Announcement

Collapse
No announcement yet.

Marshall Valvestate 8080 with unwanted distortion

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

  • Marshall Valvestate 8080 with unwanted distortion

    Hi there!

    I got a Marshall Valvestate 8080 combo amplifier which is acting weird with a strange fuzzy distortion that is not normal. I tested the preamp and the speaker and both are ok. When pluggin the guitar into the FX Loop Return the strange distortion is there again, so my guess is that the problem must bem from here on. I changed the following components without luck: Capacitors C17, C20, C40, C41, C53, C55, C56, C59, C61, C62 and resistors R101 and R102. I took a few measurements on the power supply and here are the readings:
    Mains voltage: 127 Vac
    Transformer output voltage: 38,5 Vac
    Voltage to the power amp (VE on schematic): +50 Vdc / -50 Vdc
    Voltage to pre amp = +16,2 Vdc / -17,3 Vdc
    No ripple was shown on my measurementes.

    The output transitors are not heating much and I got these readings on their legs (from left to right):
    TR8 (T65): 1,5 V / 50 V / 0,0 V
    TR10 (T64): -0,5 V / -50 V / 0,0 V
    TR9 (TIP29C): 0 V / 1,5 V / -0,5V

    There is no Vdc on the speaker and the solder joints are ok.

    IC8 (TL072) is reading 500 mVdc on output (pin 7). I also changed this IC with no luck. All others IC's don't show any DC voltage on their inputs or outputs.

    Any tips on what else should I check on the amp?

    Here's the schematic:
    Marshall Valvestate 8080 8100.pdf

    Thanks in advance!

  • #2
    Why did you change all those parts?

    What does the distortion look like? Do you have a scope? If no scope, make yourself a "signal tracer probe" so you can listen to the signal as it passes through the amp.
    Education is what you're left with after you have forgotten what you have learned.

    Comment


    • #3
      I think a whole volt difference between your + and - 16V is a bit much (you said +16.2/-17.3)
      Your TR8 and TR10 base voltages should also be evenly above and below 0, so more like +1 and -1 at emitter and collector of TR9.
      "Everything is better with a tube. I have a customer with an all-tube pacemaker. His heartbeat is steady, reassuring and dependable, not like a modern heartbeat. And if it goes wrong he can fix it himself. You can't do that with SMD." - Mick Bailey

      Comment


      • #4
        I would suggest crossover distortion if the buzzing noise is there at low levels but to check that you need;
        No input signal and no speaker connected.
        Measure the voltage between the emitters of the output transistors, (BDV65/4 or TIP142/7) you need about 5mV.
        If that is correct, it is probably not crossover distortion.
        I have had totally open circuit main smoothing capacitors that cause that as well. The amplifier runs without any smoothing quite happily but there is an underlying buzz.
        DC offset on the ic is down to the input voltage. I wouldn't worry about that for now.
        I take it you have tried another good speaker.
        Support for Fender, Marshall, Mesa, VOX and many more. https://jonsnell.co.uk

        Comment


        • #5
          I saw an YT video where a dude was facing a similar problem as mine and solved it replacing all the elecrolytic caps on the amp so I tried the same approach but chose to replace just the ones that were on the path from the FX loop return to the speaker because the preamp is working ok. Also, as it's an old amp that wasn't turned on for a long time, I thought it was a good idea to replace them.

          Yes, I tried the amp with another good speaker and the problem definitely lies within the circuit, not the speaker.

          I changed the caps and resistors on the power supply because the voltage was showing the difference of +- 1Vdc. Only later occured to me that the Zenner diodes could be the problem. Changed both ZD1 and ZD2 and now I read +15,8 / - 15,7 Vdc on the power supply. This also removed the DC voltage I was reading on pin 7 of IC8.

          I don't have a scope and searched a bit on how to make and use a signal tracer. I improvised one according to this video:
          https://www.youtube.com/watch?v=IEcyDzYAsuU

          I used my cellphone to play some music through the FX loop return jack and followed the signal. Seems to me that the signal is already coming out of IC8 a lit bit distorted but I'm not sure. The leads I've used to make the signal tracer are not shielded and I'm afraid that they are picking up other noises. Are shielded cables a must for this?

          I removed TR9, TR11 and TR12 from the circuit and all three tested OK. I changed TR4, TR5, TR6 and TR7 for new parts and the problem persists.

          After all this, here are the voltage readings referenced to ground, no input signal and with speaker disconected):
          TR8 (T65): +1,43 Vdc / +50 Vdc / -10 mVdc
          TR10 (T64): -0,50 Vdc / -50 Vdc / -10 mVdc
          TR9 (TIP29C): 60 mVdc / 1,54 Vdc / -500 mVdc

          Why are the base voltages on TR8 and TR10 so offset? Can this be the problem? Also, which are the smoothing capacitors? Can they be causing this? What should I look next?

          Thank you very much!

          Comment


          • #6
            The emitter voltages are the important voltages to record because they are set with the negative feedback loop to pin 5 of IC8.
            There are four junctions to consider and 4 x 0.6 is 2.4 so I would expect a little over 2.4 volts between the bases.
            We have just barely 2volts which means we have crossover distortion. The distortion is fed, via the nfb loop to IC8, (the voltage at pin 5 is seen as very low almost undetectable without a scope), IC8 amplifies it and that is why it seems to be originating from IC8 when it isn't.
            Check R75 and R85, they set up the constant current across the voltage bias amplifier TR9, (when the current goes up the voltage across TR9 increases and the bias voltage should stay stable set by R73 & R74). If they measure within the tolerance on the coloured band, probably 10% then check R73 and R74.
            If they are OK adding a 220R preset in series with R74 will remove the distortion for testing purposes.
            If that is indeed the issue and the components mentioned are within specification, the question is which component is at fault ... is it TR9 with too much gain, TR7 or TR4 gain too low ... your call but at least you will have a working, distortion free amplifier while you ponder.
            Hope that helps.
            Support for Fender, Marshall, Mesa, VOX and many more. https://jonsnell.co.uk

            Comment


            • #7
              Hello!

              The voltage on pin 5 of IC8 is -6mVdc (no scope, just the multimeter reading, a Fluke 112).

              Lifted one leg of each resistor (R73, R74, R75 and R85) and all tested within specs.

              Here are the readings of the transistors from the power amp:
              TR4 (C / B / E) = -500mV / -48,7V / -49,2V
              TR5 (C / B / E) = -48,6V / -1,55V / -0,97V
              TR6 (C / B / E) = 48,6V / 1,63V / 1,07V
              TR7 (C / B / E) = 1,77V / 48,5V / 49,1V

              Tried to add a 220 Ohms resistor in series with R74 and the amp still sounded distorted. The readings with this resistor in series are:
              TR4 (C / B / E) = -500mV / -48,7V / -49,3V
              TR5 (C / B / E) = -48,8V / -1,54V / -0,95V
              TR6 (C / B / E) = 48,7V / 1,62V / 1,05V
              TR7 (C / B / E) = 1,96V / 48,6V / 49,2V

              I tested R98, D4, D5 and C47 and they are OK. Also tried to change TR9 because I had a spare laying around and had no change in the sound.
              Where should I go next?

              Thanks again!!

              Comment


              • #8
                Increasing the voltage potential of TR9 b/c would normally increase the c/e voltage but it doesn't.
                That means there is insufficient current supplying TR9 and therefore TR4 and TR7.
                Id R85 open circuit I wonder or is there a problem with TR6's bias perhaps.

                The problem with replacing random components is you may well be compounding the original fault by adding new man made faults.
                Not the sort of thing that any engineer would do.

                Look for the correct current flow from TR7 before you go any further.
                Support for Fender, Marshall, Mesa, VOX and many more. https://jonsnell.co.uk

                Comment


                • #9
                  Hi there! Been a little bit busy with other stuff these days. In the meantime I changed TR8 and TR10 for the TIP 142 and 147 and the amp worked fine, with none of the undesired distortion! YAY!!! Kept playing it for a little while, then unplugged the guitar and went away and left the amp on. Came back about 30 minutes later and when I plugged the guitar cable to the input, the amp gave a short whistle and both power transistors shorted out. So I replaced them and now the amp is working, but with a series lamp, as I'm affraid of removing it and the transistors shorting again. With more volume, the lamp flashes more or less according to the strength of the signal being fed into the amp (this is normal, right?)

                  So, with a 100W series lamp, here are the voltage readings (referenced to ground):

                  TR8 (B / C / E): 1,08 V / 38,3 V / -0,028 V
                  TR10 (B / C / E): -1,13 V / -38,4 V / -0,030 V
                  TR9 (B / C / E): -0,53 V / 1,08 V / -1,13 V

                  Also note that these readings are not changing after a few minutes of playing through the amp, and the transistors are not getting hot.

                  What else must I check? Thank you very much!

                  Comment


                  • #10
                    When you were playing through it, did you have all the power transistors and heatsinks fully mounted?
                    The whistle could have been high frequency oscillation that killed the output transistors. Is R119 (10 ohm 5W) still ok?
                    "Everything is better with a tube. I have a customer with an all-tube pacemaker. His heartbeat is steady, reassuring and dependable, not like a modern heartbeat. And if it goes wrong he can fix it himself. You can't do that with SMD." - Mick Bailey

                    Comment


                    • #11
                      Yes, everything was fully mounted. Later, after I was swapped the transistors and plugged the amp with the series lamp, I kept playing and couldn't reproduce this whistle by plugging or removing the input cable.

                      Measured R119 out of the circuit, it is OK (9,9 Ohms) and it shows no visual signs of stress.

                      Comment


                      • #12
                        Hi. Any more suggestions on what to look for? Thanks!

                        Comment

                        Working...
                        X