Summary of DSRI JEM-X Meeting, Thurday 19th march 2004
	
	
Those present: Niels Lund, Jerome Chenevez, Carl Budtz-Jørgensen, Carol Anne
	Oxborrow, Ib Lundgaard Rasmussen, Soeren Brandt, Niels Joergen
	Westergaard

-This was an ad hoc meeting called by SB to discuss one aspect of why our
    spectral flux determinations are consistently too low, and why spectra below
    7 KeV are generally unreliable
- SB showed that folded into our data is the effect of the on-board  rejection
	criteria.
- Currently our corrected spectra are conconstructed thus:
	S(PI) = INTEGRAL{ RDM(E,PI). ARF(E). f(E).dE}
	
	where f(E) is the actual spectrum of the source we want to observe.
	RDM(E,PI) is the ReDistribution Matrix that makes the PHA->PI conversion
	ARF(E) is the Response Matrix that gives the instrument sensitivity at
	each energy
	
- However, on-board there are rejections that are performed according to the
	PHA (or even pre-PHA signal) value of the event. On-board this
	appears as a nice smooth cutoff that's easily descernible as such.
	However, what actual energy this appears as depends on the gain of the
	detector at the time of the individual event. Hence the `clear'
	cutoff becomes a smeared out artifact of indeterminate shape.
	The attrition of events is therefore both energy and gain dependent
- NL suggests that spectra should be determined using both the PHA and PI values
	of each event, rather than just the PI values as is currently the
	case. CAO thinks this is a good suggestion since both these quantities
	are readily available in the ISDC science window groups, and would not
	require new output from the j_cor_gain, or j_calib_gain_fitting
	routines, the latter of which would be hard to accomodate in the
	current ISDC architecture.
- The status of instrument calibration in general was discussed, though with no
	specific outcome. Rudin and co. are clearly worried about the lack of
	good calibration for JEM-X
- NJW is currently writing a document to discuss calibration in general and it
	will be available from this page soon.


   CAO 22/3/04