PDS_VERSION          = PDS3
RECORD_TYPE          = STREAM
OBJECT               = TEXT
PUBLICATION_DATE       = 1996-12-07
NOTE                   = "MGS RST Instrument Health Report #16"
END_OBJECT           = TEXT
END


Engineering Data
================

Engineering Data Statistical Summary from QUERY2PLOT:
First SCET: 1996-341T08:20:24.120
Last SCET:  1996-341T20:39:48.119
CHANNL    TITLE      EU      EU      EU      EU    DN  DN    DN     DN RECRDS
                    LOW    MEDIAN AVERAGE   HIGH  LOW MED  AVERAGE  HI
------ ---------- ------- ------- ------- ------- --- --- -------- --- ------
L-0200 USO_REG_V   4.0600  4.0800  4.0791  4.1000 203 204 203.9536 205 10429
L-0201 USO_OVEN_V  2.6000  2.6200  2.6214  2.6400 130 131 131.0710 132 10429
T-0315 USO_T      23.4724 24.3575 24.4216 25.2531 134 131 130.7883 128  2607

     Nominal values per recent experience.  Approximately 96 MB of
channelized engineering data was transferred.


Open Loop Data
==============

  Data Retransmission
  -------------------

     Data from USO Test #1 were retransmitted from SPC 10 early today
(01:00 to 05:00 PST).

     The GIF Record Creation Time for both these data and the data
transmitted in near real time are identical.  Which means there is NO way
to distinguish between the original packets and the new packets by looking
at the packet contents.  I had hoped that the GIF Record Creation Time
might have been related to the transmission activity.

     On the other hand, the retransmission provided us with 121 minutes of
data (at 5000 samples per second) WITH NO MISSING RECORDS.  This is a major
improvement over the earlier transmission from which we lost 96 records
over 122 minutes.  The difference of one minute in total data results from
the fact that the original delivery covered 04:59-07:01 whereas the
retransmission was for only 04:59-07:00.

     This retransmission test also suggests that the retransmission of
data from day 323 was not actually executed.  After the supposed
retransmission we found exactly the same records missing as in the original
delivery.  The new test shows that all missing records CAN be recovered by
retransmission.  It narrows the problem to the communications system and
reduces suspicion of the DSP-R.

     Quick-look processing of the retransmitted data shows less of the
ringing associated with lost records (see Figure 1 in report of 1996-11-30).

  Data Collection
  ---------------

     Data have been delivered from an open-loop test conducted while the
Ka-Band system was being exercised.  Parameters of interest are:

YYYY/DDD DSS        DSP        SAMP  BW   DSP  BIT  TRK  REC  RECS RECS FIRST
               START    STOP   RATE      MODE RESN MODE  LEN       MSSG  SNR
-------- --- -------- -------- ---- ---- ---- ---- ---- ---- ----- ---- -----
1996/341  15 19:47:00 19:39:51 5000 2000   2   12  1-wy 1666 33841   19 40.16


  Data Anomalies
  --------------

     A total of 33841 records were delivered from day 341; 19 records were
missing.  Based on the results of the DOY 332 retransmission, we will request
that these data also be retransmitted.  The link time has already been
scheduled by Trish Priest for next Tuesday.

     Other anomalies include features seen in other tests such as ringing
in the power versus time plot.  Spurs at approximately 665 Hz on either
side of the carrier, seen at DSS 45 and to a lesser extent at DSS 65, do
not seem to be present in data collected at DSS 15.  This needs to be
confirmed with more detailed analysis, however; the plot scales at the
"quick look" stage are not good for this type of diagnostics.

  Frequency Analyses
  ------------------

     We will re-do the frequency analysis for the USO Test, using the
retransmitted data.  We will also do a frequency analysis on the data
collected during the KaBLE exercise, with better results expected after
those records have been retransmitted 3 days from now.


Closed Loop Data
================

     No new data.