Gain Notes for Revolution 239
	
JEM-X1: This is a Crab calibration revolution. It includes an exercise in
running at a decreased gain with a lower than optimum High Voltage value.
However, despite this jump, the gain smoothing is fine, and all data from this
revolution should be useable.
The first two data points in curve are the partial switch-on at the beginning
of the revolution where the unit is initially activated with HV set 3 steps
below its nominal value.

IMPORTANT NOTE: This Crab Calibration revolution had heavy grey filtering in the
first half of the revolution, that shows up as null values in the gain history
table. The version of j_cor_gain found in OSA 5.1 and later can smooth over 
these null values, but earlier versions (6.3 and earlier of j_cor_gain) could
not. It is therefore essential to use OSA 5.1 or later, and re-analyse the data
from the COR level onward. Previous versions of the software could not handle
the null values or the sudden drop in voltage/gain, and must not be used to
analyse this data. Results shown on this webpage were produced with j_cor_gain
7.4. The gain history table was reprocessed in April 2006 to make a fuller
table using a newer version of j_cali_gain_fitting, and it is necessary to
make sure that it is this reprocessed version (rev.002, NOT rev.001) that is
used for science analysis.

JEM-X2: The otherwise dormant unit was activated for the Crab calibration, and
we clearly see the strong gain increase associated with the rapid aging of the
unit after its long rest. During the rest period the ion migration in the
glass of the microstrip plate relaxes and the gain of the detector returns to
its launch value. However, as soon as the HV is switched on again the plate
undergoes in a half day, or so, the ion migration and gain increase that had
otherwise taken the whole mission to occur. Otherwise, this is a perfectly
ordinary revolution with fine gain fitting, and all data from this revolution
should be useable.

The beginning of the revolution where gain and SPAG are varying very rapidly
is now designated a Bad Time Interval by ISDC and science data from this time
period will be disregarded by the analysis software. The BTI is: 
IJD=1731.20 to 1731.70.

CAO 10/11/2006

JEM-X2 Xe line analysis added to archive during creation of new IMOD epochs and
reference channels (epochs 12 and 13) for this unit

CAO 19/10/2007