Minutes
                Einstein Observatory Users' Committee Meeting
                          September 19, 1980

Attendees: C. Canizares (MIT), P. Charles (GO), R. Giacconi (CFA) , D. Harris
(GO), J. Hutchings (GO), L. Kaluzienski (NASA), K. Long (CAL), F. Marshall
(GSFC), E. Schreier (CFA) , F. Seward (CFA/GO), H. Tananbaum (CFA)

Also Present: M. Elvis (CFA) , W. Forman (CFA) , P. Henry (CFA), T. Markert
(MIT)

I. Status of Observatory

Satellite operations proceeded smoothly until 27 August when a major failure
occurred. A commanding or on-board computer failure caused excess thruster
firings which, in turn, triggered the automatic shutdown of most experiment and
spacecraft systems. During the reactivation of these systems, two of the four
operating gyros did not turn on. The remaining two good gyros did turn on, as did
two noisy gyros which had previously been taken off line. Normal operation of
the Observatory is impossible with two gyros, and the satellite was placed in a
sun acquisition mode which uses about l~lbs. of propellant per day.

Since then, an operating plan has been defined, in which the fine sun sensor
would supply the equivalent of a third gyro rate during orbital day, and the
control system would rely more closely on the star trackers during orbital night.
The modifications to the on-board software are underway at TRW, and should be
ready for trial soon. Operationally, this mode implies viewing only those targets
which are unocculted during orbital night, and maneuvering only during orbital
day. This mode will necessarily involve some visibility restrictions, a decrease
in the number of targets which can be observed, and an increase in real time
activity.

The star trackers and detectors were activated and checked out to the extent
possible.  No anomalies were encountered.   One of the noisy gyros (#5) has,
since being turned on, been well behaved.

We are developing a safe way of operating using gyro 5 as long as it remains
well behaved. A method of protecting the gyros from being shut down was
developed, and attitude initialization using three gyros was attempted on
Saturday, 13 September. Although a solution was obtained after one day, maneuvers
proved unreliable, and the on-board attitude reference was unable to update
autonomously. The emergency shutdown had apparently changed not only the biases,
but the individual gyro scale factors. Despite these problems, the satellite
could now operate in either a controlled slow scan mode or in pointing, using
very little propellant. By 16 September, an attitude reference was uplinked which
allowed the on-board system to update autonomously. A controlled series of
maneuvers was undertaken on 17 September to allow calibration of the scale
factors.

As of 18 September, we have entered a mode of limited scientific data taking.
Several targets from the OCA have been uplinked to the satellite, based primarily
on momentum buildup considerations (RCS is disabled, momentum is currently
dumped manually, and gravity gradient torque must be minimized) and need for long
observing intervals (only real time maneuvering). Additional targets have been
made ready at the control center. This mode of operation is expected to continue
for at least several days.

We are currently working toward re-establishing normal operations as soon as
possible and lasting for as long as gyro 5 remains stable.. To this end, a
meeting with MSFC, GSFC, TRW, and SAO is scheduled for 23 September to discuss
normal operations, sun-sensor software operations, and further desirable software
modifications.

Since the failure on 27 August, we have expended some 20 lbs. of RCS propellant.
At our best normal usage, this would correspond to about 15 weeks. We have 104
lbs. left and are optimistic that reasonable scientific operations will continue
for over a year.

A greater ground-support effort will be needed in future operations. More time
will be needed in observatory command and control. In particular, more money will
be needed for operational problems and, if additional funding cannot be managed,
this might have to come from data reduction and analysis.

As soon as we have definite information concerning a new operating technique,
target restrictions will be communicated to Guest Observers. We anticipate that
pairing will be difficult and that bad latitude targets, 20° < |delta| < 60°,
will be unlikely to be observed.

II. CFA Computer Problems

Significant problems have developed with the Data General computers at CFA.

1. A malfunction during aspect determination processing.

When an aspect solution was run twice, different answers were obtained
a significant fraction of the time implying computer errors. These errors were
traced to "data late" problems. The data was coming from the disk faster than the
computer could handle it. The computer is supposed to recognize this situation
but apparently does not always do so. This has delayed production running of the
new aspect solution. The new star tracker calibration is yet to be done using
bore-sights and fiducial lights. After this is accomplished, we estimate the new
aspect processing will be on line approximately mid-October.

2. New air conditioner.

Some shutdowns have been necessary for air conditioner installation (supposed to
take one week) which has taken six weeks. Data General representatives have
found several small but vital problems probably connected with environment
perturbations caused by this installation, such as loose couplers between
computer system components and, in one instance, a piece of metal foil on one of
the circuit boards.

III. Production Processing

As of 19 September, data have been processed through 1980 day 144 (23 May). Tape
is on hand through day 224 (11 August). We are currently processing data at a
rate close to real time. The S230 computer was down approximately 20 days last
month causing delays in processing and a cancellation of our ability to do any
special processing at this time. We plan to use the M600 (the user computer) for
quick look, special processing, SSS production, and calibrations. This would
relieve some of the burden from the S230, but requires a third disk (which has
not yet been operational) for the M600. Production processing is presently
finished 14 weeks after data are acquired, 8 weeks to get NASA tapes and 6 weeks
for CFA processing.

IV. IPC Software

P. Henry presented a status report on IPC spectral analysis. He has obtained
good results for 3C273; IPC-derived spectral parameters agree with those obtained
from the MFC. He emphasized that this could only be done with special processing
using sources within 2' of the center of the field. The IPC team is preparing a
report giving the status of and present uncertainties in IPC spectral analysis.
This report will be widely circulated.

V. Guest Observer Program and Sharing of Observatory Time

Since the last meeting, a new Yellow Book containing the July 4 contents of the
Einstein Observing Catalog has been printed and distributed to all Guest
Observers. A new User's Manual is now in the hands of the printer and will be
distributed when printing is complete. This is in answer to the request that the
User's Manual be made available to people before they visit CFA to process their
data.

We have reviewed the status of all Guest proposals and have notified many
principal investigators that their observations have been completed. We have also
informed others of the status of their programs. We are now routinely making a
hard copy picture of each field as the data are processed. these pictures are
mailed within a few days of processing so Guest Observers may know that
particular observation has been processed and also roughly what is contained in
the field.

The NASA Proposal Review Committee met September 9 and 10 and approved 58 out of
79 Guest Observer proposals submitted. The time approved, as usual, was about
three months Guest observing time. The current observing program now contains
between 6 and 9 months Guest observations. If Einstein takes data for another
year, there is still time for one more proposal submission, but it may be the
last normal proposal review. The Review Committee will continue to function to
provide opportunity for follow-up observations to new discoveries and for new
proposals which may result in outstanding science.

If future observations involve severe target restrictions, many of the remaining
observations will have to be changed or deleted, so the amount of Guest
observing time remaining is uncertain. We hope to soon have a better defined
observing strategy which can be communicated to Guest Observers and to the NASA
Review Committee. In the meantime, another proposal review is planned for
December unless October developments make a delay seem desirable. Deadline for
proposal submission is now 1 November.

The observing time charged to different organizations is amazingly close to the
allotted observing time and is tabulated below.

              Total Integrated Observing Times
                  (6 Jan.'79 - 27 Aug.'80)

                                    Allotted
         Useful Time     Fraction   Fraction
         --------------  ---------  ---------
CFA      80.7 x 105 s    0.34      0.352
CAL      28.5 x 105      0.12      0.117
MIT      40.2 x 105      0.17      0.157
GSFC     34.0 x 105      0.14      0.155
GO       52.0 x 105      0.22      0.215

The attached graph shows the use of the Einstein Observatory by various
organizations on a monthly basis. Dashed lines are expected usage. The Guest
Observer fraction is 20% through March 1980, increases linearly to 50% at the end
of March 1981, and remains at 50% until the end of the mission. Monthly
fluctuations about the expected level are a factor of 2. Guest Observer usage
during the last three months has been 36%.

VI. Other Business

We had planned to add new Consortium observations (for the time period 1 April
'81 to 1 October '81) this month. The uncertain future of the spacecraft has
delayed this. We are currently adding up time remaining in the observing catalog
for each organization and will base future allotments on these results. These
observations will be added in the same way as Guest observations. There will not
be another issue of the Red Book.

A complaint was received from two Guest Observers that data concerning one of
their targets was included in a graph of Consortium results contained in a paper
submitted for proceedings of a conference. Consortium papers submitted for
publication in journals are routinely reviewed in order to prevent such
occurrences. We plan now to also review papers submitted for conference
proceedings and to specifically check that results of Guest observations included
have the approval of the Guest Observer and that the work is properly
referenced.

The next Users' Committee meeting will be held December/January, exact time to
be decided.

FDS:jk