Help diagnosing subwoofer amp

[bkspero]

My son has asked for my help in learning to build and repair electronic devices, so after 35 years away from a soldering iron I need help getting started. Our first project is to try and repair a JBL TLXPS10 powered subwoofer amp that started to roar while the stereo to which it was attached was playing relatively quiet background music. I've looked through the web and this site and saw a few threads discussing issues with this amp. The most detailed was a few years ago on diyaudio:

Discrete Subwoofer Amplifier Repair

Many of the symptoms are similar to ours. Darkened spots on the power amp board darkest next to the leads of a resistor. In our case its R74 and not R60. I've attached a link to a photo of the board (in an effort to save server space here) and the circuit board diagrams and schematics for both the preamp and power amp. If they load too slowly, let me know and I'll post the files here.

http://pws.prserv.net/divergence/Bur...OnAmpBoard.JPG

http://pws.prserv.net/divergence/JBL...Schematics.pdf

You can see that there are some capacitors and diodes in the darkened area as well. We tried to test them as well as we could (recognizing the limitations of testing in a board). My presumption is that if we get a normal response it suggests the component is ok and if we get an abnormal result it is ambiguous due to possible interactions with other parts of the circuit. Results so far are that the capacitors seem ok. With a DMM on a low ohm scale they started out at a low reading and the reading slowly increased to a plateau. The resistors in the area also checked out with values within about 5-10% of the nominal listed on the schematic (including R74).

The diodes in that region were ambiguous, showing continuity in both directions on the diode test scale. The diode test scale seems ok, as we got a normal diode response on a diode in another area of the board (continuity in one direction and open circuit in the other direction).

The power transisters (Q6 and Q7) tested ok with the DMM. Both behaved like diodes from the center connnectors to the side connectors, and have relatively low resistance in both directions between the pair of side connectors. Q4 and Q5 gave ambiguous results.

We see no sign of discoloration on either the preamp board or the relatively fine wires connecting the amp and preamp (which connect at P1 near the darkened area of the power amp board).

Any suggestions as to the cause of the failure and which device or devices I should remove and test? My guess would be ZD6 failed and shorted to ground. Is that reasonable? Or could it be C34 and our DMM test is misleading me? I'd rather not get into a situation where we're pulling component after component from the board without a clear idea that we're on the right path. My resoldering skills are not good enough that we could resolder several devices with trimmed leads successfully.

Thanks in advance for your help.

[anatech]

Hi bkspero,
It wouldn't surprise me if you found that ZD6 was shorted. I'd change those capacitors in the overheated area (4) and have a good look at all those zeners. Mount the replacements off the board. Consider going up one size in wattage (1/2W --> 1W or 1W --> 5W).

Good luck!

-Chris

[Wavebourn]

bkspero;

I hope nothing died in your amp, except one leaking capacitor that spit an electrolyte that shorted the board conductors!

Take it off and wash carefully with a dish soap and a toothbrush, then carefully rinse and dry. Replace the capacitor that is leaking out electrolyte, it should be well visible which one. I had very similar problem with a car radio, you may see my photos:

http://www.wavebourn.com/forum/viewtopic.php?t=476

[jaycee]

I personally would check that R73/74 are still in value, replace the capacitors C34 and C35, and the zeners ZD5/ZD6. I'd use higher wattage zeners.

You may want to check the voltages at pins 4 and 8 of U6, to make sure there is the correct 15 volts symmetrical there. If this has gone high, the opamps will be toast.

[bkspero]

Thanks for the quick response. We'll try replacing the diodes with larger ones and set them off of the board. And also replace C28, C29, C34, and C35.

What do you think about also replacing the large capacitor that just shows along the lower left edge of the photo? The yellow/white hardened glue-like material that's underneath that capacitor is black along its edges near R74.

My concern about changing the large capacitor is what should we do with the yellow/white stuff. Do you think that we should try to scrape the old stuff off of the board, or might that cause more harm than good. And should we use more white material under the new capacitor? If so, can you recommend a place where we can get some?

As I read this and put myself in your place I think about the saying that no good deed goes unpunished. Sorry about all the extra questions. Please know that I appreciate the help.

Barry

[bkspero]

Thanks for the suggestion.

See the yellow-white material under the large capacitor at the left edge of the photo. Its hard and I thought that it was some sort of glue. Could that be leaked electrolyte that has dried or baked to hardness?

On the other hand, that capacitor is C36. Would its leaking electrolyte cause R74 to overheat?

Thanks again,

Barry

[clem_o]

The white stuff is just ordinary PVA-type glue, it's used to anchor the component(s). Depending on how well it's stuck to the board, you can choose to leave it or remove it. Do get some PVA glue to rebond any new large capacitors you put in, as there's much vibration expected inside a subwoofer enclosure...

Electrolyte is conductive and corrosive, so it can cause a component to fail... It will also eat into the PCB material, so make sure you clean the stuff off as best as you can....

Cheers!

Alternative to the PVA glue - hotmelt works fine, although bonding may not be as good in the long term...

[anatech]

Hi bkspero,
I don't see any leaking electrolyte. I do see glue as Clem pointed out. When you remove the components in the overheated area, have a look at the bottom of the capacitors. If one were leaking, it will be hard to miss. Cleanup is now easy since the area is clear of parts.

No biggie.

-Chris

[Wavebourn]

You can't see any electrolyte on photos, but signs of it's presence are well visible on legs of resistors and around them.

[anatech]

Hi Wavebourn,
I still can't see any evidence of capacitor electrolyte leakage. On a photo it's tough sometimes. I have seen plenty of this over the years, I'm just not seeing it here.

If it's there, bkspero will clean it as I described in my previous post.

-Chris