Marshall 8040 repair

[SimoS]

Hello everybody.
I'm a newbie in amplificators repair, but hungry to learn.
My (little) experience is more on tube amps but now I'm working on a Marshall valvestate that is an hybrid machine (attached the schematic).
If I attach any cable to the input, also without signal, the amp become very noisy, not a specific frequence (more around 350-500Hz, I've analyzed using an app on the phone), growing when I raise the volume / gain.
I cleaned all, controlled tracks, but no results.
Surely the problem is in the pre-amp part and I concentrated my attention to the TL072 op (IC3 in the schematic) and particulary to the IC3a part.
The power is correct (-12V/+12V to pin 4/8), but with no signal the pins 1/2/3 are all aroung -11.5V, while on the other side pins 5/6/7 are 0V.
I don't know about OP but it seems strange to me, also thinking on how works a tube.
What do you think about? May be an issue of the TL072? To be substituted or before I've to check something other?
Thanks in advance, guys.
 

[orbanp]

Hi SimoS,

It is the voltage on the + input on this op-amp that sets the voltages of the output and the - input.
As you mentioned, it is at about -11.3V.
Do check why is it so!
It should get 0V through R29, 1 meg, resistor. Is the resistor OK? Is the lower half of that resistor at ground level (as it should be)?
And yes, the op-amp also could be faulty. But I would start with R29.

Good troubleshooting so far!

Good luck, Peter

[coromonadalix]

you may have ground problems ... if you connect an cable, it may act as an antenna ...  check all the ground plane(s) and shielding ...

[b_force]

This could also be just a fried opamp

[mzacharias]

If you plug in an "open circuit" wire or cable, noise will be the normal result. That is why the input jack has a shorting switch built in.

If instead you are plugging in an instrument or a signal generator, using a good cable, there should be only normal noise and signal from the source.

[SimoS]

I checked R29. The ground side is Ovdc, the side connected to Pin3 ofTl072 is -11.2 Vdc. In your opinion is not strange to have a Vdc value here? I expect only the signal coming from the input jack. does it means that the Op is fault?

[duak]

First, is the opamp in a socket or soldered in directly?  There is a small chance that if there is a socket that the opamp is not making a proper connection. Try pulling the opamp out slightly (1 mm) with a screwdriver and pushing it back in. 

I think the opamp is damaged and must be replaced.  A high voltage could have been applied to the input circuit and I don't see any protective devices.

If this could happen again, a simple change is to add some diodes connected back to back across R29.  The bottom line of the attached file shows three different options depending on the types of diodes chosen.  (Please disregard figures on the upper line.)  If interested I can explain further.

[SimoS]

I'm going to search a TL072 to substitute the existing one. I hope is not so difficult to unsolder the existing one and replace it.
Surely is interesting to introduce a protection to avoid the same problem in the future.
I red on many forum that the problem may be generated by the insertion/extraction of the input jack with the amp on; this generate some spikes possibly dangerous for the TL072. What do you think about? The protection can be built around the jack and not on the board? It seems difficult for me to operate on the board where the space is so limited.
Thanks for your help.

[duak]

SimoS, I am sorry for taking so long to reply.

Yes, I think the opamp can be damaged by inserting the plug.

The easiest way to remove the opamp is to cut the leads right at the body.  The leads may then be removed one at a time.

The TL072 is the most common type of opamp that will work in that circuit because it has FET inputs.  Some other common opamps such as LM358 have bipolar inputs and will not work well.

Two Zener diodes may be connected back-to-back across the input connector - see "input clamp" attached (please excuse the poor quality).  1N5231 500 mW 5.1 volt diodes are common in North America but other types may be used.  Note that the Zener diodes may change the timbre of the sound slightly due to their capacitance.

I see that Marshall adds protection circuitry in the MG10. This requires adding two 1N4148 diodes to the circuit board.  This should not affect the sound.  Please note that the diodes must be connected correctly or the amplifier will be damaged.

Cheers,

[SimoS]

Thanks Duak.
I substituted the TL072 and now the amp is OK :-)
I cut the pins as you said and solded the new Op on the pins without unsolder them from the board; much more simple ;-)
I found in my stuff 4 zener diode with impressed C5 PH. Do you think I can use them for the input protection?

Ivan

[b_force]

I would rather use the method with the diodes instead of zeners, since zeners are a lot more noisy.
Which is not a very good solution because of the huge gain in the amp.

[duak]

Ivan, I don't know the characteristics of your zener diodes C5 PH.  Do you have more information or a picture?

b_force, I believe the noise generated by zener diodes only occurs when they are conducting significant amounts of reverse current.  In normal operation, the signal level is so low that virtually no reverse current flows.  However, it would be interesting to try this out.

Cheers,