Marshall Mode 4 (mf-350) IC's blowing up

[52 Bill]

Thanks glen, noted!

But with all four new ICs connected, no load, variac low and limiter operating, one of the ICs is getting hot and the others not. Could that be a main board problem? Or must it be somethingon the IC board - or the IC?

I hesitate to suggest this, as these things are so flakey, but if you move the one module that is getting hot to another position, does the same module get hot or does the same position bet hot?

[Enzo]

I cannot over emphasize how sensitive this stupid circuit is to anything out of the ordinary.

NEVER ground any part of the output, including connecting a scope probe ground lead.

The output of this amp is a bridge, with two modules per side. But the two on each side are not parallel, one is a follower of the other. When you leave one out or operate one at a time, it or they find themselves in an incomplete circuit. Imagine trying to service a conventional SS power amp by removing one driver or the outputs on one side. The resulting operation and voltage/current readings won't be meaningful, and could be harmful to the remaining circuitry.

In my experience, there is not much on the main board that hurts them, the power modules live and die on their own. One thing they DO NOT LIKE is being connected when there is any charge in the power supply. Of your filter caps are charged, even with power off, if you plug in a module cable, and you get a little spark, that is a very bad thing. All the contacts in the connectors do not make and break at the same instant, so voltages may appear at some pins before others. The net result of that can be what amounts to a reverse voltage between two pins on a module for a brief moment. But that can be enough to burn out a diode inside an IC. About the only main board trouble that arises is the occasional blown mute/standby part. That doesn;t hurt the power modules, it just leaves the amp muted.

[Alex R]

Sounds to me like I took out one of the ICs when I fitted it/plugged it in. Sigh. Good job they're pretty cheap.

[Mars Amp Repair]

yeah,
I would suggest very strongly against swapping the connecting plugs between the ic's...esp between heatsinks as each pair of ic's share the same B+ or B- depending on the heatsink.
Oh yeah BTW, the heatsinks are electrically HOT.

At this point you will definitely need to replace all four ic's again. Remember after replacing the ic's to never power it up or variac it up with a load connected. It does a good enough job of blowing itself up without trying to do it into a load.

I outlined a method for powering these things up when you have a really flakey situation like yours. you'll have to do a search on the MF350 in this forum. If things are wrong, even 5-10vac from the variac is enough to blow the ic's.

Note: if you have removed & replugged those 15pin ic board connectors very many times, you may have already created an intermittent problem over and above your original problem. those connectors are not available. I had to hard wire all four of them on both ends on the one nightmare MF350 I fixed once. that's 30 connections per cable times four, but it was worth it to finally get the thing working & make back the time & many IC's I blew working on it.

Just realized I didn't actually explain the method...I'll look in my symptom/remedy file & try to get it on here...


glen

[Alex R]

Thanks Glen. I've done the ICs on three or four of these now and I reckon most of what could go wrong has gone wrong.. but my heart's in my mouth the whole bloody time with them, I've lost so many ICs and then all that soldering starts again... it would be good to have a method that works better than mine, definitely!

[Steve Conner]

Someone should make an aftermarket power board for these things that replaces the modules with a bunch of good old-fashioned TO3s. I'd do it if I thought it would sell enough to make it worthwhile.

I don't actually understand what makes them so temperamental. As other posters said, two of the ICs are driven by the preamp, and the rest are just power followers. Theoretically, you should be able to remove all but the two master modules and it should still work, as long as you don't connect a low impedance load and drive it hard. If you reconnected the speaker unbridged, it should work with only one module.

But as we all know, theory is the same as practice, in theory. My gut feeling would be that the modules go into push-pull parasitic oscillations under some circumstances, and blow each other out.

[icefloe01]

...Note: if you have removed & replugged those 15pin ic board connectors very many times, you may have already created an intermittent problem over and above your original problem. those connectors are not available...


glen

Any information on these? Or at least a pic of them?

Eric.

[Mars Amp Repair]

sorry, computer took a d$%p for a few days. These danged ic's are so intertwangled, it is difficult to say why one would get hot like that.
My take is it's already too late & the IC is bad.
My records inidcate from the one MF350 I had so much trouble with:

1. Monitor voltage at speaker output with NO LOAD.

2. Also monitor with another voltmeter, one of the rails (pin 12 or 15 on the IC board plugs).

3. Bring up the variac so that you only get about 2v on the rail (yes only 2 VOLTS) while looking for dc offset.

4. mine had issues with the jumper wire connections, so at this point I wiggled those wires to try & figure out where the bad connection was without blowing up the ic's everytime.

5. If no offset, switch off the variac leaving the dial in the same position that rendered 2v on the rail & move your rail meter to the next test point (unless you have about 3 meters ;-])

6. Move your meter to pin 2 on plug con6 or 7 to the main board on the left most IC of the pairs (looks like you could do this on any of the ic's). You should have nearly 0vdc here. Once you have any appreciable dc here, raising the variac will result in POW!

7. Wiggle the connection wires to all 4 IC boards watching closely to this voltage.

8. As mentioned, my issue was with the connector wires. I think this method just allows you to do some fundamental checks without the IC's blowing themselves to smithereens everytime.

It turned out that the previous servicer had created a problem on the main board while trying to replace the power IC's & in an effort to fix THAT self inflicted issue, they had worn out the IC board plugs pulling them on & off as they replaced a bunch of IC's on the main board. The 3 stooges then created yet the 2nd self-inflicted problem by tearing a clad on the main board trying to fix the 1st self-inflicted problem.

I'm fairly certain the only problem this amp had to begin with was blown IC's.

I'm not certain how much help it might be for you, but cha know...It vootent hoit...glen

[icefloe01]

Any chance of getting a pic of those connectors?

Thanks,
Eric.

[Mars Amp Repair]

here's the schemo...no pix, tho...glen

[teemuk]

here's the schemo...

...And if anyone gets a headache from following the signal path this little simplification might be useful:

[Steve Conner]

Yikes! Oh well, I guess there's nothing wrong with it that throwing out of a 9th floor window wouldn't fix.

The theory behind it is quite sound, and similar to the other Valvestate amps. The left-hand pair of amp chips have their output impedance raised by current feedback, to approximate the poor damping of pentode tubes without feedback. Essentially the signal from the preamp now controls output current, instead of voltage. A feedback network like the one in a classic tube amp, with resonance and presence controls, is then placed round the power amp to lower the impedance again.

The right-hand pair of chips are set up as ordinary voltage amps with a gain of -1, feeding off the output of the left-hand pair, so they just do the opposite of the left-hand ones, quadrupling the power output into a given load impedance.

[Alex R]

thanks guys. I gave in with the iCs when the second lot popped and ordered four whole new IC boards from marshall, not to dear. When they arrive I will follow glen's methods TO THE LETTER...

[Rick Erickson]

[If you plug one of these into your dummy load and you've hooked your scope up across the load, the scope ground will of course ground the screen and blow the ICs. Oh yes. Been there.]

Which is exactly why my scope, my soldering iron and my signal generator all have the ground connections lifted. Of course I don't recommend you do that, it wouldn't be safe.

RE

Big Brother IS watching you...

[Enzo]

Does that work here? Even if all your test gear floats, are any of them connected together? Plugging a jenny into the input would put the scope at ground potential if they are commoned. And can't let the test gear touch each other.

I prefer using an isolated power outlet for the unit under test instead. Whatever is succesful is fine, different strokes and all.

I find it easiest here to ground my scope to chassis and probe either side of the load.