3 months ago my sister told me that her receiver - a Yamaha CR-640 – started to stop playing music (no output to the speakers). After checking speakers and all the “environmental issues”, I took it to the bench.
In my hands the amp worked flawlessly, no mater what I did to it.
I had a good look inside, nothing obvious.
The safety relay was operating normally.
I had the bright idea to replace the heat conducting paste transferring the heat from the power amplifier chips to the heasinks in the hope that it was an internal thermal sensor operating inside the chip.
For this I had to dismantle all the sides of the chassis. The only screws that were not removed were these that attach the two PCBs to a steel subframe.
I brought it back to shape, tested it, all OK. It was delivered back, where it belonged.
4 days ago the “customer” called me complaining that the same problem started occurring again, but now very frequently.
Well this time I could notice the symptoms. The safety relay was operating disconnecting the speakers. The DC voltage rails measured OK and the output DC offset wasn’t more than 20mV.
After spending many hours DC measuring every node of the PSU, Amp., and monitoring circuits and recording them on the schematic, I sat my butt down to understand what was happening. I even simulated the monitoring circuit.
The solution sprang out of the drawing while comparing the ground lines of the schematic with the measured potentials on the pcb tracks.
The ground track of the monitoring circuit was measuring –2.5Vdc. Bridging this with a wire to chassis wiped the problem out.
Well as you may guess, it was 2 of the PCB’s screws that I had not initially removed that were not pushing firmly the ground track to the steel subframe.
Lessons I have learned:
Not to assume that a ground is a ground. I have to measure to verify.
Not to underestimate the importance of physical contact
Regards
George
In my hands the amp worked flawlessly, no mater what I did to it.
I had a good look inside, nothing obvious.
The safety relay was operating normally.
I had the bright idea to replace the heat conducting paste transferring the heat from the power amplifier chips to the heasinks in the hope that it was an internal thermal sensor operating inside the chip.
For this I had to dismantle all the sides of the chassis. The only screws that were not removed were these that attach the two PCBs to a steel subframe.
I brought it back to shape, tested it, all OK. It was delivered back, where it belonged.
4 days ago the “customer” called me complaining that the same problem started occurring again, but now very frequently.
Well this time I could notice the symptoms. The safety relay was operating disconnecting the speakers. The DC voltage rails measured OK and the output DC offset wasn’t more than 20mV.
After spending many hours DC measuring every node of the PSU, Amp., and monitoring circuits and recording them on the schematic, I sat my butt down to understand what was happening. I even simulated the monitoring circuit.
The solution sprang out of the drawing while comparing the ground lines of the schematic with the measured potentials on the pcb tracks.
The ground track of the monitoring circuit was measuring –2.5Vdc. Bridging this with a wire to chassis wiped the problem out.
Well as you may guess, it was 2 of the PCB’s screws that I had not initially removed that were not pushing firmly the ground track to the steel subframe.
Lessons I have learned:
Not to assume that a ground is a ground. I have to measure to verify.
Not to underestimate the importance of physical contact
Regards
George
Attachments
Last edited:
On another forum I and others were helping someone fix an old Leak FM tuner which seemed OK until he added a stereo decoder, when it stopped working. To cut a long story short, it turned out that when he replaced the valves after "cleaning the valveholders" the retaining spring for one had moved slightly and shorted out the grid of the local oscillator.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.