A turn on thump (assuming thats what it is) will look a lot worse into a high impedance load of a multimeter than what it will look like across 8 Ohms.
Maybe it's not as bad in reality as initially stated.
After unplugging the louder left channel, it actually sounds like the right channel is pretty distorted. I shut the TT off and re-measured DC offset on the outputs and they're holding steady. Is this something that can be tweaked with R3 on the FE, or am I looking at component failure somewhere? Turning the volume up results is absolute garbage on just that right channel.
Swapping inputs and outputs results in no change. The problem follows the right binding posts. I thought at first maybe it was one of my speakers, so I crossed the cables and it followed the right channel. I have 3 sets of inputs and switching between any of them results in no change.
To measure input vs. output, I need an AC signal generator or something?
To measure input vs. output, I need an AC signal generator or something?
Well I feel stupid, yet again. This morning before work I was racking my brain and I realize that I never tried swapping the cables that I built. I made the second one a few months back, and since I made them, immediately ruled them out as a problem. Sure enough, a continuity check of the new cable shows a short somewhere. That would pretty much exactly explain my problem. The cables are made from hand woven cat5e, with the colors separated at the ends. I must have missed one. Will rebuild the cable tonight.
...... tough bugger , that BA3 ........ 
No,
download the test frequency put up by Pano.
Use that for your measurements.
There are many over downloads you can use for test signals.
Question for the front-end only:
How much bias are you running? Schematic says 45mA (1V/22R), Nelson hints that 100mA is possible with more heatsinking. Anyone tried this and how did it sound? Unfortunately the boards are not really heatsink friendly but I have some 1 inch x 1/2 thick aluminum bar stock I could use. I've noticed at 45mA the mosfets the mosfets are fairly cool. More bias sounds like a good idea.
How much bias are you running? Schematic says 45mA (1V/22R), Nelson hints that 100mA is possible with more heatsinking. Anyone tried this and how did it sound? Unfortunately the boards are not really heatsink friendly but I have some 1 inch x 1/2 thick aluminum bar stock I could use. I've noticed at 45mA the mosfets the mosfets are fairly cool. More bias sounds like a good idea.
I have only biased to just a hair under 1, sounds glorious! I have ordered a bunch of larger heatsink, the ones I have now are not that large and get fairly toasty, on both BA-3 pre and power amps.
Russellc