AMB Laboratories DIY Audio

[ekano]

Hello!

I'm building a 4-channel beta22. I've got 3 boards working perfectly, but the last one is giving me trouble. It's my V- board.

When I try to set the VR1-trimpot for 4,5v across R9, I cant get a stable reading. I can set it at 4,5V, but if I wait a bit, suddenly it will start varying between ~3,9 and ~4,7V. The voltage jumps up and down, and then all of a sudden it will be fine again for a while (stable 4,5V) until a moment later it will start jumping up and down again. This is done with inputs shorted.

This in turn affects the bias at VR2, which shows a similar behaviour.

I tried connecting a source and headphones; When the voltage is stable, it works like it should, outputting the sound perfectly. Then when it randomly starts fluctuating again, I hear a buzzing noise in the left headphone.


I've checked the board multiple times, and all components seem to be the correct value and polarity. I populated all 4 boards at once, one component at a time on all the boards, so there should be no difference. I have also reflowed all joints. The sigma22 outputs a steady +-30V, so no problem there.

Any ideas whats causing this instability? I guess some component is bad, but where should I start looking?

[amb]

My guess is that one of the JFETs Q1-Q4 has developed an intermittent problem. If you have a spare matched quad, try replacing them to see if it cures the problem.

[ekano]

Thanks, I will order a new set from your shop. Is there any other replacement I could order at the same time? Some other part possibly causing this?

[amb]

Thanks, I will order a new set from your shop. Is there any other replacement I could order at the same time? Some other part possibly causing this?

I can't think of any other problems given your descriptions.

[ekano]

Hi again amb,

Replacing the JFET's did not solve the problem. Voltages are still unstable.

What should I try next?

[amb]

Sorry that didn't fix it, it was my best guess.

Unfortunately, a problem like this is hard to track down. I believe it's one of the transistors, and since you have 3 working boards, you could use one of them to help you debug. Swap transistors (BC550C and BC560C) from the good board to the bad board, one at a time, starting from the input side of the amp and work towards the output and then re-test after each swap. When the problem changes to the other board, then you've located the noisy/unstable part. This is tedious, but I can't offer a simpler but sure-fire way to identify a part that misbehaves in this manner.

I hope you have good desoldering tools/equipment to avoid damaging the board and parts.

[ekano]

I think its solved now! Switched BJTs between the boards, and on switching Q25/26 (did two at a time) the problem vanished. Oddly enough, it didnt appear on the board I switched with.. but replaced them and voltages seem fine.

Thanks for your help! =)

/ekano