Ad Widget

Collapse

Announcement

Collapse
No announcement yet.

Marshall Haze 40 - Low Volume from Pre-Amp section

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

  • Marshall Haze 40 - Low Volume from Pre-Amp section

    A buddy just dropped off his Marshall Haze 40 because it was very quiet. The problem is in the pre-amp section (if I put my signal generator into the FX Return the power amp section works great).

    Schematic: https://www.thetubestore.com/lib/the...-Schematic.pdf

    With my signal generator at .5Vpp and the clean volume maxed, the signal going into the phase inverter (V3B pin 7) is about 5VPP but with a +20VDC offset.

    At the previous gain stage (V2A+B) at the input (pin 2) the signal is only 4VPP on clean, but around 20VPP on OD channel. The problem I see here is that the tops of the waveforms are totally cut-off, but this goes away when V2 is pulled. That is very confusing. I've tried replacing V2, V1 and V3.

    Attached is a shot from the scope. OD channel set to minimum distortion:

    YELLOW = V2 input (pin 2)
    CYAN = V3 (PI) input (pin 7)
    BLUE = speaker output (8 ohm load)





  • #2
    Here's a scope shot with V2 pulled:

    Comment


    • #3
      Originally posted by waspclothes View Post

      With my signal generator at .5Vpp and the clean volume maxed, the signal going into the phase inverter (V3B pin 7) is about 5VPP but with a +20VDC offset.
      DC offset at the PI input grid is normal. Scope the other end of C33 to see signal without DC offset if you like.


      At the previous gain stage (V2A+B) at the input (pin 2) the signal is only 4VPP on clean, but around 20VPP on OD channel. The problem I see here is that the tops of the waveforms are totally cut-off, but this goes away when V2 is pulled. That is very confusing. I've tried replacing V2, V1 and V3.
      V2A grid conduction clips positive tops at around 2V. So no clipping when V2 is removed. All normal.
      Your input signal is way too high. Reduce until you see a decent sine at V2A grid and scope the signal at pin 8.
      Last edited by Helmholtz; 07-18-2021, 12:05 AM.
      - Own Opinions Only -

      Comment


      • #4
        Originally posted by Helmholtz View Post

        DC offset at the PI input grid is normal. Scope the other end of C33 to see signal without DC offset if you like.




        V2A grid conduction clips positive tops at around 2V. So no clipping when V2 is removed. All normal.
        Your input signal is way too high. Reduce until you see a decent sine at V2A grid and scope the signal at pin 8.
        Thanks Helmholtz, I did as you said and measured a very nice 30VPP signal coming out of V2, but a very small 2VPP signal making it to the phase inverter with pretty much only a relay in-between that does not make a clicking noise when it's supposed to switch Channels. A new one will be ordered and I'll report back.

        Comment


        • #5
          Originally posted by waspclothes View Post
          A buddy just dropped off his Marshall Haze 40 because it was very quiet. The problem is in the pre-amp section (if I put my signal generator into the FX Return the power amp section works great).

          Schematic: https://www.thetubestore.com/lib/the...-Schematic.pdf

          With my signal generator at .5Vpp and the clean volume maxed, the signal going into the phase inverter (V3B pin 7) is about 5VPP but with a +20VDC offset.

          At the previous gain stage (V2A+B) at the input (pin 2) the signal is only 4VPP on clean, but around 20VPP on OD channel. The problem I see here is that the tops of the waveforms are totally cut-off, but this goes away when V2 is pulled. That is very confusing. I've tried replacing V2, V1 and V3.

          Attached is a shot from the scope. OD channel set to minimum distortion:

          YELLOW = V2 input (pin 2)
          CYAN = V3 (PI) input (pin 7)
          BLUE = speaker output (8 ohm load)
          So just an update: This has been solved. Turns out at some point the amp was 'modded' and someone attempted to replace R90 from 10M with a 1M resistor (no idea why), but actually put a 4.7k resistor in place - dragging the signal way down. Very hard to spot, I'm never looking for resistors have been replaced with incorrect ones when I'm debugging things. But I suppose I should have when my buddy told me that he thought the amp had been modded before he got it.

          Comment


          • #6
            Originally posted by waspclothes View Post
            Turns out at some point the amp was 'modded' and someone attempted to replace R90 from 10M with a 1M resistor
            ??
            Schematic shows R90 as 4M7.

            - Own Opinions Only -

            Comment


            • #7
              Hi,
              There was a raft of mods sanctioned by Marshall for the HAZE series of amps, earthing structure, relay popping.............. It may pay to check with your local agent.
              John

              Comment


              • #8
                Originally posted by John G View Post
                Hi,
                There was a raft of mods sanctioned by Marshall for the HAZE series of amps, earthing structure, relay popping.............. It may pay to check with your local agent.
                John
                Are these online anywhere? I'd be very curious to see what they are, and if that's what the previous changes were supposed to be before I undo them.

                Comment


                • #9
                  Originally posted by Helmholtz View Post

                  ??
                  Schematic shows R90 as 4M7.
                  That explains how they accidentally put a 4.7k resistor in this position - only one stripe off

                  Comment


                  • #10
                    Hi,
                    I don't have the info, but a Google did bring up.
                    Marshal documentation CATS-00163
                    Mods called ECN-010-MZ
                    KIT-00021 has the parts as a kit.
                    Cheers
                    John

                    Comment

                    Working...
                    X