Loud hum and blown fuse

[ksjazzguitar]

I'm having a problem with my guitar amp (a Polytone Mini-Brute, 13 years-old.) I was on a gig, taking a break, with the amp left on. I was out of the room, but I am told that there was a loud hum for a few seconds and then it was dead. I checked and the fuse was burned out. When I got it home, I replaced the fuse, and a loud hum came out, even though nothing was plugged in. The loud hum lasted for about 2-3 seconds but seemed to be unaffected by the volume nobs. The fuse burned out again.

Any ideas?

One source said that I should check the filter caps. I assume that they are the big electrolytic caps down by the power supply. They seem to be intact (there is a slight bump on the top, but not much) and the solder joints seem OK.

What do you think?

[guitician]

I wouldn't assume that the caps are bad. I failed transistor/IC could draw an excess current, if it has failed, and cause those same symptoms.

[52 Bill]

My first guess on that amp, would be a shorted output transistor. The hum that you are hearing is probably DC voltage on the speaker.

I don't know what your skill set is, but it will require basic skills and at least a multimeter to troubleshoot it.

[Jag]

Also, disconnect the speaker until you have the issue resolved. If your speaker keeps getting blasted with DC, it won't live very long!

[ksjazzguitar]

I have a "basic" skill set: I can use a DMM, I know how to solder, I know what a transistor is, etc.

Any advise on how to trouble shoot this?

******

OK, I hooked up it back up, but without the speaker. I did measure a small DC current (5mA) before the fuse blew. It may have been more, but it happened so fast, the DMM may not have registered it.

These are the schematics:

http://www.murchmusic.com/Polytone%20Info/Schem1.JPG
http://www.murchmusic.com/Polytone%20Info/schem2.JPG

Any ideas?

[52 Bill]

Start by checking for dc on the amp output. Replace the power fuse, unhook the speaker wires, noting how they were attached. If I remember correctly there are two sets of wires going to each terminal, one set from the power amp and one set to the extension speaker socket.

Take your meter and set it up to read dc volts. With the amp turned on, carefully take a reading across the two wires coming from the power amp. Be careful not to short the two wires together. If there is any DC voltage there, you will have to remove the power amp chassis from the bottom of the case.

Let us know what you find out.

[Enzo]

I would head straight to the power transistors and start looking for shorted ones. And while I was about it, I would check the main rectifiers or bridge if it has that.

[ksjazzguitar]

OK,

It's hard to see if there is a DC voltage across the speaker terminals as the fuse burns out in 1/2 a second and the DMM takes longer than that to react, but there did seem to be a blip of DC voltage.

Forgive me, because my knowledge of of electronics is more theoretical than practical (3.5 years of electrical engineering before switching to music). I assume that the output transistors are the MJ4502 and the MJ802 (Q5 and Q6 in the linked schematics) - the big metal cans (TO-3 packaging) that are attached the the heat sinks?

Assuming that these are what we are talking about, I took them out and tested them with my DMM and it's diode setting.

For the MJ4502, it got the following readings:
(red to black, then black to red)
BE .513 -
BC .510 -
CE .003 .004

For the MJ802, it got the following readings:
(red to black, then black to red)
BE .659 .477
BC .661 .477
CE .004 .004

So, the continuity on the CE, that alone tells me these are shot, yes? I should swap them out?

Even if the answer is yes, it still begs the question, were these the cause or the symptom? After a decade and a half of hard use, will they just blow out? I saw one of the guys in the restaurant fiddling with the thermostat nearby, could a power surge have caused this?

[Enzo]

They are shorted anad must be replaced.

One failed. And when one fails, it usually takes the other with it.

[Steve Conner]

Yup, whack a new pair of power transistors in there and get back to us... If you can't find the originals, the MJ15024 and MJ15025 would exceed the original spec. (edit: no, they don't meet the minimum current gain spec)

Don't forget you will probably need to "rebias" the thing just as if it was a tube amp getting new power tubes.

Also search the forum for "light bulb limiter", this gadget is cheap and easy to make, and if there's anything else wrong (you asked about cause vs. symptom) it'll save your new transistors from blowing out again as soon as you power it up.

I think those Polytone amps have poor heatsinking, so maybe the output transistors just gave up after 13 years of running too hot.

[52 Bill]

You should also check the driver transistors while you have the amp open.

Steve's point about heatsinking is a good one, as the power amp is enclosed inside the sealed speaker cabinet. I've always thought that the worst case scenario for a Polytone amp is a long outdoor gig, on a hot summer day, with the amp sitting in full sunshine.

[ksjazzguitar]

OK,

Thanks for all the advice so far. I think we're making progress.

Forgive my ignorance, but how do I bias the amp. I understand the basic principle, just not the practice. Is it explained somewhere in the forum already? I've found the "bias adjust", just need to know what to measure and it should be.

And Bill,

How would I identify the driver transistors? (Sorry again, I',m a bit of theory and little practice.)

[52 Bill]

The drivers are the ones on the pc board that connect directly to the outputs.

The trim pot in the power amp section allows you to set the bias/idle current. After the repairs are done you will need to perform this adjustment. There are a few of different ways to do this. You can set it by monitoring the AC current draw of the entire amp, or you can read the voltage drop across the ballast/emitter resistors to figure current draw, or you can drive the amp with a signal and adjust for the least amount of crossover distortion as viewed on an oscilloscope, etc.

As I'm sure you will soon find out, everyone here has their own favorite method.

[ksjazzguitar]

OK, thanks for everyone's help so far.

Here's where we are. I replaced the output transistors. The electronic supply store didn't have the MJ4502 and MJ802 - they suggested that I use NTE180 and NTE181. I checked the NTE web site and they were listed as substitutes so I used them. I replaced them and plugged everything in and voila, no hum, no fuse blown after .5 seconds. I plugged in the guitar and actually got a sound. Of course, the sound was a little fuzzy, I assume that it needs to be biased.

That brings me to my last (I hope) problem: How do I bias this thing? Unfortunately, the internet is full of info on how to by tubes, but nothing about biasing solid state amps. I've found the bias adjust trim pot, but, gee, wouldn't some test points and the proper settings written on the board be nice? No such luck.

Bill mentioned 3 methods: AC current draw of the entire amp, voltage drop across the ballast/emitter resistors to figure current draw, or drive the amp with a signal and adjust for the least amount of crossover distortion as viewed on an oscilloscope.

Since I have no access to an o-scope, that rules out the third option.

Would someone be so kind as to describe a method for either of the first two, how to do it, and what the reading should be?

Thanks again, almost there.

[Enzo]

wouldn't some test points and the proper settings written on the board be nice?

All the solid state amps are the same. Unlike tube amps, solid state amp bias is not a matter of taste. The bias setting for one of these is when the crossover distortion just disappears. When you see some note like "adjust for 10mv across R1234" that is not really setting the bias, that is just adjusting numbers to some average the factory uses to get close.

And test points? ALmost always it will be one of the output ballast resistors.

Everyone cringes when PV says "set the bias for -55v." "No no no" they cry, "we must set it to 70% of whatever." Just so in SS land. if you get some spec like Xmv across some resistor, it is only an approximation.

If you don;t have a scope, you can't easily watch the distortion notch. But if you understand what the amp is doing, it is simple to adjust.

These amps are essentially running in class B. Each side - positive and negative - should turn off at the exact moment the other side turns on. The two sides should hand off seamlessly one to the other. If you adjust the amp cold, one side gets to zero too early and it has to wait a moment for the signal to start the other side going. That results in the crossover notch distortion. And going the other way, if the second side turns on before the first side turns off, then both are on together, and that results in current flowing from one powr rail to the other through the outputs, and excess current flows. When that happens the amp runs hot.

SO consider that the crossover notch just disappears at the point both sides are about to conduct at the same time. That means that if you set the bias any hotter, then that excess current draw will start. So a simple trick is to watch the mains current draw. Start with the bias at its coldest, and slowly turn it up. The mains curent draw will stay low until at some point all of a sudden it starts to rise with the adjustment. The point at which it starts to rise is the point at which the two sides start conducting at the same time. BY adjusting the amp right at the point of rising mains current, you have it set on the threshold, where it belongs. Back off just a hair from current rise and you are done.