oliver@vf.jsc.nasa.gov (05/23/91)
Several recent posts have expressed interest in STS-51F, the flight which had
the Abort to Orbit. Here is some information as well as a narative of the
launch which I hope answers most of the questions.
STS-51F - Shuttle mission 19
crew: CDR: Col. C. Gordon Fullerton
PLT: Col. Roy D. Bridges, Jr.
MS1: F. Story Musgrave, MD
MS2: Anthony W. England, PhD
MS3: Karl G. Henize, PhD
PS1: Loren W. Acton, PhD
PS2: John-David F. Bartoe, PhD
Ascent/Entry Flight Director: Cleon Lacefield
Capcom: Capt. Richard (Dick) N. Richards
Flight Dynamics Officers: Brian Perry, Bruce Hilty
Booster Systems Officer: Jenny Howard
Primary TAL site: Zaragosa, Spain
Liftoff: 85/210:21:00:00
Landing: 85/218:19:45;26
Duration: 7:22:45:26 127 orbits
Payload: Spacelab 2 - Contained several experiments to conduct research in
plasma physics, solar astronomy, and astrophysics. Included the
deployable/retrievable Plasma Diagnostic Package (PDP)
Weights: Orbiter at Liftoff: 252,628 lbs
Cargo: 34,400 lbs
PDP: 628 lbs
Pre-Deorbit Burn: 225,398 lbs
Orbit: 142.9 x 108.7 nautical miles, 49.5 degrees inclination, following
the OMS-2 burn. The highest apogee on orbit was 173 n. mi.
The following is an excerpt from an article written by Brian Welch which was
printed in the August 9, 1985 issue of the JSC Space News Roundup:
The Challenger launched at 5pm EDT on July 29 after having been delayed for
more than one hour so that a software patch to one of the ship's solid rocket
booster gyro assemblies could be properly executed.
Weather was good at the launch site and onlookers later said the hot Florida
summer afternoon seemed particulary conductive to relaying the sound and
vibration of a Shuttle liftoff.
At the Flight Dynamics console, in the Mission Control Center, Perry and
Hilty were among the busiest people in the room. Hilty, monitoring the output
from the Abort Region Determinator, the ARD, was seeing displayed before him
continuous updates on the various abort options available during any launch.
The ARD cranks out guidance solutions for each of the four about cases, updating
those solutions every second.
Perry meanwhile was monitoring Challenger's trajectory, watching for lofting
or depression as she made her way to space. As he watched the total vehicle
velocity change, he also was watching the ground guidance model. "We model the
guidance solution on the ground completely independent of the vehicle," he
explained. "We watch the relative compare between the ground and the Shuttle,
and depending on what situation you are in, you then choose which solution to
take."
Three tiers up at the Booster console, Howard was watching the parameters of
the main engines as they were sent down via telemetry from onboard sensors to
the ground. The sensors, two on each engine, are actually very thin platinum
wires. Resistance measurements are made which provide an indicator of pump
temperature for each engine. As the temperature increases, so does the
resistance. The controller logic aboard Orbiters is able to tell, within
certain limits, a valid reading from one which indicates the sensor wire is
broken or has otherwise failed.
The rationale behind the engine sensors is to protect against worst case
shutdowns - to react with the speed of a computer to prevent an engine from
going beyond redline temperatures and exploding. The problem, however, is that
sensors are metal, and have to withstand very high temperatures but still allow
an electronic system to sense thermal changes. In some cases, temperatures can
surpass the melting point of these metals. The technology of the transducers
has to be pushed to the limit in order to provide sensory data to the main
engine controller and to the ground.
At 3 minutes, 31 seconds, the high pressure fuel turbopump turbine discharge
temperature B transducer failed. This put the center engine on a single string
for redline shutdown redundancy. Two minutes, 12 seconds later, the other
sensor failed, taking the center engine down with it. Automatically, sensors on
the two remaining engines were inhibited by the system, to guard against the
posibility of other sensor failures and the loss of the other engines.
"Center engine down," Howard called on the Flight Director loop.
"Houston, we show the center engine thermal," Commander Gordon Fullerton
radioed to the ground almost at the same moment.
"We copy, stand by," said capcom Dick Richards.
At this point, Perry, Hilty and their back room support were well into their
drill. Challenger was at about 58 nautical miles in altitude, about 275
nautical miles down range of KSC. The ship had already passed a point in the
launch phase that allowed safe return to KSC and was within a 45 second window
where it would be necessary to dump orbital maneuvering system fuel if one
engine went down. At that point, Challenger was capable of making orbit on two
engines. One critical calculation which had to be made, however, was to figure
the underspeed and determine the ever changing impact point of the external
tank.
It took only a matter of seconds for Perry and his flight dynamics team to
analyze all of this data and assess all of the ramifications. "Flight, we're
ATO," Perry said.
There followed a very quick succession of events, only seconds after the
center engine had gone to zero thrust. Lacefield told Capcom to abort ATO.
"Abort ATO. Abort ATO," Richards radioed Challenger.
Fullerton responded by turning a rotary dial to a setting marked "ATO" and
then punched a button to execute the command. Challenger responded by immedi-
ately dumping 4,400 pounds of Orbital Maneuvering System (OMS) fuel through the
OMS engines, igniting the fuel as it went. This had the effect of providing
some additional thrust, but that effect was small compared to the loss of the
weight. The setup was now complete for the proper disposal of the tank. The
total time from shutdown of the center engine to analysis and decisions by two
flight controllers to instructions from Richards to Fullerton for selection of
the ATO mode and finally, to dumping the fuel: 25 seconds.
"It doesn't take long to make the call," Perry later said. "It depends on
where we are in the launch phase. In this case, we were fairly close to the
boundary where we wouldn't have to dump fuel. It took a few seconds to confirm
that and then decide to go ATO with a fuel dump on the fly."
At 7 minutes into the flight, Challenger reached the single engine trans-
Atlantic landing (TAL) capability, meaning she could make a landing in Europe
even if another engine were to go down. "After that point, we could absolutely
avoid the water, even if a second engine failed," Perry said. "There was never
a time when we didn't have an acceptable place to land. The problem was where
the tank would go."
"Challenger, Houston," Richards broadcast. "Single engine TAL capability."
"Roger, single engine TAL," Fullerton said.
"Roger that, and main engine limits to enable, Gordo," Richards replied.
That last call to the crew was based on Boosters desire to again engage the
sensors on the two remaining engines. "When we got to single engine TAL,"
Howard said, "we then reenabled the sensors because we could make TAL on only
one engine. We don't want to inhibit (the sensors) unless we really have to.
One of our most important jobs is to figure out where that switch should be."
Howard's hardest job was still ahead of her. Between 8 minutes and 8 minutes
10 seconds into the flight, Challenger flew beyond the point where she could
absorb another engine loss and still dump the tank safely. "She had flown out
of single engine TAL capability with ET protection," Lacefield said. There was
no choice but to keep climbing uphill, to make for space.
Five seconds after that, a transducer failed on engine No. 3. "At that point,
the parameters from the engines looked good, but I now had lost 3 out of 6
sensors," Howard said.
The remaining sensor on No. 3 engine then recorded a temperature excursion
near the engine redline. Like all such sensors, it was taking a sample at the
rate of once each second. In order to shut the engine down, the sensor had to
record temperature violations on three successive samples, and then pass a logic
test from the onboard controller. If the controller makes an electronic
decision that the sensor is telling it the truth, it will shut the engine down
to prevent an explosion.
Jenny Howard, within the scope of a few seconds, was seeing exactly the same
data as the engine controller. She had 3 sensors down out of 6. She had one
main engine down and another that seemed to be wavering. Her one remaining
sensor on that engine seemed to be behaving like the ones on the center engine
had. Challenger was, in those seconds, not capable of a TAL with safe ET
disposal. They had to go to orbit, and they had to have two engines. It was
time for a decision.
"Flight, limits to inhibit," she called.
There is a economy of language on the loops in the flight control world that
approaches a practiced art. To the layperson, a terse phrase such as "limits to
inhibit" would not mean much. To the flight controllers, the call for inhib-
iting the sensors was an immediately recognizable signal carrying with it many
ramifications. It was a call for quick action. Flight Director Cleon
Lacefield, now standing at his console, turned to Richards and repeated the
instruction for the crew. "Capcom, limits to inhibit."
"Challenger, Houston," Richards radioed immediately. "Main engine limits to
inhibit."
In those few moments, a gamble was underway. The team was gambling - perhaps
not against great odds, but who could say at the time - that a human being could
monitor the engine parameters, armed with an additional data point that
Challenger's onboard controller could not possess - the knowledge that there was
only one acceptable way to go, and that was up. They were banking on the faith
that a highly trained human being could monitor the data and make a decision as
fast as a computer could if the engine really did go past the redline.
At 9 minutes, 42 seconds, Challenger reached main engine cutoff. Her inertial
velocity was 25,760 feet per second, 110 feet per second short of her target,
but that could be made up with an OMS burn. After MECO, she came off the
external tank at 4 feet per second, then translated away at another 5 to 10 feet
per second. Tumble valves opened on the external tank, giving it a tumble of 2
revolutions per minute. The tumble helps controllers model its entry character-
istics. The tank then traveled halfway around the Earth and landed in the
Indian Ocean south of Australia.
Richards radioed Challenger, "No OMS-1 required. APU's off on time," he
said. There was no immediate response.
Richards called back with a radio check.
"Okay, you're loud and clear," Fullerton responded cheerfully. "No OMS-1 and
APU's off on time. Sorry, we were busy with the tank DTO."
Only a few seconds after experiencing the first in-flight abort case in the
Shuttle program, Fullerton and Pilot Roy Bridges were cooly fulfilling a minor
test objective, to photograph the tank as it separated, in order to help
engineers get an idea how tank insulation weathers the launch phase.
A side note: The last abort simulation practiced by the crew and controllers
just three days before the flight involved an abort to orbit with the center
engine out, very similar to that experienced on the launch day.
--
Pat Oliver - Lockheed Engineering and Sciences Company at NASA JSC
2400 NASA Rd One, Houston, TX 77058 (713) 483-3323
OLIVER@vf.jsc.nasa.gov