Summary:

I set up a free-space beat on theNW side of the PSL table between the IR beam from the PSL and from EX, the latter brought to the PSL table via ~40m fiber. Initial measurements suggest very good performance, although further tests are required to be sure. Specifically, the noise below 10 Hz seems much improved.

Details:

Attachment #1 shows the optical setup. 

• I used two identical Thorlabs F220APC collimators to couple the light back into free space, reasoning that the mode-matching would be easiest this way.
• Only 1 spare K6Xs collimator mount was available (this has the locking nut on the rotational DoF), so I used a K6X for the other mount. The fast axis of the Panda fibers were aligned as best as possible to p-polarization on the table by using the fact that the connector key is aligned to the slow axis.
• I cut the power coupled into the PSL fiber from ~2.6mW to ~880uW (using a HWP + PBS combo before the input coupling to the fiber) to match the power from EX.
• The expected signal level from these powers and the NF1611 transimpedance of 700 V/A is ~320 mVpp. After alignment tweaking, I measured ~310mVpp (~ -5dBm) into a 50 ohm input on a scope, so the mode-matching which means the polarization matching and mode overlap between the PSL and EX beams are nearly optimal.
• To pipe the signal to the delay line electronics, I decided to use the ZHL-3A (G=27dB, 1dB compression at 29.5dBm per spec), so the signal level at the DFD rack was expected (and confirmed via 50 ohm input on o'scope) to be ~19dBm.
• This is a lot of signal - after the insertion loss of the power splitter, there would still be ~15dBm of signal going to the (nominally 10dBm) LO input of the demod board. This path has a Teledyne AP1053 at the input, which has 10dB gain and 1dBm compression at 26dBm per spec. To give a bit of headroom, I opted on the hacky solution of inserting an attenuator (5dB) in this path - a better solution needs to be implemented.
• The differential outputs of the demod board go to the CDS system via an AA board (there is no analog whitening).

Yehonathan came by today so I had to re-align the arms and recover POX/POY locking. This alllowed me to lock the X arm length to the PSL frequency, and lock the EX green laser to the X arm length. GTRX was ~0.36, whereas I know it can be as high as 0.5, so there is definitely room to improve the EX frequency noise suppression.

Attachment #2 shows the ALS out-of-loop noise for the PSL+X combo. The main improvements compared to this time last year are electronic. 

• The failed experiment of making custom I/F amplifier was abandoned and Rich Abbott's original design was reverted to. 
• New power splitter was installed with 3dB less insertion loss.
• According to the RF path level monitor, the signal level at the RF input to the demod board is ~10dBm. Per my earlier characterization, this will give us the pretty beefy frequency discriminant of ~15uV/Hz.
• I estimate the frequency noise of the detection electronics + ADC noise now translate to 1/3 the frequency noise compared to the old system. With some analog whitening, this can be made even better, the electronics noise of the DFD electronics (~50nV/rtHz) is estimated to be <10mHz/rtHz equivalent frequency noise. 
• Note that the calibration from phase-tracker-servo to units of Hz (~14 kHz / degree) was not changed in the digital system - this should only be a property of the delay line length, and hence, should not have changed as a result of the various electronics changes to the demod board and other electronics.

Next steps:

• Improve pointing of green beam into X arm cavity.
• I plan to recover the green beat note as well and digitize it using the second available DFD channel (eventually for the Y arm) - then we can simultaneously compare the the green and IR performance (though they will have different noise floors as there is less green light on the green beat PDs, and I think lower transimpedance too).