Marshall MG50DFX

[lwn9186]

My son has one that Marshall replaced once under warranty, now it's out of warranty and the thing won't change channels, the led stays lit weather the switch is in or out. Any help for a newbie????

Thanks

[Amp Kat]

footswitch jack

[lwn9186]

Mg50dfx

I unplugged the harness connector for the rear jacks and the light is still lit either on or off. Thanks

[lwn9186]

Any other ideas???

[Enzo]

Unplugging the rear board is guaranteed to kill the switching circuit. The circuits run THROUGH the footswith jack. The cutout contacts in it complete the circuit when it is not in use.

The panel switch itself must close for the FS to work. See that it does.

There are two zeners near it that are across the control lines, either one shorted?

[lwn9186]

I desoldered the jumper and took the cap out to check the switch and it is ok. Both zeners check ok. I took the 2n7000, TR1 out of the circuit and have no light, but do when the switch is on. I'm not real sure how to check it. Thanks

[Enzo]

OK, without TR1, the FS turns the panel light off and on? We need that to happen. Does the panel switch also do this when TR1 is gone?

Look at page 2 - the switching circuit. Oh wait, first: Well on page 3, note the footswitch normals pins 2 and 3 of CON1 together with ZD1 protecting against overvoltage. On page 1 note pins 2 and 3 of the connector are FS B and FS A. NOW look at page 2.

From those notes, we know now that the FS jack shorts FSA to FSB. SO the LED is connected to the panel switch through the FS jack on the other page. If the FS jack and the wires to and from it are OK, then the panel switch SW1 should turn the LED off and on. The LED is wired to +5V through 470 ohm R76.

If those do not control the LED, thenyou have a wiring problem somewhere.

If they do, then apparently TR1 is screwing things up.

TR1 is a simple inverter. Whe the LED is on, the cathode is grounded through the FS etc. That means that R81 and the gate of TR1 are grounded. SO R81 has zero volts on ot, and that means control signal SW_B is "low." When the FS opens - or the panel switch, the LED is no longer grounded, current no longer flows through it, it goes dark. That also means the +5v comes through it and on through R81 to SW_B, bringing it "high." High/low, on/off, 1/0 - all that digital logic stuff.

Now TR1. If the gate of TR1 is grounded, TR1 is off, won;t conduct. So it does nothing to stop +5v from coming through R79 and on through R82 to SW_A. SW_A and SW_B are control signals to the DG212 ICs that switch channels.

In the absence of a grounded FS, R78 is free to pull R81 and the gate of TR1 up to +5v.

If the gate of TR1 is put at +5V, it will turn on and conduct, and that will pretty much ground R82, bringing SW_A down to ground, or "low."

SO whenever SW_A is low SW_B will be high, and vice versa. Hence TR1 is an inverter.

If TR1 is screwing things up, then I'd say it was either shorted or installed wrong.

Note that without TR1, SW_B should still go high and low with the switch.

Looking at page 1, you can see that two of the channel switching elements are controlled by SW_B. So if the LED goes off and on, partial channel switching should occur.

[lwn9186]

Yes with the rear harness plugged in and TR1 out of the circuit, the light switches on and off like normal except for a small blip of the light when the switch is first switched. Which I believe is cap 75 being charged. TR1 was installed correctly so I would have to assume it is bad. This amp has had little playing time as my son is away at college and only gets a chance to play when is home for the summer.
Thanks for all the help.

[Mars Amp Repair]

It is verrrry common for the mosfet to be the culprit causing all these switching issues. For simplicity and because of a solid history of issues with this switching mosfet, I'd first replace the little bugger first.

As an aside, I have one customer who has a music studio for guitar training & he has 2 of these Marshall amps with the mosfet switching. I have now replaced the mosfet in each of the 2 amps 3 times.

I had him do all kinds of countermeasures to reduce static electricity in his studios as we are very dry here in Denver, etc. to no avail.

I finally found a replacement Mosfet that is stronger & used that. Within 4 months it fried again (BS170 & BS270) from Mouser.

The owner's sidekick finally leaked that they have power issues in the location as well as A CELL TOWER ON THE ROOF RIGHT ABOVE THE STUDIO!!!

The amps are going out of warranty soon, so I suggested he just trade them for another amp that hopefully won't be sensitive to this issue. For all the questions I asked the owner about local interference etc, he never mentioned the biggest source less that 10ft away on the roof...big surprise...

g

[Enzo]

My first thought when I spotted that MOSFET a long time ago was - wouldn't something like an MPSA13 pop right into place there?

[lwn9186]

Ok guys, your way above my level, but I went ahead and ordered some 2n7000, some bs170's and 270's from Newark, Marshall parts online was out of them. We will see if thats the culprit when they come in. Thanks way much again..........

[Mars Amp Repair]

OK,
I'd just try the 2N2007 as other than that one place, I've never had a repeat repair on the same unit for that issue. ALSO IMPORTANT!!! the 'BS' mosfets have a different leg arrangement...you'd need to re-orient the 'BS' mosfet 180deg from the 2n2007 to get the pin-out correct. see attached (if it makes it through: glen

[Mars Amp Repair]

OK, so the attachment didn't get through...anyway, just use the 2n7000 & I'm sure your problems will be solved. If you want to see the basing & datasheets for any semiconductor, you can go to this link and just type the part number in the search box on top. Then you can pick the manufacturer or the part you want to use:

Datasheet - Part name start with

glen

[Enzo]

What do you think of my MPSA13 idea Glen? Or other similar little darlington? ANy reason it wouldn't work?

[lwn9186]

Enzo, I was thinking the same..................Thanks for the thought.......