Subject: JEMX Monday Meeting 2005-08-08

Present: CAO, SB, JC, NL, ILR, CBJ, NJW

JEMX status, SB: JMX1 active again in rev342 after prolonged period of too
    high particle background.
    A hot spot has appeared around position approx. (118,30). The average
    count rate from it is about 1.6 cts/s, but it is quite irregular in its
    intensity variation. The counts from it extends up to a PHA value of
    about 100. Reports from ISDC that QLA does not function well - images
    have strange features (REMARK: this problem seems to be fixed with
    OSA5 software that is used from now (Monday))
    
    There was a discussion if the on-board discriminator value could be
    used to reject such a hot spot, because its spectrum is very much
    inclined towards the small signals. It is a worry that such a hot
    spot becomes so intense that the telemetry is filled up with that
    kind of events. Currently that is not a problem.
    
    At the beginning of revolution 343 the DV was stepped one down and
    the new start-up procedure (begin right on the nominal voltage in
    stead of spending about half an hour 3 steps down and then move to
    nominal) is now the default. At reactivation during an orbit the
    old procedure is still used.
    
    JC: The overshoot around position 118 is noticed, but the anode
    at position 172 is now dead, the first one in 2005.
    
SVR, NL & NJW: We are late with the delivery. There are several problems:
    - The positions reported by j_ima_iros in SRCL-RES are worse than
      in OSA4.2. The reason for this will be analyzed by NL.
    - The spectra extracted with OSA5 for the Crab are not at all
      satisfactory. Is there a problem with the ARFs used ? This is
      analyzed by NJW
    - The new gain correction for the Xe peak places the peak at too
      low values rather than the opposite. This is investigated by CAO
      and SB.

An analysis by SB shows that the photon detection efficiency depends on
the size of the original signal i.e. before the application of the 'curtain'
correction. This happens because the discriminator cuts a fraction of
the events. At a given photon energy this fraction depends on the current
local gain and local corrections. NL has written a note about the implication
of this effect on the image making and source property extraction.

SDAST meeting: The next SDAST meeting could be end of September or early
October. Perhaps it should be combined with the consultation of
'spectrum extraction' experts.

/NJW
----------
Niels J. Westergaard
Danish National Space Center
Juliane Maries Vej 30, DK 2100 Copenhagen O
Phone: +45 35325705; FAX:  +45 35362475