Thanks for the kind words... its really appreciatedand great to hear that your amp is all up and running now.
I'm not sure what to think of that modification list, you would need to look very very carefully at what is involved. Black Gate caps have been out of production for well over a decade... and there is a lot of scope for things to go wrong when undertaking something like that.
As per my customers' request I have replaced C801 and C802 with:
https://www.audiohobby.eu/en/mundor...ytic-ag-glue-on-22000-uf-63vdc-125c-2pin.html
I have reset the balance to centre and the volume to minimum. Also I have adjusted R764 to the lowest bias.
I have powered on the amp via the series lamp limiter and the buib is very bright and flickering. Also the power led on the front of the amp is NOT on.
Do I need to snip R754 again and adjust the value again?
Last edited:
Replacing the main reservoir caps will not alter any of the DC conditions.
There are really only a few possibilities.
1/ You have fitted the caps incorrectly (polarity).
2/ Some solder has dropped onto the board somewhere causing a short.
3/ There is some physical problem with all the previous work such as a poor joint or crack in the print and the act of working on the board to replace the caps has shown up such a problem.
Recheck all your work very carefully before doing anything else.
The bulb flashing brightly and no LED means there is a near short on the power supply.
If you can see significant voltage across the 0.1 ohm resistors in either channels output stage then that shows where the current is going. If there is no voltage across those resistors then the short is elsewhere.
There are really only a few possibilities.
1/ You have fitted the caps incorrectly (polarity).
2/ Some solder has dropped onto the board somewhere causing a short.
3/ There is some physical problem with all the previous work such as a poor joint or crack in the print and the act of working on the board to replace the caps has shown up such a problem.
Recheck all your work very carefully before doing anything else.
The bulb flashing brightly and no LED means there is a near short on the power supply.
If you can see significant voltage across the 0.1 ohm resistors in either channels output stage then that shows where the current is going. If there is no voltage across those resistors then the short is elsewhere.
Yes the C802 cap was fitted in reverse polarity.
Yesterday (11/7/19) I made an attempt to to sort out the problems. As I said above C802 was fitted wrong. in regard to C802, i matched the neg side of the cap with the pin marked ground on the underside of the PCB and NOT with the side marked with the dot on the upper side of the pcb.
Didn't make any difference; the amp still makes the series lamp limiter bulb v. bright and flickering. Also the power led is off.
I think I have managed to find where the short is: I am not sure if the short is just on C802 or is on C802 (Neg/-19V leg) and collector pins of Q764, Q760, Q756 and Q763.
I cleaned around the area of C802 and C801 caps with IPA and I also brushed the PCB several times (to remove stray solder).
After many rounds of de-soldering and re-soldering C802 (with short persisting), ic. cleaning pads and legs of caps with IPA and braid; then swapping the cap in C802 for the one in C801 - the short has gone!
I powered on the the amp. The power led is off. The bulb is v. bright and flickering.
shorts to ground:
C802 (Neg/-19V leg)
Q764 Collector
Q760 Collector
Q756 Collector
Q763 Collector
At 20.53 above shorts have disappeared.
20.55 Power on the the amp. No power led. The bulb is v. bright and flickering.
POWER OFF
20:57 - shorts to ground have reappeared.
12/7/19
Shorts to ground are still there.
at some point (i think whilst holding the positive probe on the NGE/-18V pin of C802), the short disappears.
19.00 Power on the amp. No power led. The bulb is v. bright and flickering.
POWER OFF
19.03 Retest C802, short has reappeared.
NOTES:
I am a bit flummoxed, sometimes the short is there and sometimes it isn't.
From memory, I don't think there is an voltage across R768. I haven't tested the 0.1x2 resistor on the other channel yet.
Even swapping the cap from C801 to C802 hasn't made the short go way permanently. It worked at first, then the short came back.
WHAT DO I do???????
I'll double check for voltage across both 0.1x2 resistors and report back.
Somehow i've made the short reappear. This was a while after I managed to blow fuse F901 while adjusting the trimmer resistor (then trying to remove the test leads from the 0.1x2 resistor) on the L channel. Easy mistake, easily fixed.
Whilst connected to the series lamp limiter:
Last night I had both channels set just perfectly (i.e. 8mV across each 0.1x2 resistors).
I had a friend come round last night and wanted to show him the amp working. While the amp was connected to the series lamp limiter, I powered on the amp and lamp glowed at medium brightness for a a second or two before the lamp went very dim and 'click-clicked' (as it normally does when it is working).
After that I though i should put some insulation (foam mat) underneath the pcb/heatsinks in order to prevent shorts (while testing). So I removed the board and added some foam sheet underneath the PCB, then replaced the PCB.
Ever since I have done this, the lamp of the series lamp is very bright and flickering; indicating a short symptom somewhere.
Last night and today i have tried several times to clean the board with IPA, check for shorts on most of the transistors that I replaced on the right channel; and removed what I thought could be excess solder on any pin of any device that I know that I have worked on.
How do I completely disable one of the channels so I can try and narrow down where the fault is?
If i snip R724 and R726 will this disable the right channel?
If i snip R723 and R725 will this disable the left channel?
Last time I had this problem; I just adjusted the metal clamps on Q762 and Q764 and this solved the problem. This time, this was the first thing I tried; but this time it didn't work. Also I have tried removing the metal clamps from, and
lifting, all the transistors on the heatsink for the right channel. Didn't help either.
I need a little help to SYSTEMATICALLY narrow down where the fault is?
Last edited:
Use your meter on the resistance range to check that no part of Q762 or Q764 has continuity through un- insulated contact with the heat sink which will be connected to earth. Do that with the power turn-off.
We eventually got as far as you getting the amp working, and then you wanted to mod it:
Following all that there were problems such as reservoir caps fitted with incorrect polarity... we sorted that and...
Why were you adjusting the bias again? It all reads as though you have zapped the output stage again
Every possible scenario for working on this has been covered at length in this thread. Remember... if you link the bias generator out then the output stage can draw no current provided the output and drivers are OK. That doesn't include physical shorts of course.
Following all that there were problems such as reservoir caps fitted with incorrect polarity... we sorted that and...
Why were you adjusting the bias again? It all reads as though you have zapped the output stage again
Every possible scenario for working on this has been covered at length in this thread. Remember... if you link the bias generator out then the output stage can draw no current provided the output and drivers are OK. That doesn't include physical shorts of course.
The click-clicking of your amplifier when your friend visited is not the sign of a fault. When an amplifier is powered up a 10,000uF capacitor will draw a huge surge charging current which will disappear once this is fully charged. That is normal and explains why your lamp glowed briefly.
The next step you took was to put foam under the heat sink to insulate this from the chassis which is when your present trouble started.
It is not normal to isolate the heat sinks from the chassis and it is possible this has caused one or other of the channels or both to oscillate even if there is an isolated fault in only one of them.
The instructions in the service manual might be seen as strategic however at a practical level one has to take a tactical approach with the methods used. Had you done that you might have avoided a mishap with shorting out some components.
As Mooly had suggested/implied you need to determine if the standing current can be adjusted by R756.
He has suggested shorting the collector -emitter of Q752 which in theory would reduce the standing current to zero and the voltage drop where the target is 8mV.
If the damage has already been done you could set R756 to the maximum resistance to do much the same without risking any more butter fingers outcomes.
I suggest you invest in a set of parrot clip leads since the claws of these are tiny and with care they can be laid in the chassis attached to the the components to be measured without fouling anything else.
That will leave you with two hands free to turn the power on or off and with a comfortable position to see the meter and the light bulb you will be able to turn the power on or off.
If the signs are good then it will be possible to adjust the R756 so it can be seen if this produces a variation on the meter reading. It it doesn't there is a fault and it is time to turn the power off.
You should never test an amplifier with the power transistors removed from the heat sink since heat increased the current drawn in a process called thermal runaway.
Without the light bulb failure is a definite option and fingers crossed you don't have this due to that protection.
One can take measurements with a normal set of test probes. The way to do this is to put one lead through the hole in the black binding post and use the red to measure the differences in voltages between two points in the circuit.
It would not be a bad idea to cut a strip of electrical insulation tape and roll this around the red test probe leaving just the tip exposed. It is sticky stuff but removable.
Reverse connection of capacitors is neither healthy for the capacitor, the environment or people. When they fail they go off with a bang like a shotgun cartridge and discharge their chemical contents everywhere including into the air.
You might have avoided this through the protection bulb but that does not mean the capacitor has not gone leaky or progressed with continued use into a more serious fault. You should replace it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.