J. Kissel, K. Kawabe

Enclosure S/N S2100832
Actual Board S/N ??? (either 01 or 02 from T2100403 characterization data)
Circuit Schematic D2000592
Design Document T2000698
Whitening Chassis interface D2100630
OMC Wiring Chain D2100716

In order to ensure the new DCPD preamps (or transimpedance amplifiers, or "in-vacuum whitening," "readout-front-end") are functioning before we close out HAM6, I set up a measurement scheme (reviewed by Keita) to use the new calibration test inputs of the DCPD chain (via the whitening chassis) to measure the frequency response. 

The preamps are functional, the test inputs can be turned on, and the test input to PD output transfer function looks as expected.
See 
   - attached cartoon of the layout, 
   - some pictures of the setup and whitening chassis as the kinks in the documentation are still being worked out, and 
   - Bode plots of results, both each channel alone, and in how they'll be used -- be it the sum or the ratio. 

I don't yet have the mapping from DCPD preamp serial number (built and tested by the UK, and called SN01 and SN02 in T2100403) to the two DCPD / QPD vacuum enclosures we've made (S2100380 and S2100382), but as Keita indicates in LHO:61512, the enclosure we installed was S2100382. But, regardless, from this quick and dirty TFs, I see that the response from each looks like what we expect from T2100403, and I achieved the response using the relays on the whitening chassis (D2100630) as designed. 

I'm not so sure about the calibration of test input into current. The (differential) test input, serving as the denominator of the transfer function, goes straight into U13 on page 1 of D2000592, which I'm pretty sure is set to be a unity buffer 1 [V] differential -to- 1 [V] single-ended gain. I drove 1 V_peak, differential into this input. To turn this denominator into current, I assume the current across R1 is equivalent to current coming in off of "PhotoDiode1 pin1" or "PhotoDiode2 pin2." Thus, all I've done was take the raw transfer function and multiply by the value of R1 = 1e5 [Ohm = V/A]. This gets me the units I want, [V/A]
    Resp V =   blah             1        [V]       1       [V]     1     [V]   [V]   [V]
             ----------- *  --------  =  --- * --------- = --- * ----- = --- * --- = ---   
    Exc V  = 1 V_diff_pk    (1 / R_1)    [V]   1 / [V/A]   [V]   [A/V]   [V]   [A]   [A]
 
but conceptually I'm not sure that's the right thing to do, and all the UK test data show that the flat region has a gain of ~93 dB [V/A] not 100 dB [V/A] = 1e5. The test input was developed after the final design and design review documentation, and I've only found the schematic, without a design note helping me confirm my guesses -- so I continue to question my basic electronics skills. Ah well. Detailed understanding of how to calibrate this test input transfer function will come in due time. Good enough for the day after install, and you know everything I know if you want to compute the right thing. 

Of course, after the chaos of closing up the chamber has ceased, I'll measure these channels to much greater precision / wider frequency range / more data points so I can get an exquisite fit for compensations and balancing.