Help with DC on volume pot

[MikeBettinger]

Originally Posted by mkusan

Perhaps I should measure it from the underside, pcb side to exclude cold solder joint. Your suggestion is noninvasive, if I understand correctly one side of every resistor should read 0.0V relative to ground, main ground that would be E14?

What I would consider to be a main ground would be back in the power supply, preferably the transformer center tap.

Mike

[Aucosticraft]

mkusan,
1)Did you use MP3 player on battery while testing?
2)you said when no music is played, still some dc appears at the output. well, then was any player like CD or VCD player that runs on mains supply was connected to AMP's input? ( I wish to ask if the ground of other player that operated on mains supply was connected to the ground of this AMP?)
3) did you try physically disconnecting all inputs to the AMP?
4) also try connecting 4.7k resi at the input and ground. for both the channels. and on selector switch select the input where resistor is connected. but with no other device connected.

[mkusan]

Originally Posted by Aucosticraft

mkusan,
1)Did you use MP3 player on battery while testing?
2)you said when no music is played, still some dc appears at the output. well, then was any player like CD or VCD player that runs on mains supply was connected to AMP's input? ( I wish to ask if the ground of other player that operated on mains supply was connected to the ground of this AMP?)
3) did you try physically disconnecting all inputs to the AMP?
4) also try connecting 4.7k resi at the input and ground. for both the channels. and on selector switch select the input where resistor is connected. but with no other device connected.

1)Yes
2)No, it was easier for me to do tests with mp3 player, but I can try with standalone CD unit
3)Yes
4)Will try and post back

One more question for everyone, since there will be a lot of measuring tonight, can I use dim light bulb tester while measuring to minimize my chances of screwing something up or that will render the data useless.

Thanks
Marko

[Mooly]

Bulb tester should be OK as long as the stabilised rails don't drop out (35 volt rail to preamp and the supply to the opamp). Better to have no speakers connected too.

[amptech]

would agree mooly...mkusan use a standard 60w light bulb not them energy saver types..

[mkusan]

Originally Posted by Mooly

Very difficult working from pdf's on screen.

Measure on DC volts between these two points with the filter switch ON which places the caps in circuit.

Also measure DC voltage from ground to these two points which should also be zero.

Conducted the test with 75W in bulb tester:
across resistors started 2.3mV but dropped after 3 minutes to 1.5mV
from ground to R537 1.4mV to 1.1 after 3 minutes
from ground to R531 3.1 dropped to 2.8 after 3 minutes

Mora data will follow

[mkusan]

Originally Posted by Mooly

Look at this. The filter switch is now in the "on" position.
The two voltages shown must be zero with respect to ground and zero volts between them.

The two caps marked "Check" if fitted back to front or faulty could cause a DC fault.

I checked C517 and C527, I haven't reversed polarity according to the minus sign on the board and to the photographs of old caps I took before recapping. In circuit capacitance value gives me 4.6uF for C517 and 0.9uF for C527.

[mkusan]

Originally Posted by Mooly

The VU meters are driven from the main speaker output via an opamp and rectifier. It's all AC coupled so the only way the meters can deflect is either because of a problem around the opamp stage or because the amp really is actually oscillating and is unstable and there really is an unwanted AC signal present.

You mention in post #1 that the amp has had quite a bit of work done to it. All these problems perhaps suggest something that has gone amiss during all the mods rather than a genuine "one off" fault.

You are going to have to work on this one stage at a time and it would be useful to also check the stability with a scope, however if you remove C635 and C636 feeding the VU meter stage then the meters should show zero at all times. If they do not and still drift and show various readings then that confirms a problem with the VU stage.

If removing the caps does cause the meters to read zero again correctly at all times then that points to instability in the amp somewhere.

Removed C635 and C636 and the needles are zeroed in, moving of volume pot does not affect them anymore.

[mkusan]

Originally Posted by MikeBettinger

I agree that the resistors are the last to suspect. But the ground to the board being slightly bad, cold solder joint could possibly add a small potential and would account for the multiple symptoms as it is common to both the filtering and the bass control, also to both channels.
The image you posted was good but it is still hard to make out what is what and I assume by zooming out it will get harder to read. Measuring on the milivolt range of your meter using the main supply ground as your negative and the positive to the ground side of the resistors I mentioned see if you read anything. As Amptech pointed out it should be zero volts. Possibly during your first work you disturbed the connection and it started working, much like rocking a control back and forth to quiet it down. With time bad connections come back. Sometimes a can of freeze stray can help locate the culprit!

Mike

Mike, I think we made progress tonight hope so, at least.
I was tired already but decided to follow your advice on resistors. I measured bunch of them from R539 to R546. Multiple resistors did not have absolute zero on one leg but R546 had 25V on one leg and 12V on the other, and touching that leg (I wore headphones) made a very unpleasant, loud buzzing noise in right channel, none of the other resistors made such noise while tested. R546 is a 330K third resistor from bottom of my picture resistor right to the suspicious one.

[MikeBettinger]

Originally Posted by mkusan

Mike, I think we made progress tonight hope so, at least.
I was tired already but decided to follow your advice on resistors. I measured bunch of them from R539 to R546. Multiple resistors did not have absolute zero on one leg but R546 had 25V on one leg and 12V on the other, and touching that leg (I wore headphones) made a very unpleasant, loud buzzing noise in right channel, none of the other resistors made such noise while tested. R546 is a 330K third resistor from bottom of my picture resistor right to the suspicious one.

From looking at the schematic that would be correct as it is connected between the supply and base of Q506. You are tired . We are looking to verify that your ground to the circuit is actually ground, so the ground side of R29,30, 35, 36 should tell the tale. They should be zero and I'm just guessing but, I would think that at least 50mv or more would create the pop! If you don't see anything there with the meter ground to the centertap then it's a goose chase.

Mike