40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  40m Log, Page 137 of 337  Not logged in ELOG logo
ID Date Author Type Category Subject
  10077   Thu Jun 19 22:04:23 2014 ericqUpdateComputer Scripts / Programscaget/caput now return in reasonable time

I think I've fixed the caget/caput issue. Rana's observation that pinging the IP directly was faster than pinging the hostname set me on a path of googling which informed making the following changes to the DNS setup on chiara (specifically, informed by this thread: http://www.dslreports.com/forum/r11836974-BIND-slow-to-reply-over-LAN-Solved)

/etc/bind/named.conf.local has these lines:

zone "martian" IN {
 type master;
 file "/etc/bind/zones/martian.db";
 };
zone "113.168.192.in-addr.arpa" {
 type master;
 file "/etc/bind/zones/rev.113.168.192.in-addr.arpa";
};

The first zone command links hostnames like c1lsc to an IP like 192.168.113.62, but apparently in the second, we need to do the inverse. So, for each line in martian.db like

c1lsc           A       192.168.113.62

I added a line in rev.113.168.192.in-addr.arpa like so:

62 IN PTR c1lsc.martian

This seems kind of silly, but now if you do the host command from a workstation, it can find the hostname associated with an IP. 

controls@pianosa|~ > host 192.168.113.62
62.113.168.192.in-addr.arpa domain name pointer scipe12.martian.113.168.192.in-addr.arpa.
62.113.168.192.in-addr.arpa domain name pointer c1lsc.martian.113.168.192.in-addr.arpa.

[At this point, note that we have a bunch of duplicate entries in https://wiki-40m.ligo.caltech.edu/Martian_Host_Table  with these scipe## hostnames. What are these for?]


 
Now (edited for brevity):
 
controls@ottavia|~ > ping -c 5 -D c1sus
PING c1sus.martian (192.168.113.85) 56(84) bytes of data.
<SNIP>
--- c1sus.martian ping statistics ---
5 packets transmitted, 5 received, 0% packet loss, time 3997ms
rtt min/avg/max/mdev = 0.051/0.075/0.114/0.028 ms
controls@ottavia|~ > ping -c 5 -D 192.168.113.85
PING 192.168.113.85 (192.168.113.85) 56(84) bytes of data.
<SNIP>
--- 192.168.113.85 ping statistics ---
5 packets transmitted, 5 received, 0% packet loss, time 3998ms
rtt min/avg/max/mdev = 0.052/0.130/0.380/0.127 ms
 
controls@pianosa|~ > time caget C1:LSC-XARM_GAIN
C1:LSC-XARM_GAIN               0.015
real    0m0.039s
 
controls@pianosa|~ > time caput C1:LSC-XARM_GAIN 0.0151
Old : C1:LSC-XARM_GAIN               0.015
New : C1:LSC-XARM_GAIN               0.0151
real    0m0.054s
 
 
 

 

  10076   Thu Jun 19 17:28:19 2014 SteveUpdatePSL window device is ready for test

The window is at the north west corner of the  PSL enclosure.

The 1 mm thick BK7 window is AR coated both side R <0.08%,  W2-PW1-1004-1064-0

The PZT stack is 0.75" OD, 0.25" thick with  ~ 6 mm ID,  motion range 2.5 micron at 200V

Soft silicon rubber isolation.

 

RXA: This is the opto-mechanical phase shifter that Steve has built for diagnosing scattered light problems. We put it into the reflected light path of any of the cavities and see if it can move the scattering noise from DC up to a higher frequency. e.g.:

The paper on this from GEO

Attachment 1: softW2.jpg
softW2.jpg
Attachment 2: softW2b.jpg
softW2b.jpg
  10075   Thu Jun 19 17:11:52 2014 steveUpdatePEMit is warmer today

 The PSL HEPA was running at 30VAC-variac. It is 1 degree C warmer than usual. The HEPA fan motor turned up to 60 V

Attachment 1: warmertoday.png
warmertoday.png
  10074   Thu Jun 19 16:49:15 2014 ranaUpdateComputer Scripts / Programsautolocker running again

  We've noticed that the caget/caput and cdsutils take a long time to return the command prompt. The following two ping commands indicate that it may be related to routing or our new BIND9 DNS setup on chiara.

 

 

controls@ottavia|~ > ping -c 5 -D c1sus

 

PING c1sus.martian (192.168.113.85) 56(84) bytes of data.

 

[1403221338.383403] 64 bytes from 192.168.113.85: icmp_req=1 ttl=64 time=0.114 ms

 

[1403221343.386211] 64 bytes from 192.168.113.85: icmp_req=2 ttl=64 time=0.059 ms

 

[1403221348.387666] 64 bytes from 192.168.113.85: icmp_req=3 ttl=64 time=0.060 ms

 

[1403221353.389736] 64 bytes from 192.168.113.85: icmp_req=4 ttl=64 time=0.059 ms

 

[1403221354.390713] 64 bytes from 192.168.113.85: icmp_req=5 ttl=64 time=0.060 ms

 

--- c1sus.martian ping statistics ---

 

 

 

5 packets transmitted, 5 received, 0% packet loss, time 20007ms

 

rtt min/avg/max/mdev = 0.059/0.070/0.114/0.023 ms

 

 

 

controls@ottavia|~ > ping -c 5 -D 192.168.113.85

 

PING 192.168.113.85 (192.168.113.85) 56(84) bytes of data.

 

[1403221463.737857] 64 bytes from 192.168.113.85: icmp_req=1 ttl=64 time=0.078 ms

 

[1403221464.737353] 64 bytes from 192.168.113.85: icmp_req=2 ttl=64 time=0.059 ms

 

[1403221465.737318] 64 bytes from 192.168.113.85: icmp_req=3 ttl=64 time=0.050 ms

 

[1403221466.737316] 64 bytes from 192.168.113.85: icmp_req=4 ttl=64 time=0.052 ms

 

[1403221467.737321] 64 bytes from 192.168.113.85: icmp_req=5 ttl=64 time=0.050 ms

 

 

 

--- 192.168.113.85 ping statistics ---

 

5 packets transmitted, 5 received, 0% packet loss, time 3999ms

 

rtt min/avg/max/mdev = 0.050/0.057/0.078/0.014 ms

 
  10073   Thu Jun 19 14:52:20 2014 not ericqUpdateComputer Scripts / Programscontrol room bashrc change

Quote:

Some time ago, Rana changed the PS1 prompt codes on the control room computers. However, the exit codes of commands weren't being displayed, and there was some lingering color changing after the line. Hence, I changed it to look like this:

PS1='\[\033[0;35m\]\u'
PS1="$PS1\[\033[0;30m\]@"
PS1="$PS1\[\033[0;33m\]\h"
PS1="$PS1\[\033[0;97m\]|"
PS1="$PS1\[\033[0;92m\]\W"                                                 
PS1="$PS1\[\033[0;31m\] \${?##0}"
PS1="$PS1\[\033[0;97m\]>\[\033[0m\] "

The \${?##0} means: display the exit code if it is not zero (which means success). Thus, it only displays the exit code when its something other than what is expected.

  It's a very good plan to always inspect the exit code of your command.  Well done.

  10072   Thu Jun 19 14:41:00 2014 ManasaUpdatePSLISS disabled

I would like to measure the switching time of the AOM. So I have disconnected the modulation input to the AOM that comes from the ISS. I have also turned OFF the SR560's and the AWG that belong to ISS. 

Pics and cable connections of the state in which the ISS setup was left at, will be updated soon.

  10071   Thu Jun 19 14:12:45 2014 steveUpdatesafetyoptical table scans

 

 PSL, AP, ETMY and ETMX optical tables were scanned for stay beam.

  10070   Thu Jun 19 12:10:18 2014 AkhilUpdateElectronicsGain plots to Characterize UFC-6000 RF Frequency Counter

The goal is to characterize the Mini-Circuits RF FC (Model UFC-6000)  by plotting Gain Plots.

Work Done:

The sampling rate of the UFC-6000 RF FC is 1s (should look into making the sampling time smaller). So to satisfy Nyquist criterion, the maximum modulation frequency is 0.5 Hz beyond which  aliasing effects are seen.

The measurements taken (mentioned in my previous elog) are used to plot Gain vs Modulation frequency for carrier frequencies of  5 MHz and 25 MHz.

Calculations:

A modulated signal can be represented as X(t)= A*sin (Fc*t+D*sin(Fm*t+phase1))  where Fc and Fm are carrier and modulation frequencies respectively and D is the modulation depth.

This signal Y(t) is input to the FC and the output frequencies of the FC are recorded.

Let the output of the FC is Y(t)= A'*sin(Fc*t+D'*sin(Fm'*t+phase2));

 

Gain = D'/D;

phase = phase2 - phase1;

 

D' is calculated by subtracting the carrier frequency from the output frequency and calculating the amplitude of the resulting fitted sine wave.

The phase can be calculated if the phase of the input is known(which will be done next).

Plotting the Bode plots would give response of the FC to modulation.

 

The plots generated  will be used to estimate:

1) Transfer Function of FC to be known to build an FOL-PID loop.

2) Quantization noise from Power Spectral Density(PSD) vs Hz.

 

To Do Next:

1)Calculate the phase difference  to complete the Bode plot. This would require interfacing of the ADC on raspberry pi.

2)Estimate the quantization noise from the digital output.

 

 

Attachment 1: GainVsFreq.png
GainVsFreq.png
  10069   Thu Jun 19 10:39:22 2014 steveUpdatesafetysurf safety training

 

 Andres Medina and Andrew "Harry" Hall received 40m specific safety training. They did general safety already and their laser safety training will be this after noon.

Attachment 1: 2014surfsAA.jpg
2014surfsAA.jpg
  10068   Thu Jun 19 00:02:48 2014 JenneUpdateLSCPRMI+2 arms recovery

Arms locked in comm and diff with ALS. PRMI locked with REFL33 I&Q while arms off resonance. Having trouble reducing CARM offset, even to get to arm powers of 1.

After Manasa installed the new Xgreen PD, Koji looked at the PSL table alignment with me.  I saw only a very weak beatnote with the X BBPD, even though I could see the beatnote on the Y PD from the leakage of the X beam to the Y PD  (Yend shutter was closed, so just PSL and X greens were on the table).  I had thought that my near-field and far-field alignments were pretty good (actually, I checked them, but didn't feel that I needed to tweak them since Manasa did the alignment this afternoon).  Anyhow, it was just a matter of tweaking up the alignment a bit, and then the X beatnote got up to about -25dBm at a few tens of MHz.  I am starting to question myself if the other BBPD is broken, or if I just not well enough aligned.  Anyhow, the spare is in, we can still have a look at the previous X BBPD, but it may be okay, and it's just me embarrassing myself by not catching an alignment problem.

Anyhow, after the X beat was found, I was able to (on my first try) lock the arms using ALS comm and diff.  (I already had a nice strong Y beatnote, so that didn't need finding, other than temp adjustment of the end laser).  I ran the carm_cm_up.sh script, and it did everything nicely.  I did a quickie check of the phase tracker loop gains, but that should be redone in the morning.

PRMI was a little reluctant to lock, so I played around with the MICH and PRCL gains, but didn't really find any combination that was any better than the usual (+0.8 for MICH, -0.04 for PRCL as I had last night, although I needed to reduce the PRCL gain back to -0.02 to eliminate loop osc). 

After an arm lockloss, I relocked just the PRMI and used awggui to put a line into C1:LSC-PRM_EXC to check the RF PD phasing.  I changed REFL33 from 133.5 to 138.5, and REFL 165 from -142.5 to -152.5.  I didn't think that REFL11 needed changing, and I didn't check REFL55.  I also checked that I could lock PRMI without arms, using both REFL33 and REFL165 - they seemed about the same to me, both stable.  REFL33 has 1's in the input matrix, and I was using 0.07's for REFL165.  The demod phase adjustment didn't really improve PRMI locking while the arms were held off resonance, even if I moved the arms even farther from resonance (usually we do 3nm, I went out to 5nm to see if that helped - it didn't).  I tried REFL165 locking, but that wasn't any good either.  I tried using REFL165 with the arms held off resonance, but that didn't seem to catch at all (at least with REFL33 I was getting short lock blips). 

Anyhow, of the 3 or 4 times that I caught REFL33 PRMI lock and tried to reduce the CARM offset, only one time did I even get to arm powers of about 1 (CARM digital offset of -0.1, with CARM held on sqrt transmission signals), and then it didn't stay for more than a few tens of seconds.  The other few times, it broke lock on the way up to arm powers of 1. 

So, carm_cm_up.sh works pretty well, although perhaps the arm powers of 1 offset reduction needs to be a little slower.  PRMI doesn't catch and hold lock very easily with REFL33, and even less so with REFL165.  It may be useful to try catching lock with REFL11 or 55, and doing a transition over to 3f.  No real progress forward, but we're pretty much recovered.

 

  10067   Wed Jun 18 22:47:48 2014 ericqUpdateCDSRaspberry pi added to martian network

I set up a raspberry pi on the martian network, to be hooked up to a frequency counter for tracking ALS beatnotes. 

The instructions at https://wiki-40m.ligo.caltech.edu/Martian_Host_Table are outdated, the name server configuration is now at /etc/bind/zones/martian.db, I need to remember to update the wiki soon. 

In any case, the raspberry pi is called "domenica," is found at 192.168.113.107, and has the standard controls user, with /cvs/cds mounted in the same way as the control room machines. 

Once I'm comfortable with the configuration of the pi, I'm going to take an image of the SD card that serves as its hard drive, so that we can just image new cards for future raspberry pis on the martian network if we ever want them. 

  10066   Wed Jun 18 22:34:44 2014 ericqUpdateIOOcaget frusrtation

Quote:

 Somehow the caget/caput commands are really slow. I'm not sure if this is new behavior or not, but after changing values, it takes ~1-2 seconds to move on to the next command.

This is still happening. Specifically: on all of the control room computers, calls to caget display the result immediately, but then hang for five seconds (consistently five). We had also seen a situation where calls hang indefinitely on ottavia/pianosa, but a reboot "fixes" this.

Some observations:

  • Front end machines and the FB have proper caget/caput response times.
  • Control room machines have some odd ping behavior when targeting frontends/FB; namely the ping times themselves are ok, but each ping line takes quite some time to show up, which made us think that there is odd network routing issue happening with some network switch. 
  • Front ends and FB get epics from /opt/rtapps, whereas control room machines get epics from /ligo/apps, which has different contents. (Is this for Gentoo vs. Ubuntu? I don't really get why this is the case...). This means different environment setting scripts to be called, so maybe the control room machines are misconfigured in some way for the new name server?

I poked around the network settings on all of these machines, but everything seemed reasonable. Nothing was changed. Rossa and Pianosa have their network settings done through some Ubuntu GUI, but I don't know where the settings are written. I had expected their settings to be in /etc/network/interfaces; maybe we should change this to be consistent with other machines, and easier to administrate via the terminal. 

Despite all this, ezcaread is fine.

  10065   Wed Jun 18 21:53:48 2014 KojiUpdateElectronicsChanges to the PD frequency response measurement system

Not "hot" current but "photo" current. Is this my bad!?

It was me who told Nichin that the DC transimpedance was 200Ohm. But according to this entry I checked the RF transimpedance of AS55 before.
In my code, the DC transimpedance was defined to be 50Ohm. If we believe it, 30mV corresponds to 0.6mA.

Quote:

The multimeter readout 30mV finally. Taking the transimpedence as 200ohm approx., the hot current is about 1.5mA.

 

  10064   Wed Jun 18 21:37:11 2014 KojiUpdateIOOMC REFL investigation

[Jenne Koji]

We decided to check the situation of the REFL MC beam path.

- No resolution of the weird MC REFL DC increase
- The reflection from the PD was adjusted to hit the beam dump
- The MC WFS paths were aligned again


Detail:

We found that the reflected beam from the PD was hitting the mount of the beam dump.
So the entire MC REFL path was steered down such that the last steering mirror does not neet to steer the beam.
When the alignment was adjusted so that the reflection from PD hit the beam dump, the beam spot on the small mirror right before the PD
got a bit marginal but it seemed still OK after some tweak.

Then we looked at the reflection value. It is still about 6.5. No change.

As we messed up the entire MC REFL path, we aligned the MC WFS paths again.
This was done with the unlocked MC REFL beam. Once the cavity was locked,
it turned out that it was enough for the WFS too keep the MC locked.

  10063   Wed Jun 18 19:30:28 2014 JenneUpdateComputer Scripts / ProgramsRossa having a better day

I have modified the /etc/default/grub file, so that we're loading up the previous linux kernel version on reboot.  Now Rossa boots up (at least one time so far) without any fancy button-pushing.

(Note, if things go south again, we have to push "shift" starting after the Dell screen is gone.  Holding it down while the Dell screen is still up doesn't seem to make it register that you want the grub menu).

The grub file USED to look like:

GRUB_DEFAULT=0
GRUB_HIDDEN_TIMEOUT=0
GRUB_HIDDEN_TIMEOUT_QUIET=true
GRUB_TIMEOUT=10
GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian`
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash"
GRUB_CMDLINE_LINUX=""

but now it looks like:

GRUB_DEFAULT=2
GRUB_HIDDEN_TIMEOUT=0
GRUB_HIDDEN_TIMEOUT_QUIET=false
GRUB_TIMEOUT=10
GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian`
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash"
GRUB_CMDLINE_LINUX=""

Note that the first line, GRUB_DEFAULT has changed from 0 (first item in grub menu, if you successfully hit shift and get to it) to 2 (the third item in the list).  I think that the GRUB_HIDDEN_TIMEOUT_QUIET=false is supposed to force it to show the countdown time when you can push shift, but I didn't see any difference there.

To edit this file, you must use "sudo [text editor] grub".  I like emacs, so I used "sudo emacs grub".  After an edit, before a reboot, you must run "sudo update-grub".  Then you can reboot (sudo shutdown -r now is what I use), and hopefully it will boot as you directed.

Right now, the 0th (first) entry in the grub menu is "Ubuntu, with Linux 2.6.32-61-generic".  The 2nd (third) entry is "Ubuntu, with Linux 2.6.32-58-generic".  If we install a new kernel version, the 61 version will bump down in the list, and we'll be booting that, which I assume will fail.  We should not update Rossa until we're ready to go to Ubuntu 12, and when we do, we must ensure that the grub file first line reads GRUB_DEFAULT=0.  (As a side note, the 1st (really 2nd) entry in the grub menu is "Ubuntu, with Linux 2.6.32-61-generic (recovery mode)", which we don't want.  The 58 version has a recovery mode as the next line item, since it alternates version-N, version-N recovery mode, version-(N-1), version-(N-1) recovery mode, etc.)

  10062   Wed Jun 18 18:16:26 2014 NichinUpdateElectronicsChanges to the PD frequency response measurement system

[Nichin, Eric G, Koji]

Continuing out work from elog:10037, we wanted to check if the frequency response of AS55. Having figured out exactly how to use the Laser diode controller (LDC 3744C), we hooked up a fiber power meter to the optical fiber illuminating AS55 (that we disconnected from its mount last Friday ) and raised up the current to 150mA to get almost 0.8mW power reading.

When aligning the fiber to illuminate the PD, we found that the beam was pretty wide. So we pulled out the collimator and tweaked it to get a focused beam. The fiber was mounted back and was aligned to get a maximum DC reading. The multimeter readout 30mV finally. Taking the transimpedence as 200ohm approx., the hot current is about 1.5mA.

Network analyzer was now connected to the modulation input of the laser and the RF output from REF DET and AS55 (inputs to RF switch at rack 1Y1) were connected as input to measure the transfer function. We got just noise on the scope of NA. So, then we tried REFL33 as the Input and still got nothing (We were also not sure if this PD was properly illuminated, we did not check). However the REF DET was giving a nice response on the scope. Turns out all the PDs were disconnected form the Demodulator (D990511) on rack 1Y2.

On closer inspection the RF cable between domodulator and RF switch that was labelled AS55 had a loose SMA connector at the switch end. I will have to fix that tomorrow . For the time being Koji connected the cable labelled REFL33 to the AS55 demodulator and we finally got a response form the AS55 PD on the NA. However no readings were recorded. The power supply to REF DET was turned off in the end as Eric G claimed that it has been ON for almost a year now, which is not a good thing. Also, we removed the modulation input from NA to the diode laser and terminated the input with a 50ohm terminator.

We planned to pull out and check each and every RF cable (especially the SMA ends for faulty soldering and loose connections) and fix/ replace them as needed.

  10061   Wed Jun 18 18:00:36 2014 ericqUpdateComputer Scripts / Programscontrol room bashrc change

Some time ago, Rana changed the PS1 prompt codes on the control room computers. However, the exit codes of commands weren't being displayed, and there was some lingering color changing after the line. Hence, I changed it to look like this:

PS1='\[\033[0;35m\]\u'
PS1="$PS1\[\033[0;30m\]@"
PS1="$PS1\[\033[0;33m\]\h"
PS1="$PS1\[\033[0;97m\]|"
PS1="$PS1\[\033[0;92m\]\W"                                                 
PS1="$PS1\[\033[0;31m\] \${?##0}"
PS1="$PS1\[\033[0;97m\]>\[\033[0m\] "

The \${?##0} means: display the exit code if it is not zero (which means success). Thus, it only displays the exit code when its something other than what is expected.

  10060   Wed Jun 18 17:38:14 2014 JenneUpdateComputer Scripts / Programsautolocker running again

The MC autolocker is once again running on Ottavia, with the nohup command.  

It was hanging for a long time (at least minutes) on the cavput and the caputs in the MC2 tickle on and off scripts.  I claim that there isn't a good reason to not just use ezcawrite, or whatever the latest and greatest fully functioning function is, so I've changed the cavput to a series of ezcawrites, and all of the caputs are also ezcawrites.  Now I don't see any hanging, and the MC locks itself.

This does not solve the scripto_cron issue, so if Ottavia is rebooted, or the autolocker is otherwise killed, it will not start itself up. 

  10059   Wed Jun 18 16:44:55 2014 ManasaUpdateElectronicsBBPD installed for BEATX detection

This BBPD is the spare that we pulled out and is installed for ALSX-PSL beat note detection.

  10058   Wed Jun 18 15:25:06 2014 NichinUpdateElectronicsBBPD Transimepedence plot

 Motivation:

To measure the transimpedence of  the Broadband photodiode (D1002969-v8), using a New focus photodiode (1611) as reference. The amplitude modulated Jenne Laser (1.2mW) was used @20mA

The steps involved in getting the transimpedence:

Acquiring data

  • The following conditions were set on Network Analyzer Agilent 4395:

1) Frequency sweep range: 500KHz to 300 MHz.

2) Number of Points sampled in  the range: 301

3) Type of sweep: Logarithmic

  • Set the NA to give the corresponding transfer function value (output of BBPD over output of 1611) and also Phase response in degrees.
  • Save the data into floppy disk for processing on the computer.

Plotting

  • The matlab code attached (Trans_plot.m) will then give plots for the absolute values of transimpedence in V/A.
  • Logic involved in the code will be presented clearly in a separate Elog. 

 Results

The Plots of transimpedence obtained are attached. The data and matlab code used is in the zip file.

The transimpedence of  Broadband photodiode (D1002969-v8) was around 1200 - 1300 V/A for most of the range (2), but the value started falling as the frequency approached 200 MHz. 

Attachment 1: BBPD.png
BBPD.png
Attachment 2: BBPD_readings_06-18-2014.zip
  10057   Wed Jun 18 13:39:01 2014 ranaUpdateCDScdsutils reverted to version 238

 

 After some email consult with Jamie, I got cdsutils working again by reverting to v238. The newest versions are not compatible with our python 2.6 on Ubuntu 10. And our Ubuntu 12 machines do not have NDS2 clients that work yet.

The read/write commands work at the moment, but the NDS based ones don't yet work on pianosa due to some NDSSERVER flag/setup issue maybe, Jamie ??

controls@pianosa|~ > z

usage: cdsutils <cmd> <args>

Advanced LIGO Control Room Utilites

Available commands:

    read         read EPICS channel value

  write        write EPICS channel value

  switch       switch buttons in standard LIGO filter module

  avg          average one or more NDS channels for a specified amount of seconds

  servo        servos channel with a simple integrator (pole at zero)

  trigservo    servos channel with a simple integrator (pole at zero)


  version      print version info and exit

  help         this help


Add '-h' after individual commands for command help.

controls@pianosa|~ > z read C1:LSC-DARM_GAIN

0.0

controls@pianosa|~ 2> z avg 3 C1:IOO-MC_F

Error in write(): Connection refused

Error in write(): Connection refused

NDSSERVER variable incorrectly defined, or no NDS servers available.

controls@pianosa|~ 1> echo $NDSSERVER

fb

  10056   Wed Jun 18 13:29:48 2014 ranaUpdateComputer Scripts / Programsautolocker confusion

 

 Moral is wrong.

AutoLocker was working fine last night and we observed it to run and do the appropriate mcdown and mcup stuff. Probably something is breaking with it after several hours.

If you have suspicions about the script, best to look through the logs during the first few hours when we had it running yesterday.

  10055   Wed Jun 18 11:57:44 2014 not JenneUpdateLSCIFO alignment recovery

Quote:

I noticed today, and Rana said that he saw Saturday, that the MC refl value when the MC is unlocked is unusually high.  It typically goes to about 4.5 V, but now is going up to 6.5V.  Since the PMC output is the same as usual (max seen has been about 0.82 today), something must have happened between the PMC and the IMC. 

Late last week, EricG and Nichin were looking at things on the AS table.  Was anything touched on the AS table?  Was anything touched on the PSL table?  'Fess up please, so that we can pinpoint what the change was.

 

 Nope, we did not touch any of the PDs other than AS55. I have mentioned in  my elog:10037 what we did exactly.

We just looked at all the other PDs to check if they were being illuminated by the correctly labeled fiber. Nothing other than that.

  10054   Wed Jun 18 11:28:19 2014 ManasaUpdateComputer Scripts / Programsautolocker confusion

Quote:

I renamed the Autolocker and described it in the elog from this weekend. To run it, you have to run it from the scripts/MC/ directory (as always). I restarted the autolocker on Ottavia with nohup.

> nohup ./AutoLockMC.csh &

To figure out how to get cron to run it, we will have to debug the difference between linux and solaris pgrep, so that's in progress.

I am NOT convinced that the autolocker script is running or doing anything. The MC was unlocked as of this morning and the autolocker wasn't doing anything to any of the MC servo buttons and sliders. The autolocker control shows that it is 'alive' but it still blinks 'MC down' even after I locked the MC manually. I am suspecting that this might have to do something with the caget and caput in autolock script itself rather than the CRON. I will look into this problem later today.

Moral of the story: The autolocker is not doing its job.

  10053   Tue Jun 17 21:49:13 2014 JenneUpdateLSCIFO alignment recovery

PRMI locking with REFL33 is fine.  As it was yesterday, it's a little wobbly without the ASC (just PRM oplev), but I don't know that it's any different than it used to be.  It'll hold for long periods of time, so I feel okay about it. 

When the PRMI is locked, you can push the "up" button on the ASC screen, and it'll turn down the PRM oplev gain by a large factor, and engage the ASC.  When you lose lock, press the "down" button to undo these changes.  (Probably the ifoconfig script should include the ASC down script).  These up and down scripts for the ASC are already included in the carm_up script (the ASC up), and the watch scripts which run a down script (including ASC down) for the whole IFO when ALS loses lock.  If the ASC is engaged, I get bored of watching it before the PRMi loses lock on its own, so I think it's okay.  (Let's say that means I've watched it stay locked for at least a few tens of seconds, but it looks like it always has with the ASC - like it'll stay forever).

The only thing that seems different about the PRMI is that I've increased the PRCL gain from -0.02 to -0.04.  This is a value that it was at some weeks ago, and then we turned it down for loop osc reasons, but now it doesn't want to catch lock with the lower value, and if I turn it down after it's locked, it has trouble holding on.  I have included this change into the PRMI sideband configure script.

I haven't tried anything creative like locking with REFL 165.  I also didn't lock with 11 or 55, since 33 just worked.

  10052   Tue Jun 17 19:39:29 2014 ranaUpdateComputer Scripts / Programsautolocker confusion

I renamed the Autolocker and described it in the elog from this weekend. To run it, you have to run it from the scripts/MC/ directory (as always). I restarted the autolocker on Ottavia with nohup.

> nohup ./AutoLockMC.csh &

To figure out how to get cron to run it, we will have to debug the difference between linux and solaris pgrep, so that's in progress.

  10051   Tue Jun 17 17:14:14 2014 ManasaUpdateLSCIFO alignment recovery

 1. Recovered MC alignment and locked it manually after the ottavia cron failed to help.

2. Measured the MC spots and could not get the MC spots better looking than this.

spot positions in mm (MC1,2,3 pit MC1,2,3 yaw):
[1.6609930611758554, -1.4655939584833202, 1.3133188677545831, -1.9483764375494121, -1.6539541350121174, -0.8862744551298497]

3. Realigned the beams to the MC WFS and enabled WFS servo.

MC Trans SUM is ~17000 counts and MC REFL is ~0.5 counts.

To-do list

MC spots

MC WFS

IOO QPDS center

BBPD char

Recover REFL 33

MC REFL

MC autolocker cron

  10050   Tue Jun 17 17:04:26 2014 ericqUpdateComputer Scripts / ProgramsFB troubles

Quote:

Also, the CDS FE status screen had red lights blinking as if it required an 'mxstream restart'. I did the same and it did not fix the problem. So I tried to restart fb using the usual 'telnet fb 8087'; but could not restart fb that way.

 FB is acting strange. When ssh-ing in, certain commands cause an inescapable hang, which can't be ctrl-c'd out of. Telling it to reboot does nothing. This kind of situation was seen by me before, when we were getting all the front ends back, I eventually hard rebooted it, hoping it was a one time thing. Guess it's not. 

Looking at the dmesg output, daqd seems to be segfault-ing all over the place. This may be related... Here are some examples:

451314.730502] daqd[17339]: segfault at 7ff589ae3b30 ip 00007ff589ae3b30 sp 00007ff49931dfb8 error 15 in libmyriexpress.so[7ff589ae3000+1000]

[530516.313238] daqd[18442] general protection ip:7f3f2ce73a6c sp:7f3e29949d50 error:0

[530516.313250] daqd[18420] general protection ip:7f3f2ce73a6c sp:7f3e2a19fd50 error:0 in libc-2.10.1.so[7f3f2ce3f000+14c000]

[530516.313262]  in libc-2.10.1.so[7f3f2ce3f000+14c000]

[530516.327083] daqd[18412]: segfault at 3b04c9cd0 ip 00007f3f2ce73a6c sp 00007f3e2a4a7d50 error 4 in libc-2.10.1.so[7f3f2ce3f000+14c000]

[537695.364481] daqd[18489]: segfault at 12dbbcae0 ip 00007fa35a3b8a0a sp 00007fa298381af0 error 6 in libmyriexpress.so[7fa35a399000+28000]

[577316.821618] daqd[18758]: segfault at 7f5c4d3e9b30 ip 00007f5c4d3e9b30 sp 00007f5b5cc23fb8 error 15 in libmyriexpress.so[7f5c4d3e9000+1000]

I'm not inclined to go reboot it right now, but not sure how to address these problems...

 

 

  10049   Tue Jun 17 16:52:40 2014 ericqUpdateComputer Scripts / Programsautolocker confusion

Quote:

the MC autolocker is NOT running alright.

I'm kind of confused by the current auto locker situation. Somebody renamed the script from autolockMCmain40m to AutoLockMC.csh and did not update the crontab with the new filename. 

Furthermore it doesn't seem like  scripto_cron,a script which keeps the auto locker alive, runs ok on ottavia. When I run this on the command line, it claims that the process is already running, and returns some bunk PID that doesn't correspond to any running process. The script has a line stating setenv OSTYPE solaris , so maybe there's something funky going on with it's pgrep-ing or other command parsing.   

Lastly, if I try running AutoLockMC.csh directly on ottavia, I get a bunch of complaints about pmath and pezcabit not being found. 

  10048   Tue Jun 17 12:04:40 2014 manasaUpdateComputer Scripts / ProgramsMC autolocker NOT running, FB needs some attention

MC autolocker and Ottavia

I assume that the MC was left with a fully functioning autolocker enabled and running on ottavia last night.

But as of this morning, the MC autolocker is NOT running alright. The MC was in an unlocked state and the autolocker has been doing nothing to the servo sliders.  I think this was the state of MC since last night as seen on the stripchart.

Since the autolocker has been left to run on ottavia, I tried to look at the cronlogs to see if it running the autolocker script at all. I looked at the list on ottavia and it has the MCautlocker on it cronjobs list and yet doing nothing.

Later, I did a softreboot on ottavia when I could not ssh into it from rossa or pianosa. ssh to ottavia now works just fine.

I am leaving Ottavia at this and returning to the more important job of fixing the MC. I locked the MC manually and am now working on the alignment.

 

Framebuilder

Also, the CDS FE status screen had red lights blinking as if it required an 'mxstream restart'. I did the same and it did not fix the problem. So I tried to restart fb using the usual 'telnet fb 8087'; but could not restart fb that way.

Attached: FE status screen

Attachment 1: FE_red.png
FE_red.png
  10047   Mon Jun 16 23:15:44 2014 JenneUpdateLSCIFO alignment recovery

I noticed today, and Rana said that he saw Saturday, that the MC refl value when the MC is unlocked is unusually high.  It typically goes to about 4.5 V, but now is going up to 6.5V.  Since the PMC output is the same as usual (max seen has been about 0.82 today), something must have happened between the PMC and the IMC. 

Late last week, EricG and Nichin were looking at things on the AS table.  Was anything touched on the AS table?  Was anything touched on the PSL table?  'Fess up please, so that we can pinpoint what the change was.

 

Also, this afternoon, I touched up the MC alignment a bit, although it still needs work (I've asked Manasa to look at it tomorrow).  Rana centered the WFS to my MC alignment (this will need to be redone after the MC is truely aligned), and we turned the WFS on.  I also locked both arms individually, and locked MICH and PRMI sideband.  The PRMI wasn't especially stable unless I turned on the POP ASC.  I assume (hope) that this is just because I was doing it during the day, and not because there is something actually different about the PRMI since the computer meltdown.

 

Rana and I also took some notes on things that need to be done, starting tomorrow (the first line and the yellow line are scribbles):

Note_061614_230723_0.jpg

  10046   Mon Jun 16 21:56:53 2014 JenneUpdateComputer Scripts / Programsfstab updates, MC autolocker running, FSS slow loop alive

[Rana, Jenne]

Mc autolocker:

MC autolocker is running.  The trouble was with caput and caget and cavput on Pianosa.  Rana has switched those lines in mcup and mcdown over to ezcaread and ezcawrites, and that seems to have fixed things. For the MC2tickleON and -OFF scripts, we left the caput and caget and cavput, and saw that they do run successfully on Ottavia.  (We tried testing again on Pianosa, and now it seems to be okay with cavput, but we promise it was hanging up on this earlier this evening.)  Anyhow, it's all a bit confusing, but it seems to be running fine now.

The autolocker is now running on Ottavia, and Rana is putting it into Ottavia's cronjob list, and commented it out on op340m.

 

fstab changes:

We have removed the option "all_squash" from /etc/exports on Chiara (both lines).  We then checked that the files have ownership "controls controls" on Chiara, Pianosa and Rossa.  Ottavia still has ownership of "nobody nogroup", so we still need to figure that out. 

 

FSS slow loop:

We confirmed that the slow loop is running.  Also, since caput and caget seem to take a while, and the real PID integral gain is the value that we set times a sampling rate, the effective gain had changed.  So, Rana compensated by changing C1:PSL-FSS_SLOWKI from 0.03 to 0.1. 

 

Other thoughts:

Do we have an autoburt saver on another computer, in addition to op340m?  (It's in the op340m crontab list)  We really only want one of these at a time.

 

  10045   Mon Jun 16 21:22:16 2014 JenneUpdateComputer Scripts / ProgramsRossa having a bad day

[Jenne, Rana]

Today, Rossa has been hanging at bootup.  You get the desktop, and most of the gui things, and can move the mouse pointer around, but clicking the mouse or using the keyboard have no effect.  Once you try clicking something, the mouse pointer turns into the spinning ball, and stays like that.

If, upon rebooting (soft rebooting from another machine, through an ssh session), you hold down the Shift key, you should get to a Grub menu.  If you arrow-key down and select the next-most-recent version (not the recovery mode, but just the regular earlier version), and press Enter, Rossa starts up nice and happily.

I am not sure how to make Rossa always boot into this version of things, or how to get rid of the newest version so that the version that works is the most recent, but I'm hoping one of my Linux buddies will help me out on this one.  I think (maybe) that I need to find out what package was recently updated and could have caused problems, and then revert that one package.  (I think I can look at tail /var/log/apt/history.log to tell me what has recently been updated).

  10043   Mon Jun 16 15:32:12 2014 ranaSummaryIOORingdown recap

 

To do this better:

1) Just step the analog input to the AOM (i.e. no RF switch) and measure the PMC trans output with a fast DC coupled PD. IMC should be unlocked and disable during this part. We want the PMC ringdown to be faster than 1 us.

2) Re-install a fast PD in the MC trans path without disrupting the existing MC trans QPD setup.

3) Measure IMC ringdown and fit the data to find the cavity losses.

4) Think about how to use the Isogai, et. al (2013) technique to better measure the losses, taking into account the mirror transmissions.

  10042   Mon Jun 16 12:58:50 2014 manasaSummaryIOORingdown recap

A recap of the ringdown measurements made at the 40m in mid 2012, the hardware that was installed for the same and results from the measurements.

IMC ringdown:

Hardware installed 
A trans PD was installed (elog 7122).
This PD does not exist in the trans path anymore.

Measurements
The polarity in the MC servo was flipped with the MC WFS disabled and the PD trans signal was used to look at the cavity ringdown. 
Ringdown time = 13 us
Cavity finesse from the measurement = 453 (inconsistent with actual finesse). 

Attachment: Ringdown measurement and fits

PMC ringdown:

Hardware installed
The AOM was installed before the PMC. The AOM was driven by the driver installed on the PSL table. (RF power ~1.5W @ 80MHz @1.0V modulation input). An RF switch was installed to control the AOM driver input. ZASWA-2-50DR+ was installed.
Note: The AOM was used by the ISS crew after this. So the RF switch has been removed and the AOM is no more a part of the ringdown setup.

Measurements
No measurements were made as we could not obtain relevant TTL signals to control the switch remotely.

Attachment 1: Ringdown_815.pdf
Ringdown_815.pdf
Attachment 2: MCringdown_cum.pdf
MCringdown_cum.pdf
Attachment 3: cum_plot.png
cum_plot.png
  10041   Sun Jun 15 14:41:08 2014 JamieOmnistructureCDScdsutils re-installed

Quote:

Quote:

 CDSUTILS is also gone from the path on all the workstations, so we need Jamie to tell us by ELOG how to set it up, or else we have to use ezcaread / ezcawrite forever.

It's in the elog already: http://nodus.ligo.caltech.edu:8080/40m/9922

But it seems like things still haven't fully recovered, or have recovered to an old state?  Why is the cdsutils install I previously did in /ligo/apps now missing?  It seems like other directories are missing as well.

There's also a user:group issue with the /home/cds mounts.  Everything in those mount points is owned nobody:nogroup.

I also can't log into pianosa and rosalba.

 I also still think it's a bad idea for everything to be mounting /home/cds from an IP address.  Just make a new DNS entry for linux1 and leave everything as it was.

  10040   Sun Jun 15 14:26:30 2014 JamieOmnistructureCDScdsutils re-installed

Quote:

 CDSUTILS is also gone from the path on all the workstations, so we need Jamie to tell us by ELOG how to set it up, or else we have to use ezcaread / ezcawrite forever.

It's in the elog already: http://nodus.ligo.caltech.edu:8080/40m/9922

But it seems like things still haven't fully recovered, or have recovered to an old state?  Why is the cdsutils install I previously did in /ligo/apps now missing?  It seems like other directories are missing as well.

There's also a user:group issue with the /home/cds mounts.  Everything in those mount points is owned nobody:nogroup.

I also can't log into pianosa and rosalba.

  10039   Sat Jun 14 18:43:15 2014 ranaUpdateIOOMC Autolocker upgrades

 Somehow the caget/caput commands are really slow. I'm not sure if this is new behavior or not, but after changing values, it takes ~1-2 seconds to move on to the next command.

On op340m, once I ran autolockMCmain40m in the foreground, I could see that there were a lot of error messages that weren't being caught in the log file. On Solaris, the TDS and EZCA binaries work well, but the caget/pu stuff is missing a lot of the new flags and the new CDSUTILS python scripts are not there. So I decided to stop running on there and move over to running the MC autolocker on megatron. I also changes the mcup and mcdown scripts to BASH from CSH. There are still some PERL commands in there from Rob/Matt/Sam, so I also added the proper commands so that the pick up the scripts/general/perlmodules/ directory.

Yes, yes, I know. You would all feel warm in your tummy if we did everything in python because its the best language ever, but how about we lock the interferometer first and then we can rewrite all the last decades worth of scripts?

Looked at the trend from the last time the MCWFS were one (2 weeks ago!) and aligned the MC suspension back to that point. After that the WFS come on fine and MC looks good. I fixed the nameserver/DNS/fstab stuff on megatron, asia, and belladonna.

I've left AutoLockMC.csh (too painful to convert to BASH) running on megatron via an open terminal on pianosa. If it looks OK for the weekend, Q should move the crontab line from op340m over to megatron and then update the Wiki appropriately.

CDSUTILS is also gone from the path on all the workstations, so we need Jamie to tell us by ELOG how to set it up, or else we have to use ezcaread / ezcawrite forever.

  10038   Fri Jun 13 19:09:44 2014 KojiUpdateIOOA blown fuse found on the euro card crate at 1X2 (IOO) rack.

[Rana Zach Koji]

We tracked down the MC locking issue to be associated with the power supply problem.
Replacing a fuse which had incomplete connection with the new one, the MC started locking.

We still have the MC autolocker not running correctly. This is solely a software problem.


We went down to the IOO electronics rack to investigate the electronics there. After spending
some time to poking around the test points of the MC servo board, we noticed that the -24V
power indicator on the MC demodulator module was not lit. In fact, Steve mentioned on Wednesday
that the -24V Sorensen supply had lower current than nominal. This actually was a good catch
but should have been written in the ELOG!!!

We traced the power supply wires for the crate and found one of the three -24V supply has no
voltage on it. Inspection of the corresponding fuse revealed that it had a peculiar failure mode.
The blown LED was not lit. The connection was not reliable and the -24V power supply was flickering.

We then replaced the fuse.This simply solved all of the issues on the MC servo board. The electronics
should be throughly inspected if it still has the nominal performance or not, as the boards were exposed
to the single supply more than a week. But we decided to try locking ability first of all.

Yes, we now can lock the MC as usual.

Now the newly revealed issue is MC autolocker. It was running on op340m but op340m does not want to run it now.
It should be closely investigated.

Also turning on WFS unlocks the MC. Currently the WFS outputs are turned off.
We need usual align MC / check spot position / adjust WFS QPD spots combo.

  10037   Fri Jun 13 18:16:00 2014 NichinUpdateElectronicsChanges to the PD frequency response measurement system

 

As we had planned yesterday (ELOG 10034) I and Eric Gustafson wanted to manually measure the transimpedence for REFL33. But on closer inspection I found the RF signal cable coming from the Photodiode REF DET (mounted on the POY table), that we were supposed to plug into the network analyzer, did not have an SMA connector at the end. There was just the Teflon and metal part sticking out of the insulation. So we disconnected the cable labeled REF DET from the PD and pulled it out to fix it. (POY table and from near the 1Y1 rack)

 

 

Being unable to locate any SMA male connectors in the 40m lab [pasternack PE4025], we headed over to Downs where Rich Abbott did a quick and awesome job of soldering the SMA connector and also teaching me in the process. I will write an ELOG on how to do a clean solder of the SMA connectors to the RF cable shortly for future reference.

 

 

Coming back to the 40m we rerouted the REF DET cable from near the 1Y1 rack and into the POY table. This job was done mostly by Eric. We were also unable to locate a torque wrench to tighten the cable at the PD’s end and had to leave it finger tight. Eric is planning to buy a new torque wrench as we will need it often.

 

 

Also, I cross checked the SwithList located at /users/alex.cole/switchList with the RF switch connections at 1Y1 rack and turns out it is consistent, except that at CH2 of the first switch where MC REFL was to be connected, there is a unlabeled cable. It might belong to the correct PD, but must be made sure of. The rest of the channels that are not mentioned in the list were unconnected on the RF switch.

 

Now instead of disconnecting REFL 33 to make measurements with the NA, we had to take out AS55 from the RF switch, as the former was very hard to remove without the torque wrench. Then Eric removed the optical fiber which was illuminating the AS55 (AS table) from its mount to hook it up to the power meter. But then we were not sure of how to operate the Laser diode controller (LDC 3744C) and decided to leave stuff as it is and continue either tomorrow or on Monday. Right now we closed the optical fiber of AS55 with a cap and it remains unmounted. The RF cables of REF DET and AS55 were left hanging near the 1Y1 rack.

  10036   Fri Jun 13 11:33:55 2014 AkhilUpdateElectronicsCharacterization of UFC-6000 RF Frequency Counter

 Goal:

To characterize the Mini-Circuits RF FC (Model UFC-6000) and plot Bode Plots at varying Modulation frequencies.

Work Done:

Here are the list of measurements(files attached) taken from FC using SRS(Model DS345) Synthesized Function Generator for a Sinusoidal signal at different Modulation Frequencies ranging from 0.01 Hz to 1 KHz:

Carrier Frequency                          Modulation Depth                                                        Attached measurement Folder 

5 MHz                                                     Δ f = 5 MHz                                                                            Bode_5

10 MHz                                                   Δ f = 10 MHz                                                                          Bode_10

20 MHz                                                   Δ f = 20 MHz                                                                           Bode_20 

 

The measured data will be used to estimate:

1) Transfer Function of FC 

2) Quantization noise from Power Spectral Density(PSD) vs Hz

 

To Do Next:

1)Complete interfacing the Pi with EPICS.

2)Make bode plots (Matlab script attached) and PSD plots and estimate the control parameters for optimal design of FOLL PID loop.

 

 

Attachment 1: Bode_Plots.zip
  10035   Fri Jun 13 09:20:37 2014 SteveUpdateSUSrestored damping at PRM and ETMY

ETMX medm screen values are blank.

  10034   Thu Jun 12 16:56:31 2014 NichinUpdateElectronicsPD Inspection

I and Eric Gustafson inspected the automated PD frequency response measurement system which Alex Cole built last summer. We just lifted the tops off the tables [AS table, POY table and ITMX table] and looked at the alignment checking to see if the correct optical fibers from the fiber splitter were illuminating the correct photodiodes. We did not change anything at all and put the covers back on the tables.

The PDF attached shows the state of each PD fiber pair.  The fibers labeled REFL11 and REFL55 were reversed and illuminating the wrong photodiodes.

We will do a manual measurement of REFL33 tomorrow using the network analyzer and the modulatable laser but not the RF switch.  Afterward we will check to make sure the RF cables are connected to the correct channels of the RF switch according to the switch list (/users/alex.cole/switchList).

Attachment 1: Inspection_PD_Freq_Resonse_system_12th_June_2014.pdf
Inspection_PD_Freq_Resonse_system_12th_June_2014.pdf Inspection_PD_Freq_Resonse_system_12th_June_2014.pdf
  10033   Thu Jun 12 15:31:47 2014 JamieUpdateCDSNote on cables for talking to slow computers

Quote:

We have (now) in the lab 2 cables that are RJ45-DB9.  The gray one is LIGO-made, while the blue one is store-bought.  

The gray LIGO-made one works, but the blue store-bought one does not.  I checked their pinouts, and they are completely different.  On the sketch below, the pictures of the connectors is me looking at them face-on, with the cables going out the back of the page.  The DB9 is female. 

 There are RJ45-DB9 adapters in the big spinny rack next to the linux1 rack that are for this exact purpose.  Just use a stanard ethernet cable with them.

  10032   Thu Jun 12 12:30:50 2014 denFrogsGeneralWorld Cup Soccer 2014

Quote:

 

 world-cup-2014-mascots.jpg

  10031   Thu Jun 12 11:03:11 2014 KojiFrogsGeneralWorld Cup Soccer 2014

world_cup.jpg

  10030   Thu Jun 12 10:41:58 2014 SteveUpdatePSLPMC-T trend of 4 years

Quote:

Quote:

Also, while I was working on the PSL table, I heard noise that sounded like a bearing rolling around.  I suspected the HEPAs, since the one on the north east corner of the table has a problem when it's turned up high (we've known about this for a long time), however turning off the HEPAs didn't affect the noise.  The noise is strongest near the back of the PSL controller on the shelf above the table, and the PSL controller box is vibrating.  So, I suspect that the fan on the PSL controller box is about to give out.

EDIT:  To clarify, I mean the Innolight's controller.

 The bearing is chirping in the back of the 2W Innolight laser controller. It is loud enough to hear it. I placed 4 soft  rubber feet under the controller to avoid shaking other things on self.

The HEPA filter bearing becomes noisy at 50V

 Keep it at 20V for low noise

  The aging of the laser came up when the noisy bearing showed.  ~10% down in in 4 years. That is pretty good.

Attachment 1: 4yTrend2Winno.png
4yTrend2Winno.png
  10029   Wed Jun 11 20:09:37 2014 ZachConfigurationWikiDokuWiki situation

I have been looking into the DokuWiki situation, with no great progress so far.

The problem is with authentication. When you try to access the wiki, you get: "User authentication is temporarily unavailable. If this situation persists, please inform your Wiki Admin."

 As Evan points out, this goes away if you turn off ACL (Access Control Lists), but---as he also mentions---that just means that authentication is disabled, so the wiki becomes open. All signs point to this being a problem with the authentication mechanism. We use the 'authplain' plaintext method, where the usernames and hashed passwords are stored in a plaintext file.

Evan and I have both done plenty of testing with the config settings to see if the problem goes away. Even if you restore everything to default (but enable ACL using authplain and the existing user file), you still get an error.

I went as far as installing a brand new wiki on the web server, and, surprisingly, this also exhibits the problem. Interestingly, if I install an old enough version (2012-10-13), it works fine. After this revision, they transitioned their authentication methods from "backends" to "plugins", so this is a clue. As a side note, the other wikis on nodus (ATF and Cryo) are running earlier versions of DokuWiki, so they have no problems.

As it stood, our options were:

  1. No wiki (not even read-only, since the issue prevents logging in and also prevents anonymous reading, somehow)
  2. No ACL = open wiki. Also not good.

So, I created a temporary read-only version of the wiki using the aforementioned earlier DokuWiki release. It has a soft link to the real wiki data, but I deleted the user database so no one can log in and edit (I also disabled registration). It can be found at https://nodus.ligo.caltech.edu:30889/wiki_ro/.

I also created a backup of the real wiki as wiki_bak to avoid any potential disasters.

The simplest thing to do would be to just roll the wiki back to this working version, but that doesn't sound so smart. Clearly, it was working with the plugin structure before our snafu, and somehow our fixing everything else has broken it.

Quote:

Quote:

It looks like auth is broken on the AIC wiki (though working fine on ATF and Cryo). I did some poking around but can't see how anything we did could have broken it.

I went into local.php and changed $conf['useacl'] = 1; to $conf['useacl'] = 0; and it looks like the auth issue goes away (I've changed it back). This isn't a fix (we want to use access control), but it gives us a clue as to where the problem is.

 

  10028   Wed Jun 11 16:01:31 2014 SteveUpdateCDSc1Vac1 and c1vac2 rebooted

Quote:

 

 I have brought back c1auxex and c1auxey.  Hopefully this elog will have some more details to add to Rana's elog 10015, so that in the end, we have the whole process documented.

The old Dell computer was already in a Minicom session, so I didn't have to start that up - hopefully it's just as easy as opening the program.

I plugged the DB9-RJ45 cable into the top of the RJ45 jacks on the computers.  Since the aux end station computers hadn't had their bootChanges done yet, the prompt was "VxWorks Boot" (or something like that).  For a computer that was already configured, for example the psl machine, the prompt was "c1psl", the name of the machine.  So, the indication that work needs to be done is either you get the Boot prompt, or the computer starts to hang while it's trying to load the operating system (since it's not where the computer expects it to be).  If the computer is hanging, key the crate again to power cycle it.  When it gets to the countdown that says "press any key to enter manual boot" or something like that, push some key.  This will get you to the "VxWorks Boot" prompt. 

Once you have this prompt, press "?" to get the boot help menu.  Press "p" to print the current boot parameters (the same list of things that you see with the bootChange command when you telnet in).  Press "c" to go line-by-line through the parameters with the option to change parameters.  I discovered that you can just type what you want the parameter to be next to the old value, and that will change the value.  (ex.  "host name   : linux1   chiara"   will change the host name from the old value of linux1 to the new value that you just typed of chiara). 

After changing the appropriate parameters (as with all the other slow computers, just the [host name] and the [host inet] parameters needed changing), key the crate one more time and let it boot.  It should boot successfully, and when it has finished and given you the name for the prompt (ex. c1auxex), you can just pull out the RJ45 end of the cable from the computer, and move on to the next one.

 

 Koji, Jenne and Steve

 

Preparation to reboot:

1, closed VA6, V5 disconnected cable to valves ( closed all annuloses )

2, closed V1, disconnected it and stopped Maglev rotation

3, closed V4, disconnected its cable

   See Atm1,  This set up is insured us so there can not be any accidental valve switching to vent the vacuum envelope if reboot-caos strikes.[moving=disconnected]

4, RESET c1Vac1 and c1Vac2 one by one and together. They both went at once. We did NOT power recycled.

    Jenne entered the new "carma" words on  the old Dell laptop and checked the good answers. The reboot was done.

    Note: c1Vac1 green-RUN indicator LED is yellow. It is fine as yellow.

5, Checked and TOGGLED valve positions to be correct value ( We did not correct the the small turbo pumps monitor positions, but they  were alive )

6,  V4 was reconnected and opened. Maglev was started.

7,  V1 cable reconnected and opened at full rotation speed of 560 Hz

8,  V5 cable reconnected,  valve opened..............VA6 cable connected and opened........

9,   Vacuum Normal valve configuration was reached.

 

Attachment 1: beforeReboot.png
beforeReboot.png
  10027   Wed Jun 11 15:57:18 2014 JenneUpdateCDSNote on cables for talking to slow computers

We have (now) in the lab 2 cables that are RJ45-DB9.  The gray one is LIGO-made, while the blue one is store-bought.  

The gray LIGO-made one works, but the blue store-bought one does not.  I checked their pinouts, and they are completely different.  On the sketch below, the pictures of the connectors is me looking at them face-on, with the cables going out the back of the page.  The DB9 is female. 

06111401.PDF

ELOG V3.1.3-