PART 1: Practicing for maneuver designs and tweaks
PART 2: Propulsion work to minimize a potential valve leak

October 9, 1995
The Navigation (NAV) Team has been very busy preparing for events occurring 
during Galileo's final approach to Jupiter and the Jupiter encounter. (The
Galileo atmospheric Probe and Orbiter both arrive at Jupiter on December 7.)
There are three small trajectory correction maneuvers (TCMs) scheduled to
adjust the flight path of Galileo before Jupiter arrival: TCM-27 on November
17, TCM-28 on November 27, and TCM-28A on December 2. The large Jupiter
Orbit Insertion (JOI) burn, which slows down Galileo so it will be captured 
into orbit about the giant planet, occurs on December 7 (Pacific time), about 
one hour after Galileo is finished receiving data from the atmospheric Probe. 
Then the first orbit trim maneuver (OTM-1) will be performed on December 9 to
correct any errors in the first Jupiter orbit. The NAV Team must determine the
necessary velocity changes for each of these maneuvers.

Usually, the Galileo Flight Team has about 7 days to do the work of generating
the spacecraft commands for a maneuver (called a "sequence"); this time
includes the NAV Team work to calculate the desired velocity change. However,
the Flight Team has only 24 hours (working around the clock) to generate the
last maneuver before Jupiter arrival (TCM-28A) and the first maneuver after
arrival (OTM-1). Similarly, we will be computing a late update (called a
"tweak") for both TCM-28 and JOI on the last day before each of these
maneuvers, using the same 24-hour schedule. There will be many tired people
on the Flight Team after the Jupiter encounter period! The Flight Team must
practice these 24-hour maneuver designs and tweaks before they happen to be
sure so that there will be no problems. When the Flight Team practices a
maneuver design or maneuver tweak, it's called "test and training".

For the past several weeks, I have been heavily involved in test and training. 
I am the NAV Team coordinator for three test and training exercises. I have to
supply a schedule to the Project test and training coordinator showing how
much time each team (such as the NAV Team or the Orbiter Engineering Team)
has to do their part of the work for a maneuver design. I also have to supply 
the names of people from the NAV Team who will be doing the test. During the 
test, I monitor the work of the NAV Team. (And, if I have the time, I do some
trajectory or maneuver design work myself!) Then, after the test is finished, 
I have to give a presentation to the Project managers on how the NAV Team did,
what problems we had, and suggestions for improving the process. The first
test and training exercise was done during the week of September 18 to test
the generation of the TCM-28 tweak. The second test and training exercise was
done during the week of October 2 to test the 24-hour maneuver design used for
TCM-28A and OTM-1. Overall, both of these tests were successful, although
there were some minor problems (not unexpected) that need to be worked out.
The third and final test and training exercise will be done near the end of
October to test the JOI tweak process.

Normally, I have quite a few meetings to attend each week. Some of them have
had to do with preparing for and conducting the test and training exercises I
mentioned earlier. I also attend regular meetings that are held each week.
There is a weekly NAV Team meeting where we discuss the activities that
people on the NAV Team are currently working on and what work is coming in
the future. Each week I also attend a TCM Design Team Meeting and a Relay/JOI
Working Group Meeting. At the TCM Design Team Meeting  members of the Flight
Team discuss TCM design issues. The Relay/JOI Working Group is made up of the
people who are responsible for producing the critical sequence that controls
Galileo during the time it is receiving data from the Probe and performing the
JOI maneuver. (This is the most important sequence the spacecraft will ever
execute. Receiving data from the Probe and getting Galileo into orbit about
Jupiter are prime mission objectives that must be completed successfully
This sequence is designed to work in the presence of numerous spacecraft
failures and to continue to issue commands regardless of what is happening to
Galileo during this critical time.) There are usually several informal meetings
each week of NAV Team personnel to discuss navigation strategies for the
upcoming Jupiter encounter. Other meetings I have to attend are approval
meetings for the sequences that control the spacecraft during Jupiter approach
and the orbital tour. I must make sure that these sequences have been checked
by NAV Team people to see that the expected navigation tracking coverage,
optical navigation pictures, and maneuvers have been put in the sequence
correctly.

Last month I also assisted the NAV Team Chief in preparing a memo containing
the NAV Team budget for the remainder of the baseline Galileo mission, which
ends in December 1997. This involves figuring out the number of people needed
and the cost for the work the NAV Team has to do. 

Week of 10/9
The week started out like any other week. Tuesday we approved a high level
(not very detailed) science plan for the Ganymede encounter on our eighth
orbit (in May 1997). Our minds were focused ahead to the orbital tour phase
of the mission. We looked at the science plan for observations, when we
would do orbit trim maneuvers to target us for the ninth encounter, and the
myriad of engineering events to support getting the science data.

The next day we switched to final approval of the last Jupiter approach
sequence, and the actual sequence which will perform the Io flyby, recording
of the probe's transmitted data, and the critical main engine burn that will
place us into orbit as an artificial satellite of Jupiter. All of that. What a
switch from the previous day!  We went from thinking a year and a half into
the future, to getting ready for one month from now. Having this sequence
development concluded was a giant effort, now successfully off of our task
checklist.

Unfortunately, we could only relax for a few minutes. Shortly after picking
up burgers for the family, we were called back in for a serious spacecraft
anomaly, or problem.
  
What happened? After taking an image of Jupiter, the tape recorder had been 
commanded to rewind so that it could begin playing back the image. Like
most tape recorders, the recorder is supposed to stop when it senses that it
has reached the start of the tape. This time, the spacecraft was reporting
that the tape recorder was stuck in the rewind mode, and was attempting to
slew (or move) continuously.

After analysis, commands were planned to safely stop the recorder. Due to a
previous failure of the high power ground transmitter at the Canberra
(Australia) Deep Space Antenna site, we have to wait until Thursday morning
to send the tape recorder commands. It's *really* hard to sit and wait while
you know that your spacecraft is in trouble. In what may be an amazing
coincidence, or a possible hint of what the problem is, a similar problem has
occurred on the ground simulator (a hardware and software identical copy of
the spacecraft computer systems). Analysis of the ground failure may lead
to understanding of the flight problem.

Thursday -  The commands have stopped the tape recorder successfully. We
have organized the effort to find what the problem is and how to fix or work
around it. If you've ever had to figure out why your computer isn't working,
you already have a good idea about how we do this--you check out everything
you can. One group is searching for any hardware problem which could cause
the "signature" we see in telemetry (the signature refers to the apparent
symptoms of the problem). Another group is looking for any possible
spacecraft software problem that could cause the symptoms. Other groups
are looking in other directions.

I'm assigned to coordinate  a group to look to see if it's possible that the
way we have arranged a sequence of spacecraft commands could have caused
the problem. With a long history of successful commanding, behind us, we're
looking for *anything* we have done differently between the past
successful encounters and this sequence (a sequence is a series of
spacecraft commands loaded into the computer on board, "clocking out"
[running] at the appropriate time). All possible differences are being looked
at, even if no one believes they could be relevant.

Daily meetings are being held to give the results found that day. Today, it's 
mainly descriptions of the plans people have to do their job. The hardware 
testers have a tough job. They have to be careful that their tests don't cause
further damage, and that their tests don't accidentally make the problem go
away (on the ground simulator) before we can find out what is causing it.

Friday - Today's anomaly status meeting concentrated on analyzing the
failure of the *ground* tape recorder. We saw some plots of the telemetry
produced by the ground recorder. The particular failure possibilities were
laid out, and there was some discussion about what tests should be done to
decide which failure possibility was the real one. It's still not clear that 
the failure of the ground tape recorder and the spacecraft tape recorder are 
due to the same problem, but it's a reasonable thing to investigate. It's also
much easier for us to examine the tape recorder that's downstairs in the
testbed than the one that's near Jupiter. The tests were already in progress
during the meeting, and were scheduled to continue over the weekend.

The analysis of the ground software and sequence identified three activities 
that were linked in one way or another to the period when the tape recorder 
problem arose. None of these activities seems like a likely cause of the 
anomaly , but further investigation is needed to make sure.
                            Jim Erickson