Lick Observatory 3m Prime-Focus Camera ``PFCam''

I am also the PI (although all the clever design work and fabrication is being done by the team of Matt Radovan, Barry Alcott, Jim Burrous and Terry Pfister) for a prime focus CCD camera to be used at the Shane 3m telescope at Lick Observatory on Mt. Hamilton south of San Jose. The camera is on-axis and sits above the field corrector originally designed and built for AMOS, the Lick Observatory multi-object fiber-based spectrograph. The PFCam field is 10 arcminutes on a side and the CCD is a thinned SITe 2048 x 2048 chip with 24 micron pixels. The initial filter complement is Johnson UBV, Kron-Cousins RI, Gunn gZ and Spinrad R. Here is a link to the Lick Shop PFCam page. The first PFCam engineering run occurred March 26/27 1997.

PFCam Status:

Nov 1997: The most recent run was four nights in Oct/Nov 1997. Science images were obtained as part of a brown dwarf search in nearby young clusters and for a program to determine the luminosity function for galaxies in nearby groups of galaxies. The camera performed well, but it was determined that the telescope/new PF corrector collimation was not very good (on-axis, in-focus images were comatic).

August 1998: The first science run had to be canceled due to major problem with the PFCam dewar. The inner can sprung a leak (this is the second time this has happened to PFCam and the third failure of this particular type of inner can). A redesign of the inner can is completed and the PFCam dewar is slated for being rebuilt and completed by September 1998.

October 1998: The PFCam dewar was rebuilt and looks good. We had a baptism under fire as the first run with the new dewar was a regularly scheduled team from UCB (Basri and Agulia). There were a few problems.

1. When the science filter wheel is running in the ``dual servo'' mode, there is a large pickup noise in the CCD. For now we have disabled the servo mode. Testing the filter-wheel drift while moving the telescope all over the sky showed the worst-case movement is <12 micron motion of the filter.

2. There were some communications problems between the data computers due to too many connections being established between Orthrus and Shane (this has been resolved).

3. There was a hang when the system was initialized with the tv guider translation stage in a limit. This bug has been corrected.

4. On the first night there appeared to be condensation on the dewar window. A dry nitrogen feed has been set up.

5. There are light leaks. For now, these will be minimized by using a dark cloth.

6. It is difficult to fill the dewar at MOS landing. A pressure fill system has been put together.

January 2000: PFCam is nominally if good working order with no items on the fix list. Recent changes are:

1. The dual-servo stabilization of the science filter wheel is now automatically disabled during readout eliminating the pickup noise problem.

2. A dark shroud was added which significantly reduces the stray light problems seen before August 1999.

3. Kirk Gilmore modified the readout software and reduced the full-chip, unbinned readout time to just below a minute.

PFCam Overview:

The current detector is a SITe 2048 x 2048 thinned CCD with 24micron = 0.296" pixels. The field is 9.8 x 9.8 arcminutes. The readout noise (measured on the telescope) is 6e-. There is a remotely operated 5-position filter wheel and double-slide guillotine shutter both in front of the CCD mounted in a dewar cooled with liquid nitrogen. Guiding is done off-axis with a Lick guide camera mounted on a table that allows scanning in one dimension for guide star aquisition.

As of Feb 1998, the filter complement is Johnson U, B, V, Kron-Cousins R, I and Gunn Z. A Gunn-g and ``Spinrad'' R will be available by Fall 1998. 3-inch square filters are required to image the full field.

Based on observations of standard stars on a moonless, clear night the following numbers were derived for a star with U=B=V=R=I=20 at airmass=1. These calculations assume the KPNO atmospheric extinction terms. These numbers scale with relative telescope aperture comparing with the KPNO 4m. We come out a little ahead in the U and B bands and a little behind in R and I.

Rough exposure times to reach mag = 22 (delta mag= 0.05m) in the five filters are: