Trace Elliot GP7 bass combo breaking up / distorting with low-level signal

[HenryL]

Our venerable GP7 which has been in our lives one way or another for about 40 years - latterly just low-level use in our home studio - has started playing up since we have threatened it with a rare live gig.

If playing gently and/or with the pre-amp output down low then it is futzy, breaks up, distorts. Playing harder or just a higher level out of the preamp it produces volume still, and sounds more normal but not entirely convincing, nor confidence-inspiring obviously.

I'm trying to get some background information together, ideally a map.. , before diving in. I've posted a request in the schematic forum here as I've not managed to find one that is explicitly for this relatively early model.

This is the 4x10 model "1004 Combo". It does not have an effects loop. It has two 'Hi Z' inputs and pre and post fade line-outs from the preamp. It has a single K135/J50 output pair and according to some of my old notes the rail voltages were about +70 / -70V quiescent.
The testers signature dates it to 1983. The power amp board is marked "MPB2 Issue2".

I have cleaned the immediately accessible connectors between the pre and pwr amp without gaining any improvement.

Hopefully it may yet turn out to be a detectable bad joint or similar somewhere but any pointers as to common/potential causes of this kind of symptom are gratefully received..

The behaviour points to the problem being in the power amp stage, so my first specific question is this:

The closest-looking schematic I've so far found is from a "Series 6" with "AH200" output stage, service manual dated 1999 ( a mere 16years later! ) though the power board schematic is much closer dated at 1989.
This schematic at least shows a K135/J50-based output stage (no bipolar bears, BUZ etc.) though it has double the number of FETs in a paralleled arrangement.
The main 4700uF reservoir caps are marked alongside with '63V' in the diagram (1989) (looks like a spec rather than an operational voltage, but not sure) and specified as 80V in the enclosed "PB200" parts list (dated 1997).

Maybe suggesting that our version is running at a higher voltage (4700uF, 100VDC rated caps and +/-70V rails measured) ?

Is this circuit effectively the same as ours and therefore I could use it (to a reasonable extent) ? I'm not sure of the history/evolution or what if anything much would have had to change to double up the FETs.

Many thanks for any help with this.

[glebert]

Do you have an oscilloscope that you could visualize the output waveform? I would do it with a sine wave input with and without a speaker load. I would also look at the signal at the preamp outs (pre and post fader) and verify that the signal is clean and solid there.

[Jon Snell]

They are mostly the same circuit.
Check the bias setting; P1. Measure the current flow through the Fets. We are looking for about 15mA.
The easiest way is to desolder one of the Sources or Drains and measyre the current flow, with great care!
The smaller amplifier hasn't got the red boxed section fitted.

To get the HT voltage inline with the ratings, you may find the transformer tapping is set to 230volts. Select 240volts.


Check the large electrolytics for bad connections to the pcboard. That gives an underlying hum/buzz.
Check the bias set pot for a dirty wiper and adjust from minimum resistance first, to avoid over current and damage.

[g1]

Here is the schematic with the higher voltage filter caps. (100V)

[HenryL]

Do you have an oscilloscope that you could visualize the output waveform? I would do it with a sine wave input with and without a speaker load. I would also look at the signal at the preamp outs (pre and post fader) and verify that the signal is clean and solid there.

Yes I do have an oscilloscope. I'm going to dive in this afternoon. Agreed should aim to eliminate the preamp first.

[HenryL]

Many thanks for the tips Jon. I was hoping to keep away from the FET's at first if poss, but if needs must. The main reservoir cap joints look fine at the caps themselves, they then connect to the PCB via a plug (which I've cleaned previously). I have been peering suspiciously at the open frame bias preset as a possible culprit and was thinking I might replace it - clean it at the very least - then re-adjust. I had thought I could just dial out a crossover notch on the oscilloscope to reset it appropriately - is that a bad idea?

(BTW I'm still much appreciating the repair you did on one of my Flying Moles some years back! )

[HenryL]

Cheers all - thanks for the help - will have a poke around later this afternoon and get back with the good news (?!) in due course.

[Jon Snell]

Many thanks for the tips Jon. I was hoping to keep away from the FET's at first if poss, but if needs must. The main reservoir cap joints look fine at the caps themselves, they then connect to the PCB via a plug (which I've cleaned previously). I have been peering suspiciously at the open frame bias preset as a possible culprit and was thinking I might replace it - clean it at the very least - then re-adjust. I had thought I could just dial out a crossover notch on the oscilloscope to reset it appropriately - is that a bad idea?

(BTW I'm still much appreciating the repair you did on one of my Flying Moles some years back! )

Pleased to help but no more spares for Mr Mole and his friends, sorry.

If the pot goes open circuit, the current will increase in this circuit, so probably just needs gentle adjustment.
If you have a clamp meter, measure the current flowing through one of the DC supply wires feeding the bridge. That will save taking the FET cable off.
Maximum resistance across the pot means maximum quiescent current, these FETs are very forgiving but not bomb proof.Adjust carefully whilst listening is the easiest way to check for other faults; (R13 high value causing low voltage drop across the pot and cross over distortion).

[The Dude]

This sounds like possibly an under biased amp resulting in crossover distortion. I would not only scope the preamp, but also scope the output and see if there is any visible crossover distortion. It could also be a speaker with a rubbing voice coil that is noticeable at low volume and the rubbing is drown out by the output at higher volumes. Check the amp with another known good speaker.

[HenryL]

Hi The/Dude - just caught up with your post and I think you have hit the nail on the head with your second suggestion.

In time-honoured perverse fashion, having got the amp on the bench this afternoon and threatened it with surgery then it behaved itself impeccably. Preamp and power amp outputs all squeaky clean on the oscilloscope, no crossover distortion or other signs of things futzing out, so I rigged up an alternative speaker cabinet and the amp was sounding fine. I had to stop at that point today but it is clearly now back to checking the speakers in the combo cabinet.

The annoying thing is that this was the first thing I thought of and I did listen to each of the speakers up close but the distortion seemed to be coming from all of them so I didn't bother going to get the monstrous H&H cabinet from the other end of the house... there's a moral in this story somewhere! It could still turn out to be an intermittent problem in the amp, but I doubt it now. So the next step is go back and test the combo speakers again. Individually, I think.

[HenryL]

Thanks for your further advice/info Jon. It looks like I may not need to go there now but duly noted for future ref.

Re the moles, I picked up a couple of the knobbly BX.. driver modules (or whatever exactly is in that inline package) shortly after your repair - as emergency spares - and the other mole failed in the same way shortly after the first one - by a strange co-incidence. So I was able to fix it myself. Suspecting it may be down to a limited service life / the amount of on-hours I don't leave the amps on day in day out as I used to, but now switch them on strictly when needed. So far so good. So they are not flying completely without parachutes but they only have one addtiional life left each! (The BX modules seemed to have disappeared from the market the last couple of times I thought to look).

[HenryL]

Hello people,
So the problem turned out to be neither the main amp board nor the speakers, but, rather mundanely, was the switched speaker socket that connects the amp to the internal combo speakers evidently not making a good clean contact through when nothing was plugged into it (the normal combo case). Hence it worked fine when I finally plugged in a separate cab. I'm quite surprised it produced the kind of distorted sound it did - it wasn't a typical 'erratic contact' kind of sound but a quite regular fuzz (with the odd dropout). This is what threw me off the scent from the start - otherwise I'd have got to checking the socket much sooner. (Note to self!)

Since there are two speaker output sockets offering permutations of 'parallel with the internal speakers' and 'interrupt the internal speakers' as a precaution I've rewired them so both sets of socket contacts are feeding through to the internal speakers in parallel, providing redundancy. It means if you want to disable the internal speakers while the amp is still in the combo case then both sockets have to be plugged into, but it's not a setup I can ever see us using.

Thanks again all for the helpful input.

[Jazz P Bass]

I have encountered this issue on Roland Jazz Chorus amps.