1976 Marshall JMP 50W Combo Master Volume Squealing

Tube or solid-state, this section goes to eleven!
User avatar
turbonigro
Information
Posts: 33
Joined: 28 Mar 2010, 12:24
Has thanked: 4 times
Been thanked: 3 times

Post by turbonigro »

This beloved amp is getting old or tired of my playing.
Plugging in into "High" Input and turning preamp pot from 8 to 10 behaves weirdly
  • loud squeal if cable is plugged in without an instrument;
  • with single coils middle pickup squeals and with bridge or neck pickup it has harsh sound when distorting;
  • with single coils turning down the tones remove the squeal;
  • With humbuckers there is no squeal at all;
Power tubes have been already replaced with matched EL34, bias 8uf caps replaced.
What could be the issue? Bad filter caps, preamp pot or misbiased tubes?

User avatar
mozz
Breadboard Brother
Information
Posts: 53
Joined: 26 Jan 2016, 23:07
Has thanked: 32 times
Been thanked: 15 times

Post by mozz »

Sounds more like preamp tube problems.

User avatar
phatt
Transistor Tuner
Information
Posts: 1191
Joined: 04 Aug 2010, 05:17
Location: Morayfield SE Qld AU
Has thanked: 7 times
Been thanked: 327 times

Post by phatt »

Agree, preamp valves are likely the issue, as they age they can become microphonic.
Try tapping the preamp glass with a chop stick, that can give you a clue as to which one is failing.
Phil

User avatar
turbonigro
Information
Posts: 33
Joined: 28 Mar 2010, 12:24
Has thanked: 4 times
Been thanked: 3 times

Post by turbonigro »

So yes, the issue is located at 1st preamp tube. Weird fact v1a grid gets negative voltage (-5) with preamp pot maxed. Whenever the voltage stays positive there is no oscillation. Can be filtering caps not working properly?

User avatar
mozz
Breadboard Brother
Information
Posts: 53
Joined: 26 Jan 2016, 23:07
Has thanked: 32 times
Been thanked: 15 times

Post by mozz »

Check your input jack to ground connection. Make sure there is a 1 meg resistor from input/grid to ground. You can't really measure grid voltage if it's floating.

User avatar
CheapPedalCollector
Resistor Ronker
Information
Posts: 255
Joined: 19 Oct 2020, 15:15
Has thanked: 302 times
Been thanked: 90 times

Post by CheapPedalCollector »

Late 70s JMP-1 and early JCM-800 all have the same issues.

You can fix most stuff purely by lead dress. Sometimes the shielded wire gets microphonic. They all have ground loop issues with eddy currents. If you have voltage at the gain control/grid it's likely a leaky signal cap and an easy fix. There's loads of info on amps at valve wizards website, Psionic Audio on youtube is also a great resource for techs although he's not geared towards the beginner.

User avatar
turbonigro
Information
Posts: 33
Joined: 28 Mar 2010, 12:24
Has thanked: 4 times
Been thanked: 3 times

Post by turbonigro »

I have checked if the coupling capacitor between v1 and v2 is leaky by measuring voltages a cross it and bridging it with another cap but no luck on this.
I have also tried to replace v1a grid wire with a shielded one and nothing.

Right now I am measuring negative voltages (depending on preamp pot) also on v2a grid too and on its 470k voltage divider resistor. What can cause this? Disconnecting the r104 470k resistor from hot terminal on low output removes all negative voltages

What I have noticed is increasing resistor r104 from 470k to 1M get rid of the oscillation (reduces the input signal current so it is a silly discovery)

User avatar
bmxguitarsbmx
Resistor Ronker
Information
Posts: 387
Joined: 25 Mar 2010, 21:15
Has thanked: 511 times
Been thanked: 162 times

Post by bmxguitarsbmx »

I don't see how you are getting negative voltages on the grid of any common cathode amplification stage. I think you are instead getting oscillation that your meter is integrating to read as negative. Oscillation above hearing range is common and during some cases, that oscillation can be moved to hearing range.

messing with grid resistor values is just incidental, you want to find the problem.

Check your filter caps.

Also, check to make sure there is continuity between all ground points. A disconnected ground can cause havok

User avatar
turbonigro
Information
Posts: 33
Joined: 28 Mar 2010, 12:24
Has thanked: 4 times
Been thanked: 3 times

Post by turbonigro »

bmxguitarsbmx wrote: 24 May 2022, 19:58 I don't see how you are getting negative voltages on the grid of any common cathode amplification stage. I think you are instead getting oscillation that your meter is integrating to read as negative. Oscillation above hearing range is common and during some cases, that oscillation can be moved to hearing range.

messing with grid resistor values is just incidental, you want to find the problem.

Check your filter caps.

Also, check to make sure there is continuity between all ground points. A disconnected ground can cause havok
After having discarghed (I will use Lee Jackson way using screwdriver touching chassis and then cap pins) can I just check continuity and resistance?

On the other hand can the coupling capacitor from low input be leaky and cause this oscillation? Measuring continuity trough it, seems to not be but multimeter is detecting some interference causing intermittent noise

User avatar
CheapPedalCollector
Resistor Ronker
Information
Posts: 255
Joined: 19 Oct 2020, 15:15
Has thanked: 302 times
Been thanked: 90 times

Post by CheapPedalCollector »

How long ago were the caps changed?

Oscillations can be tricky to find. I agree with bmxguitars, it's probably a high freq oscillation you can't hear, you should have no voltage on the grids of any stage. You can't test for leakage by measuring across a cap with another in parallel, lift the leg that is away from the plate resistor and measure DC at that point. The circuit boards are cheap and easily damaged, so be careful. I also suspect a bad ground, I mentioned that already. The pots are all grounded by a buss wire that is notorious for issues, especially when the pots can no longer make proper contact with the chassis due to corrosion of the lock washers or straight poor connections. There are known mods to help with the problems with eddy currents too that cause noise and oscillations.

Do you have a scope? Pull the v2 pre-amp tube and see if it goes away, then pull v1 if not, you can isolate where in the circuit it is quickly this way, or find a bad tube.

User avatar
turbonigro
Information
Posts: 33
Joined: 28 Mar 2010, 12:24
Has thanked: 4 times
Been thanked: 3 times

Post by turbonigro »

CheapPedalCollector wrote: 24 May 2022, 23:47 How long ago were the caps changed?

Oscillations can be tricky to find. I agree with bmxguitars, it's probably a high freq oscillation you can't hear, you should have no voltage on the grids of any stage. You can't test for leakage by measuring across a cap with another in parallel, lift the leg that is away from the plate resistor and measure DC at that point. The circuit boards are cheap and easily damaged, so be careful. I also suspect a bad ground, I mentioned that already. The pots are all grounded by a buss wire that is notorious for issues, especially when the pots can no longer make proper contact with the chassis due to corrosion of the lock washers or straight poor connections. There are known mods to help with the problems with eddy currents too that cause noise and oscillations.

Do you have a scope? Pull the v2 pre-amp tube and see if it goes away, then pull v1 if not, you can isolate where in the circuit it is quickly this way, or find a bad tube.
Thanks guys for the support. Sorry to not mention, as soon you told me about leaky cap I was trying to look and measure them.

The caps are actually 50yrs old. Some of these has dried out, sure..

I checked ground continuity and there are no problems, even though the blurbs are as cold as stones

I will let you know probe results

Thanks again

User avatar
turbonigro
Information
Posts: 33
Joined: 28 Mar 2010, 12:24
Has thanked: 4 times
Been thanked: 3 times

Post by turbonigro »

All mullard caps unsoldered and checked. All of them are fine.
Actually someone has been here, probably when I bought it 20yrs ago.
100nf and 680nf caps are Mika and fine. Some pots have been replaced.

Checking the pots I have noticed that the two newer had a metal touching the chassis and every pot has a washer. Should not be only 1 solid ground for all of them?

Tomorrow I will assembly again everything and check with probe.

What a ground mess

User avatar
CheapPedalCollector
Resistor Ronker
Information
Posts: 255
Joined: 19 Oct 2020, 15:15
Has thanked: 302 times
Been thanked: 90 times

Post by CheapPedalCollector »

I suggest to read this article http://www.valvewizard.co.uk/Grounding.html

50 years old power supply cap cans? Replace them, long over due. You don't need to worry about the mustard caps, those should be fine.

Replacing the pots was probably unnecessary, that kind of sucks.

User avatar
bmxguitarsbmx
Resistor Ronker
Information
Posts: 387
Joined: 25 Mar 2010, 21:15
Has thanked: 511 times
Been thanked: 162 times

Post by bmxguitarsbmx »

Great! Grounds are checked. ..It could still be Grounds, but... Power supply caps. Is the most likely thing.

I don't recommend that you mess around removing parts or testing anything until the Electrolytics in the power supply are checked, and most likely replaced. I'm actually surprised that your power tubes are not red plating. When you have oscillations, they often are only limited by the the maximum dissipation of the amp. Which means you can smoke some expensive stuff unknowingly. That negative voltage on the preamp grids is concerning. A leaky coupling cap would cause a positive voltage on the grid and so would grid leak current.

User avatar
turbonigro
Information
Posts: 33
Joined: 28 Mar 2010, 12:24
Has thanked: 4 times
Been thanked: 3 times

Post by turbonigro »

bmxguitarsbmx wrote: 25 May 2022, 22:23 Great! Grounds are checked. ..It could still be Grounds, but... Power supply caps. Is the most likely thing.

I don't recommend that you mess around removing parts or testing anything until the Electrolytics in the power supply are checked, and most likely replaced. I'm actually surprised that your power tubes are not red plating. When you have oscillations, they often are only limited by the the maximum dissipation of the amp. Which means you can smoke some expensive stuff unknowingly. That negative voltage on the preamp grids is concerning. A leaky coupling cap would cause a positive voltage on the grid and so would grid leak current.
F&T caps are coming and will replace them for sure. Power tubes are red plating indeed and I will check bias after cap replacement, again.
Coupling caps from v1 and v2 are fine I have measured their values and are close to 22nf, no continuity

Right now I am also reading that really interesting ground article and this amp is a ground loop combo. But I am learning and happy to have these resources

User avatar
deltafred
Opamp Operator
Information
Posts: 1628
Joined: 06 Apr 2010, 16:16
Location: England
Has thanked: 721 times
Been thanked: 273 times

Post by deltafred »

I've been repairing a load of valve (tube) amps recently so built a simple valve tester. (I had all the parts in my scrap box so have no idea what it would cost if you had to buy everything.)

https://valveheaven.com/wp-content/uplo ... tester.pdf

I modified it slightly :-
Instead of building in meters I put binding posts so I could plug a couple of multimeters in for measuring grid and plate voltages.

Instead of the flying leads and sockets I brought all the connections out to a 25 way D type socket and have a range of 25 way plugs for the common valve types (ECC8x, EL84, EF86, EL34/6V6, etc).

(I have a ton of 25 way D type plugs and sockets from old RS232 cables that were being thrown out from where I worked otherwise would have used flying leads and sockets.)
Politics is the art of so plucking the goose as to obtain the most feathers with the least squawking. - R.G. 2011
Jeez, she's an ugly bastard, she makes my socks hurt. I hope it's no ones missus here. - Ice-9 2012

User avatar
bmxguitarsbmx
Resistor Ronker
Information
Posts: 387
Joined: 25 Mar 2010, 21:15
Has thanked: 511 times
Been thanked: 162 times

Post by bmxguitarsbmx »

turbonigro wrote: 26 May 2022, 12:25
bmxguitarsbmx wrote: 25 May 2022, 22:23 Great! Grounds are checked. ..It could still be Grounds, but... Power supply caps. Is the most likely thing.

I don't recommend that you mess around removing parts or testing anything until the Electrolytics in the power supply are checked, and most likely replaced. I'm actually surprised that your power tubes are not red plating. When you have oscillations, they often are only limited by the the maximum dissipation of the amp. Which means you can smoke some expensive stuff unknowingly. That negative voltage on the preamp grids is concerning. A leaky coupling cap would cause a positive voltage on the grid and so would grid leak current.
F&T caps are coming and will replace them for sure. Power tubes are red plating indeed and I will check bias after cap replacement, again.
Coupling caps from v1 and v2 are fine I have measured their values and are close to 22nf, no continuity

Right now I am also reading that really interesting ground article and this amp is a ground loop combo. But I am learning and happy to have these resources
Great! I wouldn't do anything until you re-cap the power supply electrolytics. Redplating power tubes can smoke the output and or power transformer. Don't bother measuring any voltages either- it is oscillating and no voltages will be stable.

User avatar
turbonigro
Information
Posts: 33
Joined: 28 Mar 2010, 12:24
Has thanked: 4 times
Been thanked: 3 times

Post by turbonigro »

bmxguitarsbmx wrote: 27 May 2022, 22:35
turbonigro wrote: 26 May 2022, 12:25
bmxguitarsbmx wrote: 25 May 2022, 22:23 Great! Grounds are checked. ..It could still be Grounds, but... Power supply caps. Is the most likely thing.

I don't recommend that you mess around removing parts or testing anything until the Electrolytics in the power supply are checked, and most likely replaced. I'm actually surprised that your power tubes are not red plating. When you have oscillations, they often are only limited by the the maximum dissipation of the amp. Which means you can smoke some expensive stuff unknowingly. That negative voltage on the preamp grids is concerning. A leaky coupling cap would cause a positive voltage on the grid and so would grid leak current.
F&T caps are coming and will replace them for sure. Power tubes are red plating indeed and I will check bias after cap replacement, again.
Coupling caps from v1 and v2 are fine I have measured their values and are close to 22nf, no continuity

Right now I am also reading that really interesting ground article and this amp is a ground loop combo. But I am learning and happy to have these resources
Great! I wouldn't do anything until you re-cap the power supply electrolytics. Redplating power tubes can smoke the output and or power transformer. Don't bother measuring any voltages either- it is oscillating and no voltages will be stable.
I have just finished the replacement of filtering caps and the squealing is still here. To bias correctly the power tubes I have also installed 1ohm resistors a cross pins 8 and 1. Plate voltage is 375v (it is somehow floating a bit) and first power tube is biased at 60% 40ma meanwhile the second is seeing 30ma...

I have no idea now.

User avatar
CheapPedalCollector
Resistor Ronker
Information
Posts: 255
Joined: 19 Oct 2020, 15:15
Has thanked: 302 times
Been thanked: 90 times

Post by CheapPedalCollector »

You shouldn't even have the power tubes in while you have the oscillation.

I suggest taking your amp to a qualified technician.

User avatar
turbonigro
Information
Posts: 33
Joined: 28 Mar 2010, 12:24
Has thanked: 4 times
Been thanked: 3 times

Post by turbonigro »

CheapPedalCollector wrote: 28 May 2022, 20:05 You shouldn't even have the power tubes in while you have the oscillation.

I suggest taking your amp to a qualified technician.
I just want to probe the circuit with the oscilloscope. One last try

So to recap, in case of oscillation the rules of thumb are:

- take out power tubes
- look for microphonic preamp tubes
- check lead dress
- leaky capacitors can be tested only lifting one leg from the circuit
- negative voltages could be just oscillation feeding back into multimeter
- check ground loops (thanks for the article)

Even though I was not able to repair this amp I have learned a lot. Thanks guys

But by next week I will call a qualified technician.

Post Reply