[elpechos]

I've got my first cmoy up and running

The first problem I ran into was the virtual ground was many volts off

My virtual ground circuit is resistor divider into a OPA633.

I solved this issue by reducing the 220k resistor dividers to 22k and now the ground is accurate to about 0.1 or 0.2 volts under load.

Unfortunately one channel is quite noticeably louder than the other

--I think both sides are amplifying as placing a finger on the respective input channels will produce a nice humming noise.

I measured resistance from ground to L and ground to R to ground on the outputs and they are quiet similar
they are
10860 ohm one channel
10900 ohm other channel

This seems to suggest both channels have same feedback resistors etc
I am using gain 10, 10k and 1k resistors.

I double checked the feedback resistor values for both channels.

I tried listening to sound without the potentiometer and one channel is still louder than the other.

I have double checked the PCB visually with a torch and also
by checking resistances on each channel are roughly the same to ground.

When operating there is a small 0.004 volt DC offset on one channel

The only thing I can think of is maybe one of the channels in the amp is damaged?
Or is there any other possibilities?

 

[Pars]

Did you make your own pcb? Nice job if so. For troubleshooting, I would remove the opamp (signal, not VG), and with power off, ohm out each pin of the socket to virtual ground. You should see the same resistance between like pins of each channel. Also, ohm adjacent pins out and the resistances should also be similar. If that all seems fine, then double check all wiring, etc. and solder joints. The 0.004V (4mV) offset is nothing to worry about. Barring finding anything else, swap opamps (if you have another) and see if it persists.

 

[elpechos]

Yup, I made the PCB using the toner transfer methods.

I'm fairly confident the PCB is correct as it came out looking nice

Removing the opamp and checking the resistances to the virtual ground is sounds like a really good idea, thanks. I should have thought of it, oh well.

I do have a couple of spare op-amps and swapping them in and out didn't solve the issue

However -- I also swapped op-amps earlier when I was using the 220k resistors in the virtual ground and even the small current draw from the buffer caused large amounts of clipping etc particulary on one channel.

I wouldn't have thought so but I guess tehre's some possibility I might of damaged one channel in all three. But if I blew a channel wouldn't there be no sound?

 

[Pars]

I would say that it is doubtful that you damaged one channel in all 3 opamps. I would see what you find with the measurements, and also double check your circuit layout. I always like to use Eagle for circuit layouts, and always do a schematic first and generate the board from that to ensure that everything is connected properly. Eagle will allow you to just do a PCB layout with you providing all connections, but I find that too error prone. I know other layout packages allow this as well.

Oh, and I got the ohming out the socket idea from Tangent