I would have guessed that you have to calibrate the detectors relative to each other before trying this. Its also going to be tricky if you use 2 different kinds of ADC for this (c.f. today's delay discussion in the group meeting).
I think Osamu used to look at fast transmission signals by making sure the PD at the end had a 50 Ohm output impedance and just drive the 40m long cable and terminate the receiving end with 50 Ohms. Then both PDs go into the SR785.
On Rana's suggestion I measured the trasfer function between the two photodiodes PDA255 that I'm using.
I took the one that I had on the end table and put it on the PSL table. I split the MC transmitted beam with a 50% beam splitter and sent the beams on the two diodes. (Rana's idea of picking off the beam and interposing the PDs before the ISS PDs was not doable: ISS PDs would be too close and there would be no room to install the PDA255 before them). See picture with the final setup.
The transfer function also includes the 40m long cable that I was using for the Arm Cavity measurement.
Here's what I got. It looks rather flat. Yesterday the calibration was probably not the problem in that measurement.
I'm now going to install the PD back on the end table and measure the TFs between the excitation and several points of the loop.
(Trivia. At first, the PDs were saturating so Koji attached attenuation filters on to them. Suddenly the measurement got much nicer)