Dusting Off a Prediction

Originally Posted 08/04/03--3:05pm CDT
There has been some tumult surrounding comments made by a member of the Columbia Accident Investigation Board. The comments pertain to the corporate culture at NASA and whether that culture can change enough to accommodate the suggestions that will be in the board's final report. It has been argued that suggestions made after the Challenger accident in 1986 were ignored, largely because of the corporate culture at NASA, and that NASA's failure to act on those suggestions was a contributing factor to the Columbia accident (There were no similar issues from a scientific standpoint, however from a management standpoint some believe that the same faulty reasoning that was employed during the Challenger incident was never corrected and was used again in determining a course of action prior to Columbia's re-entry.) I could open this can of worms, but instead, I offer the text of an email that I wrote a couple of days after the Columbia breakup.

As I do not have ready access to televised media for coverage of the Columbia investigation, I have been utilizing NASA's website to a large extent for information on that subject. I have read every PDF file, in varying degrees of completeness, and looked at every image and video file pertaining to the investigation that has been released. What follows is a summary of that information. I am writing this in part to inform you, and in part to enhance my own understanding and sharpen my own perceptions about what is known. This briefing will focus on the events of Friday, February 7, 2003 with respect to information disseminated by NASA through their website.

I read the Stanford report on "Risk Management for the Tiles of the Space Shuttle" that was written in 1994. This report made it into the news mid-week and was furnished to reporters on Friday. I can find no evidence that this report was commissioned, as such, by NASA, however, the report was given to that organization and they have used it to make changes in certain protocols relating to tile management. A list of those changes can be found within the conclusions of the report. The essential purpose of the report was to apply basic risk management techniques to assessing the vulnerability of the Shuttle's tiles to failure from debris impact, manufacturing defect, and installation deficiencies (Believe it or not, an example of an installation deficiency would be a documented case of a tile worker spitting in the RTV [Room-Temperature vulcanized adhesive] used to bond the tiles to the Shuttle's SIP [Strain isolation pad] to speed its curing time. The report notes that while the addition of liquid to the RTV DOES lower curing times, it also increases un-bonding rates.) The report also studied the impact of un-bonding tiles caused by debris impact, manufacturing defect, and installation deficiencies on surrounding tiles. That is to say, how does the partial damage or complete loss of one tile impact adjacent tiles. The report concluded with not only results of the analysis, but also practical suggestions that could be used to reduced the possibility of LOV/C (Loss of Vehicle and Crew) due to tile failure. The following is a list of the most significant findings with respect to what is already known about the disintegration of Columbia six days ago.

First, I think it is important to note that many of the issues talked about, in theory, within the Stanford report are the very same issues being talked about today with regard to the possible explanations for the loss of Columbia. I raise this point because it is very easy to view this document as predictive in nature as a result of those coincidences. However, I think it would be premature and illogical to draw any conclusions from the report other than the ones drawn by its authors. I do not believe it to be any evidence of NASA incompetence vis-a-vis what they should have known regarding the possibilities of tile failure. That is to say, while the report may seem to suggest that scientists at Stanford could foresee this disaster coming 8 years ago, I do not believe the theoretical scenarios developed in the report represent any better judgment or ability to predict or prevent LOV/C in this particular situation. Further, I do not believe NASA was derelict in its duties regarding tile management or contingency planning with regard to tile failure.

The most interesting conclusion the report draws relates to "unzipping" of tiles. This phenomenon is defined best as the loss of a patch of adjacent tiles resulting from the un-bonding of a single tile. This un-bonding could be the result of debris impact, manufacturing defect, and installation deficiencies. Essentially, the "un-zipping" occurs when the missing tile allows aerodynamic turbulence to shake adjacent tiles free or the missing tile permits plasma (super-heated ionized air) to enter the substructure of the vehicle. This intrusion of plasma can alter the bonding characteristics of the RTV causing the tiles to loosen and come free. The resulting patch of missing tiles will continue a cascading effect until either the aerodynamic character of the vehicle is so changed that proper flight attitude becomes impossible to maintain or plasma make further inroad to the aluminum substructure of the vehicle. In either event, one possible outcome is LOV/C. In the case of loss of aerodynamic character, the vehicle would show signs of escalating drag until flight systems were unable to correct for it. The result would either be complete loss of control of the vehicle or disintegration caused by dynamic forces exerted on the vehicle in the process of trying to correct the drag situation. In the case of plasma intrusion, the increased heat upon the substructure caused by missing tiles could cause burn-through, which in turn could cause loss of flight controls and other subsystems or disintegration.

Considering the description of events from telemetry, either of these scenarios seems to be possible with respect to the loss of Columbia. The sudden loss of sensors may be the result of tiles ripped loose during reentry or the result of damage to wiring systems caused by burn-through. The drag situation described in the telemetry is also consistent with tile loss. It is important to note, however, that all other systems indicated that Columbia was flying with proper attitude just prior to loss of signal even though data suggests that flight control systems were losing the battle to control the drag issue. This inability of the flight control system to make a halting correction to the drag situation may indicate that tiles were shed at a continuous rate as described above until it was beyond the ability of the flight control system to control the vehicle or the corrections being made themselves exerted more force than the vehicle's structure could withstand. If there was burn-through as the result of tile loss, the resulting heat could have caused the destruction of wiring and routing systems embedded in the substructure yielding the loss of data from those sensors. If burn-through continued, it could have led to a severing of communication paths from the flight control system to the flight control surfaces or catastrophic loss of large pieces of the vehicle resulting in LOV/C. Again, in either of these scenarios, the loss of one tile is the precipitating event. Although there have only been 2 documented cases of missing tiles after a mission (both cases involved Columbia), there are myriad ways they can be lost. In this case, the impact from insulation that was shed from the ET (External fuel tank) on assent could be a possible explanation for how a tile may have been lost that could have caused "un-zipping" on reentry.

The other interesting development was a pixelated photo of the Shuttle taken from Kirtland Air Force Base in New Mexico just minutes before loss of signal. The photo seems to show a trail of some sort behind the left wing, and some sort of defect along the leading edge of the left wing. However, when one views the photograph, it is hard to conclude anything. The photo itself is heavily pixelated and as such can reveal very little for certain.

This image is a view of the underside of Columbia during its entry from mission STS-107 on Feb. 1, 2003, as it passed by the Starfire Optical Range, Directed Energy Directorate, Air Force Research Laboratory, Kirtland Air Force Base, N. M. The image was taken at approximately 7:57 a.m. CST (1357 GMT). This image was received by NASA as part of the Columbia accident investigation and is being analyzed.

If we accept that the plume of material behind the right wing is plasma and "normal," then there definitely seems to be more material issuing from behind the left wing. The apparent misshapen-ness of the left wing may only be the result of the geometry of the angle at which the photo was taken. All things being equal, it appears as though this photo was shot from a position to the right of Columbia's terminal path. That, in and of itself, may account for what the untrained eye may see as anomalies. There were also rumors of another photo taken of Columbia by an amateur astronomer that showed some very strange sparks or arcs of light surrounding Columbia very early in reentry. When asked directly about this photo bother General Michael Kostelnik and Ron Dittemore responded that the authenticity of the photograph was being determined and that atmospheric scientists were being used to help explain any naturally occurring phenomena that might put the Shuttle at risk as it reentered the Earth's atmosphere. Assuming they had seen the photograph, my interpretation of their comments is this: Whatever is revealed in that photograph must be so incredible to behold that the authenticity of such a thing being possible is being questioned. That is just my take, but if I am correct, that photo may go a long way in solving the mystery that brought Columbia down.

There will be no briefings this weekend, and only one administrative briefing on Monday. Tuesday will mark the first briefing made by the independent commission assembled under Admiral Harold Gehmen's control. Gehmen was selected based upon his work on investigating the bombing of the USS Cole. In my estimation, although this commission was created as a contingency put into place to prevent another Challenger debacle, transfer of control from NASA to this commission will mark a significant drop in the openness of the briefings. I regard this commission with much suspicion and view its ability to cover up possible wrongdoing or incompetence to be as great and likely as its ability to gather the facts and offer a valid conclusion.