OMC DCPD S2300004 characterization Physical Setup https://dcc.ligo.org/LIGO-D2200215-v1 Chassis is on the bench in the EE shop, with the cover on but not screwed in It is powered by a dual channel power supply (GPE-2323) with +/- 18.1 V Current draw on positive leg is 180 mA, 100 mA on negative leg (with respect to common). Consistent with Test Report Using SR785 Signal Analyzer Accessory Suite D1900068. Once powered on, all expected lights turn on. A-Channel: Connected DB25 breakout to front of chassis "Preamp Chamber Interface". DCPD A channel: +leg = pin 5, -leg = pin 18, BNC shields both connected to GND = pin 13. Connected DB9 breakout to back of chassis "To AA/ADC Whitened Output". DCPD A channel: +leg = pin 1, -leg = pin 6, BNC shields both connected to GND = pin 5. B-channel: Connected DB25 breakout to front of chassis "Preamp Chamber Interface". DCPD B channel: +leg = pin 1, -leg = pin 14, BNC shields both connected to GND = pin 13. Connected DB9 breakout to back of chassis "To AA/ADC Whitened Output". DCPD B channel: +leg = pin 2, -leg = pin 7, BNC shields both connected to GND = pin 5. Remember, 1:10 whitening filter is ON with the Whitening Disable relays OFF. SR785 Setup (by hand, didn't want to waste time figuring out GPIB setup) Freq Repeat: Single Shot (don't continuously repeat sweep once done) Start: 102.4 kHz Stop: 0.1 Hz (Sweeping down allows visualization of the measurement process on SR785 faster than sweeping up - quicker corrections for any mistakes) Type: Log (log-spaced frequency vector) Auto Res.: OFF # of Points: 200 Display Options Display: Live Format: Dual RPM Frequency: OFF Grid: ON Grid Div: 10 Phase Suppress: 0.0e+00 d/dx window (%): 0.5 Display Setup TOP TRACE Measure Group: Swept Sine Measurement: Freq. Resp View: Linear (Magnitude, i.e. not in dB) Units: dB Units: OFF Pk. Units: OFF Phase Units: deg dBm. ref impedance: 50 BOTTOM TRACE Measure Group: Swept Sine Measurement: Freq. Resp View: Phase Units: dB Units: OFF Pk. Units: OFF Phase Units: deg dBm Ref. Impedance: 50 Source Auto Level Ref: OFF Amplitude: 1.0 V_pk 2023-01-23's adventure needing 3V was apparently because the SR785 accessory box's DB9 power input wasn't *really* plugged in. After plugging in securely, can get great data with only 1 V_pk. Also scaled down ADC input range accordingly, Scaled back settle *cycles* from 1 to 2 to save time. Source Ramp Rate 1 V/sec, but Source Ramping OFF (Since we're driving down to 0.1 Hz, and our *start* frequency is 0.1 Hz (i.e. we're sweeping *up*, we don't want to waste any time ramping this *too* in addition to the settle time between sweep frequency changes.) Offset 0.0 V (hmmm... maybe we should? Later...) Input Setup Input Source: Analog Input Config: CH 1 (Both Channels) Mode: A-B (positive legs into A, negative legs into B) Coupling: AC (0.16 Hz) AA filter ON Wt filter OFF Autorange UP ONLY Input Range (for 1.0 V source amplitude, 50 ms / 5 cycle integration): CH1: +4 dBVpk CH2: +22 dBVpk Transducer Params: (all as default; unused in this style of measurement) Tach Input: (all as default; unused in this style of measurement) Playback Config: (all as default; unused in this style of measurement) Auto Offset OFF Average Setup Step / Impulse response of a basic 1:10 zpk filter is 0.1 sec. (there's probably some math that would predict that for me, but I tossed it into matlab and asked it to do it for me.) Adding a 10 kHz pole *shortens* the impulse *time*, but increases the impulse *amplitude*. So, a settle time of order 0.1 seconds is in order. Frequencies higher than 0.1 sec = 10 Hz should use settle time, and below 10 Hz should use cycles (and one cycle should be plenty). Settle Time = 100 ms (101.56 ms) Settle Cycles = 1 Integration Time = 50 ms (50.781 ms) Integration Cycles = 5 with freq = [0.1 102.4e3] and 200 points, this takes 911 seconds = 15 minutes Output Hard Copy Button ASCII Dump Bitmap Printer PC X 8 Bit Bitmap Area Graphs Vector / Plotter PostScript (doesn't matter, unused) Destination Disk File File Start Number 0 Starting with DCPDA Whitening ON. Finishing with DCPDB Whitening OFF.