Observing with IRCAM at Leuschner
Marshall Perrin, Adam Leroy, and Jason Wright
With the Leuschner 30" newly collimated, it was high time to get down
and dirty and do some observin'. And so Marshall, Adam, Jason and Don headed
out into the (luckily, clear) May night. Dick Treffers met us
at Leuschner to get us started, and managed to save the day with a manual
override when the dome refused to open. Due to some troubles with the cooling
system, the detector was pretty hot (~ 110 K) and thus the image quality
wasn't the best, but we did what we could.
First task was to check the pointing. This turned out to be pretty
darn good - everything we looked at showed up in the CCD field of
view basically on the first try. At first things were a bit off center,
probably due to the new collimation, but after some offseting and a TX ZERO LAST command,
we were all set. There were a few oddities - at one point it seemed as
though a star had shifted from one part of the field of view to another
during a series of ON/OFF pointings. However, we tested this again by writing
a script to slew ON/OFF by a degree a number of times, and the star ended
up in basically the same position every time, modulo seeing fluctuations.
The following two images show the difference between 5 seconds vs. 2 minutes
on some star in Bootes. Note both the blurring due to seeing, as well as
the chip flaws that become apparent: the crack between rows 128 and 129
where the individual component quarters abut, and the dark region in the
upper left. That region showed up on several of our images, though the
stretch here was used intentionally to bring it out. Both of these images
are POST sky frame subtraction, though no CR rejection was done.
We found the best focus for the instrument was around 590 mils, give
or take a couple. It was actually somewhat hard to quickly determine
the best focus, since the seeing conditions were changing pretty
rapidly. The best we saw was a FWHM of about 3.5, but more typically
it was around 4.5-5 arcsec, for 5 second integrations.
There is a visible astigmatism when not in focus; inside of focus you
get elliptical blobs at a 45 degree angle, and outside of focus you get
elliptical blobls at a 135 degree angle. (Or was it the other way
around?) Even at the optimum focus, things aren't *quite* perfect, but
rather a point source appears about 10% wider in one direction than
the other. Some of that might have been seeing, though; more careful
measurements are needed to pin this down.
We could easily split the double star Bootes Mu-2. This frame has had CRs
rejected as well as the sky subtracted, cleaning it up substantially. Probably
a fair number of these "CRs" are really hot pixels due to the high chip
temp, but the effect was the same so QZAP in IDL killed 'em off.
We next turned our sights on the globular cluster M3. We observed M3 in
both the J and K bands. The image here is a two-color composite, with 5s
K in red and 20s J in cyan. The K band image is noticably sharper than
the J, due most likely to the shorter exposure time, which was necessary
in order to prevent the detector from saturating at the higher count rates
for K band.
Our last and best target was the famous Whirlpool galaxy M51. Well, turns
out we missed it and only got its smaller companion galaxy NGC 5195. At
the time of observations, we thought we had shots of the cores of each
galaxy, but we were somewhat confused and upon later reflection it became
clear that we hadn't shifted the telescope nearly enough to get both cores.
This would be a fun observing project to do-over again right once IRCAM
is back up; the two galaxies are 270 arcsec apart, just barely within IRCAM's
280 arcsec field of view, so it ought to be possible to get both on one
frame. Now, without any further ado, here are our two images, which
were each built up by adding 7 20sec exposures less 7 20 second sky frames.
Both of these images have had CRs removed and are displayed with a highly
stretched linear scale; the reason it looks like there's still a fair amount
of garbage present is that the color table is scaled way down in order
to bring out the still-quite-faint features of the galaxy. We probably
would have done better with a good deal more exposure time. .
And, courtesy of Kitt Peak, I found some context for the above, proving
to my own satisfaction that we really were looking in the right place;
you already saw up at the top of this page an overlay of our observations
over the KPNO image, visible also below; Click here for
the original image. This image, which I found just by doing a google
search on M51, turned out to be far more useful than the amazingly poor
and low-signal images I was able to get from 2MASS... Note particularly
the two small clusters visible to the left of NGC5195, and also how well
the galaxy's shape in IR tracks the visible. Image scale is 1 arcsec/pixel,
10% smaller than the above. The dust lane linking the two galaxies is not
visible at all, which is somewhat surprising, but then again we're not
down in the thermal IR where the dust is presumably radiating.