PIs for A2174: Lewis Knee and James DiFrancesco. Project is commensal with GALFACTS. Here are observation instructions (includes vnc instructions for Linux and Mac computers) and how to prepare each day's run.
First day of A2174 observations. Joined by C. Salter and M. Lerner in the control room. GALSPECT had been running for TOGS so I got the TO to stop it around 0345 AST. Copied script 'freddie' from /x108 test area. Observing scripts are in /share/obs4/usr/pulsar/a2174.
Everything ran smoothly. SFS run began around 0246 LST. Basketweave began
on perpuls_06_00, meaning today is day 06.
Another very smooth run. C. Salter present from commensal group. Today was
day 07, basketweave began on perpuls_07_00 (first Lambda of day 07). We decided
that on Sunday (Sept 10) we would try a remote session.
K. Douglas did observations solo. I did a second SFS run at the start, with
winking=on, for Carl Heiles. This took about 5 extra minutes but still caught
the first Lambda, for day 18 this time (perpuls_18_00).
C. Salter and K. Douglas experimented with remote observing with vnc. Very
successful. Basketweave began on perpuls_00_01 (Lambda 1 of day 0) - missed
Lambda 0 of day 20 by
about 5 seconds. So only 4 Lambdas done today. So maybe day 0 will get redone
on October 16, if we get Lambda 0 for all other days. Too soon to tell...
Another flawless night. K. Douglas in the control room at AO. The script missed
day 05 by a few seconds, so it had to jump to day 8; perpuls_08_00 to be precise.
K. Douglas remote observing from Berkeley. VNC Viewer caused a few problems,
including some dropouts as well as very slow service to collaborators looking in
on the vnc session. Problems were sorted out over the course of the night. Today
we began on Lambda 0 of day 9. GALSPECT didn't close nicely at the end, so had to
kill gdiag processes. This may have corrupted fits file 0009 - no mh file was
written automatically...will look into possible recovery.
Another remote session by K. Douglas @ SSL. VNC sessions ran much more smoothly
this night, and we began on Lambda 0 of day 11. KD needs to write a document about
checking the spectra with GSR code.
L. Knee & J. Di Francesco did the remote observing from HIA in Victoria. No problems
reported, and basketweaving began on Lambda 1 of day 1. 2 more runs before a 2-week
hiatus from observing.
J. Di Francesco remote observed from HIA. Began on Lambda 1 of day 10. So far this
day is the best candidate for a redo. KD is making qckfiles for all days so far. An
example of how to do this (for tomorrow's run) is right here.
J. Di Francesco reports a smooth night of observations. Day 14 was done, and started on Lambda 0, getting all 5 Lambdas for the night. Now we don't observe again until October 5. Summarizing days done so far, in order: [6,7,18,0,8,9,11,1,10,14]
for future IDL purposes: dates=['20060910','20060918','day2','day3','day4','day5','20060907',$
'20060908','20060911','20060915','20060919','20060917','day12,'day13','20060920','day15','day16',$
'day17','20060909','day19','day20']
During the break the GALFACTS group discovered that the 100-MHz filter wasn't set up, so
they were recording the full 300 MHz bandwidth of ALFA, probably getting more RFI that they
wanted as a result. So the .gui file was changed to filter out L-band signals outside
the range 1390-1490 MHz. There was a much more significant and immediate issue that we
heard about on the 4th. ALFA needed to have a cryopump changed, so it could not be rotated
out of the 19-degree rotation angle it was set at (optimal for drift scans). We require it
to be at 0 degrees, so today's beam spacings are not very good. Lewis Knee took the
observations, getting data for day 12 - but we are NOT going to take day 12 out of our
scripts since we want to redo this day. Began on Lambda 00 too.
James did this day's observations. Beam spacings are back to normal, and day 02 was done, starting with Lambda 00. More problems with VNC are cropping up, but thankfully the observations themselves got done properly. James thinks it was an HIA-based connection problem.
Let's look at some diagnostic plots. First, a qckfile plot of
fits file 0005 from October 6. It shows a very typical 600-second HI spectrum diagnostic
for A2174. Small RFI spikes near 1418 MHz, and a fairly strong spike at 1420.0 MHz. The
strength of the 1420 spike varies, but it's almost always there. Next, a plot of the feed diagnostics. This shows all 7 beams, and it's evident that
the feeds are working well most of the time. A few extra diamonds for rx 4 on the second
night (Sept 8) points to a problem Jeff Dever has warned us about for pol B on day 53986.
Here is a sample radar plot that looks for periodic power signals
in the data. This one is the worst offender, and we see 12-second radar quite strongly on
Sept 9 and 10. Finally, look at some receiver plots. They
all look about the same, but rx's 6-9 look a little different. These are beams 3 & 4, both
polarizations. They like like the others except maybe a vertical scaling?
Lewis had a good night of observations, getting all 5 Lambdas for day 03. No problems
reported, and the qckfiles I made look good too.
Hector has notified us that our make-up day for October 5 (ALFA stuck at 19 degrees) will
be November 5. Great!
James completed day 04 of observations, beginning on Lambda 00 once again. With the
15-minute earlier start than the night before, it took a while for the basketweave to start.
This is typical as we get to the end of a basketweave run, since the first few scans are
done so the start LST time drifts later as we get the early scan numbers.
Lewis had a good night of observing, hitting day 5 and getting all 5 Lambdas too.
Here is the sky coverage to this point. Looking good!
Lewis reports another smooth night of observing. This time day 13 was done,
beginning on Lambda 00 again.
James completed day 15, all 5 Lambdas. Went a little overtime to do this, thanks to
Chris Salter for giving us some extra time.
Lewis completed day 16, again getting all 5 Lambdas. Only three nights to go before we're
done our original allotment. For the Nov. 5 make-up day it looks like day 10 is still the
best candidate for a redo.
James took care of day 20 this morning. A few strange displays on the Snap windows
that monitor the continuum data, but this was seen only at the end of the run. Could've
been the telescope was being prepared for the next experiment. The HI qckfiles continue
to look very normal.
Lewis was on deck, and he completed day 17. The aostatus command wasn't working unless
the whole path was specified. I noticed a few other things not working the way they usually
do on AO computers, but at least the data look fine again. Tomorrow is the last night of
the original time allotment. It WILL be day 19. Oh yes, it will.
James finished off day 19. All 5 Lambdas as usual, thanks to the generosity of the
following group allowing us a little more time. So we're essentially done, though I suspect
day 10 will be redone on November 5, so we can get its first Lambda and fill in that pesky
empty spot starting near RA=3h, DEC=25d.
James redid day 10, and got Lambdas 00-05. Observing for this project is done! The
data reduction is well under control. I can't wait to read the paper for this one!
