jeffrey.kissel@LIGO.ORG - posted 13:38, Monday 15 April 2019 - last comment - 14:47, Monday 15 April 2019(48505)
Long-duration PCALX Line Sweep Errantly Injected at 501.3 Hz
J. Kissel I've found this morning, that PCALX has been errantly injecting a calibration line at 501.3 Hz, since 2019-04-15 05:36:08 UTC (1239341786). This error is a result of a poor minimum frequency definition in the guardian code which manages the high frequency, long duration sweep of the sensing function (HIGH_FREQ_LINES). We'll fix the code, install it, and change the calibration line frequency back in to the high frequency region when we go out of observation mode this afternoon for Danny's OMC scans. Currently, this is the collection of data we've gathered so far: UTC Time of Change GPS Time of Change Changed to Freq (Hz) Duration at Previous Freq (Hours) 2019-04-15 05:36:08 UTC 1239341786 501.3 24.52 2019-04-14 05:04:45 UTC 1239253503 1001.3 12.42 2019-04-13 16:39:41 UTC 1239208799 1501.3 28.47 2019-04-12 12:11:29 UTC 1239106307 2001.3 17.13 2019-04-11 19:03:16 UTC 1239044614 2501.3 I've also checked the frequency regions surrounding the above for features like violin modes, and the only one that comes close is 1001.3 (which lies close to an ITMX mode at 1001.84 Hz and and ETMY mode at 1000.44 Hz, but 0.5 Hz away should be far enough away, especially since we're driving on ETMX).
P. Covas, J. Kissel HIGH_FREQ_LINES guardian code was modified, class CHANGE_FREQ(GuardState): index = 20 request = False def main(self): CUR_FREQ = ezca['CAL-PCALX_PCALOSC1_OSC_FREQ'] if (CUR_FREQ - 500) > 1000: # Changed conditional to check that *NEW_FREQ* will be > 1000 Hz NEW_FREQ = ezca['CAL-PCALX_PCALOSC1_OSC_FREQ'] - 500 ezca['CAL-PCALX_PCALOSC1_OSC_FREQ'] = NEW_FREQ DICT_TIME['last_timestamp'] = ezca['FEC-124_TIME_DIAG'] return 'GETTING_READY' else : return 'INIT' # Returning INIT so that line freq loops back to initial value instead of stopping. Then as soon as we went out of observation ready, it was loaded and re-initialized, restarting the PCALX line frequency at 3001.3 Hz. Because of unrelated activity, we lost lock at virtually the same time, after a failed attempted at AC coupling the ISS 2nd loop in support of Danny's measurement of OMC scans. Thus, - The analysis ready stretch from Apr 15 2019 07:31:44 UTC (1239348722) to Apr 15 2019 21:27:12 UTC (1239398850) has the errant 501.3 Hz line from PCALX. - You should be able to subtract this line using the witness channel H1:CAL-PCALX_RX_PD_OUT_DQ. Ask the calibration group for the details on how to calibrate this channel, if you wish to do so.