[Date Prev][Date Next][Thread Prev][Thread Next] - [Date Index][Thread Index][Author Index]

ArcJet Update

Dear Folks,

I'd like to try to answer a few questions and clear up some misconceptions
made in comments on the Amsat-BB about the ArcJet outgassing.

1. Was the old orbit stable?
Yes it was.  I did the orbital integration and it WAS stable, at least in
the sense of not suffering re-entry or melt down.  However, as we all know,
perigee was very close.  In fact in some simulations, it was as low as
160 km.  There was considerable drag at perigee, which continuously changed
the mean motion and probably contributed to the perturbation in ALON (the
mystery effect). The keps on the S/C had to be frequently updated. The
rapidity of the perigee flyby made magnetorquing very tricky and required a
lot of time to determine attitude.  To improve these things we wanted to
raise perigee height.  That doesn't mean the old orbit wasn't stable, but
it certainly wasn't optimal, and it was close enough at perigee in June to
be frankly scary.

2.  Was the command software controlling the arcjet faulty?
No, it most certainly was not.  I wrote it, the rest of the command team
checked it.  It functioned perfectly in all the simulations.  There is not
a shred of telemetry evidence to suggest a software error.

3.  Are we out of ammonia?
As far as we can tell, yes.

4.  Where did it go?
We don't have all the answers, and hence we have not yet posted a full
report.  It appears that the TMFC (thermal mass flow controller) failed,
passing far more gas than it should have.  There was NO INDICATION of this
in the telemetry.   Spin rate changes and telemetered flow rate, etc., all
appeared in a safe range.  The software properly set the TMFC to 50% flow
rate.  This should have been ~25-30 mg/s of flow.  Even fully open, the
TMFC should not have exhausted more than 100 mg/s, which should have left
~60% of the ammonia unused, even after 58 hours of run time.  The telemetry
suggests that the gas ran out after approximately 1575 minutes (26.25 hrs.).
If the tanks were full, that's 560 mg/s, far in excess of what the TMFC
should have allowed.  Again, we're still looking at this for better answers.
As has been noted, we really only needed about half of the fuel to get to
the orbit we're in, so it is quite possible that there was also a slow leak
in the ammonia tanks such that we started with only approximately half of
the fuel. If this were the case, then the flow would have been ~280 mg/s,
still almost three times nominal maximum flow rate.  Looked at another way,
the new and old orbits indicate an apogee velocity change of 40 m/s.
Accelerating a 500 kg mass through 1575 minutes indicates a force (thrust)
of 0.213 N.  That's over twice as high as would have been achieved if the
motor were used in a proper powered ArcJet burn.  Unlike the helium tank,
the ammonia tanks have a liquid/gas interface in equilibrium, so the
pressure in the tanks is not proportional to the amount of liquid they
contain.  Pressure is a function only of temperature as long as liquid
ammonia is present.  Thus, there was no way to tell how much "fuel" we had.
If we did have a slow leak it is very fortunate we did not wait any longer
to use the remaining fuel.

5. Why wasn't the orbit checked after the first test outgassing?
Because of changing solar angle, the mystery effect and precession moving
ALON, as well as the rapidly decreasing perigee height in late June due to
solar/lunar forces (which might make attitude even harder to maintain), it
was important to complete the maneuver as soon as possible while conditions
were right and then begin the move back to 0/0.  A nominal single test burn
should have produced so little change in perigee height as to be hard to
detect against background "noise".  In addition, waiting for good keps
following the test could have required more than a week (current case in
point).  There was nothing in the test telemetry to suggest a problem, so
we proceeded with the full maneuver. From a pragmatic point of view it was
always clear that any radical departure from expected would make it highly
unlikely that we would be able to maintain a stable, low-volume gas flow as
needed for a powered ArcJet burn, so loss of extra ammonia would likely
have little effect on the powered ArcJet use.

6.  Do the command stations know what they're doing?
We like to think so.  There is ABSOLUTELY NO EVIDENCE that the ammonia
problem was due to a command error.   I cannot state that any clearer.
Please bear in mind that we all take this task extremely seriously, and we
spend a HUGE amount of time on it.  Just getting AO-40 into the 270/0
position for the outgassing took several weeks of effort and calculation.
When there is a problem, the amount of time required goes up
exponentially.  The first requirement is to maintain the integrity of the
spacecraft, then to gather critical information while it is still
available, then to analyze the information, and then to report the
findings.  I freely admit that my first thought when there's a problem is
NOT to post a message to the amsat-bb.  This is not because we want a
"cover-up," although I have been personally accused of this at least once
in quite objectionable terms.  The reality is that we are busy trying to
figure out if we still have a safe spacecraft, and then determine what's
going on.  When we are sure that we have the best possible information, we
fully report our findings.

7. What's going to break next?
The short answer is that we don't know.   I sincerely hope that nothing
else malfunctions for a long, long time, but this is after all, rocket
science.  Nothing is guaranteed.  However, there has been the suggestion
that the command stations are rushing to test things without considering
failure scenarios.  Nothing could be further from the truth.  Weeks of
discussion were devoted to the ArcJet and possible failure modes.  These
discussions involved the designers and builders of the ArcJet, the
management team, etc.  Out of these discussions came the decision to "cold
outgas" before trying hot outgassing, and to orient to raise perigee, since
a faulty or stuck valve in this orientation would not destroy the orbit.
Consider that if we had been oriented differently we could be at -350 km
rather than +850 km perigee height!!

The momentum wheels have not yet been tested.  Rest assured that when they
are, it will be after lengthy discussions regarding failure possibilities,
recovery modes, etc.  I do not regard this as a particular risky test,
unlike the ArcJet, but it will still be approached very very carefully.
Finally, the decision to deploy the solar panels will be much more
involved.  This is a "no return" event and will in all likelihood be a
considerable time in the future after prolonged demonstration that the
momentum wheels are fully functional and that 3-axis control software is up
to the task.  If there are ANY problems in this regard, we will stay in
spin mode!!! Our primary goal at the moment is to get AO-40 back into an
orientation where the transponders can be operational and RUDAK testing can
be completed.

8. And finally
I hope this clears up at least a few points.  I know this is a frustrating
issue for the users and supporters, but believe me, it's also very
frustrating for the command stations.  A lot of software written and tested
for the "hot" ArcJet, will now never be used.  And my frustration is
totally insignificant next to that of the wonderful folks who put years of
effort into building the ArcJet, and now will not see it function.  I must
admit that my disappointment is compounded by some of the comments that
appear on the amsat-bb questioning the abilities and motives of the command
stations.  I try not to take these personally, and I realize that you would
all like answers.  So would we.  We do have a very stable orbit and we are
moving back towards 0/0 to make AO-40 available for your use.  All in all,
it could be a lot worse, a whole lot worse.  I'd like to thank all of you
who have posted supportive comments on the amsat-bb.  I'd even like to
thank those who have questioned our abilities.  At least your comments show
that you're interested.  We can withstand your critical scrutiny!   ...if
we can't we need a new hobby.

  Stacey E. Mills, W4SM    WWW:    http://www.cstone.net/~w4sm/ham1.html
    Charlottesville, VA     PGP key: http://www.cstone.net/~w4sm/key

Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org