out. A sine wave may well be distorted as yours apparently was, but it could also be clean up to a point, and that also tells us something.

            You were clipping, but were you running beyond the rails? In other words, if you have +84v rail, your waveform cannot go to +85v. SO determine whether it was clipping early or if it was clipping because it ran out of room. It ought to get within a couple volts of the rail.

            You should monitor the mains current draw and also heatsink temperature. If it starts heating way up, turn it off, if the mains draw is huge, turn it off.

            If Mouser and Digikey don't have a transistor, I look at B&D or MCM often as not.
            B&D Enterprises - Electronic Components Distributor
            MCM Electronics: Home and Pro Audio/Video, Security and Test Equipment

            Those are 180v 15A 150W transistors. I wouldn't mix them, but if the originals are not available some other pair like 2SC5200 and 2SA1943 will work or even 2SC3281 and 2SA1302.

            And you can always call Samson and ask what they are using if the original types are no longer on the market. I see B&D has the C5200 pair at three bucks and change.
            Education is what you're left with after you have forgotten what you have learned.

            Comment


            • #7
              Oh, and set your bias cold to start.
              Education is what you're left with after you have forgotten what you have learned.

              Comment


              • #8
                Ok will do. Am I correct in that there is not much else I can do until output transistors and 2 diodes are replaced, besides testing components?

                Thanks,
                nosaj
                soldering stuff that's broken, breaking stuff that works, Yeah!

                Comment


                • #9
                  Probably not much.

                  But please read what I wrote about the diodes. The one on the right is just across the relay coil, so unless you unsoldered it, your test was meaningless. Also it is not in the signal path and so won't affect your sound. If the speaker relay turns on the speakers, that diode is OK.

                  The two diodes over by the power amp input I think are schottky types, so they have a very low junction voltage naturally. I suspect they are OK as well. I could be wrong. If you can use that offset adjustment to set the output to zero volts DC under no-signal conditions, they those diodes are4 probably also OK.
                  Education is what you're left with after you have forgotten what you have learned.

                  Comment

                  Working...
                  X