[MonkeysAteMe]

I built a cmoy today and one channel is much louder than the other (95% of the time). Sometimes they are of equal volume, but I do not know why.

I checked the DCV output and on R5 there is some discrepancy. On the louder channel it outputs 4.5v. The softer channel outputs 0. I think that is my problem.

I checked all the wiring and it looks good, no bridges or anything.

Also sometimes there is voltage discrepancies. It is sometimes 4.5/4.5. But sometimes it is more like 6/2 (the higher voltage being from V+ and ground).

What could be the problem?

 

[Logistics]

As far as the solder side is concerned, dude, clip those leads, first of all. Then do something about all that flux! Get a can of flux-off or find something to scrape in-between all the traces to make sure there are NO bridges. Flux can be conductive, too!

 

[xmokshax]

first: yes, the bottom of that board is rather messy - some of the lead ends look like they could even be contacting other leads. also, several of the joints look a little (or a lot) weak. some or all of them might be cold, and if not, they'll certainly be more prone to failure than a properly formed joint:

(red circles are examples, not necessarily comprehensive)


another thing that jumps out at me is that it doesn't even look like R2 on the right side of the board is even connected through to ground:

also, nice big 1/2W resistors... but they're only +/-5%, rather than the 1% that's recommended. i don't know that this could be solely responsible for your problem, but figured it was worth mentioning. you could have resistor pairs separated in actual resistance by as much as 10%.

i think a good start would be to reflow all of your joints (and add solder in the cases where you can still see any of the hole around the lead from the bottom of the board).

 

[MonkeysAteMe]

Well, resoldering the weak connections seemed to have worked. Don't I feel silly.

Well thank you everyone, I guess I just have to be neater.

 

[xmokshax]

glad to hear your problem is solved. i was actually curious exactly what lack-of-connectivity was causing that problem... i guess we'll never know.

out of curiosity, was R2 actually disconnected from ground, or is that pic just misleading?

 

[MonkeysAteMe]

It was in the hole, but it was shallow. So it may have not been conencted(I shoved it in further).

 

[holland]

Hmm, you may want to lay those resistors flat against the board in a cleaner and tighter configuration. It's just a suggestion, it's kind of messy.

I remember my very first electronics project ages ago (L.E.D. stopwatch) for "circuits 101". I had wires looping everywhere. I waved my hand over it and the counter jumped....yep, crosstalk. The TA looked at me funny, and I had to tear it down and redo the entire project because there was no way I could debug it, I learned very quickly to lay things down tight and to layout in a manner that is easy to debug and test. Fortunately the Cmoy is small.

Since that day, "design for test" has been etched in my brain as a fundamental that should never be skipped. Granted verilog design vs. physical design is different, but the fundamental ideas still hold.