Ok this is fixed. Turns out the CD4013 at U9 on the SB-100 at was bad. U8 was bad and replaced earlier. I should have suspected this but the schematic doesn't show a relationship to U6. Now I know better.
Thanks to Quench for helping me find the solution. As much as a pain that this was I think I learned some valuable diagnostic skills here.
Here's some detail for the next guys:
Pin 25 of the interconnect comes to R93 (on Rev C, unlabeled on Rev B). This is a 47K resistor. The other leg of the resistor has a 3.3uF tantalum cap soldered to it that is also connected to the top hole of C52 (ground). C52 is otherwise unused on a Rev B board.
U6 Detail (resized).jpg
Looking at the back of the board it is easy to miss that there is a jumper from the left leg of the resistor to the top of C50. C50 was not shorted so this would appear to be a dead end except that the trace from C50 continues on to a leg on U22 which is a 78L05 in a TO-220 package. This also test fine...
U6 Detail Back (resized).jpg
Goin up the trace comes to U8-14 and U9-14. Replaced U9 and back in business.
U9 Back Detail (resized).jpg
NOTE: this pics are from my parts board. I will clean and repair now that I have an idea what to do. Please don't mind the flux!