IRS Throughput Data

The data on throughput provided here was taken on 21 Dec 96. Since
shortly before that date the spectrometer has been using a new 256X256
InSb detector.

The data were taken with a 2mm slit to avoid slit-edge losses. The data
presented here are extracted by summing all of the pixels in the
cross-dispersion direction with significant signal.  (Typically,
depending on seeing, the light will be distributed over ~3+ pixels).

The graphs here are normalized to the case of a 7th magnitude star (in
each band) and divided by the integration time and spectral bandwidth
per pixel to provide a measure of throughput which is independent of
dispersion.

The bias of the detector was in all cases 0.55 volts, (the recommended
value for all conditions except for very high background).

A summary of all the data is useful and compact:
winsum.ps. A clean laser print of that postscript file is
mostly legible. The illegible exponent which occurs many times in 
the graph is 6!

The individual graphs are also available separately;

210 line grating (4.2 micron blaze):
K band (2nd order)
H band (3rd order)
J band (3rd order)
IJ band (4th order)

75 line grating (4.5 micron blaze):
K band (2nd order)
H band (3rd order)
J band (4th order)
IJ band (4th order)
IJ band (5th order)