MIMF

Warning, sensitive viewers might want to avert their gaze.

Before I disassemble this amp, is there any chance that the pcb traces could still be intact or decipherable?
It was free to me.

Hi David, that pcb looks pretty charred, looks like most of it hasn't burned all the way through, so you should be able to decipher the traces, but I'd recommend excising all the carbonated pcb and adding some perfboard or similar to replaced the damaged area, carbon deposits can have all sorts of nasty effects on circuits, even low voltage/impedance ones like output amplifiers.
The tricky bit will be determining parts values without the aid of a schematic if you are unable to find one. Some of the burnt components may still read with their correct value, the coating will burn off before the part is completely ruined usually.

This is the type of project I'd tackle for myself, but would be unlikely to for a customer, they'd probably not be too happy with the bill, it's going to be a labour of love.

Ah, looks like the schematic is commonly available (although apparently there were a couple of runs with some circuit variations). I've never needed a GK schematic, but apparently they're receptive to schematic requests.

Just an aside, it can sometime be extremely frustrating working in music electronics repair in Australia, companies that are quite happy to provide schematics to consumers in the USA will forward my requests to the domestic distributor, who all have vested interest in not providing support except their own internal repair division, even if I'm qualified and have a number of years of experience. It's almost impossible to get any sort of "Authorised Service" title, and I don't want to work for one of those companies at barely more than minimum wage.
Anyway, vent over!

Thanks for the words of encouragement Greg. From the huge folder of info on these I was able to download from music electronics forum, all I'd need is there.
Apparently GK simply substitutes a newer power amp section that's "linear" to replace these older switching amps. I don't know exactly what that means. I suppose that could be a cost effective option if the preamp/eq sections and the power supply are intact. The problem is that I can probably buy working examples of these amps all day for less than $300.
I owned an older version of this amp for many years and mine was eventually gifted inadvertently to some church after I loaned it out to a friend. I don't remember it being particularly good sounding but it was stunningly reliable. Most bass amps don't do anything for me with the exception of my beloved GK 200MB.

Well I'll use it as a learning experience if I get that far with it. I've never studied amps, I don't really know how they work even but I can finally figure that out now.

Hmmm, not sure about the info on linear vs switching, the schematics I've looked at are all linear, power supply and power amp both.
The output power amplifier is of the "fully complimentary emitter follower" topology, and is biased to operate in class AB.
I'd be more surprised if they'd changed from a class D (switching) amplifier to a class AB (linear) at any point, there would be other circuit changes that would need to be made (to the power supply mostly). They may have switched from a switch mode power supply to linear at some point, but that seems highly unlikely.

Anyway, good luck with your project David!

Btw, here's a good link that may help: OzValveAmps.org - solid state amp repair.

You'll have to start by finding a schematic and a picture. I can't see any part ids on the board silk screen so a schematic alone won't be enough.

The pattern of carbon looks like there was a fire followed by an explosion. The heat might well have ruined the copper traces too and possibly made the PCB material partially conductive. You need to clean away all the carbon. You'll have to use wire to replace any tracks. You might also have to cur out some of the carbonised PCB material and replace it with blank stuff, then wire up the parts. From a photo of a good amp, you should be able to replace the parts. I would think about replacing everything that has been carbon tarnished and most of the transistors int he drive stage even if they look un-burnt. Excessive heat and higher current draw may well have damaged, albeit partially.

Might also be worth adding a fuse to guard against future troubles.

These repairs take a lot of time and most paying clients won't pay, but I have have some that will.

If you have a good photo and have to replace sections of PCB and you have some PCB blank with copper, you can mill the copper away to make the traces. For a small section, this might be worth it.

If this is a switching amp, the layout may be more important since the switch speeds are higher and layout become important, but based on the picture it is hard to tell but I would come done on it not being hence writing the above.

Hi Veronica,

I'm pretty sure this is just a class AB amp (early 80s design). I judge it has 3 rails and this section is the 100W treble side. The 300W bass side looks totally intact.
There is only one track that seems to have vaporized. Someone suggested scraping off the carbon and dousing the damaged section pcb with Corona Blue (Which i happen to have).

I found a schematic and the component layout artwork but no track artwork. There are about 6 resistors, 2 caps and 3 small pnp transistors in toto so hopefully a cheap fix.

Once the carbon has been removed, I would use a PCB lacquer.

You should be able to infer most of the track layout from the component layout and schematic but it sounds like you won't need to.

Have fun and good luck.

Yep, should be relatively easy to infer the trace layout based on the schematic, the layout, and the remains of the charred components and pcb. Layout won't be particularly critical, you're correct that it's a class AB amp, so near enough should serve well, you shouldn't have any unexpected oscillations. Once you've figured out the trace layout and location of the components, I would recommend just cutting out the carbonized section of pcb, it's no good for holding components anymore anyway. You could etch a new section of pcb, but perfboard or veroboard will work just as well, just epoxy it in over the removed section, this will be much easier to connect with the rest of the circuit than leaving the charred section in place and having to use standoff's or something to keep it above the rest of the pcb.

Veronica is correct, you should test all of the transistors (particularly the MJ15022/MJ15023 power transistors) before powering the unit. One or more of them is likely shorted, this is the most likely cause for the failure. Do you have an in-circuit transistor tester? They're a very handy piece of kit to have, and can be build quite easily with a couple of 555 timers and a few passives (<$10 worth of parts if you don't bother with an enclosure). BJT's are much easier to test in-circuit than fets, so if you have one, this amp should be nice and quick to test, as it's all BJT's.

Good luck, let us know your progress.

Thanks fellows,
I have just disconnected the power supply wires to the high frequency channel of the amp and am glad to say that the low frequency channel is working just fine. That limits any and all damage to the part of the amp I don't have much use for anyway. <g>
Does anyone know what the equivalent transistor to an MPSA06 might be? I'd need it in a TO-92 case if possible.

This one seems like it would be identical and is available at Mouser
http://www.mouser.com/ProductDetail/Fai ... Re53wTk%3d

In the old days, the by word for the books was rather vulgar, but that's another story.

For future reference, http://ia700804.us.archive.org/15/items ... lector.pdf is what you need but it is old, so any new transistors won't be covered, but this would give you a good start.

If you have a BC387 or a 2N4401 on hand, they would work.

Veronica, Do you think the 40V max is going to be enough? The MPSA06 is 80V. I think the supply rails are around 42V but I could be looking at the wrong rails.

Probably not. These were right out of towers and I didn't look further. It would depend on how it is used. Although some amps do power the driver stages from the main rail, not all do. I'd need to look at the schematic to get a better idea.

Schematic is here:

The 100W rails are at +-60V

I think the lower rating will be ok until there is no load connected, so best use the high rating.

I also noticed that the output transistors and those 0.33 5W resistor are only just about ok rating wise for the rated output power but it would be possible to drive the output hard enough to fry them. I.E. for the voltage rails and load, the output transistor rating can be easily exceeded and the 5W rating on those resistors also.