Help with a Marshall 1987x

Discussion in 'Amps/Cabs Tech Corner: Amplifier, Cab & Speakers' started by Dr. Jimmy, Jan 21, 2008.

  1. Dr. Jimmy

    Dr. Jimmy Member

    Messages:
    3,982
    Joined:
    Oct 31, 2006
    Location:
    Connecticut
    Hi, I was wondering if anyone had any ideas that could help me with some issues I am experiencing with my Marshall 1987x 50W reissue.

    It all started at a gig Saturday in which I popped the mains fuse. Putting the amp on my bench yesterday I noticed one of my power tubes was bad and was the culprit. I put in a new set of Winged C's and biased 'er up.

    Issue #1: The plate voltage on these new tubes is 430V, according to my notes from the last tube change (which were Electro-Harmonix tubes) the plate voltage was 460V. Why the decrease?

    Issue #2: After getting the bias all set (these tubes are matched really tight, 16.42W and 16.48W) and the amp warmed up, I ran 1k into it and checked the output on my scope. With the signal applied to ch.1 I can barely crack the volume knob before the signal clips horribly, with a high pitched squeal emitted from around the tube area. Putting the signal into ch. 2 gives me a damn good signal on the scope, and I'm able to turn the vol. up to a good level before clipping. I've replaced V1 in hopes that V1B was suspect, but no dice. Any ideas?

    Issue #3: While I was under the hood with this amp, poking around measuring voltages, I measured 176V across R11 which is the slope resistor (and appears to be tied to the cathode of V2b on the schematic), is this normal?

    Thanks in advance!
    Bill
     
  2. Dr. Jimmy

    Dr. Jimmy Member

    Messages:
    3,982
    Joined:
    Oct 31, 2006
    Location:
    Connecticut
    Here's a long winded update :lol:
    Last night I decided to pull out my Furman Voltage Regulator and plug the amp into that (and the regulator into the variac) so I can get a more consistant line voltage coming into the amp. Big difference, the plates were now reading 460V. I redid the bias and now we're at 16.7W and 17.5W (not as closely matched as before, but close to what my notes indicated for the last pair of tubes that I am replacing).

    Again I put the 1k into ch. 1 and noticed the instant clipping, and the clean signal when applied to ch. 2, so I removed the cap from across the vol. 1 pot. Wow! The signal cleaned right up and I was able to open the pot a lot more before the signal clipped (I measured approx. 60W at clipping). I did notice the signal jumping about a bit as I turned it up, so I plugged the signal generator into the Furman and that corrected it.

    My next questions are:
    1) What can I expect tonally now that I removed that cap?
    2) I use a Weber MASS to help control my volume at gigs. I jumper the channels and have both volumes set to roughly 1:00. Are the tubes biased too hot for this? The amp sounded fine like this when I used the EH EL34's are the Winged C's sturdier?
     
  3. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    1 - much more even volume taper on the pot, but a bit less bright at lower settings. Actually the bright cap is a good idea, but the stock value is far too large, by about a factor of 10. Try a 500pF cap instead.

    2 - yes, too hot. Aim for a maximum of 15W per tube and maybe a tiny bit cooler. Marshalls don't like being biased too hot when you're going to really crank them, they tend to overload the tube screens, which can cause failure.
     
  4. Dr. Jimmy

    Dr. Jimmy Member

    Messages:
    3,982
    Joined:
    Oct 31, 2006
    Location:
    Connecticut
    1. Excellent, I believe the one I pulled out was 47pF. Will a 500pF improve the taper as well?

    2. I will try that as well, will I still get some good "beef" out of the amp with it set cooler like that? Also, since the tubes aren't exactly matched there is a slight difference between the two, so one may end up at 15.5W and the other at 16.5W, should I aim to get the larger one to 15W? (Sorry for the awkwardly phrased, dumb question).

    Thanks for the reply!
     
  5. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    The stock one is 4700pF (4n7) - basically it's not a bright cap, it's a treble-and-mid cap, and lets so much through it totally changes the taper of the pot to 'off to on'. With a smaller cap, you'll still get a nice sparkle at lower volumes, but the taper will be much more natural, and what gets through will be just the top-end, so it doesn't sound as harsh as before. I have no idea why Marshall used such a big cap.

    Not a dumb question :). Yes, the tone should be more even and less strident with it biased a bit cooler (this may be part of the problem with the over-sensitive bright channel too).

    Assuming the tubes are reasonably well matched, I would keep the hotter of the two at 15W and the cooler one slightly below it. If they start to get really mismatched, you'll probably need to keep the cooler one from getting too cold or crossover distortion will occur (usually if it goes under about 50%), by raising the hotter one a bit. But if you're only off by a watt or so it should be fine at about 14 and 15W. (That's still 60% of tube dissipation BTW, which is correct - EL34 Marshalls sound best biased around 55-60% in my experience - and allows for the screen current too.)
     
  6. Dr. Jimmy

    Dr. Jimmy Member

    Messages:
    3,982
    Joined:
    Oct 31, 2006
    Location:
    Connecticut
    Thanks so much for the advice here. I can't wait to get home and put all of this into action!!! :BEER
     
  7. gtrnstuff

    gtrnstuff Member

    Messages:
    2,282
    Joined:
    Oct 19, 2005
    Location:
    Middle TN
     
  8. Dr. Jimmy

    Dr. Jimmy Member

    Messages:
    3,982
    Joined:
    Oct 31, 2006
    Location:
    Connecticut
    Last night I dropped the bias down a hair to approx. 15.5W per tube.

    I put a 390pF cap across vol. 1 pot, I wanted to do 500pF, but didn't have any kicking around. I put the 1k signal into the #1 input (the 1k was at approx. 70mV), and again as soon as I cracked the ch. 1 volume the signal on the scope looked nasty. I measured the voltage across the 8 ohm load to get a rough idea of power and it was 59.5W. This was with the ch. 1 volume at 1. Does that seem high? I also heard the high pitched "eeeeeeeeeee" as I turned the vol. up.

    For laughs, I removed the cap and the ch. 1 vol. again worked like a normal volume pot. Again, I measured close to 60W across the load but this time the volume was around 5 or so (and no "eeeeeee" :) )

    Maybe I'm just overthinking this????
     
  9. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    If you're using the scope to tell you if you like the volume pot taper, yes.

    If you prefer the tone and taper with no cap by listening to it with a guitar as the signal source, leave the cap off. If you prefer it with a cap on, try several values until you find the one that sounds best.

    That's really all you need to know :).


    A scope is a good troubleshooting tool for when odd things are happening, but for normal stuff just use your ears... they're surprisingly useful ;).

    I'm going to be out of here for a while, so have fun :).
     
  10. Dr. Jimmy

    Dr. Jimmy Member

    Messages:
    3,982
    Joined:
    Oct 31, 2006
    Location:
    Connecticut
    I guess my point was that by viewing it on the scope, it looked like it was getting hit awfully hard for the volume only being at 1. I am a bit concerned that there may be something wrong, and I will be causing further damage by playing it.

    I have rehearsal Friday and I'll be able to really wind 'er up and see how it sounds. :RoCkIn

    Thanks for all the advice!!!
     
  11. Dr. Jimmy

    Dr. Jimmy Member

    Messages:
    3,982
    Joined:
    Oct 31, 2006
    Location:
    Connecticut
    Here's the latest:

    Again tried it with a 390pF cap on the vol. 1 pot, as soon as I turned up the knob just a hair the signal clipped, I heard the oscillation, I measured 59W across the load and the 1k input signal was magically changed to 10k on the output.

    Turned off the signal generator, left it plugged into ch. 1, turned up the vol. on the amp and still saw/heard oscliiation, clipping etc...

    Plugged in my strat and banged a few chords, could hear the oscillation happening as I hit the strings harder. Vol. on ch. 1 around 9:00. My theory is that this is what caused me to blow the tube at the gig Saturday, I was plugged in and hit a few chords when the amp just died. The oscillation must've redplated the tube and it finally quit.

    Removed the cap, and I was able to turn the vol. 1 knob to around 9:00 before it clipped, no oscillation heard, measured approx 55W across the output, output was 1khz.

    Ch. 2 showed none of these problems.

    I've always run this amp into an attenuator, jumpered channels with both volumes at approx. 1:00. I wonder if I have damaged it by doing that???

    Any more suggestions/recommendations/references for somewhere to get this fixed. I'm getting rather fed up with it at this point. [​IMG] [​IMG] [​IMG]
     
  12. Dr. Jimmy

    Dr. Jimmy Member

    Messages:
    3,982
    Joined:
    Oct 31, 2006
    Location:
    Connecticut
    Bump for the morning crew.
     
  13. ChickenLover

    ChickenLover Member

    Messages:
    1,696
    Joined:
    Sep 7, 2006
    Location:
    Florida
    Sounds like it could be oscillation to me. Adding the bright cap allows a lot more highs through so it's more susceptible to oscillation. Check your layout...I'm not familiar with the 1987x and what is PCB mounted vs. what isn't. Have you done any other mods or moved any wires

    In general, the grid wires (i.e. the wires connected to pins 2 and 7 on your preamp tubes) are most sensitive to picking up stray signals and the plate wires (wires connected to pins 1 and 6 on your preamp tubes) are most likely to give off these stray signals (that can then be picked up by the grid wires). The earlier it is in the preamp the more sensitive it is to picking up signals (because the signal is small and will be further amplified so much) and the later it is in the amp the more likely it is to give off these signals (because the signal has already been amplified a lot and the signal is large). The wire from the treble cap to the treble pot is sensitive and should be kept away from everything except the chassis...but in your amp the pot is probably PCB-mounted. Try to keep the NFB wire away from grid wires. Keep the OT wires away from grid wires.

    If you want a quick and dirty way to test for/get rid of oscillations (usually) you can add a .01uF cap in parallel with the plate load resistor on the preamp gain stages (just try one stage at a time) and when you put it on the correct plate resistor (trial and error) that should get rid of it...but it will also affect tone and ideally you shouldn't need it in a NMV Marshall...it's just a quick way to test the theory and perhaps isolate 'where' the oscillation is occurring. HTH.
     
  14. Dr. Jimmy

    Dr. Jimmy Member

    Messages:
    3,982
    Joined:
    Oct 31, 2006
    Location:
    Connecticut
    Excellent advice! Thanks!

    The pots are all chassis mounted, and there have been no mods (haven't moved any wires around either).

    Since it is most noticable on ch. 1 is it safe to assume that it could be the plate resistor associated with that tube, V1b? Also, is this oscillation the cause of the excessive power (leading to the blown tube) that I am seeing across the output?

    Thanks again!
     
  15. ChickenLover

    ChickenLover Member

    Messages:
    1,696
    Joined:
    Sep 7, 2006
    Location:
    Florida
    Not necessarily...there is so much more high-end in the Treble channel that it's more susceptible to oscillation. And it's not so much the plate resistor that might be the problem...but by putting that cap across the plate resistor you are shunting a lot of the ultra-high-frequency stuff at that plate to ground (your B+ supply is AC ground...it looks like ground to any signal). It will usually prevent the oscillation and most of the tonal change will be in the ultra-high-frequency range...but there will often be a noticable change to the high end (and it's not necessarily a bad thing...high-gain amps use these snubber caps all the time).

    Yes, the oscillation could be that power you see at the output. You might be able to stop the oscillation just by turning your Presence and Treble controls all the way down (again, just as a test).
     
  16. ChickenLover

    ChickenLover Member

    Messages:
    1,696
    Joined:
    Sep 7, 2006
    Location:
    Florida
    BTW, I didn't see it mentioned...how big of a 1kHz signal are you putting into the amp? Should be on the order of 100mV...or 0.1V.
     
  17. Dr. Jimmy

    Dr. Jimmy Member

    Messages:
    3,982
    Joined:
    Oct 31, 2006
    Location:
    Connecticut

    I measured it at 70mV. I also plugged my Strat into it and could hear the oscillation as I hit the strings harder......
     
  18. Dr. Jimmy

    Dr. Jimmy Member

    Messages:
    3,982
    Joined:
    Oct 31, 2006
    Location:
    Connecticut
    Thanks for the explanation, I did notice a slight reduction as I turned the presence down (I normally have it at 9:00).
     
  19. Dr. Jimmy

    Dr. Jimmy Member

    Messages:
    3,982
    Joined:
    Oct 31, 2006
    Location:
    Connecticut
    Bump for the afternoon crowd....
     

Share This Page