Need help fixing a 3886 based Behringer Studio Monitor amp!

[sylk]

One of my Behringer Truth B2031 Studio monitors has started overheating and making popping noises with no input at all. I am a fair hand at soldering but don't know enough about amplifier circuits to figure out what exactly is wrong.

The following are the speakers symptoms:

Speaker turns on and play music fine for roughly 20 minutes. It will play perfectly for about that time at any volume even quite loudly.

After 20 minutes of being on the speaker starts popping and the amplifier heatsink gets extremely hot to the touch. So hot you would not want to touch it for long, however will still play.

If you have no inputs connected to the amplifier it will still overheat and start popping the speaker after about 20 minutes.

If you disconnect the speakers and without any input connections it will still overheat, but of course no popping as there is no speaker connected.

I have attached pictures of the amplifier circuit board. I have focused on several areas that look like they might be troublesome. Please let me know if you have any other questions or tests you would like me to do. I have a good multimeter on hand if needed and can take pictures or video of anything.

I have searched the internet extensively for problems with this studio monitor and several people were having problem with the 7915 voltage regulators overheating. I attached small heatsinks to those just to see if it would help and it did not. I have found the replacement parts at Parts Express and will be replacing them anyway as they are only $1 a piece, and seem to be a trouble spot.

However please let me know if the capacitors need to be replaced or how to test them as they do not look blown as some capacitors ive seen on motherboards, but might be leaking? If they do need to be replaced, can you point me to the right replacement parts? Or better versions?

[AndrewT]

During the 20minute warm up time, check the PSU supply voltages. Check the chipamp temperatures. Check the output offset.

[sylk]

The LM3886T chips go from room temperature to about 107 degrees in the first 60 seconds. They then slowly climb at maybe 2 degrees per minute from there. However they never stop getting hotter. They eventually pass the 7915 in temperature and climb to 250+ degrees Fahrenheit, around which the popping starts.

I have modified a picture to show the temperature readings at which the popping starts.

The LM7915CT chip and the 7815 chip, which are known to go bad on these boards and the ones I put heatsinks on rapidly climb at a rate of about 1 degree every 5 secs until they get to 140 degrees and then slowly climb to 170 where they seem to stay.

I didn't know what you meant by output offset so I measured the voltages of the chipamps at first turn on and once they start popping. At start the pins I have identified in the picture are at 38v DC.

Once the popping starts the voltage changes rapidly between 34 and 39v DC. I could not detect varying voltages anywhere else.

What I believe are the PSU wires measure 30v AC (60mv DC) for each yellow wire and 0 for the brown. Once the popping starts the PSU voltages start jumping between 28 and 30 volts.

I don't know what you mean by check output offset, could you explain that for me?

[Th3 uN1Qu3]

Originally Posted by sylk

I don't know what you mean by check output offset, could you explain that for me?

Put your DC voltmeter on the speaker outputs of the 3886 and see if you get any DC (you shouldn't). Anything over 100mV is bad - though if the problem is like you described it's gonna be a lot more. It would also be wise to disconnect the woofer to prevent damage to it.

250 F is 120C - that's WAY hotter than an amplifier chip is supposed to run. The chips have probably gone bad if they get that hot. Are they screwed to the heatsink tight enough? 170 (77C) is alright for the 7815.

[sylk]

The PSU yellow cables each measure at .56 amps while the amplifier is on using a clamp meter.

[Dilby©]

The DC voltage appearing at the speaker terminals.. Should as close to zero as possible.

[sylk]

The output or pin 3 as I discovered on the schematic for the chip started with a dc voltage of 8mv on the hottest 3886 in my picture and slowly rose to 30.5mv DC

The output of the second hottest 3886 in my picture started at -12mv and rose to -35.2mv DC.

The output of the third stayed at 1.1mv the entire time.

They are all screwed in quite well with thermal paste in between them and the sink, and the entire heatsink gets very hot.

Also all measurements in this thread have been taken with no load (speakers) attached and no inputs attached. Let me know if you need any with a load.

[sylk]

As the amp heats up the DC out put of pin 3 continues to rise on two of the chips. At 100c it is at 43mv DC and rising steadily.

[sylk]

Also all my replies are very delayed as I am still in moderation just fyi. Thank you all for your help!

[abraxalito]

Its looking like you may have parasitic oscillation in one or more of the LM3886 chips. 0.56A as a quiescent current is way beyond what would be expected for a no-signal condition. To check if this is the case, you'll need an oscilloscope. The third LM3886 doesn't seem to have a problem though.

Do you have a schematic?