Marshall 9004 Rackmount Preamp Negative Voltage Supply

[bobloblaws]

Hi, I have a Marshall 9004 rackmount preamp here (schematic attached) which is not producing sound output, apparently on account of the negative voltage rail not getting the required -12V. I replaced D2 and REG2 but that did not help. I measured ~20V on both sides of D1 but on D2 I measure ~20V on the incoming side and + 0.6V on the REG2 side, and I did make sure the diode is oriented correctly. The output of REG2 is about +0.7V where it should be -12V. I'm relatively new to solid state electronics, any suggestions on further troubleshooting? Could C48 be the culprit? I haven't investigated that yet because I haven't had the PCB out of the unit yet, only worked from the component side so far. Thanks.

- B.L.

Marshall-Rackmount-Preamp-9004-Schematic.pdf

[g1]

Yes, C48 could be the problem. Do you measure a low resistance across it?

[olddawg]

Or any other hard short after the regulator. Is anything hot to the touch? An old trick is if you have a camcorder or camera with an IR (low light) mode. Hot components will glow through the viewfinder.

[bobloblaws]

Yes, C48 could be the problem. Do you measure a low resistance across it?

I was able to measure resistance across C48 in circuit by attaching my DMM leads to the connected end of D2 and the positive terminal of C45. The reading I get is ~16Mohm. What sort of resistance would we be expecting?

[Enzo]

A shorted part will have a low resistance, but it could also just be super leaky, and so acts as a short at anything over (I'll make it up) say 4 volts. Your meter cannot see that. I doubt the cap is bad.

Um, how about D2 is open? Or the solder is broken or a trace is cracked. If C48 were shorted, you would blow fuses. It would mean D2 was across the AC voltage and ground.

If C48 were shorted you would have zero volts on it, or maybe a small negative voltage. But you have a positive 0.6v, and that positive voltage is leaking in from the +12 supply through the circuits. That makes an open D2 my prime suspect.

[bobloblaws]

Um, how about D2 is open?

When you say "open", do you mean that it doesn't conduct any current? Anyway, I snipped it out and tested it and also swapped in a brand new diode.

Or the solder is broken or a trace is cracked.

I checked the continuity of everything I could in the vicinity, but C48 and C43 are radially oriented and flush with the board so I guess I need to dismantle this unit so I can get the PCB out and check those solder joints.

[Enzo]

Unplug unit from wall, but turn power switch ON. So unit is not powered but in ON condition. Measure resistance from cathode of D2 to ground. Also measure resistance from Anode of D1 to ground, they SHOULD measure identical, do they? That will verify a good connection to the transformer.

Cathode is the line end.

Alternatively one could measure resistance between Anode of D1 and cathode of D2, should be zero ohms as they should be wired together.

If you don't have the -20v or whatever it is right at D2, then it won't be there further down the line.

[bobloblaws]

Unplug unit from wall, but turn power switch ON. So unit is not powered but in ON condition. Measure resistance from cathode of D2 to ground. Also measure resistance from Anode of D1 to ground, they SHOULD measure identical, do they? That will verify a good connection to the transformer.

Cathode is the line end.

Alternatively one could measure resistance between Anode of D1 and cathode of D2, should be zero ohms as they should be wired together.

If you don't have the -20v or whatever it is right at D2, then it won't be there further down the line.

They measure the same. In fact, they are physically connected (Edit: you already mentioned that, missed it the first time, sorry), the continuity tester confirms it and I can see both diodes are connected to the same trace on the PCB. It did indeed seem odd to me that D1 had +20V on both ends while D2 had +20V on the incoming side and the aforementioned +0.6 on the other side. My understanding is that since the only thing different between D1 and D2 is the reverse orientation that I should measure -20V at that point (instead of +0.6V). Since a fresh diode did not alter the situation I speculated that it had something to do with loading by components further downstream which is why I swapped out the REG2 transistor. If you are convinced that C48 is not the culprit then I am at a loss.

[Enzo]

Oh convinced is a strong term, always check anyway.

[Mick Bailey]

These voltage regulators have built-in short/thermal/overload protection and if the output load is too high (or shorted) they won't produce any output voltage. However, the unregulated input will read good. So I'm thinking your output loading is OK. Anything that will drop a 20v supply down to zero (or as you read it, +0.6v) has to be pulling a lot of current and something will blow or get hot. Now here's something to consider - the absence of a voltage doesn't imply 0v potential. So in this case if C48 was a dead short you'd read 0v across it, or if D2 was open you'd also read 0v across C48. The difference is in the first case you're reading 0v as a short and the other 0v as an open. Of course, you can have both conditions simultaneously; shorted cap and open diode.

I would lift D2 at the regulator side and make sure you have a voltage present. Also, triple check you have the orientation correct and the regulator is a 79L12 (not 78L12). I still suspect D2 is open, though. Not impossible for the replacement part to be bad.

One other thing to check - make certain your DMM battery is good. When I get odd voltage readings that's my first suspect.

[bobloblaws]

Oh convinced is a strong term, always check anyway.

Point taken, will do.

[bobloblaws]

Arrrgghhhhh! I am such an idiot! The problem was I initially read "15V" at the transformer input jack and feeling giddy that I just happened to have a 15V step down transformer in my possession saving me hunting one down in the vast expanse, I missed the fact that it said "15V AC". And of course my transformer is DC output (obviously I should have known by virtue of the design of the circuit, but like I said, I'm a relatively newbie, it just never jumped out at me). *Sigh* Sorry to waste everbody's time, but at least I may have picked up some valuable information from your responses anyway, so thank you.

[g1]

One more little tip.
~ symbol is often used to denote AC volts. Look back at your first post. When I first read it, I assumed you had AC voltage there. If you had said DC, one of us probably would have caught it sooner.
(or another possibility is you were using the AC volts range and have a meter that doesn't block DC when on the AC range?)

[Enzo]

That is how I read it.

In fact that is one reason I suggested reading the resistance to ground. An AC source would be a transformer and would have continuity, a DC source would face a rectifier.

[bobloblaws]

One more little tip.
~ symbol is often used to denote AC volts.

Yes, in fact that symbol is printed on the back of the unit underneath 15V AC. I was just being an airhead and not paying attention or using my head. Gotta learn to slow down a bit and work smarter.