Orbit glitch consumes propellant on new satellite

so who messed up?

Orbit glitch consumes propellant on new satellite
Posted: Sun, May 16 11:52 AM ET (1552 GMT)
A Japanese communications satellite launched last month has reached its final orbit but
with far less propellant onboard than planned, apparently because of an unforeseen
aspect of its transfer orbit. Space News reported late Friday that the Superbird 6 satellite,
launched on an Atlas 2AS in mid-April, has reached its planned position in
geosynchronous orbit but used far more propellant than planned. According to
the report, the Atlas placed the spacecraft into the planned transfer
orbit, but that orbit failed to properly take into account the
gravitational effects of the Moon and Earth. As a result, the
spacecraft's perigee dropped perilously low, from 200 to 100
kilometers, forcing the spacecraft to use additional propellant to raise its orbit.


Herm
Astropics http://home.att.net/~hermperez

In article ,
Herm wrote:
...the Atlas placed the spacecraft into the planned transfer
orbit, but that orbit failed to properly take into account the
gravitational effects of the Moon and Earth...

so who messed up?

(First, a correction: they probably meant "the Moon and the Sun".)

Probably the satellite's owners, unless the contract specified delivery in
final orbit (in which case the satellite builder is to blame). As far as
orbits go, the launcher people basically give the customer what he asks
for. Getting from the transfer orbit to the final orbit is his problem.
Their responsibility pretty much ends at separation.

Superbird 6 did use a most unusually high transfer orbit, making it
especially vulnerable to lunar and solar perturbations. Possibly the
maneuver planners treated it as a straightforward extension of more normal
transfer orbits, not realizing that there were new issues. If my guess
about the detailed nature of the hitch is correct, it's one that *is* well
understood, but is minor enough in the usual transfer orbits -- for the
short duration of the usual stay in such orbits -- that it's not covered
in the usual discussions of the problem.

Lunar and solar perturbations make various parameters of the orbit
oscillate somewhat around nominal values. On average this has no
long-term effect, and it's easy to dismiss it as unimportant... but even
short-term effects on the orbit's eccentricity are dangerous: higher
eccentricity makes the apogee go up *and the perigee go down*, which is
problematic in a highly elliptical orbit where the perigee is none too
high to begin with. If the perigee gets lowered into the atmosphere by an
eccentricity oscillation, the fact that in theory it would eventually come
back up again is not comforting.

For short stays in normal transfer orbits, lunar/solar oscillations aren't
important. But the higher orbit is more strongly affected, and with its
much longer period (which means, e.g., much less frequent passes through
the Van Allen belts), quite possibly they weren't in a hurry to start
maneuvering.

Also, tracking of very high orbits isn't so good. This would give them
incentive to delay, for more accurate determination of the starting orbit.
And it could mean that they wouldn't notice the problem until it had
become critical... which would explain significant fuel expenditure.
Slightly raising the perigee of such an orbit is very cheap if it's done
at apogee, but if you notice a few hours before perigee that perigee is
going to be Much Too Low, fixing that is expensive.
--
MOST launched 30 June; science observations running | Henry Spencer
since Oct; first surprises seen; papers pending. |

"Henry Spencer" wrote in message
...
In article ,
Herm wrote:
...the Atlas placed the spacecraft into the planned transfer
orbit, but that orbit failed to properly take into account the
gravitational effects of the Moon and Earth...

so who messed up?

(First, a correction: they probably meant "the Moon and the Sun".)

Probably the satellite's owners, unless the contract specified delivery in
final orbit (in which case the satellite builder is to blame). As far as
orbits go, the launcher people basically give the customer what he asks
for. Getting from the transfer orbit to the final orbit is his problem.
Their responsibility pretty much ends at separation.

Superbird 6 did use a most unusually high transfer orbit, making it
especially vulnerable to lunar and solar perturbations. Possibly the
maneuver planners treated it as a straightforward extension of more normal
transfer orbits, not realizing that there were new issues. If my guess
about the detailed nature of the hitch is correct, it's one that *is* well
understood, but is minor enough in the usual transfer orbits -- for the
short duration of the usual stay in such orbits -- that it's not covered
in the usual discussions of the problem.

Lunar and solar perturbations make various parameters of the orbit
oscillate somewhat around nominal values. On average this has no
long-term effect, and it's easy to dismiss it as unimportant... but even
short-term effects on the orbit's eccentricity are dangerous: higher
eccentricity makes the apogee go up *and the perigee go down*, which is
problematic in a highly elliptical orbit where the perigee is none too
high to begin with. If the perigee gets lowered into the atmosphere by an
eccentricity oscillation, the fact that in theory it would eventually come
back up again is not comforting.

For short stays in normal transfer orbits, lunar/solar oscillations aren't
important. But the higher orbit is more strongly affected, and with its
much longer period (which means, e.g., much less frequent passes through
the Van Allen belts), quite possibly they weren't in a hurry to start
maneuvering.

Also, tracking of very high orbits isn't so good. This would give them
incentive to delay, for more accurate determination of the starting orbit.
And it could mean that they wouldn't notice the problem until it had
become critical... which would explain significant fuel expenditure.
Slightly raising the perigee of such an orbit is very cheap if it's done
at apogee, but if you notice a few hours before perigee that perigee is
going to be Much Too Low, fixing that is expensive.
--
MOST launched 30 June; science observations running | Henry Spencer
since Oct; first surprises seen; papers pending. |


Boeing (formerly Hughes Space and Communications) typically runs the
transfer orbit mission for its satellites. It also has a great deal of
orbital transfer experience so I don't believe that they would have been
caught by suprise by orbital anamolies caused sun and moon. Otherwise I
can't speculate on what might have gone wrong in this situation.

In article . net,
no_one wrote:
so who messed up?
Probably the satellite's owners, unless the contract specified delivery in
final orbit (in which case the satellite builder is to blame)...

Boeing (formerly Hughes Space and Communications) typically runs the
transfer orbit mission for its satellites.

Indeed, I'm told that this *was* a "delivery in final orbit" contract, so
the satellite still belonged to Boeing and this is 100% their problem.
The finger would tend to point that way anyway if they were running the
transfer, but with ownership still resting with them, there is no room for
doubt about ultimate responsibility.

It also has a great deal of
orbital transfer experience so I don't believe that they would have been
caught by suprise by orbital anamolies caused sun and moon.

Apparently they were, although this may perhaps have been another MCO
(where an otherwise-minor error repeatedly escaped being caught and dealt
with, and by an unfortunate combination of factors, happened to have
disastrous consequences).
--
MOST launched 30 June; science observations running | Henry Spencer
since Oct; first surprises seen; papers pending. |

"Henry Spencer" wrote in message
...
In article . net,
no_one wrote:
so who messed up?
Probably the satellite's owners, unless the contract specified delivery
in
final orbit (in which case the satellite builder is to blame)...

Boeing (formerly Hughes Space and Communications) typically runs the
transfer orbit mission for its satellites.

Indeed, I'm told that this *was* a "delivery in final orbit" contract, so
the satellite still belonged to Boeing and this is 100% their problem.
The finger would tend to point that way anyway if they were running the
transfer, but with ownership still resting with them, there is no room for
doubt about ultimate responsibility.

It also has a great deal of
orbital transfer experience so I don't believe that they would have been
caught by suprise by orbital anamolies caused sun and moon.

Apparently they were, although this may perhaps have been another MCO
(where an otherwise-minor error repeatedly escaped being caught and dealt
with, and by an unfortunate combination of factors, happened to have
disastrous consequences).
--
MOST launched 30 June; science observations running | Henry Spencer
since Oct; first surprises seen; papers pending. |


The Delta-V to drop the perigee from 167 to 100 km is remarkably small.

V^2= 2GM(1/r - 1/(2a))
where GM= 3.986x10^5 km^3/sec^2
r= 6378 + 122343 = 128721 km (apogee radius)
2a= 167 + 2x6378 +122343 = 135266 km (nominal perigee)
= 100 + 2x6378 + 122343 = 135199 km (low perigee)

Vapogee = .5474 km/sec (nominal perigee)
= .5447 km/sec (low perigee)
Delta-v = .0027 km/sec, or 2.7 m/sec

For a normal GTO we have
r=6378 + 36000 = 42378 km (apogee radius)
2a = 167 + 2x6378 + 36000 = 48923 km (nominal perigee)
= 100 + 2x6378 + 36000 = 48856 km (low perigee)

Vapogee = 1.586 km/sec
= 1.579 km/sec
Delta-v = .007 km/sec, or 7 m/sec

The conclusion is that the satellite should always have its perigee raised
during the first apogee passage, as a matter of common prudence. It's
possible that they've gotten away with not doing it for normal GTO (when
solar panels don't deploy properly at first?), so it became a well-known
"fact" that the perturbations can't do any harm, sort of like falling foam
on wing edges.

Murray Anderson

"Murray Anderson" wrote in message
...

"Henry Spencer" wrote in message
...

It also has a great deal of
orbital transfer experience so I don't believe that they would have
been
caught by suprise by orbital anamolies caused sun and moon.

Apparently they were, although this may perhaps have been another MCO
(where an otherwise-minor error repeatedly escaped being caught and
dealt
with, and by an unfortunate combination of factors, happened to have
disastrous consequences).
--
MOST launched 30 June; science observations running | Henry
Spencer
since Oct; first surprises seen; papers pending. |


The Delta-V to drop the perigee from 167 to 100 km is remarkably small.

Delta-v = .0027 km/sec, or 2.7 m/sec

For a normal GTO we have

Delta-v = .007 km/sec, or 7 m/sec

The conclusion is that the satellite should always have its perigee raised
during the first apogee passage, as a matter of common prudence. It's
possible that they've gotten away with not doing it for normal GTO (when
solar panels don't deploy properly at first?), so it became a well-known
"fact" that the perturbations can't do any harm, sort of like falling foam
on wing edges.


For GTO it is not common to raise the perigee at first apogee. Firstly
because its not necessary as the effect of lunar perturbations on that orbit
is not significant and secondly because first apogee is only about 5 hours
after launch and so there isn't time to determine the orbit and configure
the spacecraft before it gets to apogee. Apogees 3 and 4 are more likely
times for the first manoeuvre.

SSTO is different because now the time to Apogee 1 can be, as in this case,
more like 24 hours so the nominal strategy would probably be to fire then.
If you don't go on A1 then you have to wait another 48 hours but
nevertheless this would normally be planned for as a contingency in case the
nominal firing was missed for any reason.

When I worked on an SSTO mission some time ago, we noticed shortly before
launch that the perigee height was doing strange things in some simulations
of the contingency case. For some launch dates it rises every orbit and for
some it falls with roughly a two week period. We realised that this was
caused by the Moon and after some analysis we closed the launch window on
days where the perigee height would drop too much in the contingency case.

It sounds like the launcher was targetting a minimum residual shutdown where
all the useable propellant is put into raising the apogee so this would
complicate the launch window analysis as you would need to cover the full
range of possible apogee heights.

Possibly in previous missions Boeing has always managed to fire at A1 and so
the effect would never have been seen in flight but in this case something
caused them to go to a backup strategy. If the lunar effects had not been
foreseen then it would be down to luck whether the perigee height was rising
or falling.

With a rough guess at the injection parameters for this case I got a 30-40km
drop in perigee height per rev.

Jon Marshall

"Jonathan Marshall" wrote in
message ...
"Murray Anderson" wrote in message
...

"Henry Spencer" wrote in message
...

It also has a great deal of
orbital transfer experience so I don't believe that they would have
been
caught by suprise by orbital anamolies caused sun and moon.

Apparently they were, although this may perhaps have been another MCO
(where an otherwise-minor error repeatedly escaped being caught and
dealt
with, and by an unfortunate combination of factors, happened to have
disastrous consequences).
--
MOST launched 30 June; science observations running | Henry
Spencer
since Oct; first surprises seen; papers pending. |


The Delta-V to drop the perigee from 167 to 100 km is remarkably small.

Delta-v = .0027 km/sec, or 2.7 m/sec

For a normal GTO we have

Delta-v = .007 km/sec, or 7 m/sec

The conclusion is that the satellite should always have its perigee
raised
during the first apogee passage, as a matter of common prudence. It's
possible that they've gotten away with not doing it for normal GTO (when
solar panels don't deploy properly at first?), so it became a well-known
"fact" that the perturbations can't do any harm, sort of like falling
foam
on wing edges.


For GTO it is not common to raise the perigee at first apogee. Firstly
because its not necessary as the effect of lunar perturbations on that
orbit
is not significant and secondly because first apogee is only about 5 hours
after launch and so there isn't time to determine the orbit and configure
the spacecraft before it gets to apogee. Apogees 3 and 4 are more likely
times for the first manoeuvre.

SSTO is different because now the time to Apogee 1 can be, as in this
case,
more like 24 hours so the nominal strategy would probably be to fire then.
If you don't go on A1 then you have to wait another 48 hours but
nevertheless this would normally be planned for as a contingency in case
the
nominal firing was missed for any reason.

When I worked on an SSTO mission some time ago, we noticed shortly before
launch that the perigee height was doing strange things in some
simulations
of the contingency case. For some launch dates it rises every orbit and
for
some it falls with roughly a two week period. We realised that this was
caused by the Moon and after some analysis we closed the launch window on
days where the perigee height would drop too much in the contingency case.

It sounds like the launcher was targetting a minimum residual shutdown
where
all the useable propellant is put into raising the apogee so this would
complicate the launch window analysis as you would need to cover the full
range of possible apogee heights.
From ILS web site:
a.. Apogee altitude (km) achieved 122,343 (mission requirement of 86,882 to
123,622)

a.. Perigee altitude (km) achieved 167.1 km (mission requirement 167.0 plus
or minus 2.0)

a.. Inclination (deg) achieved 26.25 (mission requirement less than or equal
to 28.1)

I guess that's minimal residual.


Possibly in previous missions Boeing has always managed to fire at A1 and
so
the effect would never have been seen in flight but in this case something
caused them to go to a backup strategy. If the lunar effects had not been
foreseen then it would be down to luck whether the perigee height was
rising
or falling.

With a rough guess at the injection parameters for this case I got a
30-40km
drop in perigee height per rev.

Jon Marshall


The sun was in about the same position at that time - new moon occurred on
April 19. So that would have increased the effect.

Murray Anderson

In article ,
Jonathan Marshall wrote:
Possibly in previous missions Boeing has always managed to fire at A1 and so
the effect would never have been seen in flight but in this case something
caused them to go to a backup strategy...

I looked back at the relevant issue of Jonathan's Space Report, and
apparently they *did* do at least one apogee burn at A1 or not long
thereafter, boosting perigee to 1137km by the 21st (the launch being on
the 16th).

With a rough guess at the injection parameters for this case I got a 30-40km
drop in perigee height per rev.

For a guess, they thought they were safe after the initial perigee-raising
burns, and delayed further action until it was convenient in some way.
High-altitude spacecraft aren't easily tracked, and since they thought
they were safe they didn't put a high priority on further tracking, and
nobody realized that the perigee altitude was steadily eroding.
--
MOST launched 30 June; science observations running | Henry Spencer
since Oct; first surprises seen; papers pending. |

"Henry Spencer" wrote in message
...
In article ,
Jonathan Marshall wrote:
Possibly in previous missions Boeing has always managed to fire at A1 and
so
the effect would never have been seen in flight but in this case
something
caused them to go to a backup strategy...

I looked back at the relevant issue of Jonathan's Space Report, and
apparently they *did* do at least one apogee burn at A1 or not long
thereafter, boosting perigee to 1137km by the 21st (the launch being on
the 16th).

With a rough guess at the injection parameters for this case I got a
30-40km
drop in perigee height per rev.

For a guess, they thought they were safe after the initial perigee-raising
burns, and delayed further action until it was convenient in some way.
High-altitude spacecraft aren't easily tracked, and since they thought
they were safe they didn't put a high priority on further tracking, and
nobody realized that the perigee altitude was steadily eroding.
--
MOST launched 30 June; science observations running | Henry Spencer
since Oct; first surprises seen; papers pending. |


The SpaceWarn Bulletin #606, dated May 1, gives "apogee 120,679 km, perigee
1,138 km, and inclination 25.5°.", and presents this as the initial transfer
orbit (plainly false, of course). Is it possible that Space Command got the
orbit wrong, JSR assumed the number was post-apogee-burn, and Space Warn
didn't notice the problem at all?
This looks interesting.

Murray Anderson

In article ,
Murray Anderson wrote:
I looked back at the relevant issue of Jonathan's Space Report, and
apparently they *did* do at least one apogee burn at A1 or not long
thereafter, boosting perigee to 1137km by the 21st (the launch being on
the 16th)...

The SpaceWarn Bulletin #606, dated May 1, gives "apogee 120,679 km, perigee
1,138 km, and inclination 25.5°.", and presents this as the initial transfer
orbit (plainly false, of course). Is it possible that Space Command got the
orbit wrong, JSR assumed the number was post-apogee-burn, and Space Warn
didn't notice the problem at all?

JSR says the initial orbit was 167km x 122343km x 26.3deg; I'd guess those
numbers are from the launch provider. It says Space Command was slow about
issuing orbital data, coming out with an element set on the 21st showing an
orbit of 1137km x 120678km x 25.48deg "following the initial apogee burns".
--
MOST launched 30 June; science observations running | Henry Spencer
since Oct; first surprises seen; papers pending. |