To aid Jenny's valiant attempt to finish her SURF project, I did some things with the front end system over the last couple days, largely tricking Jamie into doing things for me lest I ruin the 40m RCG system. Several tribulations have been omitted.
We stole a channel in the frontend, in the proccess:
- Modified the C1GFD simulink model (now analog) to be "ADC -> TMP -> DAC" where TMP is a filter bank
- C1GFD_TMP.adl (in /opt/rtcds/caltech/c1/medm/c1gfd) is the relevant part which connects the ADC to the DAC in the frontend
- Confirmed that the ADC was working by putting a signal in and seeing it in the frontend
- Could not get a signal out of the anti aliasing board
- Looked sad until Kiwamu found a breakout board for the SCSI cable coming from the DAC
- Used SR560 to buffer DAC output
- drove a triangle wave with AWG into the TMP EXC channel (100 counts 1 Hz) and looked at it after the ~25 ft of BNC cable running between the DAC and the NRPO driver
- wave looked funny (not like a triangle wave), maybe the DAC is not meant to push a signal so far, so added buffer
- Took the control signal going to the fast input of the NPRO driver (using the 500 Ohm SR560 output - see Jenny's diagram) and put it into the anti aliasing board of the ADC
- Added switchable integrator to filter bank with Foton
- I couldn't get the names to display in the filter bank, so I looked sad again
- Jamie and Koji both poked at the "no name displayed" problem but had no conclusions, so I decided to ignore it
- I confirm that when the two filters were toggled "on" that the transfer function was as expected: simple integrator with a unity gain at ~10mHz - agrees with what Foton's Bode Plot tool says it should be (see attached DTT plot)
- I got Jamie to manually add the two epics channels from the TMP model to the appropriate .ini file so they would be recorded
- C1:GFD-TMP_OUTPUT (16 Hz)
- C1:GFD-TMP_INMON (16 Hz)
- RefCav heater servo seems to still be set up, so we can use existing channels:
- C1:PSL-FSS_RCPID_SETPOINT (temp setpoint - will do +/-1C steps about 35 C)
- C1:PSL-FSS_MINCOMEAS (In loop temp sensor - in C)
- C1:PSL-FSS_RCTEMP (out of loop temp sensor - in C)
- C1:PSL-FSS_TIDALSET (Voltage to heater - rails @ +/- 2V)
- Closed loop on the control signal for the NPRO driver with an integrator, saw error signal go to zero
- Turned up gain a little bit, saw some oscillations, then turned gain down to stop them, final gain = 2
- Left system on for Jenny to come in and do step responses
|