RHIC polarimetry (p-Carbon)

From RHIC Spin Group
Jump to navigationJump to search

Last modified by Schmidke on 8-03-2022

 

Run History

Overview of p-Carbon polarimeters configuration: Run 09 Run 11 Run 12 Run 13 Run 15 Run 17 Run 22 Run 24
The most significant events during the run:   Run 11 Run 12 Run 13

Material concerning Run13 detector instabilities is here

 

Online Operations

The current online logbook for the p-Carbon polarimeters is available at:

There is a handful of useful links explaining how to operate the p-Carbon polarimeter:

 

Online machines

The online machines are yellowpc.rhic.bnl.gov and bluepc.rhic.bnl.gov. For proper operation the following programs and packages have to be installed on the online machines:

  1. camac driver
  2. CDEV libraries
  3. ROOT, CERNLIBs binaries and libraries
  4. mc, gv, DHCP, tftp

Below one can find the current disk configuration in the online machines.

 

Online tools

LogView is a powerful tool to display values of any "online variable" saved by the loggers. However, sometimes it is useful to dump the data in a text file. One can use the tool called exportLoggerData to get values of an online variable ("cell") for a given period. Here is an example:

exportLoggerData -logger RHIC/Polarimeter/Yellow/YelPolarData,RHIC/Polarimeter/Polarization \
-start 04/04/2011 -stop 04/07/2011 -cells "polarimeter.yel:runIdS,YelPolar" \
-excluderowswithholes -timetolerance 600000

Sometimes one need to match a LogView alias to an actual device name. This can be achieved with the following command:

WhatsBeingLogged -findrequest "PP_Voltage_Monitor_BarGraph" -history -listcells

More information on these useful tools is available at http://www.cadops.bnl.gov/Controls/doc/FillData/

 

CAMAC Test

 

Running a Test

Dima and Igor developed a program that tests memory in WFD modules in the CAMAC crates. A test is performed by writing random numbers into memory and reading them back. One can start a test on either of the online machines:

cd /usr/local/polarim/bin
./cmcmtest <module_id> <crate_id>

where <module_id> and <crate_id> are the module and crate id's

An example output without any errors should look like this:

 

Reinitializing Xilinx Chips

After powering CAMAC crates most of the modules are not initialized, so memory test will fail. To initialize modules 2, 4, 6, 8, 10, 12, 14, 16, 18, and 20 in crate 5 do:

./cmcmtest
Command>cr 5
Command>bur
BUR>a9 w8 n2
BUR>n4
BUR>n6
BUR>n8
BUR>n10
BUR>n12
BUR>n14
BUR>n16
BUR>n18
BUR>n20
BUR>exit
Command>q

To scan only 1Mb of memory one can issue the following command:

Command>top 0x100000
Command>run

 

Offline Operations

The offline analysis of the p-Carbon data is done with the cnipol package developed by the group over the years. Visit the offline analysis page to learn how to run the programs.

You can find the README on how to process the offline data at the following

https://github.com/rhicspin/cnipol/blob/master/README.md

 

Data Location

For polarimetry dedicated analysis one can use data stored on disks mounted to RCF machines. The data is located in the following directories:

rlnxsp03.rcf.bnl.gov:/data1/run11

Run 7, Run 8, and Run 9 data can be found in:

rlnxsp04.rcf.bnl.gov:/data1/bd1
rlnxsp04.rcf.bnl.gov:/data1/bd2
rlnxsp04.rcf.bnl.gov:/data1/bd3
rlnxsp04.rcf.bnl.gov:/data1/yd1
rlnxsp04.rcf.bnl.gov:/data1/yd2

Also on the online machines check out the following directories:

 

How to back up pC polarimeter data?

The data collected by the pC polarimeter is temporarily stored on the online DAQ machines yellowpc.rhic.bnl.gov and bluepc.rhic.bnl.gov. For the long term storage we utilize the High Performance Storage Service (HPSS) provided by the RHIC and Atlas Computing Facility (RACF). The HPSS provides an interface for transferring large files of at least few gigabytes to tape. An RACF account is required to follow the steps outlined below.

To transfer files to HPSS one needs to copy them first to any RCF machine (e.g. rftpexp.rhic.bnl.gov). Then from that machine, you may use the hsi commands. The following is a generic example that should work:

cd /dir/with/files/
nohup hsi -q -l your_login_name "mput * /dir/hpss/"

If your kerberos credentials on RCF have expired the system will ask you to enter your kerberos username and/or password. One can chose to use hsi interactively by issuing typical linux commands in the hsi shell:

cd /dir/with/files/
hsi
pwd
ls -la
cd /home/cnipol/cni_runXX
ls -la
mput * .

Normally, the archived data from different runs is located in /home/cnipol. The AGS data after 2006 is stored in /home/kinyip. Use the put command to transfer single files, mput for multiple files. The system will transfer the files to the tape according to their robot schedule. For mput to avoid being asked to transfer each file do prompt off.

Since the typical size of our data files is already hundreds of MB it is recommended to simply copy the individual files without adding them to a larger tar archive. In fact, it will make the life easier if you decide to get a specific single data file in the future. However, for log and hbook files it is advantageous to first tar them in a single file for efficient transfer.

 

Common Problems and Troubleshooting

 

No pulser signal in the data

  1. Check that the pulser signal is not blocked in the configuration file.
  2. Try to take data with the internal WFD clock. Use -C option for rhicpol.
  3. Check if the pulser signal really comes from the generator to V124. As we learned the hard way the signal can be off if the crate in rack R12A17 is turned off. This rack is located in the room next to the counting house at IP12.

 

The downstream PC inside the tunnel won't boot

First check if the DHCP server is running on yellowpc. To restart the daemon execute the following commands as root:

>/etc/init.d/httpd status
>/etc/init.d/httpd start

 

PolarControl Error: E-Can't access RHIC ADO

It looks like this error appears in PolarControl when the end of measurement is not properly reported by the cnipol daemon. Because of this the online analysis does not run and the plots are not produced. [1]

The current solution to this is to run the analysis manually from the online machine. Login as mcr to either online machine (yellowpc/bluepc) and execute the script:

acnlinc>ssh mcr@yellowpc
yellowpc>/usr/local/polarim/bin/analyze.sh <runId> <pol_name>

where <run_id> is the measurement run and <pol_name> is one of the following names: polarimeter.blu1, polarimeter.yel1, polarimeter.blu2, polarimeter.yel2

For example:

/usr/local/polarim/bin/analyze.sh 15172.007 polarimeter.blu1

 

No carbon data while adjusting timing parameters

  1. Make sure the pulser signal is visible in the window
  2. Check that the target is not broken
  3. Another useful thing is to increase the integration window so it spans over both alive and empty bunches (WinBegin=0, WinEnd=100 or more, BZDelay=0)

Last modified by Schmidke on 8-03-2022