Announcement

Collapse
No announcement yet.

MARSHALL DBS 7400 CLIPPING + signal flow chart

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

  • MARSHALL DBS 7400 CLIPPING + signal flow chart

    Hi,
    It's been a looong time i came here, but i had to do this one...

    I had to service my DBS 7400, it made i clipping sound on higher volumes where it normally shouldn't, so... to the bench it is.
    On my scope i saw the positive signal clipping at around 50V but the negative signal went way up (down) to -65V without clipping.
    I already had the schematics, but they're not made for beginners! And googling for "DBS 7400" doesn't give many useful results... as if no one had to repair these? Also no service manual available.
    I had a hard time reading through the schematics, to a point i got the crazy idea that the engineers in the 70's-90's had one goal and that is to make their schematics look like riddles or puzzles!
    I recapped a couple SANSUI AU717 (70's hifi amp) and have a SANSUI AU919 and MARANTZ PM6A coming up... they're a pain to work on (schematics & physically)!
    Luckily in the 90's they started building decent units, flexible to work on, as is definitely the case on the DBS 7400! The build Q is astonishing... build for the road.

    But before i continue, a little word about this GREAT amp... This (2 ohm capable!) amp has a CLASS H output stage, meaning that there are 2 power supply rails for each side of the amp circuit (+56V, -56V, +112V, -112V).
    The TR4-8 do the work for the positive signal up to ca. 50V and TR21-25 for the negative part, from here TR1-3 & TR18-20 take over via the mosfets TR12,13 and TR15,16 to add the higher voltage.
    It adds it a few volts above the output signal instead of adding the full extra 56V at once like CLASS G output stage.
    CLASS G and H are a very nice topology to get more efficiency (less power losses) out of your amp, it's been used in many brands with linear high power amps. This way the power TR don't have to deal with the full voltage swing but it is divided in 2 steps (in this case, more steps is possible).

    My problem was that the +112V rail didn't "jump in" so it clipped to the +56V rail, this was caused by a broken TR12 (IRF530), so that was an easy fix!
    But there's a catch... This amp has 2 compressors built in, one adjustable with on/off on the front and one (exactly the same but no potmeter) inside. The 2nd compressor has a slow release/attack and is ALWAYS in circuit, meaning that even when you turn the volume up to max, it will NEVER clip. You could put a bypass switch between p6 & p8 of con2 on the tube pcb to bypass the compressor, but it will probably make pop sounds. The output signal is limited to approx. 55V, way below the clipping point of the second voltage rail, it has just kicked in a few volts because of the voltage sag from the first rail (tested on 4ohm, it doesn't even get hot!).
    Only when you apply a high signal from zero it will swing much higher, but less than a second later it falls back to the compressed level. So the string release or slap (peak levels) pass through with high amplification and this works very well. You could also see it on a scope by turning the volume to max very fast while applying a high signal on the input.

    Because i lost most of my time with the schematics and i'm thinking of recapping this 90's amp one day (it's my main amp on gigs), i decided to make a flow chart for the signal path, because it keeps jumping from one pcb to another passing dozens of connectors & wires, this results in continuous browsing pages.. i just don't like it. So i tried to make a clear overview on 1 page, with color codes to recognize instantly what connector/wire you're working on.
    Also the fact that there are errors in the schematic:
    * The pre/post switch is nowhere to be found on the amp
    * connector 3 on the output pcb (pg 8 of 8) refers to P6_CON3_A3, it should be P4_CON3_G2!
    I marked the signal path in green on my schematics but i didn't had the chance yet to scan them again, i will do that later.

    I added the original schematics too, maybe this might help you out one day...
    Attached Files
    Last edited by Stoffie1974; 04-27-2020, 01:11 PM. Reason: Forgot important information

  • #2
    Glad you got it fixed. That's a very pretty connection chart . Schematics do get hard to follow when there are multiples PCBs and even more connectors.

    Originally posted by Stoffie1974 View Post
    But before i continue, a little word about this GREAT amp... This amp has a CLASS H output stage, meaning that there are 2 power supply rails for each side of the amp circuit (+56V, -56V, +112V, -112V).
    The TR4-8 do the work for the positive signal up to ca. 50V and TR21-25 for the negative part, from here TR1-3 & TR18-20 take over via the mosfets TR12,13 and TR15,16 to add the higher voltage.
    It adds it a few volts above the output signal instead of adding the full extra 56V at once like CLASS G output stage.
    It seems to me that, looking at the positive side, that the gate of TR13 is about 49V. The source tracks the output so TR13 will turn off when the output goes above about 47V, thus turning on all the high voltage rail transistors. There's nothing linear about it so it's rail switched class G operation.
    Experience is something you get, just after you really needed it.

    Comment


    • #3
      Originally posted by nickb View Post
      Glad you got it fixed. That's a very pretty connection chart . Schematics do get hard to follow when there are multiples PCBs and even more connectors.



      It seems to me that, looking at the positive side, that the gate of TR13 is about 49V. The source tracks the output so TR13 will turn off when the output goes above about 47V, thus turning on all the high voltage rail transistors. There's nothing linear about it so it's rail switched class G operation.
      My bad, i thought it was a class H because on my scope i see the voltages of the mosfets follow the output linearly a few volts above/under, so that's why i assumed it's a class H (i forgot to take a picture when i measured & compared both sides).
      But thanks for the correction though

      Comment


      • #4
        Hi, it's me again... I noticed a small error in my flow chart, but i'm unable to edit my thread or replace the attachment.
        The first compressor should be BEFORE the signal switch, that's it.
        Maybe somebody can explain me how to re-edit my thread?

        Comment


        • #5
          Just post the updated one is easiest, I think.
          Experience is something you get, just after you really needed it.

          Comment


          • #6
            Originally posted by nickb View Post
            Just post the updated one is easiest, I think.
            But there's no attachment button, where can i upload it?

            Comment


            • #7
              The attachment icon is the 3rd from the right. (Insert Image)

              Click image for larger version

Name:	Attachment Icon.GIF
Views:	1
Size:	8.0 KB
ID:	857626

              Comment


              • #8
                Here it is, thx a lot guys !
                Click image for larger version

Name:	DBS7400 signal flow chart.jpg
Views:	1
Size:	190.8 KB
ID:	857628

                Comment

                Working...
                X