I have started to solder together the amp now, but I´m wondering how to test the power supply. Can I just power it up before I solder the amp section and see if it lights up? I know many regulators get ruined if you turn them on without a load.
I´m also a little confused on how the options I have on the input. Should I use the pot and c1/c2/r1/r2?
Hope someone can help me out! Thank you!
The regulators works perfectly without load since the voltage reference and opamps are quite enough as load. Good idea to build the power supply first.
I´ve made a real mess now. I was missing some parts for the psu, and impatiently soldered the amp while I was waiting for the parts.
Now I have finished the board and put voltage on it, but its failing. The lights on both positive channels are flashing, then the negative for one channel is flashing, and then its nothing. I´ve tried to resolder most of the missing negative channel, but I see no difference.
I have been having nightmares about searching for an error on this board.. I hope you have some input on where to start looking.
I'll suggest that you do a real good visual inspection first.
I could not find any misplaced parts on the left side of the board, but I´ll try soldering some of the parts again. On the right channel, it seems like I have placed a capacitor on E28. Could this have caused a fault on the right side as well?
I´m all out of 0805 470k resistors now, and it´s too expensive to place another order from Elfa just for this part. Can I fit a regular resistor temporarily, just to see if I can get it running?
Sure you can test with a hole mounted resistor.
I found a new resistor for e28, and found some other errors I made on the boards. I soldered capacitors on c/g41 and 42 instead of leaving them open. I removed them and fired it up again, but the result was the same.
As mentioned, its blinking for half a second, and then the leds go black. The negative on the left channel does not fire at all.
I realise that its almost impossible to troubleshoot this board. Do you have any hints on where to start? I know its likely to be binned, but I´m willing to put some time into this. If some components might be damaged, I´ll order some new one.
I recommend that you once again check all parts.
You can remove R3, then the amp will be turned off. Interesting to see if you'll get a proper stabilized voltage.
You should pay attention to pnp/npn transistors are on the right spot.
Well, I´ve made some progress! I had switched two of the transistors on both amps. t13 and t14.
Now three of the leds are stable, but the left negative is still not lighting at all. All the channels look identical to me now. I´ve been looking at the boards several times a day for a while now, so the channels are symmetric. I have 1.8v drop over the working leds, and only 1.0v over the dark one.
I´m hoping for a new clue here. I obviously have a voltmeter, and also an old oscilloscope, if that helps. I´ve tried to remove r3, but it does not seem to make any difference.
Thank you for your help!
Do you have proper voltage in one channel now?
Yes, one channel seem to be working. I will test it with sound later today.
...and when I soldered in r3 again and connected a source, the negative left channel just lit up. Everything is working now. You never wrote anything about the self-healing properties of this amp!
Thank you very much for the help! I have only tried it from an mp3 player, but it sounds really promising!
Good top hear that you have succeeded. A small review later on would be nice to hear.
How do I disconnect the servo to measure 0v over E1? I think I might have done this wrong the first time around.
You don't have to disconnect anything. Just measure the voltage across R1 and adjust with P1. Let J4, J5 be open. Turn the pot to zero and see what you got. A negative voltage means that the input bias current is flowing into the amp. In that case put a positive voltage on the pot (J4).
The reason I was asking, is that the amp was acting up when I last used it. Its been of little use in the summer anyways.
Now when I open it to measure, I see that the leds for the two regulators in the middle is disappearing 5 seconds after power up. One negative, and one positive. I´m sorry to bother you again, but do you have any suggestions on where to start?
I really want this to work correctly!
I must have more details in order to help you. Which supply voltages are OK?
These are the supply voltages:
T28 and V27 are also +-13,86 at startup, but then the fuses goes after five seconds. It makes a hissing sound before the fuses cut the supply.
Just tell me what else you might need.
Some background: Both channels worked perfectly and now suddenly you have a problem with the both channels?
First I'll advice you to visually inspect the board.
If you remove R3, no current at all should flow. If it stills consumes current then check all transistors with diode test. Start with output transistors and go backwards.
I sort of gave up on troubleshooting this, and put it in a box somewhere. I found the amp again two weeks ago, and still cannot find any faults. I have replaced some op-amps, desoldered and tested some transistors against new ones, and still cannot find any defect parts.
I have visually inspected the board a lot of times now, and can not find any mistakes. It looks worse now than before the troubleshooting because of all the desoldering. The amp was working for a few weeks before it broke down. It's hard for me to say if it was working perfectly, since i did not bother to measure anything with my oscilloscope. It seemed ok.
I have a small humming sound from the board itself (transformers, I guess) that quits immediately after the fuses kick in. The weird part, and hopefully where the solution lies, is that the fault is in the positive side of one channel, and the negative side of the other channel. And its almost completely synchronized.
Is there any way i can test the parts while still on the pcb? The resistances around the board seems to be the same on a lot of random test points.
I really could use a good headphone amplifier to make my days at work more enjoyable. I hope you can help me out!
Have you removed the R3?
Yes. Same result with r3 removed.
I tried making this amplifier work again. I replaced some parts and looked over everything again once more. I really cannot find the fault here.
I tried filming the amplifier with a thermal camera, and BR1 and BR4 are running really hot. None of the resistors or silicone in the middle seems to heat up, except E29 and R29.
Does this help at all? Any other input on what I could try? I really want this thing running.
The R29 should be 470 kohm. IS it really this value? Is the opamp soldiered in the right direction?
Yes, the resistors are 470k, and the opamps are in the right direction.
I'm trying to get the picture. The amp worked good and then after some time switched off it doesn't and the both channels seems to have the same fault?
I recommend that you cut the supply lines to both channels. Check that you have the correct voltage. 13.93 V is the typical value.
The amp was working for a while before it went silent.
Should I disconnect the supply to the psu or the amp? I would guess the regulators behave badly with no load?
Are all electrolytic caps mounted correctly? Both channels are dead? If yes, then it's not a random error.
Yes, the electrolytics are mounted correctly.
The supply lines seems to go to a lot of components in the amplifier, could you tell me the part numbers to remove and where to measure the voltage?
Thank you again!
I'll suggest that you try to get a working power supply first. Then you'll have to cut four traces, the power lines to each amp.
Ok, the power supply is not working when i cut the four lines. Where do I go from here?
Which power supply doesn't work?
The two in the middle. The leds stop shining after about five seconds.
What is the values of the output voltages?
These are the values. I noted the fuse number on the rails i measured. (In parenthesis the values after some seconds, when the amp fails)
F1 +13,8v (5v)
F2 - 13,8v
F4 -13,8v (-5,5v)
Is this the output from the regulators? Is the voltage at each amplifier zero(should be since you have cut the connection)? Can you notice any heated component?
This is at the output of the lines from the psu that i have cut previously.
I will have a look with a thermal camera.
One positive regulator works the other doesn't afters one time. It means that most parts are in the right place. The two halves of the pub's are indentical so compare especially how electrolytic caps are mounted. You have probably made same mistake for the negative regulator.
I have measured all the electrolytics in the psu, and all the resistors. When still on the pcb, they all measure the same.
When I watch the amp power up with a termal camera, the only thing really different is the rectifiers. The two in the middle (the two failing ones) gets really hot. When I measure resistance across these, I get almost a short between the ac and the positive on the two failing ones, but the working ones show several megaohms.
Should I simply order new ones, or do you think these are only symptoms for another error?
Thank you for following me through this so far!
Switch over to "diode" on your DVM and measure the diodes in the rectifier bridges. If you read near zero, the you probably have broken diodes => change the parts.
With new rectifiers, the psu seems ok. I soldered the amp section in, and it does not shut down like before. It makes a mechanical 50hz humming sound, and e29 lights up with the thermal camera.
What should I measure next to make sure it is ok? I sort of assume that this is a temporary thing until the new rectifiers brake down.
R29 should be 470 kohm. Impossible to get it heated for any value more or less. Check the resistance and the voltage across it.
Ok, it could be something close to it that is heating up, but I will have a look.
E29 is fine. Have a look at a thermal photo of the amp. Does it look fine?
What can I measure to confirm that it is working properly now, amd how do I adjust the pots correctly?
Thank you again, it would have been really great to get this amp working properly.