After further testing I'm a little more confused.
The voltage seems to be fine coming off the CRC board (+/- 26V ish). The large caps before the +/- 15 regulators were tested and swapped to no effect. Thinking the 7815 was bad, I replaced it with a lm317 circuit that behaves fine UNTIL it is actually placed in the circuit with the -15V. When i use it in place of the 7815 I find a similar phenomena of it starting at an ok voltage and then dropping to around +10V. What is particularly odd about this is this will happen when I don't have the output of the lm317 connected to anything. I have the CRC board output coming into the main GSSL board. I run a wire from the 7815 In pad to the In on the lm317 circuit, connect the ground pad, and do NOT connect the lm317 output back to the board... and the voltage at the lm317 out still drops.
Furthermore, if I connect only the +15V to the board (with the lm317 circuit) and leave off the -15V side, the +15V is fine and stable. As soon as I connect both +15/-15 regulators the positive side drops. So, I thought there has to be something wrong between the power input on the board and the regulators. I already eliminated the rectifier diodes, so I removed the large 1000uF caps between the power input on the main board and the regulators thinking something must be wrong with those... now the +15V is fine and the -15V rail is -12V... If I put one of the caps back on the negative side nothing changes. Both caps in brings the positive side down and the negative side is fine.
Any idea what might cause something like this in the circuit? The only thing I can think of is that this has to do with the interplay of the -15V and +15V on the board which I believe only happens at the VCA distortion trim pots... (I'm actually using fixed resistors there temporarily) but those seem fine and are correct values - 25k+25k-ish per side.
This is all without the switch control board connected and no chips in.