40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  40m Log, Page 326 of 346  Not logged in ELOG logo
ID Date Author Typedown Category Subject
  1102   Thu Oct 30 20:39:47 2008 carynConfigurationPEMtemperature sensor
We attached the temperature sensor box to the MC1/MC3 chamber with a C-clamp. We connected the temp sensor to a 2nd box with a short BNC. Bob set up a power cable coming from the X-end towards the MC1/MC3 chamber(Thanks, Bob!) We soldered the end of Bob's power cable to a plug and attached it to the 2nd box (The power supply enters through the 2nd box). A ~20ft BNC cable connects the output signal of the 2nd box to the tall thing by the PSL where all the signals go labeled 1Y2. Once we had everything connected, we put in the fuses for the power supply. So, now the temperature sensor is receiving power. We checked that the power supply was working (we measured +15.08V and -14.95V, and we wanted 15V and -15V so it's OK for now). Tomorrow we will modify C1IOOF.INI file and reboot the frame builder.

About sensor-
There is an LM34 (looks like a transistor) glued w/ epoxy and thermal paste to the inside of a Pomona box ~1"x"1.5"x2". The lid to the box is covered with a 1-2mm thick piece of copper and a little thermal paste is sandwiched between the Pomona lid and the copper piece. A C-clamp attaches the copper piece to the chamber. A BNC is connected to one side of the box (the side with less copper)

About power supply box-
There is a power regulator and an op-amp inside a Pomona box ~2.5"x4"x2". The power regulator is attached to the center of lid of the pomona box with a screw and washer. There's a power plug on the front of the box
Left:+15V:red wire
Center:GND:white wire
Right:-15V:black wire
There are 2 BNC connections on the sides of the box. The left BNC connection is for the output signal and the right BNC connection is for the temperature sensor (if the power plug is coming out of the box towards you).

Sensor location-
Chamber which contains MC1/MC3. On the door facing towards the Y-end. On the bottom-left side. Behind the door. Attached with a C-clamp.

Power supply box location-
Chamber which contains MC1/MC3. On some metal leg thing near the floor facing towards the Y-end. Attached with a zip-tie

Power supply-
Coming from the X-end from a tall thing with all the fuses labeled 1X1
Fuse 160:+15V:red wire
Fuse 171:GND:white wire
Fuse 172:-15V:black wire

Signal-
Going towards the PSL to the tall thing labeled 1Y1 on the rack labeled SN208
ICS-110B
J12 (which we believe corresponds to 50-51 and channel number 13650)
Temperature sensor is connected to J12 with a ~20ft BNC attached to a BNC2LEMO connector we found lying around
  1104   Sun Nov 2 20:21:58 2008 ranaConfigurationloreHP 5550dtn (Grazia) set up on allegra
I set up printing to grazia from allegra. The CUPS interface was not as straightforward as Tobin had made it seem in the Wiki. I had to type in the IP address and port number by hand.

The steps (AFAIR):
1) Goto http://localhost:631/
2) Click on "Add Printer"
3) Choose HP JetDirect
4) Use the correct address (socket://131.215.115.220:9100)
5) Choose HP and the 5550 postscript driver as the options
6) Try to only print useful stuff and not kill too many trees.
  1109   Mon Nov 3 19:18:47 2008 AlbertoConfigurationGeneralnew elog
Phil Ehrens kindly poured our elog's content in a CD that now is here at the 40m.
I've been trying to install the 2.7.5 version of the elog on Nodus, a Sun machine, but the installing procedure is different from linux and I dont' know it. I tried to recompile the elog from the source code but the way gcc is called must be wrong because I get this error message:

nodus:elog-2.7.5>make
gcc -DHAVE_SSL -o elog src/elog.c -lsocket -lnsl -lssl
src/elog.c:45:25: openssl/ssl.h: No such file or directory
src/elog.c:329: error: parse error before "SSL"
src/elog.c: In function `ssl_connect':
src/elog.c:331: error: `SSL_METHOD' undeclared (first use in this function)
src/elog.c:331: error: (Each undeclared identifier is reported only once
src/elog.c:331: error: for each function it appears in.)
src/elog.c:331: error: `meth' undeclared (first use in this function)
src/elog.c:332: error: `SSL_CTX' undeclared (first use in this function)
src/elog.c:332: error: `ctx' undeclared (first use in this function)
src/elog.c:340: error: `ssl_con' undeclared (first use in this function)
src/elog.c:341: error: `sock' undeclared (first use in this function)
src/elog.c: In function `retrieve_elog':
src/elog.c:383: error: `SSL' undeclared (first use in this function)
src/elog.c:383: error: `ssl_con' undeclared (first use in this function)
src/elog.c: In function `submit_elog':
src/elog.c:631: error: `SSL' undeclared (first use in this function)
src/elog.c:631: error: `ssl_con' undeclared (first use in this function)
make: *** [elog] Error 1

Joe, Yoichi, anyone else knows how to do that?
  1110   Mon Nov 3 21:38:32 2008 YoichiConfigurationGeneralnew elog

Quote:
I tried to recompile the elog from the source code but the way gcc is called must be wrong because I get this error message:

nodus:elog-2.7.5>make
gcc -DHAVE_SSL -o elog src/elog.c -lsocket -lnsl -lssl
src/elog.c:45:25: openssl/ssl.h: No such file or directory
src/elog.c:329: error: parse error before "SSL"


The location of ssl.h is a bit strange in the sunfreeware version of OpenSSL. Since elog does not use configure script, you have to
edit Makefile and add an appropriate -I option to an appropriate variable definition (probably LIBS or CFLAGS, because the elog Makefile does
not use INCLUDES).
If you don't understand what I'm saying, just wait for me.
  1119   Thu Nov 6 22:07:56 2008 ranaConfigurationComputersELOG compile on Solaris
From the ELOG web pages:

Solaris:

Martin Huber reports that under Solaris 7 the following command line is needed to compile elog:

gcc -L/usr/lib/ -ldl -lresolv -lm -ldl -lnsl -lsocket elogd.c -o elogd

With some combinations of Solaris servers and client-side browsers there have also been problems with ELOG's keep-alive feature. In such a case you need to add the "-k" flag to the elogd command line to turn keep-alives off.
  1133   Thu Nov 13 15:50:44 2008 AlbertoConfigurationGeneralGPS 10MHz clock
The schematic of the 1Y7 rack that Alan pointed out (see attachment) don't represent anymore the actual rack.
First, with Yoichi we found that the GPS receiver for the 10 MHz is in a different position,
on the other side of the VME computer. It seems to output 1 kHz, which also appears in some modulated way.
This signal is then passed to a board on 1Y7 that seems make just copies of the signal. One of these goes
to an other board in 1Y6 that looks like a GPS receiver but has actually no GPs antenna input. Here it is
not connected to anything, but on its same crate is a the awg computer, so it might be providing the clock
to that by the crate.

We also checked the clock monitor output on the boards in the PSL that provides for the clock to the Penteks
and it seems that these are actually getting 4MHz.
Attachment 1: rackstuff.pdf
rackstuff.pdf rackstuff.pdf rackstuff.pdf rackstuff.pdf
  1146   Wed Nov 19 10:32:11 2008 AlbertoConfigurationElectronicsAll the Marconi Set to the Rubidium Frequency Standard
I placed the SRS Rubidium FS275 over the PSL rack, next to the frequency counter. This one and the Marconi on the PSL rack have been connected to the 10MHz output of the frequency standard. I set also the first Marconi, the one that used to drive the others, to external, direct frequency reference. Now it reads 166981718 Hz versus 166981725 Hz measured by the frequency counter: 8 Hz difference.
  1147   Wed Nov 19 18:02:18 2008 ranaConfigurationElectronicsAll the Marconi Set to the Rubidium Frequency Standard
Not sure what was going on before. I changed the frequency counter to use an AC coupled input, and had it average
for 50 seconds. The output now agrees with the Marconi front panel to less than 1 Hz. Its still not 0.000 Hz,
but not bad.
  1148   Wed Nov 19 18:12:35 2008 ranaConfigurationIOOnew channel for MC drum modes
I set up the lockin to take the MC Demod Board's Qmon signal, demodulate it at 27.5 kHz, and
put the output into a DAQ channel (I think its either MC_DRUM1 or MC1_TEMPS). However,
the MC_DRUM channel doesn't look like its getting anything in the DTT although it looked fine
on a scope. I used the 'sensitivity' setting of the lockin to make the demodulated signal
large enough but not so large that it would saturate the ADC (+/- 2V).
  1158   Sat Nov 22 10:55:51 2008 CarynConfigurationIOODrum modes Lock-In settings changed
I unhooked the MC Demod Board's Qmon signal from the Lock-In. Set the demodulation frequency to 31.11Hz with 1V amplitude, and
put the output into MC_DRUM1. DTT showed a ~30Hz peak. Dataviewer showed signal with amplitude ~20,000.
Otherwise the settings were as Rana had them: Time Constant-100us,24dB/Sensitivity-200us/Low Noise
Want to check if Lock-In frequency drifts.
  1159   Mon Nov 24 16:43:34 2008 ranaConfigurationComputersAlex and Jay took away some computers from the racks
I was over at Wilson house and saw Jay and Alex bring in 3 rackmount computers. One was a Sun 4600 and
then there were 2 3U black boxes. I got the impression that these were the data concentrators or
data collectors or framebuilder test boxes. They said that they got these from the 40m and no one was
in the lab to oppose them except for Bob and he didn't put up much of a fight.

Everything looks green on the DAQ Detail and RFM network screens so perhaps everything is OK. Beware.
  1161   Mon Nov 24 19:15:16 2008 rana, alberto, johnConfigurationEnvironmenttemperature
The PSL Room Temperature was alarming because it had gone above 23 C. This set off an unfortunate chain of events:

We found that the PSL HEPA was set low (20%). This is a fine setting for when no one is working in there but it
does raise the temperature since there are heat sources inside the blue box.

We tried to change the office area temperature to compensate and also the westmost sensor inside the lab area by 2 deg F.

The office area one was problematic - there was so much dust in it that the gas valve nipple was clogged. So we've
now blown it all clean with a compressed air can. We're now tuning the calibration screw to make our new
digital sensor agree with the setpoint on the controller.

Expect wild temperature swings of the office area for a couple days while Alberto and I tune the servo.
  1163   Tue Nov 25 19:29:15 2008 rana, alberto, johnConfigurationEnvironmenttemperature

Quote:
The PSL Room Temperature was alarming because it had gone above 23 C. This set off an unfortunate chain of events:

We found that the PSL HEPA was set low (20%). This is a fine setting for when no one is working in there but it
does raise the temperature since there are heat sources inside the blue box.

We tried to change the office area temperature to compensate and also the westmost sensor inside the lab area by 2 deg F.

The office area one was problematic - there was so much dust in it that the gas valve nipple was clogged. So we've
now blown it all clean with a compressed air can. We're now tuning the calibration screw to make our new
digital sensor agree with the setpoint on the controller.

Expect wild temperature swings of the office area for a couple days while Alberto and I tune the servo.


This morning Bob found 92F in the office area and in the control room of the lab. He turned down the thermostat and when I got in at about 9 I found 65. After a few hours of adjustment of the thermostat's calibration, I could stabilize the room temperature to about 72F. I also turned down the thermostats inside the lab of a couple of degrees F.
  1164   Thu Nov 27 22:56:42 2008 ranaConfigurationEnvironmenttemperature
8-)
Attachment 1: mc.png
mc.png
  1166   Tue Dec 2 17:56:56 2008 Alberto, RanaConfigurationPSLMC Alignment
In the attempt to maximize the Mode Cleaner transmission and minimize the reflection from the steering mirrors of the MC periscope, we could not get more ~2 V at the MC Trans PD and ~ 0.5 V at MC REFL_DC. As it turned out from the SUS Drift Monitor, the reason was that the MC optics had been somehow displaced from the optimal position.

After restoring the reference position values for the mirrors and tweaking again the periscope, we got ~3V at the MC TransPD and 0.5V at the reflection.
The beam was then probably clipped at the REFL PD so that we had to adjust the alignment of one of the BS in the transmitted beam path on the AS table.
We also zeroed the WFS PDs, but not before reducing the power from the MZ, for their QPDs not to saturate.

After relocking, the transmission was 3V and the reflection ~0.3V.

The beam isnow centered on the Trans PD and REFL PD and the Mode Cleaner locked. More details on the procedure will follow.
  1171   Wed Dec 3 19:21:09 2008 ranaConfigurationPEMRanger move
I looked at the Ranger signals. Somehow it has a relative transfer function of 'f' between it and the Guralp.
      Ranger
i.e.  ------ ~ f
      Guralp

which is strange since according to their manuals, they should both be giving us a voltage output which is proportional
to velocity. I checked that the Ranger only has a load resistor and then an SR560 low pass at 300 Hz. Jenne assures
me that the Guralp breakout box shouldn't have any poles either (to be double checked). Its a mystery.

We made sure that the SR560 now is DC coupled, G = 100, & 1-pole low pass at 300 Hz. I moved it over next to the Guralp
(went through the mass recentering procedure after forgetting to lock it before moving). It is behaving as it was
before.

Attached is a 2 page PDF of the comparisons. The 'MC1' channels are Guralp and 'MC2' is Ranger.

The second attachment compares our seismometers (in counts) with the LHO Guralp seismometers. There's no high frequency
rolloff there like what we see here so I bet a dollar that there's a pole in the Guralp box somewhere.
Attachment 1: c.pdf
c.pdf c.pdf
Attachment 2: wsnb.pdf
wsnb.pdf
  1175   Thu Dec 4 16:29:20 2008 josephbConfigurationComputersError message on Frame Builder Raid Array
The Fibrenetix FX-606-U4 RAID connected to the frame builder in 1Y7 is showing the following error message: IDE Channel #4 Error Reading
  1177   Fri Dec 5 01:41:33 2008 YoichiConfigurationComputersMEDM screen snapshot now works on linux machines
As a part of my "make everything work on linux" project, I modified 'updatesnap' script so that linux machines can update MEDM screen snapshots.
Now, all 'updatesnap' in the subsystem directories (like medm/c1/lsc/cmd/updatesnap) are sym-link to /cvs/cds/caltech/medm/c1/cmd/updatesnap.
This script will take a window snapshot to a PNG file, and move the old snapshot to archive folders with date information added to the filename.
For compatibility, it also saves JPEG snapshot. Right now, most of 'view snapshot' menus in MEDM screens are calling 'sdtimage' command, which cannot display PNG files. I installed Imagemagick to op440m. We should change MEDM files to use 'display' command instead of 'sdtimage' so that it can show PNG files.
I've already changed some MEDM screens, but there are so many remaining to be modified.

PNG is better than JPEG for crisp images like screen shots. JPEG performs a sort of spacial Fourier transformations and low-pass filtering to compress the information. If it is used with sharp edges like boundaries of buttons on an MEDM screen, it naturally produces spacial aliasing (ghost images).

I also created several sym-links on the apps/linux/bin directory to mimic the Solaris-only commands, such as 'sdtimage', 'nedit' and 'dtterm'.
For example, nedit is symbolic linked to gedit. Many MEDM buttons/menus, which used to be incompatible with linux, now work fine on the linux machines.
  1178   Fri Dec 5 01:58:58 2008 YoichiConfigurationASCtdscntr.pl now works at 40m
Tobin gave me the perl version of tdscntr some time ago.
Pinkesh and I modified and tested it at LHO.
I further modified it today and now it runs fine on the linux machines at the 40m. I haven't tested it with the Solaris machines.
My modifications include changing channel names to 40m ones, and using tdsavg to get QPD data rather than ezcaread.
The use of tdsavg is intended to avoid aliasing problem.
tdscntr.pl is installed in /cvs/cds/caltech/apps/linux/tds/bin

Now, the alignX runs on linux up to the centering of the QPDs.
However, ezcademod seems to behave wrongly on linux. I plan to investigate on this problem tomorrow.
I may try tdsdmd instead.
  1192   Thu Dec 18 12:52:00 2008 AlbertoConfigurationSUSMode Cleaner Cavity Alignment
This morning I found the MC locked to the 10 mode. When I locked it on the 00 mode, it was unstable and eventually it always got locked to the wrong mode.

I looked at the Drift Mon MEDM screen, which shows a reference record for position, pitch and yaw of each mirror, and I found that the MC optics were in a different status. Moving the sliders of the mirrors' actuators, I brought them back to the reference position. Then the lock got engaged and it was stable, although the MC reflection from the photodiode, with the wave front sensors (WFS) off, was about 2V. That's higher than the 0.5V the it could get when we aligned the cavity and the input periscope last time.

With the WFS on, the reflection dropped to 0.3V and, so far, the the cavity has been stably locked.
  1193   Thu Dec 18 19:15:54 2008 Alberto, YoichiConfigurationSUSMode Cleaner Cavity Alignment

Quote:
This morning I found the MC locked to the 10 mode. When I locked it on the 00 mode, it was unstable and eventually it always got locked to the wrong mode.

I looked at the Drift Mon MEDM screen, which shows a reference record for position, pitch and yaw of each mirror, and I found that the MC optics were in a different status. Moving the sliders of the mirrors' actuators, I brought them back to the reference position. Then the lock got engaged and it was stable, although the MC reflection from the photodiode, with the wave front sensors (WFS) off, was about 2V. That's higher than the 0.5V the it could get when we aligned the cavity and the input periscope last time.

With the WFS on, the reflection dropped to 0.3V and, so far, the the cavity has been stably locked.


This evening the mode cleaner was again locking on a higher mode so we tweaked the mirrors' actuators by their sliders on the MEDM screen until we improved the reflection to 0.3V.

Then we went inside and, on the AS table, we centered the beam on the wave front sensors.

Now the mode cleaner is locked, the reflection is less than 0.3V and the transmission about 3V, tha is it is in ideal conditions. We'll see if it holds.
  1194   Fri Dec 19 11:18:52 2008 AlbertoConfigurationGeneralMode Cleaner Temperature Monitor
I reduced from 10 to 5 the gain of the SR560 that Caryn has set up after the lock-in amplifier nest to the PSL rack because the overload LED was flashing.
  1195   Fri Dec 19 11:29:16 2008 Alberto, YoichiConfigurationMZMZ Trans PD
Lately, it seems that the matching of the input beam to the Mode Cleaner has changed. Also, it is drifting such that it has become necessary to continuously adjust the MC cavity alignment for it to lock properly.

Looking for causes we stopped on the Mach Zehnder. We found that the monitor channel:
C1:PSL-MZ_MZTRANSPD

which supposedly reads the voltage from some photodiode measuring the transmitted power from the Mach Zehnder, is totally unreliable and actually not related to any beam at all.

Blocking either the MZ input or output beam does not change the channel's readout. The reflection channel readout responds well, so it seems ok.
  1207   Mon Dec 29 21:51:02 2008 YoichiConfigurationComputersWeb server on nodus
The apache on nodus has been solely serving for the svn web access.
I changed the configuration and all files under /cvs/cds/caltech/users/public_html/ can be seen under
https://nodus.ligo.caltech.edu:30889/

The page is not password protected, but you can add a protection by putting an appropriate .htaccess
in your directory.
For the standard LVC password, put the following in your .htaccess
AuthType Basic  
AuthName "LVC password"
AuthUserFile /cvs/cds/caltech/apache/etc/LVC.auth
Require valid-user
  1208   Tue Dec 30 18:51:18 2008 rana,yoichiConfigurationElectronicsIlluminator Power Supply reset
We noticed that none of the illuminators were working.

The switches were off on all the ports. After turning them on it still didn't work.

The +24 V Sorensen power supply which powers all of the illuminators had its OVP light on.
We turned it off, ramped the voltage to zero, turned it back on, and then went back to +24 V.

We then checked the operation of the illuminators; ETMY is still MIA.

Each of the illuminators sucks ~0.6-0.7 A when the (unlabeled) rheostat knob panel is set
to the "25" setting.

It seems pretty unwise, in the EMI sense, to be sending Amps of unshielded, high current,
switching supply outputs for 40m down the arms. This creates a huge antenna for radiating
the switching noise. I hereby assign minus 5 points to whoever designed this system.

Illuminator Upgrade:
- Use LEDs of a wavelength that the OSEMs don't see. LEDs are also cool so that the
  Suspension won't drift in alignment.

- Use 2 power supplies so that the power is balanced.

- Make is +/-12 V twisted AWG 14 wire so that the EMI is contained. Should also
  be shielded cable.
  1217   Thu Jan 8 16:49:37 2009 ranaConfigurationloreHP 5550dtn (Grazia) set up on allegra

Quote:
I set up printing to grazia from allegra. The CUPS interface was not as straightforward as Tobin had made it seem in the Wiki. I had to type in the IP address and port number by hand.

The steps (AFAIR):
1) Goto http://localhost:631/
2) Click on "Add Printer"
3) Choose HP JetDirect
4) Use the correct address (socket://131.215.115.220:9100)
5) Choose HP and the 5550 postscript driver as the options
6) Try to only print useful stuff and not kill too many trees.


It ought to be root to do that.
  1224   Tue Jan 13 11:10:42 2009 robConfigurationComputersconlogger restarted
unknown how long it's been down.
  1230   Thu Jan 15 22:30:32 2009 ranaConfigurationDMFDMF start script
I tried to restart the DMF using the start_all script: http://dziban.ligo.caltech.edu:40/40m/280

it didn't work Frown
  1232   Fri Jan 16 11:33:59 2009 robConfigurationDMFDMF start script

Quote:
I tried to restart the DMF using the start_all script: http://dziban.ligo.caltech.edu:40/40m/280

it didn't work Frown


It should work soon. The PATH on mafalda does not include ".", so I added a line to the start_DMF subscript, which sets up the DMF ENV, to prepend this to the path before starting the tools. I didn't put it in the primary login path (such as in the .cshrc file) because Steve objects on philosophical grounds.

Also, the epics tools in general (such as tdsread) on mafalda were not working, due to PATH shenanigans and missing caRepeaters. Yoichi is harmonizing it.
  1236   Fri Jan 16 18:45:20 2009 YoichiConfigurationIOOMC_L gain increased by a factor of 2
Rana, Yoichi

Since we fixed the FSS AOM double-pass, which used to be a single-pass, the MC_L gain was too low for
making the cross-over at 100Hz.
Rana increased it by a factor of two. Now it seems that the cross over is ok (attachment 1).

We also noticed that the MC_F spectrum is noisier than before (attachment 2).
The reference is from 6/24/2008.
Attachment 1: MC_F-MC_L-xover.pdf
MC_F-MC_L-xover.pdf
Attachment 2: MC_F.pdf
MC_F.pdf
  1244   Thu Jan 22 11:54:09 2009 AlbertoConfigurationPSLMach Zehnder Output Beam QPD
I rotated by 180 degrees the 10% beam splitter that it is used to fold the beam coming from the Mach Zehnder (directed to the MC) on to the QPD.

The alignment of the beam with that QPD has so been lost. I'll adjust it later on.

The rotation of the BS had the (surprising) effect of amplifying the Absolute Length experiment's beat by 9 times. Maybe because of a polarizing effect of the Beam Splitter which could have increased the beating efficiency between the PSL and the NPRO beams?
  1252   Sat Jan 24 11:50:24 2009 AlbertoConfigurationElectronicsPhotodiode Filters' Transfer Functions
I found an elog entry by Jenne with the measurement of the transfer functions of the filters of some of our photodetectors. Since it might turn useful to us these days, while I'm thinking about posting them on the wiki sometime, I also copy the link here:
Jenne's on the PD's TF

If we still had the data for those plots, it would be great. Do we have it?
  1253   Mon Jan 26 14:51:54 2009 josephbConfigurationVAC 
We need a new RS-232 to Ethernet bridge in order to interface properly with the new RGA. The RGA has a fixed baud rate of 28.8k, and the current bridge (which used to work with the old RGA) doesn't have that baud rate as an option. Currently looking into purchasing a new bridge, and trying to make sure it can meet the communications requirements of the RGA.
  1258   Thu Jan 29 16:50:53 2009 josephb, albertoConfigurationComputersMegatron fixed
The warning light on megatron and the fans at full speed were fixed by not just power cycling, but completely unplugging megatron from power, waiting for a minute, and then reconnecting the power cables.

Apparently, the Sunfire X4600s at Hanford have also had intermittent problems, which required unplugging the machines completely. In their case, they were unable to access the machine normally, nor could they access the the Intergrated Lights Out Manager (ILOM).

Here, we could interact normally with megatron, but were unable to connect to the ILOM. We were able to get BIOS, but unable to change any of the setting for the ILOM connection. Since the ILOM is a seperate processor and effectively always on, even when the power light is off, a normal shutdown won't reset it. Hence the need to completely disconnect the power supply.
  1261   Fri Jan 30 17:30:31 2009 Alberto, JosephbConfigurationComputersNew computer Ottavia set up
Alberto, Joseph,

Today we installed the computer that some time ago Joe bought for his GigE cameras. It was baptized "OTTAVIA".

Ottavia is black, weighs about 20 lbs and it's all her sister, Allegra (who also pays for bad taste in picking names). She runs an Intel Core 2 Quad and has 4GB of RAM. We expect much from her.

Some typical post-natal operations were necessary.

1) Editing of the user ID
  • By means of the command "./usermod -u 1001 controls" we set the user ID of the user controls to 1001, as it is supposed to be.

2) Connection to the Martian network
  • Ottavia was given IP address 131.215.113.097 by editing the file /etc/sysconfig/networ-scripts/ifcfg-eth0 (we also edited the netmask and the gateway address as in the Wiki)
  • In linux1, which serves as name server, in the directory /var/named/chroot/var/named, we modified both the IP-to-name and name-to-IP register files 131.215.113.in-addr.arpa.zone and 131.215.11in-addr.martian.zone.
  • We set the file /etc/resolv.conf so that the OS knows who is the name server.

3) Mounting of the /cvs/cds path
  • We created locally the empty directories /cvs/cds
  • We edited the files /etc/fstab adding the line "linux1:/home/cds /cvs/cds nfs rw,bg,soft 0 0"
  • We implemented the common variables of the controls environment by sourcing the cshrc.40m: in the file /home/controls/.cshrc we added the two lines "source /cvs/cds/caltech/cshrc.40m" and "setenv PATH ${PATH}:/cvs/cds/caltech/apps/linux64/matlab/bin/"
  1263   Mon Feb 2 12:35:22 2009 AlbertoConfigurationGeneralNew Elog 2.7.5 in Service on Nodus
I moved the 40m, the Adhikari Lab and the SUS elogs from Dziban (located in Millikan's 6th floor) to our gateway server Nodus. In this way we should the complete control of it. I also updated the elog manager from the 2.6.5 version to the 2.7.5. Some smoothing of its interface might still be needed these days. We'll be testing it for a while before killing the old one. from now on everybody is invited to use only the new elog address since there will be no record of entries posted in the old one. Let me know of any possible difficulty in having access to it.
  1264   Mon Feb 2 17:25:44 2009 AlbertoConfigurationGeneralSome little problem with the new elog
For some reason the new elog does not look exactly like it should. 1) Some of the editing features are not available. 2) The Reply option opens the HTML of the message by default. I think this is happening because Nodus is a Sun of platform and things are a little bit different from linux. I'm working to fix these things. If I make any change and need to restart the elog, I'll try to be very quick. I apologize for the inconveniences.
  1265   Mon Feb 2 18:32:54 2009 AlbertoConfigurationGeneralSome little problem with the new elog

Quote:
For some reason the new elog does not look exactly like it should. 1) Some of the editing features are not available. 2) The Reply option opens the HTML of the message by default. I think this is happening because Nodus is a Sun of platform and things are a little bit different from linux. I'm working to fix these things. If I make any change and need to restart the elog, I'll try to be very quick. I apologize for the inconveniences.

 I think I solved the problem (as you can probably see).

The cause was that this WYSIWYG interface for HTML is provided by an independent text editor called FCKeditor which is included in the elog. Although the elog installer has a bug and does not unzip properly the relative package. One has to do it by hand. (going to /elog/scripts/ and unzipping fckeditor.zip by hand in the same directory).

  1266   Mon Feb 2 18:51:02 2009 AlbertoConfigurationGeneralNew Elog 2.7.5 in Service on Nodus

Quote:
I moved the 40m, the Adhikari Lab and the SUS elogs from Dziban (located in Millikan's 6th floor) to our gateway server Nodus. In this way we should the complete control of it. I also updated the elog manager from the 2.6.5 version to the 2.7.5. Some smoothing of its interface might still be needed these days. We'll be testing it for a while before killing the old one. from now on everybody is invited to use only the new elog address since there will be no record of entries posted in the old one. Let me know of any possible difficulty in having access to it.

 As a reference. The elog runs on background in nodus.

To kill the process:

1) pkill -3 elogd

2) rm -f /var/run/elogd.pid

To restart it:

elogd -p 8080 -c /export/elog/elog-2.7.5/elogd.cfg -D

  1287   Mon Feb 9 19:50:48 2009 YoichiConfigurationPSLISS disconnected
We are doing measurements on ISS.
The ISS feedback connector is disconnected and the beam to the MC is blocked.
  1292   Wed Feb 11 10:52:22 2009 YoichiConfigurationDAQC1:PEM-OSA_APTEMP and C1:PEM-OSA_SPTEMP disconnected

During the cleanup of the lab. Steve found a box with two BNCs going to the ICS DAQ interface and an unconnected D-SUB on the floor under the AP table.  It seemed like a temperature sensor.

The BNCs were connected to C1:PEM-OSA_APTEMP and C1:PEM-OSA_SPTEMP.

Steve removed the box from the floor. These channels can be now used as spare DAQ channels. I labeled those cables.

  1293   Wed Feb 11 11:39:17 2009 steveConfigurationGeneralvac envelope ground
The vacuum envelope was grounded from the IOO chamber to master ground north (under electric breaker panel L)
with 4GA  25ft red battery cable.

 

  1294   Wed Feb 11 15:01:47 2009 josephbConfigurationComputersAllegra

So after having broke Allegra by updating the kernel, I was able to get it running again by copying the xorg.conf.backup file over xorg.conf in /etc/X11.  So at this point in time, Allegra is running with generic video drivers, as opposed to the ATI specific and proprietary drivers.

  1299   Thu Feb 12 18:35:10 2009 KakeruConfigurationPSLPA current limitter


I added a PA current limiter.
It is only a voltage devider (composed with 3.09k and 1.02k resiste) between DAC and PA current adjustment input.
The output range of DAC is +/- 10[V] and  the conversion factor of PA current adjustment is 0.84[A/V] (measured value), so the PA current adjustment is limited +/- 2.1[A] ( 10[V]*1.02k/(1.02k+3.09k)*0.84[A/V] ).


Actually, the manual of the PA tells that the conversion factor is 0.25[A/V].
There is 3 possibility.
1) There are some mistakes in channels of digital system.
2) The PA manual is wrong.
  2-1) The conversion factor of current adjustment is wrong.
  2-2) The conversion factor of current monitor is wrong.
I measured the signal of current adjustment and current monitor directly, and confirm that they are consistent to the value monitord from MEDM.
Hence the PA manual must be wrong, but I don't know which factor is wrong (or both?).
If the suspect 2-2) is guilty, it means we adjust PA current with very small range.
This is a completly safety way, but a wast of resource.


Now, the slider to control current adjustment indicate the output of DAC.
I will improve this to indicate  current adjustment input, but it takes some time for me to learn about EPICS.

  1302   Fri Feb 13 16:30:49 2009 steveConfigurationGeneralstatus quo is disturbed

I have been getting ready for the annual safety inspection in the past 2-3 days.

Meaning cleaning up and disturbing the status quo on the floor  mostly under the optical tables and their surroundings.
For example: pd power supplies, He/Ne laser ps. and their positions.
BNC cables and ac power line positions can be different.
The new rule: no electronic equipment on the floor.
 
All electronic equipment were moved-placed into a plastic dish or tray.
  1303   Sat Feb 14 16:15:19 2009 robConfigurationComputersc1susvme1

c1susvme1 is behaving weirdly.  I've restarted it several times but its computation time is hanging out around 260 usec, making it useless for suspension control and locking.  I also found a PS/2 keyboard plugged in, which doesn't work, so I unplugged it.  It needs to be plugged into a PS/2 keyboard/mouse Y-splitter cable. 

  1312   Mon Feb 16 20:47:48 2009 ranaConfigurationIOOMode Cleaner WFS Loop Gain change
I found the MCWFS gain slider down at 0.012. In this state the UGFs are probably around 10-30 mHz
and so there is no reduction of seismic noise. It is mainly a DC alignment tool in this state.

We often have reduced the loop gain thusly, to prevent the dreaded "MCWFS eating CM loop gain" disease.
That disease is where there are CM loop instabilities at ~5-30 Hz because of loop cross-couplings
who's exact nature has never been understood (TBI).

Today, I implemented a 4th order, 7 Hz low pass (RLP7) into the loops and turned up the gain by a factor
of 30 to 0.3. In this state, the damping time constants seem to be ~0.5-2 seconds as shown in the first
PDF. I didn't have enough patience to do the interminable swept sine measurements down to 0.1 Hz.

The second PDF shows the Bode plot of the RLP7 filter compared to the pre-existing but unused ELP10.

The third PDF shows my estimate of the OLG TF. This is made by just putting a "Pendulum" filter into the
MCWFS bank and then plotting all the filters together using FOTON. The BLUE curve shows the old TF but
with the new high gain and the RED curve shows the new TF with the new gain.

With this new filter, I bet that we can get away with the higher WFS gain, but if there's any problem during the
handoff, the gain should be reverted to the low value.


In the 4th PDF file, I plot the spectra of 4 of MC2's control signals so that you can see what is bigger than what.
ASCPIT is the one that has the feedback from the WFS's in it. These are all just in units of counts and so to compare
them in some sort of displacement units you have to take into account the pitch moment of inertia, the mirror mass,
and the mis-centering of the beam from the center of rotation of MC2...
Attachment 1: pmc-pzt-cal.pdf
pmc-pzt-cal.pdf
Attachment 2: a.pdf
a.pdf
Attachment 3: olg.pdf
olg.pdf
Attachment 4: mc2.pdf
mc2.pdf
  1314   Mon Feb 16 22:58:51 2009 rana, yoichiConfigurationSUSHysteresis in SUS from Misalignments
WE wondered if there was some hysteresis in the SUS alignments. When we leave the optics misaligned for a
long time it seems to take awhile for the optic to settle down. Possibly, this is the slow deformation of
the wires or the clamps.

The attached PNG shows the plot of the bias sliders for a few days. You can see that we misalign some of the
optics much more than the others. This must be stopped.

Kakeru is going to use his nearly complete optical lever calibrations to quatify this by stepping the optics
around and measuring the effect in the optical lever. Of course, the misalignment steps will be too large to
catch on the OL, but he can calibrate the align-sliders into radians to handle this.
Attachment 1: a.png
a.png
  1320   Wed Feb 18 19:13:20 2009 ranaConfigurationComputersSVN & MEDM & old medm files
Allegra had a 2 year old version of SVN installed and CentOS (yum) couldn't upgrade it, so I did an 'svn remove subversion'
and then a 'svn install subversion' to get us up to the Dec '08 version (1.5.5) which is the latest stable.

I also removed all of the old ASS medm directories without backing them up. There's a new RCG script version which is
fixed so that it no longer dumps these old medm directories in there; there's no need since there's already an
medm archive area.

I also removed the medm/old/ directory, did an svn remove, and then copied it back. This is the only way I know of
removing something from the repository without removing it from the working directory.
  1333   Mon Feb 23 16:42:08 2009 josephbConfigurationCamerasCamera Beta Testing

I've setup the GC650 camera (ID 32223) to look at the mode cleaner transmission.  I've also added an alias to the camera server and client for this camera.

To use, type: "pserv1 &"on the machine you want to run the server on and "pcam1 &" on the machine you want to actually view the video.  At the moment, this only works for the 64-bit Centos 5 machines, Rosalba, Allegra and Ottavia.

Note, you will generally want to start the client first (pcam1 or pcam2) to see if a server is already running somewhere.  The server will complain that it can't connect to a camera if it already is in use.

I've setup the GC750 camera (ID 44026) to look at the the right most analog quad TV.  This can be run by using "pserv2 &" and "pcam2 &". 

If the image stops playing you can try starting and stoping the server to see if will start back up. 

You can also try increasing or decreasing the exposure, to see if that helps.  The increase and decrease buttons change the exposure by a factor of 2 for each press.

  Lastly, the button "Read Epic Channel" reads in the current value from the channel: "C1:PEM-stacis_EEEX_geo" and uses it as the exposure value, in microseconds (in principle 10 to 1000000 should work).

For example, to exposre for 10000 microseconds, use "ezcawrite C1:PEM-stacis_EEEX_geo 10000" and press the "Read Epic Channel" button.

ELOG V3.1.3-