Subject: APO 3.5m users committee minutes, 1/31/11

From: Suzanne Hawley

Submitted: Fri, 11 Feb 2011 11:04:57 -0800 (PST)

Message number: 1216 (previous: 1215, next: 1217 up: Index)

           APO 3.5-m Users Committee Phonecon, 1/31/11

Attending:  Suzanne Hawley, Mark Klaene, Michael Strauss, Bill Ketzeback, 
Russet McMillan, Remy Indebetouw, Sean Moran, Jon Holtzman, Rene Walterbos, 
Bruce Gillespie

Absent: Scott Anderson, John Bally, Al Harper


User feedback, comments from institutional representatives:

  o Chicago (Al Harper) - no report.

  o New Mexico State (Jon Holtzman) - Jon had nothing to report.

  o Washington (Suzanne Hawley) - Suzanne (for Scott) had nothing to report.

  o Colorado (John Bally) - no report.

  o Princeton (Michael Strauss) - Michael said the PU users are happy.

  o Johns Hopkins (Sean Moran) - Sean had nothing to report.

  o Virginia (Remy Indebetouw) - Remy said that his users are generally 
happy.  There has been an issue with synchronizing the guider images and 
spectra from Triplespec.  The names in the file headers seem to be off by 
one image.  Remy agreed to ask Matt Nelson at UVa about how the headers 
are written and report back.


Discussion of telescope/instruments report - Klaene:

Mark mentioned the forecast for the first major winter storm this winter, 
expected to hit us this week.  The matplotlib memory leak mentioned below 
has been mitigated, but won't be eliminated until the software vendor 
fixes their code.  The telescope has generally behaved well, but pointing 
errors are slowly creeping upward.  On the instruments, the condensation 
on the SPIcam chip has been removed by heating the detector and pumping -- 
we are running the chip heater whenever the instrument is not being used 
for observing.  Agile is experiencing spontaneous aborts which is due to 
the software drivers from Princeton Instruments.  TripleSpec has been 
occasionally giving zero-filled frames, and resetting the controller seems 
to fix it temporarily.  Low-level banding seen in echelle images was 
looked into by Dale Sandford from Yerkes, one of the developers.  No 
solution is yet obvious, but it seems the problem may be related to 
instrument and/or telescope grounding.  The Goddard Fabry-Perot made its 
final run prior to its upgrade that begins next week.  The echelle shutter 
is not functional, and is being investigated today.  Mark's detailed 
report for the reporting period follows:


           3.5-m Telescope, Instruments, and CIF Projects Highlights
                          12/2/10 through 1/26/11

0) Overview

Weather turned slightly worse this period with winter storms bringing 
clouds but very little precipitation.  Cloud cover kept us closed a few 
whole nights but many nights we were able to open for a few hours.  We 
also managed to get through the holidays with very few issues that 
required day staff to come in during their break.

We started filming for the new APO promotional video and "B-roll" footage 
in high definition. The primary effort was during the recent lunar eclipse 
when we acquired a couple very cool images of the APOLLO laser in 

1) Telescope

Virtually no telescope problems were reported during this period.  We are 
seeing an occasional aborting of some of the guiders (Na2 and TCam) that 
is being investigated.  The restart is quick and the new TUI guiding plots 
make noticing the aborts much easier.  New TUI 1.8.5 and 1.8.6b versions 
were issued.  Note - memory leak is still present but at a much reduced 
level.  Changes are:

 	- Updated for backward-incompatible StripChartWdg

 	- The Mac version now includes matplotlib 1.0.1 and unix users may 
wish to upgrade as well.  This offers two important benefits:  TUI will no 
longer exit at startup if matplotlib's font cache is outdated; and 
Matplotlib's memory leak in canvas.draw is reduced by a factor of 10.

 	- Added target axis position and UTC date to the Status window.

 	- Added Calibration offsets to the Nudger window.

 	- Renamed Object offsets to Object Arc offsets in the Offset, 
Nudger and Status windows and reorganized the associated help files to 
better consolidate information about types of offsets.

 	- Enclosure window:  added eyelids summary status (which tells you 
if the eyelid for the current port is open) and Open All and Close All 

        - fixed issue with unwanted ctrl-click in slitviewer windows.

2) Instruments

DIS has performed well this period with only minor issues with the 
slit-viewer shutter after its recent replacement.

SPIcam was discovered to have some condensation on the detector again. 
After last month's vacuum pumping we noticed that the CCD was running 10 
deg colder than normal when the condensation appeared.  We ran the chip 
heater to get to the nominal setpoint and the condensation then 
disappeared.  This allowed us to continue observations for the time being. 
The chip heater when running at greater than 50% does induce some 
artifacts so the heater is shutoff prior to observing.  We are 
investigating a current limit such that the heater could stay on 
continuously at a low level.  Agile spontaneous aborts are being 
investigated.  Despite repeated attempts, the camera vendor (Princeton 
Instruments) does not seem to be able to offer a solution for this in 
Linux.  We are investigating some software updates and a possible move to 
a Windows platform.  A decision will be made after this week's observing 
as to the seriousness of the problem.

TripleSpec is not reporting slit position in the headers.  This is on the 
list of things to fix.

Echelle continued showing some banding in bias images.  Dale Sandford from 
Yerkes was on site (week of Jan 24-28) troubleshooting this.  The problem 
does show up in the lab at about a 1/2 count level but not repeatedly, 
much like at the telescope.

NIC-FPS was returned to CU to address the readout problem when rotated. 
We also hope the new fanout board will be installed during this trip.

GFP had a design review for the new detector and the IFS and will be taken 
off line after their run next week to start the upgrade.  Initial 
commissioning using their control system is scheduled for March.


NIC-FPS update - Klaene:

The instrument is currently at Colorado, Mark said.  The good news is that 
the new fan-out board was installed and has cut the read noise by a factor 
of two, almost to the level we saw before the 16-channel read conversion 
was done several years ago.  We are still expecting the instrument to 
return to APO in time for the March bright run.  The problem with the 
instrument malfunctioning seen when rotating has yet to be diagnosed.


TUI for windows machines - Hawley:

Suzanne said that Mac OS10 v.10.4 will soon be incompatible with the 
current version of TUI.  There is also the question for TUI users who use 
Windows machines -- maintaining TUI for the older version of Mac OS and 
Windows is becoming very difficult, so we need to assess our need for 
supporting TUI for Windows platforms:

===> ACTION:  Users Committee members should poll their users to find out 
how many are using TUI with Windows machines and report at the next Users 
Committee meeting.


ReSTAR proposal - Holtzman, Hawley:

ReSTAR Phase 2 is starting, and Jon attended a kick-off meeting in Tucson 
in November.  The pre-proposal paper that we crafted in the past weeks 
will be submitted tomorrow.  This is to be construed as a "foot in the 
door", and enables NOAO to work with us to negotiate observing time for 
instrument(s) and/or cash, and other forms of short and long-term support 
and collaboration, if we choose to go this route.  One thing we need to 
address more fully is our prioritization of what we want our 
instrumentation suite to evolve into over the coming years.


Class visits - Klaene:

For everyone's information, there is a worsening housing shortage at 
NSO-Sunspot due to the many staff they are hiring for their ATST project. 
This means for example that some SDSS observers have been displaced and 
now have to be in the dorms while they are observing.  In mid-March, we 
are expecting a class from Colorado with up to 19 students and faculty; 
this will be hard to accommodate between the available housing at APO and 
Sunspot, and other liens in the period for APO housing.  For future 
reference, we believe that class visits of less than 10 people total would 
be the preferable group size, at least for the next few years.



[open from previous meeting] -- none

[new actions from this meeting]:

===> ACTION:  Users Committee members should poll their users to find out 
how many are using TUI with Windows machines and report at the next Users 
Committee meeting.


Next meeting:  The next Users Committee phonecon will be on Monday 28 
February at 8:30 AM Pacific Time.  The agenda and other materials will be 
sent to the committee members during the preceding week.

APO APO APO APO APO  Apache Point Observatory 3.5m  APO APO APO
APO  This is message 1216 in the apo35-general archive. You can find
APO  the archive on
APO  To join/leave the list, send mail to
APO  To post a message, mail it to