Gain Notes for Revolution 666
	
JEM-X1: 
This is the Crab calibration revolution, but other than this it is a
quite normal revolution with fine gain fitting. All the data from 
this revolution should be useable. The Xe analysis confirms that the energy
determination for this revolution is good, within about 2% of the ideal. 
However, as for all revolutions, the first few science windows should be
avoided in any analysis requiring very good gain determination. Since the
temperature variation of the gain has tripled since the beginning of the
mission, future offline calibrations, including temperature dependency, could
possibly benefit this revolution when they become available in the coming
year.


JEM-X2: 
This was the first revolution where the instrument was switched on since
the last Crab calibration revolution 605, 6 months ago. This means that the
first 20 SCWs were spent letting the instrument stabilize in its aged, charged
state. During this warmup period the surface charge and internal ion migration
in the microstrip plate arranges itself to reflect the combined revolutions of
use the instrument has experienced. This involves changes in the absolute
gain (sharp steady rise in calibration spectra positions), the spatial gain
variation (unmappable changes in the SPAG table) and changes in the gain of
the plate at the heavily irradiated calibration areas compared to the rest of
the plate (reference Xe channel). While these changes are occurring the
spectral calibration of the instrument is unreliable (see the strong changes
in the corrected Xe line position during the first part of the revolution).
For all these reasons, the first 20 or so SCWs should not be used for
energy-sensitive analyses. Also, due to some kind of ionic changes in the plate
the automatically generated ISDC gain history table cannot be used (average
Xe line position appears at 27.9 keV). Use the IC gain history table generated
offline and delivered to ISDC by DNSI. If you have a newly updated IC package
the OSA script will find the necessary file automatically. Otherwise, download
the file from the last column in this table and set the hidden OSA input
parameter gainHist to:

gainHist=<Path to downloaded file>/JMX2_gainHistoryIC_rev0666_003.fits[JMX2-GAIN-OCL,1]

When this corrected gain history table is used, the average Xe position for
the revolution is 29.64keV, which is the ideal value and energy corrections
are good within 2% for all science windows after the first 20.
Future improvements in the temperature dependence of the gain might also be
possible in the coming year, bringing the energy corrections to within 1% of
the ideal, so be sure to keep your IC files updated.

Even using the IC table, a Bad Time Interval has been declared to force the OSA
software to disregard the earliest part of the revolution.
BTI period is IJD = 3008.6 to 3008.9


CAO 11/4/2008


JEM-X2: A new IC table was created and delivered to ISDC that includes a small
temperature effect so that the Xe line position remains more stable during
the course of the revolution, and improving the gain correction for all except
the first 20 science windows to within
1-2% of the ideal value of 29.6 KeV. Be sure that you have the newest set of
IC tables from ISDC to benefit from this improvement.

CAO 22/1/2009