The probabilistic risk assessment for a foam shedding event causing a
crit1 failure happening in future flights is not quantifiable, simply
because the base sample for analysis has been ever changing, without
revisions reflecting those changes. The data set for analyzing the
dynamic loading events during the shuttles ascent is over 100, but the
actual missions that are similar is very few, and therefore a true
reliable base may not be achievable. ISS support missions such as
sts-121 are much different from the early missions of the program
standard weight tank SWT, and light weight tanks LWT, and therefore
many of the previous missions even sts-107's ET#93 (LWT) must be
eliminated from a data set. The original external tank design or
standard weight et was modified primarily for weight reduction, with
the light weight tanks starting flights on sts-6. But this weight
reduction was achieved mainly by a reduction on the safety factors,
from 1.4 to 1.25 for well defined loads based on flight data (thrust
loads, internal pressure and inertia loads), and kept at 1.4 for all
aerodynamic and dynamic loads. This means a 0.15 reduction for some of
the orbiter stacks elements meant the sts-1 safety recommendations were
converted from reduction in payloads, to a reduction of safety margins.
The implementation of the performance enhancements (angle of attack,
early q throttle up, 1st and 2nd stage gimbal change) for iss support
has increased loading on the orbiter stack, and previous risk analysis
must be updated to reflect the increased risks associated with such
loading increases. Updating a quantitative risk assessment with flight
data, in conjunction with the increased loading on the orbiter stack
caused by performance enhancements may give a clearer picture as to the
cause of foam shedding events, and the risk associated with sts-121.

STS-51L Challenger
The push by nasa to make the shuttle stack lift more payload weight was
implemented with sts-8, by modifying the solid rocket boosters. A srb
thrust modification was implemented to increase payload lift capability
resulting from a stronger solid fuel propellant coupled with a lighter
rocket casing.
The rogers commission concluded the information demonstrating a pattern
of "O ring" burn through was available, but this information was
not correlated, and then eventually disregarded by nasa officials in
launching sts-51L. STS-61a, sts-61b and specifically sts-61c the
programs 24th flight srb post flight inspection completed on January
12, demonstrated a pattern of obvious problems with srb casing burn
through and "o ring" failures (NSTS-22301). STS-61C flight landed
January 18, 1986, just 10 days prior to the last flight of challenger,
and the death of the sts-51L crew. The two records set that ill fated
launch day of Jan 28, 1986, still stand today, the commonly known
coldest launch temp of 36 f deg, and the lesser known fact that sts-51l
utilized a lightweight srb casing and still was the heaviest shuttle
stack to launch at 4,529,681 lbs. The rogers commission concluded the
lightweight SRB casings aggravated the "joint rotation", a spacing
in the O ring seal area that would allow the hot gases a path to the
rocket casing if filler putty had suffered blow through, a common
problem. The O ring failure occurred after nasa managers clearly
disregarded the Morton Thiokol engineer Roger Boijoly's
recommendation to not launch, demonstrating humans erred in the
decision making process a failure mode not demonstrated in the stated
risk analysis. But another the fact is the lightweight srb casings
utilized for challengers ascent jan 28, 1986 launching the heaviest
shuttle stack ever used in flight history were concluded to being
"aggravating" to the O ring failure which resulted in the death of
the sts-51L crew (Rogers commission report chapter VII Casing Joint
Design)

STS-107 was the 5th heaviest shuttle stack to launch in flight history
at 4,526,034 lbs (Caib report vol V page 105).

STS-121 will be the 7th heaviest orbiter stack in the programs
lifetime at 4,523,850 lbs (sts-121 press kit). But Given the fact
STS-121's SLWT #119 has had the (-Y) protuberance air load ramp
removed, and this ramps original purpose was to redirect undesirable
aerodynamic flow, and where flow problems can cause greater loading,
the removal of the ramp may cause more greater loading on the
orbiter/ET interface than previous missions, therefore this mission may
be at a greater risk for failure than other missions. But even after
concluding sts-121 is performing a test flight to assess the risk
associated what the caib determined was the cause for the death of the
sts-107 crew and loss of Columbia, nasa is still performing an
operational mission in iss support. Objectively performing the flight
test with sts-121 to determine true risk associated with a foam loss,
with the 7th heaviest launch in program history, could be seen as
pushing the envelope the 2nd flight after a death of crew and loss of
vehicle. A true objective test would not have a necessity to perform
an operational task such as iss support, where weight and launch window
are driving the test parameters, and therefore biasing the test data in
a risk analysis.

Establishing a mission is safe based on the fact the total remaining
missions in the programs lifetime is less than the projected failure
rate is not based on sound principles, as the mission failure can occur
in bin #1 or bin#100 and therefore does not provide a clear picture of
the true risk to the sts-121 crew or discovery. The sts-51l, and
sts-107 tragedies both demonstrate areas were safety was compromised
for performance, increasing the risk of death of crew and loss of
vehicle to occur at a higher frequency than risk analysis predicted.
Understanding what safety compromises in the space shuttle program have
been made to bring back preprogram expectations, demonstrates the
probability of failures increased due to human subjectivity pushing the
system, instead of objectively working with the systems limitation for
better or worse. Ethical engineering, science, maintenance
practices/standards in designing and operating space vehicles and their
programs should not allow for acceptance of astronaut deaths, or should
anyone provide an illogical, circular excuse such as "it's a
dangerous profession" when tragedies happen. Every life is precious,
and risks associated with the loss of life whether it is your own or
somebody else's should not be trivialized. As we are all members of
the "space interested community" it is important for all of us to
understand why the previous tragedies happened, because we cannot allow
them to be repeated.
Tom

Tom

wrote:
"columbiaaccidentinvestigation"'s argument seems to be that any changes
to the shuttle, even ones made with the objective of making it safer,
put it into a flight regime that has not been tested. The only way to
address this concern is to never change anything ever.

Is this the way to do anything new in space?

The point of the launch decision is that even in the case of foam
shedding from the tank impacting the thermal protection system, the
effect will be on re-entry-- the launch itself is safe up through the
time the shuttle gets to orbit. As noted, if there is a foam impact
event, the astronauts transfer to the space station, and shuttle is
inspected on orbit to determine the nature and extent of damage. It's
then either reparied, or the astronauts are returned by a different
mission. Uncerrtainty about the foam is a question about a possible
loss of the vehicle, and not loss of astronauts.

Since "columbiaaccidentinvestigation" is arguing that any changes
whatsoever put the shuttle into a regime where the only way to validate
the change is by flying ("columbiaaccidentinvestigation" claims not to
trust wind tunnel results), this seems the only sensible thing to do.

--
Franklin Jefferson
***My blog: Jefferson's Democracy***
http://franklinjefferson.blogspot.com