Announcement

Collapse
No announcement yet.

Marshall TSL:100 bias drift problem

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

  • #16
    I've worked on lots of these amps and what's difficult to rationalize is that I see early revision, well-used amps that have been in regular service without any problems. Equally, I also see amps with the same board revision that have/are suffering from bias drift and other board issues. You'd think that within a production run all boards would exhibit the same characteristics.

    I guess it's like the Mini cars with CVT transmissions. They all catastrophically failed at very low mileage. Except the few identical ones that did 200,000+ miles.

    Comment


    • #17
      Originally posted by mhuss View Post
      Do they have different revision/production code/... numbers? (e.g. Z03-14) Can't really tell from the pictures.

      FWIW, at least one of the failing boards looks like the darker color.

      --m
      I think the revision numbers only relate to the circuit and not the parts unfortunately. Sorry about glare on the photos; it was just meant to show the colour difference.

      I have a box of old drifty boards and they are all a much paler green. So perhaps my observation has a few doubts. Usually I look for tubes with heat damaged bases and the board colour and if I am suspicious then I measure the bias and watch. Usually it starts drifting within a few minutes and I always keep an eye on them for a good while. I have seen a least one amp where the overheating got so bad that the PT was also impacted.

      When I have spoken to Marshall about this they seem well aware of the issue. I had thought it was a materials problem where the board was absorbing something from the atmosphere and then changing its insulation properties hence my belief that the workarounds might not be good long term.

      Mike

      Comment


      • #18
        Job sorted now got a new board
        BBB

        Comment


        • #19
          How much $$ did you have to shell out?

          Comment


          • #20
            Hi jazz bass. The job was from a guitar shop who are Marshall dealers. They brought the board. £40 + tax. £40 is about $60. Marshall UK are really good on parts and not expensive

            Comment


            • #21
              They don't deal directly with the public, though. You have to be a trade account holder and the trade prices aren't published.

              Comment


              • #22
                Hey folks.... got one in with this problem. It's a TSL 122. But as mentioned early on in this thread... the bias only drifts (high) if the phase inverter is in place. Which leads me to suspect an oscillation of sorts.

                Anyone come across this or fixed it? BBB? How did you fix those that you had this issue with?
                Last edited by lowell; 09-09-2019, 01:05 AM.

                Comment


                • #23
                  Additionally: It can't be an oscillation, because upon pulling the PI, the bias slowly goes back down, about .1ma every 3 seconds. Quite strange... I can't think of anything that could cause this. Unless the coupling caps from the PI to the power tubes are charging when the oscillation is present, and then slowly discharging when the oscillation is gone.

                  There is also a noticeable hum that seems to increase over time when the PI is installed. All pots are at zero.

                  Comment


                  • #24
                    It sounds like a faulty PCB.

                    Comment


                    • #25
                      Hi,
                      I have had C36,C37 go leaky.
                      This is over and above the " bias drift " problem that has been well documented and has numerous fixes.
                      John

                      Comment


                      • #26
                        Maybe the hum is down to the increased current draw on the power tubes as the bias is shifting. Scoping the output tubes should show any oscillation. If there was any DC leakage through the coupling caps then I'd expect this to be present whether or not the PI is installed. Do your voltages look OK on the socket with the tube removed? Also, when I get unusual problems with these amps I replace the 22pf snubber cap C46. This can be the source of many mysterious problems. I use a 2Kv rated part.

                        If the PCB has become conductive there may be an issue where there's a voltage leakage that in itself isn't critical, but when the tube is inserted the tube's resistance is paralleled with the board resistance to give a greater leakage. Having said that, I've worked on a lot of these amps and never encountered leakage around the PI. But this reminds me of Blackstar amps that use MOSFETS instead of a PI tube. These commonly conduct between the pins, but I've had them where I've had to chase the leakage all over the board - including between pins of the preamp tubes.

                        Comment


                        • #27
                          easy way is to replace the board. Ive done a fair few and every time i try and repair the board it never works well
                          i get the board from marshall uk.

                          Comment


                          • #28
                            Agreed, If the board is faulty, then here in the UK it's a better cost option than spending too long on a defective one and it has a better long-term prospect. I've had good success with repairing these boards using a special router bit I made that easily isolates the power tube pins. But its only any good if the leakage is localized to the power tubes. Once you get voltage leakage elsewhere the board is pretty much finished. I would always make certain the board is the problem before replacing it and not a component failure. Even the early revision boards don't all fail because they conduct - I have a 1998 board here right now with no leakage problems - it got changed out because the owner had read about the failures and wanted it swapping.

                            Comment


                            • #29
                              I may have found a fix for drifting bias in these amps. Tomorrow will be the true test. I found another forum where someone mentioned C46. This cap is across V8 EL34 Plate-screen. My guess is it's there to reduce ultra high frequencies. I cut one leg of the cap. Then I measured the bias for a while. For about 30 minutes it continues rising Which I admit is not ideal... It finally stopped drifting. I readjusted the bias pots after 30 mins to set both at 90mv. I left in on for another hour and It did not rise any further for the following hour.

                              I will try it on in the morning and measure it again when cold. Will also play the amp as well. If it sounds good and the bias isn't so far off that there's any bad hum I'm calling it fixed.

                              Will report back.

                              Comment


                              • #30
                                FWIW: Both "the C46 problem" and the "bias drift problem" are known issues for this amp and are separate issues.
                                "I took a photo of my ohm meter... It didn't help." Enzo 8/20/22

                                Comment

                                Working...
                                X