Subject: JEM-X1 status & calibration planning


Dear friends:

This is a brief report on the results from yesterdays software 
update and functional test of JEM-X1. The very short version is
that everything worked as expected and the instrument condition 
is almost nominal - and this has led us to propose a few changes 
to the plan for operations prior to the coming Crab calibration.


1) Software update and performance.

The software update - only on JEM-X1 - was executed flawlessly. 
The data from the few hours of JEM-X1 operations after the update 
shows that the correction to the on-board event selection 
algorithms works as expected, but the data are still too sparse 
to allow a quantitative evaluation of the performance improvement.

We note that one of the important updates was to set the on-board 
protection limit for HV-commanding to a more relevant level - since 
the large reduction of the operational voltage shortly after launch 
there has been too much slack in the on-board limits. This is now 
corrected.


2) JEM-X1 functional test

JEM-X1 is in fine condition. The detector gain is very close to the 
value seen during the Crab calibration of August 2003. We have not 
noted any degradation of the microstrip anodes. A "hot spot" - not 
very intense - is apparent in the data, but like seen on previous 
occasions this hotspot is associated with one of the "blacklisted" 
anodes.

The software update included an update to the definition of the 
calibration areas on the detector plane. Events from these areas 
are routed to the HK data. Since launch we have had an appreciable 
leak of calibration events into our science data stream from two 
of the four sources. This is now significantly reduced. We can 
see in the new data that a third calibration region also needs to
be updated, this can be done using an "integer parameter" command
 - a code patch is not required.


3) Adjustment of JEM-X1 nominal high voltage.

The gain of JEM-X1 is today only about 60% that of JEM-X2. 
With this gain the low energy response is significantly
affected, and the reliability of the on-board event selection
is marginal. The low gain of JEM-X1 is a consequence of an extra
reduction of the high voltage, specifically for this unit, by 
two voltage steps decided in December 2002, when the concerns 
about the anode erosion were very serious. The experience we now 
have from JEM-X2 shows that the anode erosion rate is rather 
slowing down than accellerating as time goes by. 

We therefore propose, prior to the Crab calibration, to raise the 
JEM-X1 HV level by one step, corresponding to a 20% gain increase. 
This will not bring JEM-X1 to the level of the JEM-X2 gain, but 
we anticipate that the gain will increase slowly over time as seen 
in JEM-X2. 

I will prepare the necessary change request for this HV change.


4) Postponement of the software update on JEM-X2

We proposed to execute a corresponding software update on JEM-X2 
just prior to the Crab calibration. With the knowledge that JEM-X1 
is in fine condition and ready for use we now propose to postpone 
the software update on JEM-X2 until after the Crab calibration. 

We think that a calibration of JEM-X2 in the condition in which 
it has been used during the last year is much more valuable than 
obtaining a calibration with a new software configuration - which 
is not going to be used for real observations during the coming 
6 months. We also note that the gain of JEM-X2 is now much higher 
than during the August calibration and we consider a calibration 
at this high-gain state to be very usefull in order to provide 
the best response matrices for the observations already done.

The software update for JEM-X2 can then take place at any convenient 
time after the Crab calibration - to be agreed with ISOC and MOC.


With best regards - and thanks to everybody in ISOC, MOC and ISDC
which helped us carry out this important test.

niels