CREWSURVIVAL

CREWSURVIVAL

¶Issue

¶Was the accident of STS 51-L on January 28, 1986, survivable? Findings

¶In the case of the tragic loss of the Space Shuttle Challenger and her crew on January 28, 1986, the Committee is convinced that the accident was not survivable. Discussion

¶During the first two minutes and eight seconds of Shuttle flight the two Solid Rocket Boosters provide approximately three million pounds of thrust. That thrust is transferred to the External Tank through the forward Solid Rocket Booster/External Tank attach- ment structure. The thrust is then transferred to the Orbiter through the External Tank. While NASA has established a "Fast- separation sequence'' to allow the Orbiter to separate from the Ex- ternal Tank and Solid Rocket Boosters, the engineering analysis in- dicates that if separation was attempted while the Solid Rocket Boosters were still firing, the Orbiter would "hang-up" on the for- ward attachment structure. This would lead to a violent maneuver which would greatly exceed maximum aerodynamic loads on the Orbiter with resulting structural failure and loss of Shuttle and crew.

¶During the course of hearings before the Committee on the acci- dent the question of survivability was frequently raised. The Com- mittee accepts the view of Captain Robert L. Crippen, who in- formed the Rogers Commission:

¶I've said this before publicly, and I'll say it again, I don't think I know of an escape system that would have saved the crew from the particular incident that we just went l b NASA, Leslie Kampschror, "Space Transportation System Mission 51-L Structural Recon- struction and Evaluation Report,''Enclosure 8 in "STS 51-L,Data and Design Analysis Task Force Search, Recovery and Reconstruction Team Report, Volume IV, Rev. A, May 8, 1986, p. 3.

77

through. I don't think it is possible to build such a system.42

¶Specifically, the Committee finds that the Space Shuttle System was not designed to survive a failure of the Solid Rocket Boosters during the first 2 minutes of flight; that is, until all the solid rocket propellant fuel has been expended. There were no corrective actions that could have been taken once the boosters ignited. The Challenger was not equipped with any means for separation during the first two minutes of flight. In addition, the crew did not have any means to escape from the Orbiter during this first-stage ascent. Neither the Mission Control Team nor the 51-L crew had any warning of impending disaster. Even if there had been warning, there were no actions that could have been taken to save the crew.43 (The issue of launch abort and crew escape is discussed in Section VI .A.2.d.)

¶Joseph P. Kenvin of NASA's Johnson Space Center summarized the circumstances in a memo to Rear Admiral Richard H. Truly, Associate Administrator for Space Flight. The undated memo read as follows:

DEARADMIRAL TRULY: The search for wreckage of the Challenger crew cabin has been completed. A team of engi- neers and scientists has analyzed the wreckage and all other available evidence in an attempt to determine the cause of death of the Challenger crew. This letter is to report to you the results of this effort. The findings are inconclusive. The impact of the crew compartment with the ocean surface was so violent that

¶evidence of damage occurring in the seconds which fol-

lowed the explosion was masked. Our final conclusions are: The cause of death of the Challenger astronauts cannot be positively determined; The forces to which the crew were exposed during Orbiter breakup were probably not sufficient to cause death or serious injury; and The crew possibly, but not certainly, lost conscious- ness in the seconds following Orbiter breakup due to in-flight loss of crew module pressure. Our inspection and analyses revealed certain facts which

¶support the above conclusions, and these are related below:

The forces on the Orbiter at breakup were probably too low to cause death or serious injury to the crew but were sufficient to separate the crew compartment from the for- ward fuselage, cargo bay, nose cone, and forward reaction control compartment. The forces applied to the Orbiter to cause such destruction clearly exceed its design limits. The data available to estimate the magnitude and direc- tion of these forces included ground photographs and measurements from onboard accelerometers, which were lost two-tenths of a second after vehicle breakup.

¶42 Rogers Commission Report, Volume V. p. 1431.⁴⁵⁴⁵ Ibid,. Volume 11, p. K-23. 4 6 Ibid. 4 7 NASA, The Apollo Spacecraft, A Chronology, 1978, Volume IV, p. 319, Appollo 12. bid., Volume I, p. 180.

78

Two independent assessments of these data produced very similar estimates. The largest acceleration pulse oc- curred as the Orbiter forward fuselage separated and was rapidly pushed away from the External Tank. It then pitched, nose-down and was decelerated rapidly by aerody- namic forces. There are uncertainties in our analysis; the actual breakup is not visible on photographs because the Orbiter was hidden by the gaseous cloud surrounding the External Tank. The range of most probable maximum ac- celerations is from 12 to 20 G's in the vertical axis. These accelerations were quite brief. In two seconds, they were below four G's; in less than ten seconds, the crew compart- ment was essentially in free fall. Medical analysis indi- cates that these accelerations are survivable, and that the probability of major injury to crew members is low.

¶After vehicle breakup, the crew compartment continued its upward trajectory, peaking at a n altitude of 65,000 feet approxi- mately 25 seconds after breakup. It then descended striking the ocean surface about two minutes and forty-five seconds after break- up at a velocity of about 207 miles per hour. The forces imposed by this impact approximated 200 G's, far in excess of the structural limits of the crew compartment or crew survivability levels.

¶The separation of the crew compartment deprived the crew of Orbiter-supplied oxygen, except for a few seconds supply in the lines. Each crew member's helmet was also connected to a personal egress air pack (PEAP) containing a n emergency supply of breath- ing air (not oxygen) for ground egress emergencies, which must be manually activated to be available. Four PEAP's were recovered, and there is evidence that three had been activated. The nonacti- vated PEAP was identified as the Commander's, one of the others as the Pilot's, and the remaining ones could not be associated with any crewmember. The evidence indicates that the PEAP's were not activated due to water impact.

¶It is possible, but not certain, that the crew lost consciousness due to a n in-flight loss of crew module pressure. Data to support this is:

The accident happened a t 48,000 feet, and the crew cabin was at that altitude or higher for almost a minute. At that al- titude, without a n oxygen supply, loss of cabin pressure would have caused rapid loss of consciousness and it would not have been regained before water impact. PEAP activation could have been a n instinctive response to unexpected loss of cabin pressure. If a leak developed in the crew compartment as a result of structural damage during or after breakup (even if the PEAP's had been activated), the breathing air available would not have prevented rapid loss of consciousness. The crew seats and restraint harnesses showed patterns of failure which demonstrates that all the seats were in place and occupied at water impact with all harnesses locked. This would likely be the case had rapid loss of consciousness occurred, but it does not constitute proof.

¶Much of our effort was expended attempting to determine wheth- er a loss of cabin pressure occurred. We examined the wreckage carefully, including the crew module attach points to the fuselage, the crew seats, the pressure shell, the flight deck and middeck floors, and feedthroughs for electrical and plumbing connections. The windows were examined and fragments of glass analyzed chemically and microscopically. Some items of equipment stowed in lockers showed damage that might have occurred due to compres- sion; we experimentally decompressed similar items without con- clusive results.

79

¶Impact damage to the windows was so extreme that the presence or absence of in-flight breakage could not be determined. The esti- mated breakup forces would not in themselves have broken the windows. A broken window due to flying debris remains a possibili- ty; there was a piece of debris imbedded in the frame between two of the forward windows. We could not positively identify the origin of the debris or establish whether the event occurred in flight or at water impact. The same statement is true of the other crew com- partment structure. Impact damage was so severe that no positive evidence for or against in-flight pressure loss could be found.

¶Finally, the skilled and dedicated efforts of the team from the Armed Forces Institute of Pathology, and their expert consultants, could not determine whether in-flight lack of oxygen occurred, nor could they determine the cause of death.44

SABOTAGE

¶Issue

¶Could the accident have been caused by sabotage, terrorism, or foreign covert action? Finding

¶The Committee is convinced that there is no evidence to support sabotage, terrorism or foreign covert action in the loss of the Chal- lenger. Discussion

¶The Committee carefully reviewed all of the evidence, classified and unclassified, to ensure that there was no sabotage associated with the loss of the Space Shuttle Challenger.

¶Committee staff met with the Director of Safety, Reliability and Quality Assurance, and the Director of Protective Services and his staff to review the National Resource Protection Plan for the Ken- nedy Space Center. The Committee is concerned with the vulner- ability of the Space Transportation System and endorses the efforts being taken by NASA to provide adequate protection to all ele- ments of the system.

ADDITIONAL OF INVESTIGATION AVENUES

¶Issue

¶Could the accident have been caused by some failure other than failure of the joint between the casings?⁴⁴⁴⁴ Ibid., Volume I, pp. 192-93. This memo was part of a package release, NASA, 86-100, draft, July 21, 1986.

80

¶Finding

¶As of September 15, 1986, the Committee has not found any cred- ible evidence to support any cause of the Challenger accident, other than the failure of the aft casings joint in the right-hand Solid Rocket Booster. Nor has there been any substantial evidence of a secondary or parallel failure on Flight 51-L. Discussion

¶After the accident, the Committee waited until the Rogers Com- mission completed its work in order not to interfere with the progress being made by the Commission appointed by the Presi- dent. By the time that work was completed the preponderance of evidence clearly pointed to a failure in the field joint of the right Solid Rocket Booster. However, the Committee was obligated to ex- plore other possibilities which could have led to the same type of failure. Among these possibilities were the following: a failure of the propellant, a structural flaw in the steel casing, and separation of the NBR insulation from the casing. In addition, the Committee was contacted by and sought additional testimony from private cit- zens who offered their concerns and hypotheses pertaining to the cause of the Challenger accident. These included the following: either a main engine fire or a fire in the main engine compartment of the Orbiter, inadvertent firing of an OMS engine, inadvertent firing of one or more thrusters on the Orbiter, overloading of the aft field joint due to excessive "moment' developed in transit of the Shuttle from the VAB to the launch pad and the use of four separate propellant casings instead of one large casing without field joints.

Main Engine Fire

¶The photographs in Volume I of the Rogers Commission report on pages 26 and 27 indicate a bright spot in the vicinity of the main engine compartment. Photographic evidence is customarily taken to be accurate. In this case, however, it must be realized that the photographs were taken from roughly three miles away and that they were enhanced by computer methods. Computer enhance- ment has the ability to highlight bright objects and subdue dull ones. In this way, the photographs become distorted, that is, the difference between light and dark becomes unrealistically pro- nounced. The bright spot in the photographs does, in fact, look like a flame. The second consideration is that the orientation between the Orbiter and the ground where the cameras were is difficult to visualize and leads to erroneous conclusions. During flight the main engines are monitored continuously for changes in pump speed, temperature and pressure. There was no indication whatso- ever of a malfunction with the main engines. A fire in the main engine compartment outside of the engines is not credible because of the lack of combustible material to support a fire of any appre- ciable magnitude. The exception would be a hydrogen leak. But, that was not supported by telemetry data. In addition, NASA has submitted photographs to the Committee from four past successful launches which show the same bright spots. The Committee, there- fore, has rejected this as a cause of the accident or as an independ- ent problem.

81

Independent Firing of the OMS Engine or Orbiter Thrusters

¶The theory that either the OMS engines or the Orbiter thrusters were inadvertently activated and fired is also based on the same photographs stated previously. Those photographs show a bright spot in the same general area where these engines and thrusters are. The Committee has received photographs from Flights 41-G, 61-A, 61-13, and 61-C, all of which show similar "bright spots') in the same location as those seen on Flight 51-L. The Committee is still evaluating the possibility of a second failure in this regard and has requested additional telemetry data from Flight 51-L. Had the thrusters been firing, however, it would have had little impact on the launch of the Challenger. The thrust from these tiny engines is insignificant compared to the thrust from the two Solid Rocket Boosters and the main engines. The inadvertent activation of the OMS engine has been ruled out on the basis of telemetry data re- ceived from NASA. NASA has stated that the bright spot seen in the photographs is a reflection from the plume of the Solid Rocket Booster motors. Neither of these possibilities contributed to the Challenger accident.

3. Overloading of the Joint

¶It is true that in transit from the VAB to the launch pad, the Shuttle system, standing erect on the launch platform and being carried by a crawler, does experience a left-hand turn. At that time, because of the configuration of the Shuttle system, an addi- tional moment, that is force times a distance, is transferred to the field joints including the aft joint on the right-hand Solid Rocket Booster that failed. However, this moment exerts a force which is only 10 percent of the force that the joint receives during other phases of the launch operations. The Committee concluded that this had no impact on the Challenger accident.

Insulation Debonding

¶The Committee investigated the possibility of separation of the insulation from the inside of the motor casings as a potential cause of the Flight 51-L accident. Had the insulation broken lose from the casing, there would have to have been a condition which would have permitted the burning propellant gases to get between the in- sulation and the casing. Furthermore, there would have to be a continuous gas flow at that point for the propellant gas to transfer a sufficient amount of heat to the casing to cause a failure. This would require an extremely large debonding of the insulation which has never been seen on any Shuttle flight when the Solid Rocket Motors were returned and disassembled for use later. When the Shuttle motors were inspected after usage, what remained of the insulation has always been in place with little damage. A de- bonding accident would have had to provide tremendous amounts of heat and again would have required a very high flow of the gas into the area where the debonding occurred. That flow of gas would have to be continuous and there is no rationale for envisioning how that could happen. In the case of the Shuttle Solid Rocket Motor design, the pressure acts to maintain the bond between the insula- tion and the casing, not to remove it. In the absence of these requirements the Committee found that debonding of the insulation was not a cause of the accident.

82

5. Crack in the Propellant

¶The Committee investigated whether or not a crack in the pro- pellant could have contributed to loss of the Challenger. A crack in the propellant would have increased the burning surface of the pro- pellant after ignition. This increase in the surface would have re- sulted in a n increase in t h thrust from the right Solid Rocket Motor. There was no evidence during the flight of 51-L of a greater thrust in the right Solid Rocket Motor. In additional, a propellant failure would have been more explosive in nature and would not have been observed as one continuous gas flame in a localized area. Consequently, it was concluded that a propellant failure did not contribute to the cause of the accident.

Crack in Motor Casing

¶The Committee was concerned that a crack in the rocket motor casings might have caused the accident if it was located in the same general area where the smoke and flame was observed during launch. All of the casings used on Flight 51-L were hydroproofed at 1.1. maximum expected operating pressure. Had there been a signficant crack in the casing it would have failed the hydroproof test. However, it could be argued that a crack developed between the test and the time the Solid Rocket Motor segments were assem- bled at the Kennedy Space Center. The failure of cracks under the pressures, such as those contained within the Solid Rocket Motors, would have been a catastrophic failure. The casings would have failed instantly at ignition because cracks in high carbon steel would propagate at a rate near the speed of sound. This is incon- sistent with the smoke seen during the early part of the launch, and the lack of smoke or flame up until 58 seconds into the launch. It is also inconsistent with the pieces of the rocket motor casings which were recovered from the ocean which clearly show the abra- sion of the hot rocket propellant gases. Consequently, a crack in the casing was ruled out as a contributing cause of the accident.

7. Joint putty temporarily holds and then releases full motor pressure

¶During the post-accident tests conducted by NASA and Thiokol, it was learned that the performance of the putty used in the joint can be quite variable. In some instances, including temperatures as warm as 75"F, the joint putty can hold back the full operating pres- sure inside the motor without transferring any of this pressure to the O-rings.45 In this circumstance, the O-rings will not "seat" and, as the joint "rotates" due to the pressure build-up within the motor, contact can be lost between the O-rings and the metal sur- faces they are meant to If the putty were then to release

¶45RogersCommission Report, Volume I, p. 64 4 6 NASA, briefings from staff.

83

¶high pressure gases into the joint, these gases could "blow-by" the O-rings, thus causing the joint to fail. This scenario is not a likely failure mode for STS 51-L, because it would produce a leak across a broad area of the joint rather than a small localized leak as ob- served in the Challenger accident.

DISCUSSION OF CRITICAL ISSUES

TECHNICAL ISSUES