posted 03-30-2013 10:26 PM            

The incidence of heat damage at the SRB joints was growing -- three of the five 1984 flights showed heat damage, eight of the nine 1985 flights, and the flight on January 12, 1986.

I find it very unfortunate that no effort appears to have been made to determine blowby damage to the remaining five 51-L SRB joints. That data, to paraphrase Feynman, "May have some relevance to our problem."

posted 04-01-2013 02:07 AM               

Now if range safety hadn't destructed the boosters, there might have been. But as it was, the remains of the boosters were mainly used to help back up the data collected elsewhere, such as from video analysis and laboratory testing.

posted 04-01-2013 06:46 AM               

quote:

---

Originally posted by PetesParkingLot:
...then I believe a case can be made that the root cause was "out of round" condition

---

posted 04-01-2013 09:19 AM               

quote:

---

Originally posted by PetesParkingLot:
Having reviewed the evidence again, I fully agree with Gene Kranz that the temperature variable is greatly over emphasized.

---

What is the source for the Kranz opinion you cite? Thank you.

posted 04-05-2013 07:22 AM            

Second, I am not attempting to prove a single factor as the cause of the 51-L failure. I am however, bothered by the historical record which states that the cold temperature was the factor which lead to failure. I believe that the cold at best was a secondary factor and may have been, as Gerry Griffin hinted, no factor at all.

I base this statement on the following facts of the 51-L flight.

1. Although all six field joints were exposed to the cold temperatures only one failed. ( The counter argument is that the cold temperature was lower for the failing joint and therefore it crossed a "magic" threshold emperature which resulted in failure)

2. Even if other joints may have been above the magic temperature and didn't fail, the Right Hand aft field joint failed at only a single point with a maximum of 18 degrees of arc. The remaining 342 degrees of arc sealed even when below the magic temperature. To me that eliminates temperature as the root cause.

3. The data I have been able to locate to date indicates that the RH aft field joint was the first time a twice flown SRB case was mated to a once flown SRB case. The more times a pre 51-L case was flown the more likely it was to have an "out of round" condition. If anyone has case use history for the first 25 flights I would very much like to have it.

4. The "out of round" condition resulted in more compression of the O-ring prior to actuation. As Professor Feynman's public demostration showed, a compressed O-ring did not return to round quickly. It would have been interesting to see him leave out the cold factor and simply show that 4 turns of compression on the C clamp had a significantly greater effect on O-ring sealing as opposed to two turns of compression.

Thiokol management was correct to reject the data that focused on cold temperatures only.

The Thiokol engineers didn't present all the relevant data. The data on temperatures did not show correlation because it measured only a secondary factor.

The primary root cause of the 51-L failure was that the RH aft field joint was in a highly compressed state at the 307 degree point. This high compression state resulted from a uniquely out of round condition that prevented it from sealing. The cold temperature was at best a contributing factor, and if we could know the state of the other field joints on 51-L, no factor at all.

Mr. Pearlman has pointed out that the Rogers Commission stated that all joints were subject to similar compression, but the data doesn't seem to support that statement. I do not believe all joints on 51-L were twice flown/once flown matings. The Rogers commision statement would be confirmed or refuted with 51-L case use data.

Mr. Behling asserts that joint rotation is the root cause. I reject that because the LH aft field joint was subject to the same joint rotation and it did not fail. If joint rotation explains the failure then one must have an explanation for why it didn't cause failure in other places. The RH aft field joint was subject to joint rotation across its entire arc and failed in only one small point.

posted 04-05-2013 09:44 AM               

quote:

---

Originally posted by PetesParkingLot:
The primary root cause of the 51-L failure was that the RH aft field joint was in a highly compressed state at the 307 degree point.

---

quote:

---

Mr. Behling asserts that joint rotation is the root cause. I reject that because the LH aft field joint was subject to the same joint rotation and it did not fail.

---

quote:

---

For a successful understanding of this technology, reality must take precedence over the historical record of the event.

---

Joint rotation with temperature effects was the cause. Period.

You fail to grasp that materials don't react exactly the same in all cases. There are micro scale effects that prevent repeatability. There will be slight material differences between o-rings, dimensional differences, placement in the groove, same with the grooves sizes, and the list goes on. It is a combination of a build up of "tolerances" that why the breech happen in this one point.

posted 04-05-2013 09:51 AM               

quote:

---

Originally posted by PetesParkingLot:
...but the data doesn't seem to support that statement.

---

posted 04-05-2013 11:00 AM            

What purpose would there be to the following commission analysis if such condition was common to all joints (failed and not):

The growth resulted in a condition at time of launch wherein the maximum gap between the tang and clevis in the region of the joint's O-rings was no more than .008 inches and the average gap would have been .004 inches. With a tang-to-clevis gap of .004 inches, the O-ring in the joint would be compressed to the extent that it pressed against all three walls of the O-ring retaining channel.

If the commission statement to the effect that all joints had similar clearance is supported by actual data then my premise may be incorrect. The gap issue then become something like the cold temperatures issue in that it was common to all joints that day and therefore isn't the root cause.

I hope to locate the tang to clevis assembly data for all joints on flight 51-l to answer this question.

As for Mr. Behling's assertion that the failure was such a complex mix of circumstances as to be just "bad luck" I believe that is a way to avoid the search for the truth. If this failure is truly too complex and mutli-varied to understand then we certainly shouldn't blame NASA or Thiokol for not anticipating such "bad luck". I do not believe that the cause is so complex that we can't identify it. I ask again if joint rotation caused the failure, why out of the 150 field joints that flew on the first 25 missions did this specific aft field joint fail at the 307 degree point? Since this was the smallest gap in the entire joint it should have been the last point to suffer joint failure from joint rotation. (Smallest gap for the O-ring to fill in the entire joint even after joint rotation.)

posted 04-05-2013 11:31 AM            

quote:

---

Originally posted by Robert Pearlman:
What data? What is the source of the data for your analysis and why do you assume it to be more comprehensive or authoritative than what the Rogers Commission had access to?

---

I am making an assumption without data that there was a relationship between case reuse and "out of round" condition. I am also assuming without data that not all the 51-L cases were in the same "used" state.

posted 04-05-2013 11:46 AM               

quote:

---

Originally posted by PetesParkingLot:
I am making an assumption without data that there was a relationship between case reuse and "out of round" condition.

---

I believe you have contradicted yourself several times throughout this thread, about whether you are trying to establish a single root cause or a prevailing cause, and whether the Rogers Commission is reliable in their findings.

posted 04-05-2013 12:38 PM            

I believe based upon the evidence I have found a condition that MAY have been unique to the 307 degree point of the aft field joint and no other joint flown prior to 51-L. As such that seems a more accurate explanation than the most of the explanations I have seen tossed around.

I am not contradicting the commission nor trying to dispute their data. I believe it is possible that the Commission allowed NASA to guide them away from the "out of round" assembly cause and toward the "cold" explanation since it allowed blame to fall on exceptional weather conditions and not on a failure to recognize a quality control issue.

posted 04-05-2013 02:25 PM               

quote:

---

Originally posted by PetesParkingLot:
As for Mr. Behling's assertion that the failure was such a complex mix of circumstances as to be just "bad luck" I believe that is a way to avoid the search for the truth.

---

quote:

---

The temperature data was not clear and a decision to scrub the 51-L mission on the basis of cold was not justified.

---

quote:

---

I am making an assumption without data

---

quote:

---

If a 747 had crashed on takeoff and one of the four engines failed from the presence of ice, would we accept ice as the "root" cause?

---

posted 04-05-2013 04:30 PM            

I have offered a logical explanation consistent with the facts as they are available as to why the RH aft field joint failed and the other five did not. I also have, at least to my own satisfaction discovered the basis for Gerry Griffin's statement that cold was not a primary factor in the accident. If seal compression from an "out of round" condition was the root cause then the accident could have happened in the July as well as January. The growing number of joint distress issues was caused not because more flights were being flown in colder temperatures, but because more flights were being flown with "used" cases.

Finally, I have a premise, subject to challenge, as to why the cold explanation has been elevated to a level not justified by the facts of the accident.

I still would like to review the case use history for STS-1 to STS-25, but just to show you how little I am seeking the "noteriety" available as a contributor to this site, I will end my contribution to this thread by repeating my paraphrase of Professor Feynman in the hope that it will trigger some readers to examine the facts and not swallow the cold temperature explanation without critical review of the facts.

posted 04-05-2013 04:48 PM               

quote:

---

Originally posted by PetesParkingLot:
I have offered a logical explanation consistent with the facts as they are available as to why the RH aft field joint failed and the other five did not.

---

quote:

---

Finally, I have a premise, subject to challenge, as to why the cold explanation has been elevated to a level not justified by the facts of the accident.

---

quote:

---

...reality must take precedence over the historical record of the event.

---

posted 04-05-2013 04:51 PM               

First of all, you don't seem to want to actually listen to what others are saying to you, simply because you don't agree with them (or, better put, they do not agree with you).

Second, you are basically saying that NASA, the Rogers Commission, etc., were incompetent but that you have all the answers.

Third, and this is really important: you are trying to get the facts to fit your ideas. That is academically wrong.

If you really wanted to have a "discussion" about this topic on this particular forum you should have approached the situation with a more open mind. Instead, you seem to want to lecture to people who probably know more about the incident than you do.

posted 04-05-2013 05:58 PM            

Each part, say in our example each SRB segment housing, is manufactured according to a specified ideal geometry, but with given tolerances. Each part is checked that the realized geometry is within the allowed bandwidth around the ideal geometry. Also the O-rings have tolerances of their hardness and behaviour as function of temperature. Furthermore, the material strength varies, so the design assumed the minimum guaranteed strength, while the real strength is usually higher but not guaranteed. The design is such that correct function is assured even for the worst possible combination of tolerances when the parts are integrated. This is valid for the specified ranges of operational conditions, of which the temperature is one important parameter. Thus, at the limit of the operational parameter ranges, a minimum margin is assured for even the worst coupling of tolerances, but usually the margin is higher since the tolerances and their coupling occur stochastically.

The engineers of Thiokol correctly stated that the 51L launch was predicted to happen under temperature for the field joints, which is out of the specified/authorized range. So for assuring the correct functioning, it would have been necessary to justify per analysis and test that for worst tolerance combination the resulting margin is still sufficient, even for out-of-authorized-range temperature.

It can be stated that the real material strengthes and the real geometry, e.g. the out-of-round conditions, for each of the field joints was different and not identical, thus he resulting margins were also different, and obviously negative for the failed field joint at one location. Not at all can it be expcted that either all field joints work, or all field joints fail simultanously.

To my understanding the failure occured due to an unfavourable combination of several geometrical, material, and operational tolerances, together with out-of-certified temperature. Perhaps the data were inconclusive whether lower temperature leads to lower margins, but the data presented by Thiokol did not at all assure that this is not the case. So it was unsure to launch and the managers took a risk, while the engineers stated the margins may be too low at predicted conditions.

I would assume that exact geometry was not measured before each launch, but that it was checked only that the geometry was within the allowed tolerance. In that case we cannot reconstruct the gap width etc. for all field joints, and so noone can demonstrate that the root cause was not connected to the temperature.

posted 05-07-2013 03:14 PM            

The NASA accident analysis team (appendix L Section 6.3) found that the failure of the aft RH field joint at approximately the 300 degree point was primarily caused by near metal to metal (a 0.004" gap) created at the time of assembly of the joint. This joint was the only joint in the 51L stack that required special assembly and the use of excessive hydraulic pressure with the “alignment tool” to join the center to aft joint (the one that failed at the 296 to 310 degree point).

This narrow gap at the 300 degree point compressed the O-ring greater than anywhere else in that joint and in the RH or LH SRB joints. This compressed state caused the O-ring to rebound slower from the compressed state to fill the gap caused by joint rotation than the 51-L O-rings portions in a more normal (0.020 inch) gap state.

Cold MAY or MAY NOT have contributed to this failure. Tests indicated that an O-ring in a normally compressed state would rebound and seal down to 10 degrees F. This means that had 51L used case segments that were not subject to the “ovality” problem, in the winter months, STS likely could have flown safely with respect to joint sealing down to 10 degrees F, a condition it never faced. This behavior makes temperature a non-relevant factor for joint sealing for flight 51-L which flew at 28+- 5 degrees. Hence, Gerry Griffin’s statement that 51-L might have failed in the middle of summer, implying that cold was not the primary factor.

Finally, it implies, for me anyway, that the decision to launch 51-L in a failed state lies at its root with the Thiokol engineers who failed to understand their joint failure modes. Had they focused on the primary factor causing the joint failure, namely “ovality” caused by joint reuse and the concurrent gap narrowing in the SRB joints, they would have been able to provide a rational explanation as to why 51-L launch should be scrubbed. As it was, by focusing on cold as a primary cause, their data did not correlate well enough to convince any intelligent engineer or manager. As a result of this mistake, they failed their managers, NASA and the astronauts.

In order to continue this line of research I need access to the case assembly data for the first 25 STS flights, something that is not readily available from sources outside NASA. Anyone who has that data I would appreciate access to it.

posted 05-07-2013 04:57 PM            

Did the Rogers Commission put a "spin" on the cause in order to shift blame away from an assembly cause (which was totally in NASA's control) to a explanation that showed a failure to deal with record cold temperatures (which was a act of nature).

When one reads Appendix L, cold is barely mentioned (outside of potential it causing ice in the joints). When one reads the Executive Summary of the Rogers Commission one is lead to the Cold temperature cause as THE cause.

posted 05-07-2013 05:16 PM               

quote:

---

Originally posted by PetesParkingLot:
Was Feynman lead to the conclusion reading cold temperatures because that is where the Rogers Commission wanted him to go?

---

posted 05-07-2013 06:58 PM            

Second, I would like to believe that Feynman had the answer to the question that has bothered me from the outset. He did not write the Commission report, the political members of the Commission did. I do tend to ask rhetorical questions that I may have some belief in to get the thoughts of other members of the site.

That question is: If cold was the cause of the failure, why did five of the six joints work when exposed to the cold and the sixth joint failed at only a single point in the 360 degrees of arc?

posted 05-07-2013 08:39 PM               

quote:

---

Originally posted by PetesParkingLot:
That question is: If cold was the cause of the failure, why did five of the six joints work when exposed to the cold and the sixth joint failed...

---

posted 05-07-2013 09:22 PM            

Cold temperature and its effect on the response of the O-ring is a sufficient explanation. It does not bother any of the readers that the cold temperature is not a variable but a constant across the O-ring and the difference in sealing outcomes of those O-ring is attributable to unexplainable, unknown factors which are beyond our ability to understand.

We simply need to accept on faith, that the cold was the cause of the 51L failure, because that is what everyone (except the NASA Accident Analysis Team) believes to be the truth. Unfortunately, for me I have a belief that the true root cause can be identified and that cold temperature is an incomplete and maybe even wrong answer.

I will not waste any more of the readers time but will leave you with one interesting fact that gives me pause: