REMOVE SEARCH FILTER
SEARCH AGAIN
Search criteria
Section: H1
Task: SEI
The CS HEPI system is planned to be upgraded to a Beckhoff PLC system. Modifications to the motor control panels to readout the VRD readback signal completed. Cabling and panel connections cleaned up. Part of ECR E2200043.
Prior to starting work:
Mitch, Jim
We got a window this afternoon to try installing L4Cs in HAM2. We just did the horizontal L4Cs and cables, because those were the most difficult, but least in the way parts. The H3 L4C in particular was very difficult, I had to go in and work under the SPI ISIJ telescope, laying on my stomach in the beam tube, while trying not to kick the modecleaner baffle out of place. Also, the gap between bottom edge of St1 and the bottom of the nozzle to the modecleaner tube is just slightly narrower than my elbow, so I could fit my arm in there, but was never sure it would come back out easily. Anyway, the hard part is done, we will finish the verticals next week sometime. I also need to get back with Fil about running the in air cables to the feed thrus. We should have the in rack electronics next week some time as well, so I will try to get the adc wiring in the model set up before interface chassis arrives.
For ADC assignment, see the green boxes on the redline posted to D1101576:
REDLINE_HAM23_Wiring_SPI_CRS_L4C_AA_ASSIGNMENT (P3)
If the Duotone channel is set on ADC0 CH31 as shown in this wiring diagram, we will have to skip the last db9 of the left AA shown on the picture above (or loose the Pod Pressure signal).
In summary, here is the proposal:
ADC0 CH24-25: L4C H1,V1 Signals
ADC0 CH26-27: L4C H1,V1 Diff Pressure/Pressure
ADC1 CH24-25 L4C H2,V2 Signals
ADC1 CH26-27: L4C H2,V2 Diff Pressure/Pressure
ADC1 CH28-29: L4C H3,V3 Signals
ADC1 CH30-31: L4C H3,V3 Diff Pressure/Pressure
[Shoshana, Ryan S] Yesterday we began building the second set of CRSs sent over from LLO (mostly just putting in heli-coils). I (Shohana) will probably continue working on them today For some of the parts sent over from LLO (where they had already gone through clean and bake) we found some black residue in some of the aluminum foil wrappings and on the parts (specifically the cross beams [D2300102] and the solid supports [D2300144]) some of which would come off when wiped, but some of the residue would not. Pictures below
[Shoshana] Continued LLO CRS build One of the CRS Legs [D2300089 SN016] has a threaded hole on the side where the threads do not go deep enough for a helicoil to be fully inserted, and ends the helicoil up sticking out by ~1/8 inch, meaning half of one of the CRSs cannot be fully built right now unless the hole is tapped deeper a shorter helicoil is used. There were no other issues and 1.5 of the extra CRS supports are built The clean and bake team at Livingston says the residue is just aluminum oxide and not an issue
Which threaded hole is it ?
If the hole depth is correct, but we're just missing the threads on the bottom part of the hole we can tap it with a clean tool.
Alternatively we can also order a 1D helicoil.
H1 ISI CPS Noise Spectra Check - Weekly Famis 39342
For some context, most of the Corner station is vented and the BS cartridge is not in the chamber which explains what is shown on the BS plots here..
To close out the week, we got stops and shroud mounts installed on the 2 CRS today. This is as far as we can go until we can get flexures or start assembling HOQIs, so probably not much more to report until later next week. We've covered the CRS in foil to protect them, and packed up the left over hardware for these 2 sensors.
Shoshana, Michael, Jim
Today we went into the PSL and started putting together the 2 complete CRS assemblies we have cleaned and ready to go. This went pretty smoothly and we now have to mostly done CRS, sans HOQI readout. We don't have flexures we can use, so we can't suspend or do any mechanical tuning, but proof masses are installed in the frames, mass adjustment hardware and picos are mounted. Tomorrow, we will probably look at what prep we can do on the other 2 CRS, but fasteners are still working through c&b for those, so I don't know there is much to do until those come out.
H1 ISI CPS Sensor Noise Spectra Check - Weekly FAMIS 39336
For some context, most of the Corner station is vented and the BS cartridge is not in the chamber which explains what is shown on the BS plots here..
J. Kissel E2600106 / IIET:37433 D2400107-v4 I've decommissioned the SPI pathfinder's ISIK Transceiver as it stands now (D2400107-v4) by - disconnecting all electronics cabling; PDs at the cable-table bracket and picomotors' quadrupus at each pico-actuated optic - disconnecting optic fibers - untangling the in-vac cabling and fibers, ensuring the - installing installation handles (LHO:90223) - setting up a new set of 1"D x 6"L class-B posts then - removed the 3x temporary class-B bolts and washers that had secured the breadboard to the posts - lifting and transferring the D2400107-v4 assembly over to the other position and - securing there with the 3x temporary class B bolts. Pictures of the move attached.
Shoshana, MichaelR, Jim
This morning we started moving parts into the lvea to start doing CRS assembly in the H2 PSL enclosure. We should have parts for 2 complete sensors right now (minus HOQIs), and should be getting the remaining parts out of clean and bake soon. We spent a lot of time trying to debug the H2 fans and hvac, but it seems not everything is fully functional yet. HEPA units are running in the anteroom however, which should be enough for us to keep parts clean while we assemble. I put a dust monitor in on a table and took counts, they stayed the in 50-100 range while I was watching. Parts are set on one of the stainless shelving units, we can use the stainless table in the anteroom for assembly for now.
J. Kissel, J. Warner ECR: E2400026 WPs: 13237, LHO:13238 Final Design Doc: T2400145 Relevant Systems Level Drawings: . Mechanical Assemblies :: ISIJ Reflector D2400102 :: HAM Table Baffles D1700335 :: HAM2 Systems Layout D0901083 (not yet-inclusive-of-SPI) . Electronics Wiring Diagram . Cable Routing In-vac Cable Routing Plan from G2401479 (pages ) . Flange Layout D1002873-v11 Executive summary: - Removed ISIJ +X-side, center, HAM2 ISI table baffle from D1700265-v4 Type 2 :: Bracket mounting bolts left screwed into ISI optical table for this will eventually become a D1700265-v4 Type 3 and remounted. - Installed ISIJ Reflector + QPD Assembly D2400102 on +X ISI side wall (D071057-v2), with its shroud (D2500030) in place :: We didn't weigh it, but the SW Assembly predicts a mass of 1.96 [kg], which we can likely round up to 2.0 [kg] with the shroud installed. - Routed and connected QPD read-out cable system to D3 flange, F10 spigot. - Routed and connected picomotor cable system to existing D1101515 quadrupus leg Cable #3, J4 :: Cable #4, J5 is connected to IO PM5 mirror, Cables #2 J3 and #1 J2 were reset in neat coil with connectors floating to avoid electrical grounding as before. - Nudged make-shift baffle system using D1700261 ballast mass baffle mounted vertically on the +X / +Y / Beam Height corner of the chamber wall upon entry into beamtube. :: the will DEFINITELY need a technically-minded reset. In at 10:30a PT and out by 12:30a PT. It's so lovely when everything goes to plan! Pictures and further info to follow in the comments below.
Removing Center ISI Table Baffle Not much to say more here -- removal was easy. We tried "just" removing the panel using the the coated/capped screws, but these didn't budget upon several attempts with Jim's fingers and grunts. So, elected to fight that battle outside the chamber and removed the whole assembly at the table mounting point. The full assembly is wrapped in dry-wipes, foil, and ameristat bagged and in Mitch's office while it waits for the Type 03 version of the panel to come out of clean-n-bake.
Installation of ISIJ Reflector itself Also not much to say, other than the great joy that the drawings of the D071057-v2 ISI Side Wall have the irregular positions of the 1/4-20 utility holes accurate enough that Bram's CAD-informed-only mounting holes for the ISIJ reflector's baseplate of the reflector lined up without issue. *phew* Also -- the D2400102 drawing doesn't highlight which length 1/4-20 bolt should be used for mounting, so we used 1/4"-20 x 0.625"L (5/8"), which was "just enough." Pictures from the main entry are the best "big picture" views, but here I attach a few more in case the need arises.
QPD Cable Routing
We didn't get dedicated pictures of this QPD cable routing but,
- it follows the plan on page 22 of In-vac Cable Routing Plan posted to G2401479-v3,
- You can see the ST1 portion of it well-enough in the above pictures,
- Jim did the routing, so I trust that there's a healthy loop in the jump from ST1 to ST0, and
- I attach a picture here of the record that we've connected it to D3-F10, and a copy of the "F-Type" (D2000225-v1) counting from page 18 of the above mentioned cable routing plan.
Picomotor Cable Routing BEFORE INSTALL Just because we knew little about the details of this cable system ahead of time given how ancient the PM5 picomotor actuated mirror system, I got a lot of good "before" pictures. Mostly, I confirm that PM5 does use the Cable #4 J5 leg of the D1101515 quadrupus, and all the other legs were neatly cable-tied up and away.
Picomotor Cable Routing AFTER INSTALL Here're photos of the routing of the two 72 [in] length D2400316 picomotor extension cables connected in series and routed to all the way around from the ISIJ reflector on the +X face to the -X / +Y corner of the table where CB-9 and the D1101515 quadrupus lives. I paid particular attention to the connectors and made sure they were left floating and not shorting to anything metal. I also re-bundled up Cable #2 J3 and Cable #1 J2 in coil similar to the before pictures, again ensuring that the connectors are floating in both space and electrical connection.
The Nudged Chamber Wall Baffle Some pictures of the baffle that we nudged that will likely need re-alignment.
tagging for photos.
On the new channel assignment of HAM2's SPI ISIJ picomotor, and why I'm confident it's CH7: Per D1000581-v13, page 13 and then D1900511-v12 page 17, I'm quite confident that the quadrupus cable and up in CH5-8 of the controller on IOT2L, called "PICOMOTOR 5" in D1900511, which I guess becomes "Picomotor B / Slot 2" on page 2 of the ECAT System Diagram D1100683-v11. This is corroborated with opening up the picomotor MEDM screen -- sitemap > LSC > picomotors > "HAM 2 + oplev" button (that has "controller 5" and "PICO B" next to it), and clicking through the channels and seeing that CH8 is called "PSL ISS QPD/PD (PM5)." Here's a labeled picture that makes things more clear.
Belated aLOG on in-vac cable routing of the ISIJ QPD A PD and serial number assignments:
Path PD Name PD SN Monopus D9-to-D25 Monopus
|----------D2600002--------|
D1600083 Type 3 D2400340 D2300128
OL ISIJ QPD A S2401092 S2500517 S2500511
Just before lunch yesterday, vac told me doors were off HAM2. I then went out and locked the ISI and added steps to the chamber.
Since we removed the cartridge from BSC2, I took the opportunity to try to remove a spring that had galled when we first installed the BS & ISI stack for Advanced LIGO. I initially tried coaxing the preload nut on the end of the spring free with alcohol and a couple of long wrenches, but eventually that wasn't enough. With Randy's help, we pulled the two cap pieces off the end of the HEPI house spring tube and lifted the cap, load cell and spring out. This spring was taken to the high bay where Randy used a porta-band to cut the galled nut off, allowing us to retrieve the other hardware off the spring. Was suprised to hear Randy say the kind of dull looking blade on the portaband had no difficulty with steel cap on the spring.
A spare spring from LLO has been installed in it's place, torqued to the crossbeam foot, the old load cell and hardware have been added above.
I didn't get any photos of the work really, but here's a shot of the de-galling method we used.
Last week I tripped the HEPI pump controller valving out the BSC2 actuators, it took me until today to try to recover it. In the process of investigating this I found that one of the 4 corner station HEPI pump carts has been more or less off since a power outage in December. The remaining pump stations have been running at about 95% drive to cover for the lost cart. I am working on some revisions to the famis and asking to Tony to add a verbal notification if there is a mismatch in the manifold pressures for the 4 corner pump stations.
Nice to know that we have overhead to lose one pump station and still supply the corner HEPI with pressure. Drive is much lower now that all 4 are running again.
Attached image shows how long we have been running with only 3 pumps. Crosshair in the upper left plot shows approximate time of the power outage. Bottom right shows we have been running at ~1900 cts drive out of 2048 total available since then. We are back around 1100 cts drive now.
I've added a message to DIAG_MAIN and to VerbalAlarms to "Check HEPI pump station pressures" if any of the four CS pump stations manifold pressures are more than 5% different than any other (this threshold may need to be updated in the future if we find it's too tight, but right now it's okay). All updates committed to svn.
The channels being used for this comparison are H1:HPI-PUMP_L0_PS{1,2,3,4}_PRESS1
As of 20:05 UTC, the cartridge from BSC2 (including the soon-to-be-decommissioned beamsplitter) has been placed on the test stand in the LVEA West bay. The cartridge was first lifted at 19:15 UTC before making its slow procession across the LVEA via the 5-ton overhead crane. More details and documentation to be posted later.
Friaday Jim and I installed the lifting pads onto BSC2. Jim also secured disconnected cables, while I assisted Ibrahim with suspension gusset removal.
J. Kissel After the install of front-end software (LHO:89777 and LHO:89919) the next step for SPI is MEDM screens -- the UI/UX for the SPI L interferometers and PY one-way optical levers. Here, I post screenshots and corresponding userapps svn file and location. Sub-screens and their future purpose will come in the comments, but make sure to also check out G2402138. In the main entry, I show off the new place that SPI lives on the LHO sitemap and the overview screen. The overview screen, and all subordinate screens use a marco file, which I also attach. All of these files are, of course, properly version controlled in the userapps SVN, here: OVERVIEW SCREEN ${USERAPPS}/spi/common/medm/ SPI_CUST_OVERVIEW.adl rev 35102 MACRO FILE ${USERAPPS}/spi/h1/medm/ h1spih23_overview_macro.txt rev 35106
POWER MONITORING
These cover the on-board power monitor SPDs. Eventually, we'll use these filter banks to calibrate the PDs into [mW] as it lands on the ISIK breadboard.
${USERAPPS}/spi/common/medm/SPI_CUST_FBR_PWRIN.adl rev 35102
LOCAL OSCILLATOR The SPI's longitudinal IFO that measures the differential displacement in the main IFO's X direction release on heterodyne interferometry, with the interference between two beams modulated at 80 MHz and (80 MHz- 4096 Hz) = 79995904 Hz, which creates a beat note at 4096 [Hz]. 80 MHz comes from the site's RF distribution system of 80 MHz. The 4096 Hz is initially digitally generated via CDS, and a copy of that is sent out via DAC to be subtracted from 80 MHz with our homegrown single-sideband mixer, a.k.a "double mixer" (D2400315). This is the screen that controls the parameters of the cds oscillator, as well as some phase rotators and filterbanks to condition the digital LO into an analog signal for the DAC. Nominally, the cds oscillator clock emits a 1 [count] amplitude wave, and the CLKGAIN, SINGAIN and COSGAIN parameters are set arbitrarily high, since its use as the LO for digital demodulation is usually entire internal to CDS. However, because this LO goes somewhere into real electronics, I've set the calibration of the signal condition filters such that the CLK, SIN, and COS gain are in units of [mV/ct] -- and I've set the LO amplitude at 5000 [mV] = 5 [V], mid-range of the DAC. Since we don't yet know the right phasing, all the phase rotators are current set to 0 [deg]. ${USERAPPS}spi/common/medm/SPI_CUST_LO.adl rev 35103
INTERFEROMETERS
Here're the screens (less commissioned) that cover the digital demodulation and conversion to differential displacement.
${USERAPPS}spi/common/medm
SPI_CUST_IFO.adl
SPI_CUST_IFO_DEMOD.adl
SPI_CUST_IFO_DISP.adl
SPI_CUST_DIFFDISP_MTRX_RAMP.adl
SPI_CUST_DIFFDISP.adl
all commited to rev 35102
ONE-WAY OPTICAL LEVERS
Here's the signal chain for the one-way optical levers. There's a good bit of confusing basis changing happening given
- the orientation of the QPD segments w.r.t. to vertical being different on HAM2's ISIJ QPDA and HAM3's ISIK QPDB,
- The usual conversion from QPD segments to pitch and yaw, as the beam flies, then
- Converting from beam pitch and yaw to ISI rotation, given that the HAM2 QPDA measuring HAM3 rotation is on the "front" +X side of HAM2, and the HAM3 QPDB measuring the HAM2 rotation is on the "back" -X side of HAM3.
So I had to make custom SPI screens (rather than use the standard ASC QPD screen) that hopefully clears all this up. Also, the QPD2CART matrix is a ramping matrix, and I've never liked the auto-generated MEDM screen for this, so I pioneered a new one that I think has a better user interface.
${USERAPPS}/spi/common/medm/
SPI_CUST_QPD.adl
SPI_CUST_QPD_MTRX.adl
SPI_CUST_OL_QPD2CART_RAMP.adl
SPI_CUST_OL_ISIOUTF.adl
all at rev 35103.
