posted 02-20-2013 11:37 AM            

Five of the six SRB field joints exposed to the cold of the launch successfully sealed themselves. Clearly, other factors were involved. Mr. Griffin seems to imply that there was a assembly defect which lead to the failure of the right SRB seal. If he is correct then the entire discussion about launching below 32 degrees is a red herring.

In addition, NASA management is not guilty of a flawed launch decision, but guilty of a lack of quality control, a much different failure. The same lack of quality control that lead to the Apollo 13 failure.

posted 02-20-2013 11:54 AM               

There is no doubt that the inability of the O-ring to properly seal the field joint was the root cause of this accident.

However, while the stiffness of the O-ring(s) was due to low temperatures, the stack also experienced the strongest wind shear on any launch to date. That meant the SRB's flexed more than what was experienced before. The combination of the stiff O-ring and flexing booster likely caused the accident to worsen, ending with the destruction of the machine.

posted 02-20-2013 12:02 PM            

There had to be more going on than simply launching at below 40 degrees. From the facts of the launch you could expect the seals to work 5/6 of the time even below 40 degrees. Was the true failure an assembly failure, as Gerry Griffin asserts?

posted 02-20-2013 12:03 PM               

quote:

---

Originally posted by PetesParkingLot:
Gerry Griffin, Apollo Flight Director, asserted in a NASA interview from 1999 that the cold was not the root cause of the Challenger solid rocket booster failure.

---

...this is based strictly on a gut feel. I think the temperature aspects were probably overdone. I have a feeling we would have lost that vehicle on a summer day.

I think there was a bad seal. We knew the design was close. It was marginal, and we were working on an improved interface between the segments. But we had flown lots of things and, gosh, clear back to — in my timeframe, Gemini and Apollo, Skylab, we had flown things that we knew design was — you know, we wish it was beefier — we wished it was — but we thought it was okay. And so I really think, you know, they had trouble making that seal at the Cape.

posted 02-20-2013 12:24 PM            

It seems that both right and left lower SRB O-rings were exposed to the same forces, one failed one did not. If the root cause of the right hand SRB failure was not the cold temperature by itself, then NASA management was not guilty of a "flawed" launch decision based upon temperature. Clearly any flight decision which results in a loss of crew is flawed, no one is arguing that fact.

posted 02-20-2013 12:30 PM               

quote:

---

Originally posted by PetesParkingLot:
It seems that both right and left lower SRB O-rings were exposed to the same forces,

---

posted 02-20-2013 12:50 PM            

If so then the focus on cold temperature launch is totally misplaced and is simply a way to complete the story and attach blame in a most dramatic way.

It seems to me that anyone who is willing to accept that the cold temperatures was the root cause of the 51L failure needs to at least offer an explanation as to why 5 of the 6 seals didn't fail in the cold.

posted 02-20-2013 01:38 PM               

Why did the other 5 field joints not fail? We don't know why. They may have partially failed but the evidence was destroyed. The one FACT not in disupute is cold weather reduced the rings' ability to seal - proven by the data collected by other flights in low temperatures. Why did the lower joint on the right side SRB fail completely - it didn't, it partially failed, seated again, then failed to fill the gap created by wind shear.

This accident had multiple, maybe thousands of small causes. What were they all? We don't know. Why do some things fail and others don't? We don't know. But that is not a reason to discount what we do know about cold and rubber.

posted 02-20-2013 02:04 PM            

That decision has never been fully questioned because we seem lost in focusing on the "flawed" launch decision of 1/28/86. The true failure of STS-51L lies with those in senior management who continued to permit flights with a an O-ring that could fail either at 40 degrees or 90 degrees.

posted 02-20-2013 02:17 PM               

quote:

---

Originally posted by PetesParkingLot:
...we seem lost in focusing on the "flawed" launch decision of 1/28/86.

---

posted 02-20-2013 02:18 PM               

quote:

---

Originally posted by PetesParkingLot:
Why did the left lower SRB seal, exposed to the very same cold temperature, seal?

There had to be more going on than simply launching at below 40 degrees. From the facts of the launch you could expect the seals to work 5/6 of the time even below 40 degrees. Was the true failure an assembly failure, as Gerry Griffin asserts?

---

Building an SRB was labor intensive and use some "craftmanship" in laying up the insulation and putty between joints.

Example: Take a bricklayer and have him build three identical walls. The three walls will have some minor differences in them. But they are not flawed and will meet the same standard "requirements." However, when subjected to extreme forces, they will fail at different values.

quote:

---

Originally posted by PetesParkingLot:
Was the choice to fly below 40 degrees truly a flawed decision or was the decision to fly any further flights with this design the "flawed" decision? Is part of spaceflight making "flawed" decisions and getting away with them?

---

posted 02-20-2013 03:45 PM               

One factor that Gerry isn't taking into account with his statement though was a set of readings taken by an ice team member with a temperature gun on the pad. He got a very cold reading from the vicinity of the aft field joint on the SRB that had the failure. At the time it was dismissed as an erroneous reading, but the IR guns had never given improper readings before, so a few years after the accident, some weather data was plugged into a computer program to simulate conditions that day. Apparently with the cold temperatures on the pad and the relatively light wind (I believe combined with a temperature inversion), the aft field joint was apparently super cooled a bit due to gaseous oxygen vapor coming from the LOX vent hood (the beanie cap) cooling the air. Rather than it blowing away, the chilled air sunk down towards the base of the pad and collected in that spot along the SRB resulting in the colder than expected temperature readings and that MAY have been a factor in why the O-ring failed the way it did (i.e. not necessarily saying that is the cause, but it likely contributed).

The Allan McDonald and James R. Hansen book "Truth, Lies and O-Rings" talks about it in a lot more detail (and with a bit more accuracy than my repeating of it) and brings up other topics from the time as well concerning the investigation into what happened. If you haven't read that book, I URGE you to do so as it is IMHO the best text to date on Challenger and explains the engineering behind what happened very well. Anyone who has studied Challenger in depth should have this in their library.

Regardless of whether the cold weather was a factor or not, the original shuttle SRB field joint design had a few weaknesses as they were susceptible to joint rotation (meaning the casings would flex more than they should have when SRB ignition pressurized the casings internally) and the O-rings were having to work harder and do more than they should have. Ultimately, the exact same O-rings were used in the SRB joint re-design while the field joints themselves were redesigned with better redundancy (with joint heaters being used to keep the O-rings within their tolerance range). Thiokol (now ATK) never utilized a replacement O-ring compound which had better cold temperature tolerance until they began work on the Constellation/Ares SRMs (which will be used in the SLS rocket I believe).

posted 02-20-2013 06:58 PM            

Clearly you could fly at 23 degrees, making the entire argument of the Thiokol engineering team from the night before not relevant. They proposed not flying below 40 degrees and it appears that you could safely fly down to 23 degrees temperature on the SRB O-rings.

posted 02-20-2013 07:02 PM               

quote:

---

Originally posted by PetesParkingLot:
So it looks like the right SRB launched at a temperature of 9 degrees and the left SRB flew at 23 degrees. Somewhere between those two numbers is the real information as to what temperature you could successfully fly an SRB. Clearly you could fly at 23 degrees...

---

NO! You cannot make that assumption! As noted above, temps were one factor out of MANY (both human and technical) that caused the failure.

posted 02-20-2013 07:15 PM            

The issue I want to nail down is what was the reality facing the launch team that day. Can one not assert that the reality was that if you launched with a SRB field joint temperature of 23 degrees or higher you would not get a breach. Five field joints successfully flew that day at 23 degrees or higher, even with the wind shear and all the other factors. One field joint failed with a launch temperature of 9 degrees.

posted 02-20-2013 09:49 PM               

quote:

---

Originally posted by PetesParkingLot:
Five field joints successfully flew that day at 23 degrees or higher, even with the wind shear and all the other factors.

---

posted 02-21-2013 07:29 AM               

quote:

---

Originally posted by PetesParkingLot:
Isn't the root cause of the Challenger disaster the launching of a SRB field joint at a temperature below 23 degrees? None of the other factors lead to a Criticality one failure that day...

---

An engineering lesson: Two "identical" objects (materials, systems, etc) will act the same way while operating within margins. The exact behavior of the object is unpredictable outside the margins.

quote:

---

Can one not assert that the reality was that if you launched with a SRB field joint temperature of 23 degrees or higher you would not get a breach. Five field joints successfully flew that day at 23 degrees or higher, even with the wind shear and all the other factors. One field joint failed with a launch temperature of 9 degrees.

---

1. Not enough data points to draw that conclusion.
   
2. Different set of boosters on the same day may have had no joint failures or a different joint fail.

posted 02-21-2013 11:08 AM            

For me, my interest is not in reliving the technical details of a engineering failure. I want to understand the choice that was faced that day by management.

I assume that NASA had the best and the brightest managers in the industry and they were all dedicated to crew safety. Yet they made a decision to fly a SRB in a temperature range that did not permit the O-rings on the SRB to seal and lead to crew death. What factors were weighed improperly to lead to this choice?

Again, it appears to me that the choice to fly below 40 degrees was not the cause of the accident. Had the right hand SRB been 39 degrees it is likely it would have sealed.

posted 02-21-2013 11:25 AM               

quote:

---

Originally posted by PetesParkingLot:
I assume that NASA had the best and the brightest managers in the industry and they were all dedicated to crew safety.

---

quote:

---

Yet they made a decision to fly a SRB in a temperature range that did not permit the O-rings on the SRB to seal and lead to crew death. What factors were weighed improperly to lead to this choice?

---

quote:

---

Again, it appears to me that the choice to fly below 40 degrees was not the cause of the accident. Had the right hand SRB been 39 degrees it is likely it would have sealed.

---

quote:

---

What information could you use on 1/28/86 to make this choice and what would the parameters of a correct choice? It is easy to criticize the bad choice, but what was the correct call that a day? Was it as simple as "We will not fly below 51 degrees"?

---

posted 02-21-2013 01:27 PM               

I am not a media basher by any means, but I recall the broadcast media just ridiculing NASA mercilessly during Challenger's many scrubbed launch attempts for the 51-L mission.

They positively went nuts with scathing commentary when a bolt securing the removable handle to Challenger's hatch got stripped. Technicians had to locate a drill and bring it up to the white room to drill the damn bolt out. When the drill's battery lost power and another drill had to be found newscasters positively went nuts. More than one talking head commented that NASA could not do anything right anymore.

posted 02-21-2013 02:06 PM            

Like everyone else on this forum, I am certain I would not have made the choice to fly on 1/28/86. I would have done so not because I knew that we couldn't safely fly in "cold" temperatures(the prior discussion shows that their is no definitive data to that effect). I would have objected to it because to do so would uneccesarily exposed 51L to a large number of temperature unknowns with no real benefit.

posted 02-21-2013 05:27 PM               

quote:

---

Originally posted by PetesParkingLot:
Like everyone else on this forum, I am certain I would not have made the choice to fly on 1/28/86.

---

The blame game involving 51-L has been played since the day it happened. To say that you, in retrospect, would do anything differently or would somehow be capable of a better decision is pretty insulting.

posted 02-21-2013 06:04 PM               

This is a very complex issue. I will just say that I hope most readers are aware of a book that is used in many university courses titled The Challenger Launch Decision written by Diane Vaughan.

posted 02-21-2013 07:06 PM               

quote:

---

Originally posted by PetesParkingLot:
Again, I make the assumption that NASA management was good and dedicated to crew safety.

---

posted 02-21-2013 07:11 PM               

quote:

---

Originally posted by Headshot:
You know that kind of crap had to mess with managers' heads.

---

NASA had, for better or worse, sold the nation on a '18-wheeler to space' that would run like clockwork. If anything, they were a victim of their own PR. When they didn't get a shuttle up regularly, the press, public and lawmakers also went bonkers. I'm just as guilty as anyone in that regard, I went to KSC to watch three launches along the causeway and all were scrubbed within a few minutes of going up. I never did see one go up from less than 100 miles away and I still get ticked just thinking of that.

posted 02-22-2013 01:02 PM            

Even after Challenger accident, when it was clear that one flight a month wasn't going to happen, NASA repeated this type of "flawed decision" with the foam problem. One cannot argue that NASA ignored the foam issue because of launch pressures. Just the opposite, by 2003, it was clear that a serious vehicle failure was more likely to end the program than a failure to launch regularly.

I recall a scene from the "Spider" episode of From the Earth to the Moon. In it a failure of the LM strut is found to have been traced to an engineering error. The guilty engineer comes to Tom Kelly and immediately reports the error. The response by Tom Kelly is appropriate pointing out that the engineer did not attempt to "cover your ass" but instead came forward with the reality. Without such complete openness with regard to the reality of failures, a complex program "won't get to New Jersey, let alone the Moon". What I see as the fundamental change in NASA was the lack of openness about the reality of the system performance.

The mission for Morton Thiokol became building and selling SRB and keeping the contract. The mission was no longer getting to the moon by the end of the decade or even flying safely.

When questions were raised about the unknowns of cold weather performance of the SRB seals, instead of passing the reality of the unknowns up to NASA management so they could truly weigh the risks, the organization "covered their ass." For me that is the lesson that I take from the "flawed decisions" of Challenger and Columbia. The lesson is that if your organization fails to pass to the decision makers a true picture of what we DON'T KNOW as well as what we do know then we should not blame decision makers when they do a poor job of managing the risk of the unknown.

What would it have hurt for Thiokol to pass the following information along to NASA Senior management?

"We know that at some temperature below 51F the O-rings will fail to seal and we will have a criticality one failure. We simply do not know what the lower limit is on temperature for the O-rings to seal. Our best estimate based upon very limited data is that it will be safe to fly the O-rings above the launch commit criteria of 31F."

posted 02-22-2013 01:15 PM               

A flight rule regarding temperatures at launch existed for a reason — in part because Thiokol and NASA had reviewed the safety margins for the solid rocket motors.

posted 02-22-2013 10:52 PM               

The books Robert mentions clearly detail the decision making processes that took place. The one core piece of data none of the engineers and managers had was the experience of a manned spacecraft failing in flight. It simply had never happened before, therefore, how could it?

The SRB's had shown a tendency to have O-ring erosion, even hot gas blow-by, yet had never completely failed or exploded. I think there was an organizational thought process (group think) to accept the damage and maybe even consider the joint design to be resilient enough to withstand the damage observed. Years later, in the discussion of the Columbia accident, the term "in family" came along to describe understood and accepted quantities of damage. The same thinking was applied on launch day of 51-L.

One of the better known quotes from the engineering meetings in deciding whether or not to launch Challenger was "My God, Thiokol, when do you want me to launch? In April?" So the pressure, real or imagined, was there. The flight previous, 61-C, with Congressman Bill Nelson aboard, was delayed a record number of times, I believe it was seven or so. Then the landing was delayed by several days on the back end of the mission. This led to the shortest time between a shuttle landing and a shuttle launch in the program's brief history. Sure there was pressure.

1986 had a record number of missions on the books. NASA's ambitious plans were on the cover of several magazines, fueled by the media frenzy over Christa McAuliffe's flight. Several high profile missions were scheduled - Galileo to Jupiter, Magellan to Venus, and the Hubble Space Telescope. All of these missions were to follow Challenger's departure from Pad 39-B. Sure there was pressure.

posted 02-23-2013 11:09 PM               

Schedule pressure was part of the situation as well, I believe. Given there were 16 flights planned for 1986, there was no margin for delays in the program. From "Encyclopedia Astronautica":

Prior to Challenger's launch, NASA was stretched to the limit. It had advertised the shuttle was providing cheap access to space, but this could only be achieved if the flight rate reached 20 per year, or one every two weeks. But the shuttle turned out to be a maintenance nightmare. Every one of the 28,000 tiles had to be checked before each flight, software for each mission had to be individually written, tested, debugged, and certified; the engines had to be thoroughly inspected and replaced or rebuilt. The result was that there were insufficient staff, time, and spare parts to sustain a rate over 10 per year. But headquarters pressed. Decision-making centered on justifying why a mission should go rather than why it should not. Challenger lifted off a record 16 days after Columbia. The ground crews were exhausted, working massive overtime and weekends. The push to achieve the necessary flight rate to make the shuttle 'economical' finally took the ultimate toll.

At the same time the O-ring issue was being looked at, another issue was rearing it's ugly head: External Tank foam shedding. Starting from STS-7 there were incidents of foam shedding. Original ET requirements were for a tank that did NOT shed foam, hence any shedding events were safety related and had to be investigated. However, NASA eventually decided that the events were "inevitable and unresolvable" with the rationale they were not a threat. Even when a piece dented a SRB near its electronics box on STS-112, they simply kept going.

posted 02-24-2013 06:57 AM            

quote:

---

Originally posted by Robert Pearlman:
...have you read "The Challenger Launch Decision: Risky Technology, Culture, and Deviance at NASA" by Diane Vaughan

---

In the case of Challenger the Primary O-ring blow-by was a known safety problem, but the assumption was the secondary O-ring would always seal. Bad assumption given the temperatures that morning.

In Columbia, "normalization of deviance" was applicable to the ET foam shedding problem which was assumed to have little damage potential, even after clear evidence to the contrary from several missions that sustained serious damage.

posted 02-24-2013 04:13 PM            

Imagine that Thiokol would have stated something like "we cannot be sure that the field joint will properly work at low temperatures. The data are inconclusive and not justifying correct operation at those temperatures. Thus, the probability of failure may be high, while the severity of a failure is immense (loss of human life) so we opt for waiting for temperatures inside the experience basis. Alternatively, dedicated tests could be run at low temperatures." Would NASA management have overruled Thiokol? To my understanding a main part of the wrong decision was rooted in the inability of Thiokol's engineers to get their concerns directly up to the relevant management and engineering decision makers at NASA. But it was not their fault, but the necessary direct communication lines did not exist. The management was obviously in-between.

Worse is that the problem was well known to Thiokol and NASA before 51L, but was not put at high enough priority to have a solution applied in good time. A solution was proposed well before, see e.g. A Fundament Flaw in SRM Joint Design.

posted 03-30-2013 08:59 PM            

Having reviewed the evidence again, I fully agree with Gene Kranz that the temperature variable is greatly over emphasized. Thanks partly to the public demonstration by Professor Feynman and to the fact that the public could easily grasp the cold temperature "explanation" the real root cause of the Challenger launch failure has been obscured.

The root cause of the failure of the aft SRB joint was an "out of round" condition which narrowed the gap into which the O-ring seal had to move through to seal.

Below freezing temperature was merely a contributing factor which was present in all six SRB field joints but failed to impact 5 of the 6 field joints.

To quote the findings of the Rogers Commission on page 70 of Volume one:

Launch site records show that the right Solid Rocket Motor segments were assembled using approved procedures. However, significant out-of-round conditions existed between the two segments joined at the right Solid Rocket Motor aft field joint (the joint that failed).

1. While the assembly conditions had the potential of generating debris or damage that could cause O-ring seal failure, these were not considered factors in this accident.
   
2. The diameters of the two Solid Rocket Motor segments had grown as a result of prior use.
   
3. 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.
   
4. 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.
   
5. The lack of roundness of the segments was such that the smallest tang-to-clevis clearance occurred at the initiation of the assembly operation at positions of 120 degrees and 300 degrees around the circumference of the aft field joint. It is uncertain if this tight condition and the resultant greater compression of the O-rings at these points persisted to the time of launch.

This is the root cause of the Challenger launch failure. None of the other joints on the 51-L flight were assembled with a .004 gap. All were exposed to temperatures below freezing. Both the left and right hand SRB aft field joints were cooled to an estimated 28 degrees +/- 5 degrees. Only the right hand SRB aft joint failed.

Even more telling is the fact the the right hand SRB sealed and did not leak everywhere but the point between 296 and 314 degrees, the exact point at which the "out of round" condition created the narrowest gap.

The temperature variable had only a marginal impact, NASA may very well have been able to fly brand new (perfectly round) SRBs in 28 degree temperatures for the entire flight manifest without any launch failure.

I wonder if the apparent inconsistent relationship between temperature and joint leakage can be more readily explained by examining the joint gap after assembly as the primary factor.

The focus on temperature while missing the major variable of the gap size at launch, left a reasonable doubt as to the relationship between temperature and joint erosion on prior flights. Had the Thiokol engineers better analyzed their data they might have convinced the management of both NASA and Thiokol.

posted 03-30-2013 09:21 PM               

quote:

---

Originally posted by PetesParkingLot:
This is the root cause of the Challenger launch failure.

---

Other joints on the left and right Solid Rocket Boosters experienced similar combinations of tang-to-clevis gap clearance and temperature. It is not known whether these joints experienced distress during the flight of 51-L.

As you are basing (at least part of) your analysis and conclusions on the Rogers Commission report, then its other findings are equally relevant, such as:

Of 21 launches with ambient temperatures of 61 degrees Fahrenheit or greater, only four showed signs of O-ring thermal distress; i.e., erosion or blow-by and soot. Each of the launches below 61. degrees Fahrenheit resulted in one or more O-rings showing signs of thermal distress.

The failure was due to a faulty design unacceptably sensitive to a number of factors. These factors were the effects of temperature, physical dimensions, the character of materials, the effects of reusability, processing, and the reaction of the joint to dynamic loading.