The Arecibo Legacy Fast ALFA (ALFALFA) Survey

A2982 Observing Checklist (for ALFALFA SEARCH MODE LBW Observations)

Version date: Sep 2015

Checklist date:	Aug 21, 2015
CIMA version :	"The Stable version (Default) (Ver 3.3.13 2014-05-08 17:14)"

Before observing, be sure to study the background document. And, if you are not familiar with it, review the description of the WAPP SEARCH mode.

If the CIMA version/time stamp is not as appears above (one or the other has changed a few times in the last week), check the CIMA User's Log to see what has happened. Presumably any changes made would not apply to us; don't take the adjective "stable" too literally.

This is a short checklist of the things to do when observing with LBW as part of the ALFALFA followup observations.
Remote observers: see the LBW remote observing checklist (Note: instructions are almost the same as for A2010/ALFA except of course when it comes to the specifics).
It is always a good idea to check with recent observers about their experiences.

Important points about ALFALFA followup observing with LBW

Fall 2015! For the APPSS, we will be using the WAPP spectrometer in a staggered, overlapping frequency band mode to search for HI signals from galaxies whose redshifts we do not know. In this "WAPP-search" mode, each of the 4 WAPP boards is set to observe both polarizatiopns of a 25 MHz bandpass with each board centered on a different frequency (instead of the whole 100 MHz ALFALFA frequency range). Be sure to understand the ALFALFA team WAPP search spectrometer mode. In previous LBW observations we used the interim correlator at specified velocity; you can find the instructions for that mode here .
Fall 2015! Don't forget to RESTART THE WAPPS at startup, before you load the configuration file. For some of us, this is blast from the past: "When in doubt, restart the WAPPs!"
Fall 2015! Sometimes it happens that one of the WAPP boards is bad. Be sure you know what to do if that is the case; see the discussion below on what to do while observing.
In 2015, the configuation file will be the same every night; only the command file containing the observing sequence will change. See the A2010 schedule page.
We still use the a2010 user account (to login, etc), but the observing program name is different. In spring 2015, the observing program is called "a2982". You can navigate, as user a2010, to the observing file directory by typing "a2982obs" in any window.
We use LBW (the L-band wide receiver; not ALFA).
We use the idl_lbw software routines to monitor the data. There are different subpackages depending on whether you are looking at correlator (ICORR) data or WAPP search data. You'll have to follow the instructions; see below!
The observing sequence is executed using a command file. Once you set up the receiver and the correlator and start to run the command file, the observing sequence will run automatically, moving from source to source and running the ON-OFF sequence for each one smoothly (but again, you need to pay attention!).
The observing program for a2982 consists normally of a series of 5-minute ON-OFF "pairs", one for each galaxy followed automatically by a CalON/CalOFF sequence which takes 20 seconds; observing each target "pair" takes about 13 minutes plus slew time. Then it's time to look at the resulting spectrum to see if there is a galaxy lurking there! Note: Sometimes observations only last 3 minutes (if the source name does not start with "S".
During each observing block, the observing sequence will be executed by running a command file called "a2982_YYMMDD.cmd", found in the observing directory "/share/obs4/usr/a2982". We also use only one configuration file: a2982.conf.
Additionally, in the "/share/obs4/usr/a2982/jan15" directory, you will find two other files for your use:
- a2982logYYMMDD: a template log file for the night. Please fill in/correct as necessary. Be sure to look at some old log files for examples, e.g. one example
- a2982_YYMMDD_nominal.txt: an approximate schedule for the night. Notice that the times here are ONLY APPROXIMATE and that the given azimuths are for the source not the dome.
NOTE THIS!!!! If there is any chance of rain or fog, ask the operator to disable the tiedowns. In March 2013, bad baselines occurred when there was some rain and/or fog because one or more of the distomats was not recording. The symptom: bad baselines with 1 MHz periodicity. Solution: disable the tiedowns. If unsure, discuss with the operator, but we are thinking that the best solution is simply to disable them. Read Phil's explanation here.
Your job includes (a) watching that the telescope does what it is supposed to; (b) making sure the timing of the observations stays roughly on schedule; and (c) making a quick check of the result for each galaxy to see if there is a detection; hence you will have to keep track of what is going on!
(Left over from the ALFALFA survey days:) We do not run TOGS; there is no commensal observing, and so (we presume), no need for a cell phone -- which is an especially good thing if we are observing at Arecibo: IF YOU ARE IN ARECIBO, BE SURE ALL YOUR WIRELESS DEVICES ARE IN AIRPLANE MODE OR TURNED OFF ALTOGETHER!

Startup:

Ask the telescope operator to move the telescope to the Az,Za of the first source and select LBW as soon as the previous observation is finished. Note that the Az of the source is given in the a2982_YYMMDD_nominal.txt, but you need to add/subtract 180 degrees for the Az of the dome. (If you are observing remotely, you should have already done this when you called the operator 30 min in advance.)
If there is any chance of rain/fog, ask the telescope operator to disable the tiedowns. Read Phil's explanation here.
Log into OBSERVER2: dtusr (once the previous observer is done). Password is taped to computer (or get it from the operator in advance).
Start CIMA on-line GUI:
- Type cima --X in an available window. Select the desired version of CIMA; this is currently "The Stable version (Default) (Ver 3.3.13 2014-05-08 17:14")"
- Project Number: a2982 (note: even if you type "A", CIMA will change it to "a" when you click "Line")
- Observer: your name (for entry into operator's log book)
- select Line and Accept
Several other windows will also open: the CIMA log display window and the CIMA observation status window. Leave them open through the night.
The Receiver selection menu will open. Choose Receiver: L-Band Wide and Accept.
Make sure that L-wide shows up in the aostatus monitor. Leave the default options as is:
- Make quick Tsys measurement: No
- Rotate receiver into position: Yes
- Retrack source if selected: Yes
Main menu / Utilities / Utilities NEW SPRING 2015!!!!!!!!!!!!!
- Restart the WAPPs
- Close when output to log window complete.
In the CIMA main menu, under Utilities ->Load configuration (file)
Load the appropriate a2982 configuration file: a2982.conf
This corresponds to the search mode in our terminology.

Then click Load and apply.
Be patient while the configuration file is applied....

If you cannot find the configuration file (or if only the a2010 files are present), then you entered the wrong project number into the CIMA opening window. You will have to exit CIMA and enter a2982
The CIMA observing menu should pop up; this particular window can also be found via the Observing modes->Spectral line observing button.
This should be set up automatically by the configuration file, so that these items should always be set as follows:

When you have checked things, you can Close this window.
Double check that the receiver IF/LO frequencies have been properly set in the Set-up->Receiver IF/LO selection menu.
This should be set up automatically by the configuration file but you need to know what correlator board configuration you are using, so it's a good idea to check.
Currently, we are using one basic observing mode (in the past, we sometimes used a second one, but not now).
Note that because some of the WAPPS boards may be bad, the rest frequency of each board may be slightly different than the picture below.
However, there should be one board at each of the four frequencies below (at approximately 1415, 1395, 1375, and 1355 MHz).
If you determine a different board is bad while observing, you'll need to return here to make changes!

When you have checked things, you can Dismiss this window.
- If anything needs to be changed, make sure you have the correct configuration file.
- If this does not help, change the settings manually, then click Apply followed by Accept.
If necessary, double check the Backend control menu
This should be set up automatically by the configuration file, so that these items should always be set as follows:

When you have checked things, you can Dismiss this window.

Double check that the right catalog is loaded in the Set-up->Pointing control menu
This should be set up automatically by the configuration file, so that these items should always be set as follows:

Source: variable (The command file will set the source.)
Catalog: a2982.cat
Epoch: Equatorial - J2000
Velocity: depends on the source
Frame: Heliocentric [Note: At startup this may be initially set to Topocentric. After running command file, recheck and confirm that it has switched to Heliocentric.]
Type: velocity(optical)
Offset info: Hide all
System: Main
Az range: Use closest in 0-720 degrees
Track tol: 10 arcseconds

The Source entry is irrelevant; it will likely be set to the last source from the previous observation.
When you have checked things, you can Dismiss this window.

And, now, you are READY TO GO!

Click on Command file observing
Click on Load
Highlight the correct command file: a2982_YYMMDD.cmd and click Accept

Click on Run and cross your fingers....

Note: If you need to skip sources that are at the beginning of the command file, Load the command file as usual. After it loads, it waits for 10 seconds (the WAIT 10 line in the command file). As soon as it starts to wait the 10 seconds, hit Stop. Quickly identify the proper line in the file to pick up the source at which you want to start observing (e.g. a "GOTO" line). Click on it or enter the line number to the left and then hit Run. Skipping or repeating sources is definitely something you want to practice using CIMA offline before your first night of observing. See also instructions below under What if the timing gets out of whack.

While Observing:

As soon as you start observe, run some immediate quick checks.
- Check that the telescope is actually at the correct declination
  1. Check the declination displayed in the CIMA observation status window Always check this!
  2: Check the "Rslt" line in the monpnt window (the upper right window on the screen below the frequency band monitor). Dec value should stay within a few arcseconds of the drift value. Always check this!
- Check the receiver is centered at the right frequency via the "Observer's On-line IF/LO Graphical Display" (the "rxview" command if you are running remotely). NEED TO ADD SCREENSHOTS FOR a2982 HERE!
  First line: L-band wide center frequency = 1420.0 MHz (this may change a very small amount)
  2nd line: Receiver filters: 1320 hipass should show 1320 to 1788 MHz
  3rd line: IF path: 10GHz to 1500 The whole thing will be green
  Bottom table: IF band setup
  
  Note: the center frequency of the band in "rxview" display will change very slightly during observing due to heliocentric correction.
  If something went wrong and you had to re-apply the configuration, this display may look wrong, with 100 MHz bandwidth windows, like this.
  Do not panic! It is just a lag in the rxview window and will correct itself once you begin integrating on the first source.
Keep the log file - it is very important to keep an accurate log of what happens, even to note "an uneventful night". Note: log files will be generated in advance for each night's observing. Use the one appropriate for your run.
- To keep the log, ssh to another machine as a2010 and then type a2982obs to get to the a2982 observing directory.
- Find the log file a2982logYYMMDD in /share/obs4/usr/a2982/MONTH15 and edit it using you favorite text editor (vi or pico work well when remote observing; emacs is too slow, unless you use the '-nw' flag to open in a terminal.)
- Check that the setup is what is indicated in the log file, i.e. the CIMA version number and the configuration file.
- Keep track of the approximate LST start, the file number and the ON-source scan number for each pair.
- Check and note any rfi in the observing band. Be especially vigilant for new sources of RFI, and know the expected locations in the band of the "usual suspects". The FAA radar is usually at 1350 MHz (although they sometimes use 1330 MHz, out of our bandpass!). Strong RFI at 1367 and 1382 MHz is due to the Punta Salinas naval base radar, which should not be on during ALFALFA run (see this link.) . If you see this, ask the operator to call Punta Salinas immediately and tell them to turn off their equipment! See observational troubleshooting list for other observing issues.)
- After each pair, take a quick look at the resultant ON-OFF spectrum (see IDL instructions following).
Keep track of the timing (but remember that the nominal timing we predict is only approximate so don't expect it to be exact!)
In the directory with the log file (/jun4 in this case), there is a file called a2982_YYMMDD_nominal.txt which contains an estimated schedule of tonight's observations. It is not precise!. For example it assumes a start time and a slew from some arbitrary position; the run may start earlier or later than that. It contains estimated slew times and scan tracking times and hence the Az and ZA of the source at the expected time of observation. Note that the Az here is the Az of the source, not the Az of the dome. The ZA of the source and dome should be close, but the Az will be off by about 180 degrees (or some multiple of that). Use this file as a guide. If you find that the actual observing times are falling behind, you may need to skip a source. Or if you are ahead or behind and the ZA is approaching 18, you may need to skip or repeat a source. Also, in general, there may be extra sources at the end. If you seem to have a gap, it is also ok to do a source twice.
What if the timing gets out of whack (e.g. you start late, something crashes etc)???
If you want to repeat or skip a source, you have two choices. You can either Stop or Abort. Stop will finish the current observation and then halt. Abort will immediately halt the observation. Once the process completes, click on the GOTO line of the source you want (or enter the line number on the left side where it says "Start from") and hit Run. Alternatively after an Abort you may reload the commmand file, make sure First Line is selected, and hit Run. After it sets things up, it waits for 10 seconds (the WAIT 10 line in the command file). As soon as it starts to wait the 10 seconds, hit Stop. Quickly identify the proper line in the file to pick up the source at which you want to start observing (e.g. a "GOTO" line). Click on it or enter the line number to left and hit Run. Skipping or repeating sources is definitely something you want to practice using CIMA offline before your first night of observing.
Determine if the current board setup is OK
It is quite possible that some of the WAPPS boards may fail, leaving less than all of them working correctly. You should be able to determine whether this is the case after looking at the first source with lbwquicklook (see below) regardless of whether it was detected or not. Note that, unlike previous LBW observations with the interim correlator, you do not need to wait until the next source begins before running lbwquicklook! In fact it is very important that you don't wait, so you can fix the boards before starting another source.
- This lbwquicklook screenshot shows a bad board 3. The order of the boards in lbwquicklook is always the same, regardless of the specific frequencies they are set to.
- If one board is bad, but is already assigned to 1355 MHz, do nothing.
- If one board is bad, but not assigned to 1355MHz, do the following:
  - Press Stop to pause observing before the second source starts.
  - Open Set-up->Receiver IF/LO selection menu and rearrange the rest frequencies so that the bad board is at 1355.405751 MHz.
  - Press Apply in the IF/LO selection menu, and wait for things to set up.
  - Continue your observing by finding the next source in the command file, and pressing Run.
- If two boards are bad, do the above, but assign one to 1355 MHz, and the other to 1415 MHz.
Check the spectrum produced by each ON-OFF pair
Part of your job is to make a quick review of whether the source was detected, at what velocity and also to note the presence of any RFI, bad baselines etc.
Open an IDL session on fusion00 (or aolc1, ...) and run @wasinit to compile the WAPPS utilities, and @lbwinit to compile the LBW observation checking program.
This observing mode takes four times as much data as the previous LBW follow-up, so these files take some time to load!
The WAPPS files are written to /share/pserverf.sda3/wappdata/wapp.YYYYmmDD.a2982.NNNN.fits
- To check a particular corfile
  lbwquicklook, file='/share/pserverf.sda3/wappdata/wapp.YYYYmmDD.a2982.NNNN.fits' where NNNN is the number of the FITS file (starting from 0000)
- To show spectral data from just one of the four boards
  lbwquicklook, file='/share/pserverf.sda3/wappdata/wapp.YYYYmmDD.a2982.NNNN.fits', board=2
- To change the number of channels used in smoothing to X (default is 3)
  lbwquicklook, file='/share/pserverf.sda3/wappdata/wapp.YYYYmmDD.a2982.NNNN.fits', smo=X
- To display the velocity scale in stead of the frequency scale in X
  lbwquicklook, file='/share/pserverf.sda3/wappdata/wapp.YYYYmmDD.a2982.NNNN.fits', /vel
- To change the x scale
  lbwquicklook, file='/share/pserverf.sda3/wappdata/wapp.YYYYmmDD.a2982.NNNN.fits', xrange=[1420,1421]
- To change the y scale
  lbwquicklook, file='/share/pserverf.sda3/wappdata/wapp.YYYYmmDD.a2982.NNNN.fits', yrange=[-30,50]
In general, we strongly suggest that you check each board individually, using the board=N option. If you suspect RFI, display the frequency scale; RFI often occurs at specific frequencies. Please try to be discriminating. And remember that continuum sources will cause a DC offset in the temperature/flux scale (either positive or negative depending on whether the source is in the ON or the OFF scan) and possibly can cause standing waves. If you see standing waves when there is not a continuum source or strong GPS, check the situation with dizzle/the tie downs (are they disabled? Consult the operator).
For further information on reducing LBW data, refer to the UAT LBW reduction instructions page.
Check here for a more thorough discussion of what is done in this script. Be sure to record in the log file all useful information about the resultant spectrum, e.g. presence of any RFI, bad baselines etc and whether (or not) the source is detected; if it is, record the approximate central channel and frequency AND velocity of the HI line. Also, make a judgement call as to whether the signal to noise is good enough or whether the source needs to be reobserved. If you want to, follow the instructions at the UAT LBW reduction instructions page to measure the RMS, HI line flux density, velocity and W50. If possible, include an estimate of the signal-to-noise ratio (just compare the peak HI line emission to the rms noise in the baseline region; a guess-timate is all that's needed!). The more info you put in the log, the better for our planning for the next observing session!

Finishing up:

If the command file finishes, then you are done. Usually we include a couple of extra sources at the end, to be done if there is time. If the time allocated for a2982 is up in less than 10 minutes, you should abort the command file.
Exit CIMA normally
Exit the observer2 machine
Exit data monitoring programs "nicely"
Exit IDL (if you were using it)
Exit out of the dataview machine
If remote observing, follow the procedures to close down the VNC connections when you are ready to do that..

After observing...

Here, we're assuming the date is 11 Jan 2015. Change as necessary.

Open a terminal on an Arecibo machine (i.e. remote.naic.edu):

Copy/move the command file and the log files (CIMA log and observer log) to the log archive and the web area:
Adjust this as appropriate for your observing date
a2982obs (If this alias fails, then cd /share/obs4/usr/a2982/)
mv a2982.cimalog_20150111 jan15/a2982.cimalog.150111 notice the name format, please; adjust date for today.
mv a2982_150111.cmd jan15
cd jan15
cp a2982.cimalog.150111 /share/a2010/obslogs/jan15/ (Note that the directory is a2010, not a2982!)
cp a2982log150111 /share/a2010/obslogs/jan15/
Determine which FITS are good, that is, which correspond to completed observations. In the same directory as above, make a list of them and write it to a file like this, named "filesYYYYMMDD.txt". The list should include all good files for the run, both detections and non-detections; it is used to determine which raw data files should be copied over to their final resting place at Cornell. To make the file, you can issue commands similar to these:
ls /share/pserverf.sda3/wappdata/wapp.20150111.a2982.*.fits > files20150111.txt
If the run crosses midnight, then you'll need to append the 2nd set of files (with the morning date) to that first file:
ls /share/pserverf.sda3/wappdata/wapp.20150112.a2982.*.fits >> files20150111.txt

You will only need the second line if you crossed midnight. Then edit the file to remove any bad files, e.g. if the observations of objects 1-10 were fine, but something failed during 11 and the run was restarted and 45 total objects were observed then it might contain only corfiles 1-10, 12-45. Be sure you get the date right; it may change during the night. Once this file is ready, copy it to the web area:
cp files20150111.txt /share/a2010/fitsfiles/jan15/
Update the log links file:
aweb (This alias points to /home/a2010/public_html/)
Open loglinks.htm and update the references to the log, CIMAlog and the list of corfiles that you made in the previous step.
Go to the loglinks page and check that your update worked.

This page maintained by the members of the Cornell ExtraGalactic Group and their friends, the other members of the ALFALFA observing team
Last modified: Fri Aug 21 12:43:55 EDT 2015 by martha