karn@eagle.UUCP (Phil Karn) (09/26/83)
DL Report Cites Failure Mechanism
After an extensive study of the circumstances surrounding the failure of
the AMSAT OSCAR 10 kick motor, AMSAT DL has released the following report.
The report is quoted verbatim except for minor editorial changes to
correct grammar, spelling, syntax, etc.
I. GENERAL
As a consequence of the collision of the third stage of Ariane L6 with
OSCAR 10, the orbit injection sequence had to be changed. This report
what has been accomplished and what problems were encountered.
II. PLANNED NOMINAL SEQUENCE
The spacecraft was designed to be separated from the launcher with a spin
rate of 10 RPM and an orientation such that a small motor-driven maneuver
could be performed on orbit 3. Following this maneuver, about four weeks were
scheduled for reorientation and calibration of the OSCAR 10 attitude
control system. When the correct attitude for a second spacecraft
maneuver was attained, the spacecraft was to be injected into an orbit
with 57 degree inclination and 1,500 km perigee.
III. ACTUAL SEQUENCE of EVENTS
After the collision with the launcher shortly after separation, OSCAR 10 was
spinning at minus 2 RPM. The sun angle was approximately 70 degrees. This
attitude resulted in the spacecraft temperatures being very much below
specifications. Moreover, the power production was extremely low. The
low spin rate in conjunction with the aerodynamic drag at perigee resulted
in a daily improvement of sun-angle of about 3 to 4 degrees. It was thus
decided to not adjust the attitude until the sun-angle had improved
sufficiently to bring power production and temperatures again into
specifications and until the sensors allowed unambiguous attitude
determination.
These conditions were met one week after launch. At that time an
emergency spin-up maneuver was intiated to allow the normal attitude
control system to take over. The spin rate and attitude were adjusted
to the first firing attitude until July 8. The first motor firing
maneuver duration was increased because of the delay in first firing
date and the consqential changes in sun-angle. The new post-first-
firing values were to be 19 degrees inclination and 1,500 km perigee.
Local weather conditions at the primary control site prevented
satisfactory telemetry reception on July 8 so the firing was moved
to July 11. The first motor firin maneuver occured at 22:29 UTC.
The motor performed flawlessly. However, it was later learned that
the burn duration was 185 seconds rather than the planned 105.
The cause of the longer than planned burn was investigated and traced
to a drawing error in the AMSAT-built Liquid Ignition Unit (LIU). The
increased burn duration resulted in an orbital inclination of 25.9
degrees and a perigee height of 3,950 km. A secondary result of the
extended burn was that the Helium bottle temperature dropped to minus
10 degrees C; again outside of its specifications.
One day later it was noticed that the low pressure Helium indication
had dropped to one half of its nominal value while the high pressure
reading remained about 200 Bars. Proceeding under the assumption the low
pressure Helium gauge was at fault, we reoriented the spacecraft to the
new firing attitude. On July 19 at 00:29 UTC the second firing was
attempted. However, ever though the magnet activation was confirmed,
no firing occured and ther were no changes in the indicated pressures.
Consequently, the spacecraft reorientation was initiated such that operations
could commence August 6 at 18:00 UTC using the existing orbit.
IV. INVESTIGATION INTO the LOSS of HELIUM
Telemetry data from various control stations worldwide was collected until
August 1 in order to recontruct the pressure history of the spacecraft.
from the ddata collected it cn be unambiguously inferred that the Helium loss
occured on the high pressure side of of the gas system and that the time
constant was 500 minutes. Furthermore, when the pressure fell below 7 Bars,
the time constant changed to about 10 days. This finding rules out any leak
in the motor or in the fluid tank system. An investigation with Messerschmitt-
Boelkow-Blohm (MBB) on August 2 left only two candidate causes for the leak:
a) the seal of the Helium bottle
b) an "AN3" connection from the Helium pyrotechnic valve leading to a check
point
Model Calculations and the observed behavior of the Helium seal in earlier
tests led to discarding the "AN3" thoery as the causeof the leak. It is
now concluded with a high degree of certainty that the Helium bottle seal
caused the gas loss.
We believe the following conditions led to this gas loss:
1) The Helium bottle used in OSCAR 10 was developed by AMSAT DL with
assistance from DFVLR [abreviation unknown- Ed] because the more conventional
gas containers did not fit into the available physical space. The composite
technology employed resulted in a bottle which could be specified for service
only for temperatures above 0 degree C. Differential temperature expansion
problems precluded hermeticity guarantees below 0 degrees C.
2) As a consequence of the poor sun angle subsequent to deployment, OSCAR 10
experienced a total of 14 cold cycles beyond its specifications early in
its life. Some temperature excursions were to as low as minus 10 degrees C.
3) Shortly before the launch campaign, an error in the termal design was
discovered. The error would have resulted in operating temperatures
10 degrees lower than the design target for OSCAR 10. Using the limited
resources available at Kourou the design could be modified only to the
extent that 5 degrees of the 10 degree error could be recovered. As
a result, the spacecraft temperature is 5 degrees lower than originally
planned.
4) The extended burn led to another low temperature cycle of the Helium
bottle to minus 10 degrees C.
Each of these conditions would not, taken separately, resulted in a leak.
However, the cumulative effect of the intial cold cycles probably reduced
the tension of the hekium bottle seal to about one half of its design value.
The temperature drop during the first burn plus the pressure change likely
reduced the seal tension sufficiently to have causec the leak.
The following evidence supports this finding:
1) The Helium high pressure indicator exhibits a shift of the zero-line by
about 200 Bars. This pressure gauge is a starin gauge affixed to the bottle.
The observed shift evidences irreversible bottle dome material deformation.
2) The change in the leak time-constant at 7 bars can be explained by the fact
that the bottle seal contains a Teflon O-ring which, from prior experience,
is known to seal between between 5 to 10 Bars. (No such ring exists in the
AN3 connection).
V. CONCLUSION FOR FUTURE PHASE 3 SPACECRAFT
a) The thermal design of the spacecraft will be changed to increase the
temperature to a nominal design.
b) AMSAT will make every effort to replace the solid propellant motor planned
for Phase 3C by the MBB liquid propulsion system. A solid propellant
motor would not have survived the thermal abuse of OSCAR 10. The liquid
propulsion system provided margins that were instrumental in saving the
mission.
c) For Phase 3C, AMSAT will attempt to resdeign the Helium bottle such that
its operating temperature can be specified to minus 10 degrees C.
VI. CONCLUSION REGARDING OPERATIONS of OSCAR 10
The present orbit of OSCAR 10 provides about 40 percent of the radio coverage
compared to the originally planned 57 degree inclination orbit.
Furthermore, the low inclination forces antenna off-pointing and/or
operational quiet times for about 40 percent of the time because of
sun-angle constraints. Consequently, operational capability of OSCAR 10 is
reduced to about one fourth of its planned value. Also, because of the
17 day periodicity of the satellite's footprint, unfortunate week-long
interruptions in East-West traffic (e.g., over the North Atlantic) results.
On the positive side it may be noted that OSCAR 10 can meet all its
scientific objectives and that the communications capability is so large that
even one fourth of its nominal capability represents a very significant
communication potential for amateur radio operators. This is all the
more true since this is the first working spacecraft of this series and the
users can take only previous OSCARs as basis for comparison. Plainly,
OSCAR 10 presents a dramatic improvement.