We have found CIRIM to be very efficient. In the K band on the 1.5m telescope at f/8 we have measured a total throughput from the top of the atmosphere to detected electrons of 34%. Expected count rates and backgrounds are given in the following table for the broad band filters with the f/8 and f/30 secondaries on at the 1.5m.
We note that the backgrounds can vary from night to night and even within a given night. The K background will be a strong function of temperature since it is dominated by thermal emission. The 1.5m f/8 data were taken on a rather average Tololo night when the temperature was 8C. For this temperature and higher, the Ks filter offers superior performance to the K filter since they have fairly similar count rates but the Ks filter has about half the background. On a very cold night or when working with very red objects the K filter may offer better performance to Ks. The H, J and I band background are dominated by OH airglow which varies in a mostly random way. At H in particular the background can vary by factors of 2.
On the 1.5m at f/8 (1.2 
/pix) we detected Ks=16.3 point
source at 3 sigma in a one minute on/off difference image. In 10
minutes we could detect Ks=17.5 sources at 3 sigma. Note, these images
were seeing limited by the large pixels. Using a slower secondary
(i.e. small pixel scale) that more fully exploits the telescope seeing
should yield higher sensitivity for point sources. We note that the
f/30 throughputs are generally lower than the f/8 since the secondaries
are undersize. Also, the f/30 backgrounds are higher since the f/30
Lyot stops do not have central obstructions in them so we view
the backgrounds of their reflecting cones. In the future we hope to
replace the current f/30 Lyot stops with micro-machined ones with
central obstructions. Also, on the 1.5m while the f/30 secondary does
place much of the light in a seeing limited core there is a fairly extensive
halo which is probably set by the telescope optics.
We have measured photometric transforms from the CIRIM instrumental to the CIT system on the 1.5m at f/30. To measure the color terms we observed CIT standards and very red stars in the Coal Sack with J-K colors of up to 2.9 from Elias et al. (1983, AJ, 88, 1027) which were measured on the CIT system. The transform from instrumental to standard magnitudes are:
The measurements of these 12 standard stars had residuals of 0.005 magnitudes from the fit. The extinction terms are the same as those that have been measured with the IR photometers at CTIO for many years. The H and K bands have no detectable color terms and they should probably be set to zero. The J band does have a significant color term in the sense that the CIRIM J band is bluer than the CIT system. Though CIRIM uses an original CIT J filter, it is bluer since the original CIT dewar had a Si lens which defined the blue edge of the band pass.