aLIGO LHO Logbook

H1 CDS

david.barker@LIGO.ORG - posted 13:58, Tuesday 24 January 2023 - last comment - 16:33, Tuesday 24 January 2023(66974)

CDS Maintenance Summary: Tuesday 24th January 2023

WP10933 Complete move of OMC to h1omc0

Daniel, Dave, Jonathan, Erik, EJ:

Daniel completed the move of h1omc and h1omcpi from h1lsc0 to the new front end h1omc0. For the past few months we have been running a hybrid system, with a reduced h1omc/h1omcpi system on h1lsc0, and an initial h1omz/h1omzpi system on the h1omc0, using the low-noise ADC which is installed on h1omc0's IO Chassis.

Today the old h1omc/h1omcpi models on h1lsc0 were moved to h1omc0, retaining their DCUIDs, and the temporary h1omz/h1omzpi models were retired. As part of the reconfiguration, IPC changes were needed to the h1iopomc0, h1lsc and h1sqz models. A DAQ restart was required.

Three different versions of the RCG was needed to compile the new models:

RGC	Models	Desc
5.05	h1lsc, h1sqz, h1omc	standard
demod_band	h1iopomc0	uses new demod band8x parts
ipc-rate-fix	h1omcpi	uses ipc_rate < model_rate

WP10940 h1oaf model change

Jenne, Dave:

Jenne's new h1oaf model was installed. It removed 9 fast channels from the DAQ (OAF-CLEAN channels) and added 3.

++: fast channel H1:OAF-CLEAN_DELTAL_EXTERNAL_2KHZ_DQ added to the DAQ
++: fast channel H1:OAF-DELTAL_NOLINES_2KHZ_DQ added to the DAQ
++: fast channel H1:OAF-LINES_SUM_OUT_DQ added to the DAQ
A DAQ restart was required

WP10941 ISI HAM8 add missing T240INF channels to DAQ

Jim, Dave:

A new h1isiham8 model was installed, which adds 3 channels to the DAQ

++: fast channel H1:ISI-HAM8_T240INF_X1_IN1_DQ added to the DAQ
++: fast channel H1:ISI-HAM8_T240INF_Y1_IN1_DQ added to the DAQ
++: fast channel H1:ISI-HAM8_T240INF_Z1_IN1_DQ added to the DAQ

WP10918 Turn on DAQ corner station data compression

Jonathan, Erik, Dave

As part of the mitigation of DAQ CRC errors from front ends to the dataconcentrators, all the corner station front end computers were modified to compress their DAQ data to both DC0 and DC1. Similarly the DCs were configured to uncompress these data. Several DAQ restarts were required between 12:02 and 12:12. H1EDC was also restarted to read the new H1EPICS_DAQ.ini with the OMZ models removed.

CDS intrastructure changes for h1omc/h1omcpi move

Jonathan, Erik, Dave:

Several CDS systems needed updating to move h1omc/h1omcpi from h1lsc0 to h1omc0 and the removal of the temporary h1omz/h1omzpi models:

testpoint.par was hand edited. H1.ipc was hand edited. Puppet changes were made to update rtsystab and H1.yaml. A new CDS Overview was generated.

Comments related to this report

david.barker@LIGO.ORG - 14:26, Tuesday 24 January 2023 (66975)

Link

Restart log:

Tue24Jan2023
LOC TIME HOSTNAME MODEL/REBOOT
09:03:45 h1omc0 h1omc
09:10:36 h1omc0 h1omcpi
09:12:05 h1lsc0 h1lsc
09:12:24 h1lsc0 h1sqz
09:26:58 h1omc0 h1iopomc0
09:27:19 h1omc0 h1omc
09:27:37 h1omc0 h1omcpi
09:31:19 h1omc0 h1iopomc0
09:32:11 h1omc0 h1omcpi
09:41:56 h1omc0 h1omcpi
09:49:26 h1omc0 h1iopomc0
09:49:47 h1omc0 h1omcpi
09:51:21 h1omc0 h1omc
09:58:30 h1omc0 h1omcpi
10:33:32 h1oaf0 h1oaf
10:34:03 h1cdsh8 h1isiham8

david.barker@LIGO.ORG - 14:29, Tuesday 24 January 2023 (66976)

Link

More RCG build details.

to build h1iopomc0

controls@h1vmboot1:~$ env|grep RTS_VERSION
RTS_VERSION=demod_band
controls@h1vmboot1:~$ env|grep RCG
RCG_SRC=/opt/rtcds/rtscore/demod_band

to build h1omcpi (and all PI models)

controls@h1vmboot1:/opt/rtcds/lho/h1/cds$ env|grep RTS_VERSION
RTS_VERSION=ipc-rate-fix
controls@h1vmboot1:/opt/rtcds/lho/h1/cds$ env|grep RCG
RCG_SRC=/opt/rtcds/rtscore/ej-dohmen-test/advligorts

david.barker@LIGO.ORG - 14:45, Tuesday 24 January 2023 (66978)

Link

Also updated the CDSRFM, RCG_VERSION, IPC_STAT MEDMs.

jeffrey.kissel@LIGO.ORG - 16:33, Tuesday 24 January 2023 (66992)CAL, CSWG, DetChar, ISC, OpsInfo

Link

Tagging CAL and DetChar.

Coupled with the installation of the new OMC DCPD whitening chassis, LHO:66980, this aLOG refers to us switching over to the OMC DCPD ADC readout / digitization system that we've been prototyping for months. To say it flippantly, "all we had to do" was to overwrite the 'OMZ' stuff onto and replacing the existing 'OMC' stuff."

Thus, after today, we have:
    - A new OMC DCPD Whitening Chassis
    - A new analog frequency response of analog AA (namely -- at the moment -- there's *no* analog AA filter)
    - A new lower noise, 18-bit ADC card, sampled at 524 kHz (instead of a hihger noise, 16-bit ADC card sampled at 65 kHz)
    - That ADC card is in its own segregated IO chassis (rather that among 8 other cards in the lsc0 IO chassis)
    - The ADC card's data is grabbed by a new IOP to User front-end model / computer system
    - It's downsampled from 524 kHz to 16 kHz in a brand new way.

For more information about the new signal chain, see LHO:66362.

Thus far we have not updated the calibration to reflect all the ways that this has impacted the DARM loop. We expect to do so in the coming few days.
Though we don't expect some huge amount of megaparsecs from this change (we weren't limited by DCPD ADC noise, so really, we shouldn't see *any* megaparsec-level improvement), I do expect this to be very discrete change in the character of the detector's noise in terms of glitches and lines -- the stuff that plagues the DetChar group and astrophysical search algorithms. I have that change will be for the better!