RCC Impact and the Boeing Study

[Craig Fink]

From Page 8 of the Boeing Study data 1/21/2003



RCC Predicted Damage at Incidence Angles Greater than 15 Degrees Based on
Ice Database

Angle Velocity Damage Depth (in)
5 720 0.11
10 720 0.18
15 720 0.23
20 720 0.28
25 720 0.33



Size =20 x 10 x 6 Density =2.4 lb/ft 3

45 ° angle of wing was taken into account

Nominal panel thickness is 0.233 in.

RCC is clearly capable of withstanding impacts of at least 15 degrees;

relative softness of SOFI (compared to ice)would indicate greater
capability

" Maximum reported angle of 21 degrees is not an problem "

Looking at using Window ice and RTV data as an analog

End Quote.



Notice that the damage depth for 15 degree impact is 0.23 inches while the
Nominal panel thickness is 0.233 inches. Essentially, this chart say that
any impact greater than 15 degrees will fully penitrate the RCC leading
edge. An impact of 25 degrees would penitrate (0.33 inches) an RCC panel
that is 41% thicker than the nominal RCC panel (0.233 inches).

After the table shows that the RCC panel will be penitrated, a bunch of
hand waving and wishful thinking is used to make the problem disappear.

Even an impact of only 5 degrees has damage depth of 0.11 inches which is
much deeper than just a loss of the coating of the RCC panel. This table
certainly does not support the statement in Summary and Conclusions on Page
13:

"RCC damage limited to coating based on soft SOFI"

A conclusion that is not supported by the data.

Craig Fink

[JGM]

Craig Fink wrote:
>From Page 8 of the Boeing Study data 1/21/2003

Actually I believe the page you quote is from the second Boeing presentation
on 1/23.

>Notice that the damage depth for 15 degree impact is 0.23 inches while the
>Nominal panel thickness is 0.233 inches. Essentially, this chart say that
>any impact greater than 15 degrees will fully penitrate the RCC leading
>edge.

This team was working with a maximum impact angle figure of 21 degrees which
came from the Debris Impact team's analysis of 2 days earlier. So 25 degrees
does represent some safety margin from the (assumed) worse case.

This apparent safety margin is meaningless however as the 21 degree max impact
angle is never meaningfully defended. The impact team defined an area coming
to within a foot or two of the leading edge of the wing (where the angles
change dramatically) as their possible impact zone. The video seems to
indicate the strike occurs *on* the lower part of the leading edge (and in fact
the first report of the impact calls out this location explicitly). This
(likely incorrect) 21 degree figure is the first in a chain of cascading
assumptions.

>Even an impact of only 5 degrees has damage depth of 0.11 inches which is
>much deeper than just a loss of the coating of the RCC panel. This table
>certainly does not support the statement in Summary and Conclusions on Page
>13:
>
>"RCC damage limited to coating based on soft SOFI"

Moreover, there is no indication that the way the Crater program works has
any meaning with respect to RCC. Crater results seem to be in the form of the
depth of "penetration" into material -- this makes sense for a foamy ceramic
tile bonded to a contiguous underlying substrate, but not for a more brittle
material mounted as a hollow shell on widely-spaced connection points. The
second hint here is buried in the caveats: there was no data available for the
results of SOFI on RCC so they used an "analog" from the existing "ice
database". It's not clear how this analog was defined -- they call out an
assumed density of 2.4 lb/ft^3 (the density of the foam rather than the ice)
but then claim that the "relative softness of SOFI compared to ice" makes these
results conservative. Did they use the data for a (much smaller) piece of ice
with a similar *mass* to the foam block? If so then the "softness" assumption
is a mistake since the area of impact and thus the structural strain will
potentially be much greater with a larger foam block. Another clue is seen two
pages earlier in the presentation when the analysis team states "flight
condition is significantly outside of test database" -- this statement appears
to apply to the SOFI-on-TPS data (since there *is* no SOFI-on-RCC data) but
it's clear that there is no real understanding of the effects of large debris
pieces at all. So now we have wildly extrapolated data fed into a
(probably inapplicable) simulation calculating the results of a (probably
incorrect) impact angle.

Note that every person in this chain probably thought that they were clearly
documenting their assumptions and the limitations of their analyses. Everyone
using that data (for further analysis or decision making) seemingly paid no
attention to these caveats.

JGM

[James Oberg]

I think you guys are right on target. Maintain firing rate.

[Joe D.]

"James Oberg" > wrote in message . ..
>
> I think you guys are right on target. Maintain firing rate.
>
We've discussed several times, as far back as March, although
then we usually focused on tile not RCC. The issue is: the Boeing
Crater *data* (not the conclusion) shows possible serious or fatal TPS damage.
Yet the conclusion drawn was "no flight risk". Exactly how they got from
this data to that conclusion, I don't know.

http://tinyurl.com/hvuv

-- Joe D.

[Brian Gaff]

| "James Oberg" > wrote in message
. ..
| >
| > I think you guys are right on target. Maintain firing rate.
| >
| We've discussed several times, as far back as March, although
| then we usually focused on tile not RCC. The issue is: the Boeing
| Crater *data* (not the conclusion) shows possible serious or fatal TPS
damage.
| Yet the conclusion drawn was "no flight risk". Exactly how they got from
| this data to that conclusion, I don't know.
|
| http://tinyurl.com/hvuv
|
| -- Joe D.

Well, I don't understand it either, as presented with what you folks just
wrote, the alarm bells say, insufficient data for a valid decision. It does
indeed imply that the figures entered were taken as gospel, and nobody
actually looked, as it were, at the small print that qualified the figures.
Maybe when the document was drawn up, some engineer should have in fact
said... Hold on, the huge variation in the possible parameters from the
various sources mean I cannot let these figures stand. Someone needs to do
the tests for foam on the actual substances in use and thus get these error
margins down.

Brian

--
Brian Gaff....
graphics are great, but the blind can't hear them
Email:
__________________________________________________ __________________________
__________________________________


"Joe D." > wrote in message
...




---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.502 / Virus Database: 300 - Release Date: 18/07/03

[Terrence Daniels]

"Joe D." > wrote in message
...
> We've discussed several times, as far back as March, although
> then we usually focused on tile not RCC. The issue is: the Boeing
> Crater *data* (not the conclusion) shows possible serious or fatal TPS
damage.
> Yet the conclusion drawn was "no flight risk". Exactly how they got from
> this data to that conclusion, I don't know.

See no evil, hear no evil? Don't rock the boat? Seeing what they WANT to see
maybe? I'm over-simplifying, to be sure, but I get the feeling - given
what's been coming out lately about the management meetings and suchlike -
that a lot of people were somehow unwilling to raise an alarm in this whole
situation.

I'm wondering if it's a group psychology issue. My father has chalked it up
to institutional arrogance already, and I think he might be right in the
sense that the people doing the work didn't want to assume that anything
might be wrong, or that the situation might be out of their control. Combine
that with a management that seems to have unquestioningly rubber-stamped
"Everything A-OK" without following up leads on engineering dissent or
asking hard questions, and I can see how we get to seven dead heroes and
another orbiter reduced to pieces on a hangar floor somewhere.

[James Oberg]

Excellent post, thanks. Yeah, even the CAIB was astonished at what the
post-disaster test data -- with a real wing structure, not JUST a single RCC
panel -- showed would happen. It would have shown the same terrifying
results last year, or twenty years ago, if NASA had thought it necessary to
run the test then.



"Patrick McConnell" > wrote in message
news:bfqjjg$cs1
> It's actually depressing how useless even the biggest and most
sophisticated
> numerical models are at predicting how strong things are. You gotta have
> test data.

[Patrick McConnell]

> Even an impact of only 5 degrees has damage depth of 0.11 inches which is
> much deeper than just a loss of the coating of the RCC panel. This table
> certainly does not support the statement in Summary and Conclusions on
Page
> 13:

I've been in rooms with groups of guys (yes always guys) trying to hash out
if something is ok or not. One thing you have to realize is that if absolute
scientific rigor was the standard for making decisions about structures,
nothing would fly. So we use engineering judgement.

At times when I was a 25 yr old kid trying to represent the stress
department, I'd get reamed out by various managment types for obfustication,
because, hey, I was inexperienced and I didn't want to say anything I
couldn't back up with a calculator. Luckily my manager was good natured
enough to take the heat when he got the call about that uncooperative
philadephia lawyer he'd sent to the meeting. I was actually called that
once. In a loud voice. By a manufacturing VP. With spit flying out of his
mouth.

After a while, through an osmotic process some call experience, I learned
that a) you have to pick your battles, and b) you can't always prove
everything you know to be true about structures. Like for example, a simple
bolt is 2 or 3 times stronger in a joint than it should be if you just look
at the bending, shear, and clamp-up stresses. If I go around telling people
they have to add 3 times more bolts than anybody else uses, I'll be flipping
burgers in no time.

So...when I look at the data in the ref post and the conclusion that
followed, I can just tell there was something going on that we don't know
about. And im not talking about dark secrets and cowardice, I'm talking
about somebody's judgement that such an impact on the RCC was
inconsequential. Why they made that judgement is unknown to us but I'd say
it was made in good faith. The supporting data presented here, if thats all
there is and who knows if thats the case, is just paperwork for bosses to
feel good about.

I would hope the call was made by someone who knew all about the design,
analysis, and testing that was done on the leading edge. If they simply made
the wrong call, as I think is the most likely thing, then a lot of people at
NASA seem to be falling on their swords for nothing. Why that would happen
is a bit mysterious, but it probably has to do with the fact that few people
realize how touchy-feely structures work can be. And that doesn't fit in
with the high-tech outer-space computerized nature of the NASA's work that
everyone, including it's overseers, like to think they do.

It's actually depressing how useless even the biggest and most sophisticated
numerical models are at predicting how strong things are. You gotta have
test data. If someone wants to pony up about a billion, i'd guess, Boeing
can generate all the test data to cover any imaginable scenario that they
might, maybe, need to cover at some time in the future for the entire
structure of the STS.

Patrick

[Jan C. Vorbrüggen]

> That's essentially correct, although I'd suggest that they both reacted as
> systems -- just differently. The relative flexibility of the fiberglass
> that allowed the impact to be absorbed over a larger part of *its* system.
>
> The RCC is so stiff that it wasn't able to transfer much of the impact
> energy to adjacent panels, leaving the target panel to absorb virtually all
> of the impact.

Indeed. My point was that after the initial fibreglass test, the thinking
was that only an integrated test would have shown problems. After the RCC
test, a unit test, as it were, seems to have been on order. And due to its
size and shape, panel 8 would have been the prime candidate (in the sense
of a worst case).

Now, what parameters one would have selected prior to 1 February 2003 for
such a test (object type, mass, size, speed, impact angle, ...) is quite
another but interesting question.

Jan

[starman]

James Oberg wrote:
>
> "Brian Gaff" > wrote
> > Well, I don't understand it either, as presented with what you folks just
> > wrote, the alarm bells say, insufficient data for a valid decision. It
> does
> > indeed imply that the figures entered were taken as gospel, and nobody
> > actually looked, as it were, at the small print that qualified the
> figures.
> > Maybe when the document was drawn up, some engineer should have in fact
> > said... ................. Someone needs to do
> > the tests for foam on the actual substances in use and thus get these
> error
> > margins down.
>
> Or failing that -- and there really wasn't time to do it once the flight
> began -- you'd BETTER assume the worst until proven otherwise.

Being the person with the responsibility of assuming the worst can be a
gut wrenching experience. It's guaranteed to make you unpopular with at
least some of your fellow employees. What could NASA do to make this
process less difficult for those involved?


-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----

[Sal Bruno]

"Patrick McConnell" > wrote in message >...

> So...when I look at the data in the ref post and the conclusion that
> followed, I can just tell there was something going on that we don't know
> about. And im not talking about dark secrets and cowardice, I'm talking
> about somebody's judgement that such an impact on the RCC was
> inconsequential.

> I would hope the call was made by someone who knew all about the design,
> analysis, and testing that was done on the leading edge.

> It's actually depressing how useless even the biggest and most sophisticated
> numerical models are at predicting how strong things are. You gotta have
> test data.

Yes, but if the person who made the call had absolute authority, would
he have opted to gather more data? Probably so.

I guess in a perfect world NASA would have had their top RCC and TPS
expert, as a deputized representative of the administrator, 'sit-in'
on this analysis or at least rigourously review it. Any doubts (and I
think their would be) as to it's probability of accuracy would have
been referenced to the administrator and steps taken to acquire
additional empirical data before rendering a conclusion.

Not practicable for every investigation, but on the major
criticality-one issues, I think it should be a future requirement,
certainly if in-flight.

I really think the administrator needs a team of deputized
representatives sitting in on more meetings. They would have the
specific job of reviewing analysis and 'finding problems', and
involving the administrator to logically take the required actions
using his authority.

[Derek Lyons]

(Sal Bruno) wrote:
>I really think the administrator needs a team of deputized
>representatives sitting in on more meetings. They would have the
>specific job of reviewing analysis and 'finding problems', and
>involving the administrator to logically take the required actions
>using his authority.

What a *wonderful* idea. Now the Administrator has to take time from
his real job in order to coordinate and manage a parallel
administration.

D.
--
The STS-107 Columbia Loss FAQ can be found
at the following URLs:

Text-Only Version:
http://www.io.com/~o_m/columbia_loss_faq.html

Enhanced HTML Version:
http://www.io.com/~o_m/columbia_loss_faq_x.html

Corrections, comments, and additions should be
e-mailed to , as well as posted to
sci.space.history and sci.space.shuttle for
discussion.