Go Back  PPRuNe Forums > Flight Deck Forums > Tech Log
Reload this Page >

Did it really happen the way they said? The Challenger revisited.

Tech Log The very best in practical technical discussion on the web

Did it really happen the way they said? The Challenger revisited.

Old 5th Feb 2003, 00:15
  #1 (permalink)  

Iconoclast
Thread Starter
 
Join Date: Sep 2000
Location: The home of Dudley Dooright-Where the lead dog is the only one that gets a change of scenery.
Posts: 2,132
Likes: 0
Received 0 Likes on 0 Posts
Question Did it really happen the way they said? The Challenger revisited.

If you really want to get to the bottom of something form a committee. That way you will get to the bottom and still not have a proper solution.
This may get a bit complex but please bear with me.

According to the Rogers commission the Challenger was lost for the following reasons. First an “O” ring that sealed the circumferential area of two mating sections of the solid rocket booster engine was compromised by being exposed to very low temperatures. This allowed hot gasses and eventually a flame to escape from the damaged “O” ring. The flame impinged on the liquid oxygen tank and when it burned through the tank catastrophically exploded. That was the company line and that is what was told to the press and the world’s public.

Here is my theory and in order to explain it you the reader need an exposure to rocket science.
There are various types of solid motor the most prevalent is employed on small missiles such as those used on aircraft. These are what are called an end burn or cigarette burn propellant grain. The propellant grain burns like a cigarette and as the grain is consumed the internal pressure drops due to consumption of the grain and an increase in the internal volume. However, as the interior volume increases the consumption of the grain decreases the weight of the missile and the speed does not drop off until the grain is almost completely consumed.

On larger solid motors the grain is cast with a geometric shape in the center. It could be cruciform in shape or possibly star shape. If you were to measure the geometric shape by starting at one point and covering the distance around the grain until you reached the starting point the measurement would be equal to the internal circumference of the containing vessel. Because of the physical size of the Challenger Solid Rocket Motors (SRM) they had to be cast in sections with one section mated on top of another until the motor was fully assembled. To maintain a positive seal between the segments an “O” ring was installed. In this type of motor the burning takes place on the exposed surface of the geometric shape and as the grain is consumed the internal pressure remains the same and the thrust is stable.

One of the characteristics of a solid rocket motor is that the relationship of the burning surface has to remain constant within allowable parameters. If at ignition a large chunk were expelled from the motor nozzle the relationship would be disturbed resulting in combustion instability or even worse, explosion. The explosion would result from more propellant burning surface resulting in higher internal pressures. This could also be caused by a crack in the grain.

Now lets’ look at the Challenger report. The report stated that the flame front escaping from the damaged seal impinged on the tank. In just about every film I have seen the flame was pointing away from the tank. There may have been two flames but I have only seen one. Lets’ assume there was a flame that impinged on the tank and liquid Oxygen escaped and flowed into the flame. If you remember back in Jr. High science they taught about the fire triangle which stated that in order to have a fire you had to have fuel, Oxygen and a fire source. Where was the fuel that would result in such a violent explosion?

Now we get back to the flame that was escaping from the compromised “O” ring seal. In order to have a flame escaping it meant that the two mating surfaces of the adjacent segments had to be on fire otherwise, the flame could not get to the seal. Now go to the second paragraph above that deals with an increase of pressure. If the two adjacent surfaces were burning it could result in a significant increase in the internal pressure of the SRB resulting in a catastrophic rupture of the motor case. This would be analogous to a crack in the grain,

When you look at the pictures of the explosion you do not see any fire. What you do see is a white cloud made mainly of liquid Oxygen and possibly liquid Hydrogen. If the Oxygen and Hydrogen combined and ignited you would not see any flame but there might be some frozen condensation resulting from any combustion.

The question is, what do you think?

I shall now retire to paint a bull’s eye on my chest.


Lu Zuckerman is offline  
Old 5th Feb 2003, 00:18
  #2 (permalink)  
Too mean to buy a long personal title
 
Join Date: Aug 2002
Location: UK
Posts: 1,968
Received 6 Likes on 4 Posts
Is this really an appropriate time for this discussion?
Globaliser is offline  
Old 5th Feb 2003, 03:23
  #3 (permalink)  

Iconoclast
Thread Starter
 
Join Date: Sep 2000
Location: The home of Dudley Dooright-Where the lead dog is the only one that gets a change of scenery.
Posts: 2,132
Likes: 0
Received 0 Likes on 0 Posts
Why do you ask?

To: Globaliser

Have you read these posts and if so why is my post so different. It simply offers an alternate theory.

STS-107, Chronicle Of A Disaster Foretold? Pages: 1 2 3 4 5 6 7


Lu Zuckerman is offline  
Old 5th Feb 2003, 08:18
  #4 (permalink)  
 
Join Date: Jan 2000
Location: 03 ACE
Age: 73
Posts: 1,011
Received 28 Likes on 18 Posts
Go for it Zuckerman.

(comments deleted) .. could we stick to the discussion topic, please ?

JT

Last edited by john_tullamarine; 5th Feb 2003 at 16:40.
El Grifo is offline  
Old 5th Feb 2003, 08:39
  #5 (permalink)  
 
Join Date: Aug 2000
Location: UK
Posts: 1,914
Likes: 0
Received 0 Likes on 0 Posts
(comments deleted) ... likewise, thanks.

JT


I think it reasonable to discuss any incident which takes lives.

Lu, I found it slightly difficult to follow the 3d geometry of the fuel layout of the SRBs. What I do recall is that they are cast in a tube shape, with some sort of hole in the middle, I think the hole expands the nearer the tail of the rocket. The SRB burns throughout its WHOLE length for the full 2 minutes, the burn gradually reaching the casing in the lower areas first. That way the casing only experiences minimal heating before the SRBs burn out. The references to cigarettes may be misleading as the burn does not start at one end and finally reach the top. I assume that as the burn is well progressed, the fuel is consumed in the lower stages and the burn reaches the casing earliest in the bottom section. This is why I expect the failure occured low down the SRB well into the burn (75secs of its 120 sec duration). The makers obviously have the fuel arrangement down to a fine art- I notice both SRBs burn out simultaneously. I recall seeing the flame that was apparently blowtorching the main fuel tank. I find it difficult to take issue with the findings. The SRBs seem to be a very reliable system for extracting very high thrust for short periods.

Last edited by john_tullamarine; 5th Feb 2003 at 16:42.
Notso Fantastic is offline  
Old 5th Feb 2003, 08:43
  #6 (permalink)  
PPRuNe Enigma
 
Join Date: Feb 2001
Location: Scotland
Posts: 427
Likes: 0
Received 0 Likes on 0 Posts
Lu; what concerns me most is that 'O' ring erosion had been noted on several earlier flights.

Instead of realising that this was indication of a design problem, NASA appears to have concluded that it was safe to continue flying because previous flights had completed successfully in spite of the 'O' ring erosion.

But the 'O' rings were not designed to work in this way. Just because you get away with it a few times doesn't mean it is safe.

Now we have similar statements from NASA that debris from the external tank had damaged tiles before, and this was not considered a problem because the flights had completed successfully. Very worrying.
Grainger is offline  
Old 5th Feb 2003, 08:46
  #7 (permalink)  
Too mean to buy a long personal title
 
Join Date: Aug 2002
Location: UK
Posts: 1,968
Received 6 Likes on 4 Posts
Lu

If this follows on from those discussions, I apologise. I haven't had time to get into that thread after it got going. This just looked like a new topic.
Globaliser is offline  
Old 5th Feb 2003, 13:37
  #8 (permalink)  

Iconoclast
Thread Starter
 
Join Date: Sep 2000
Location: The home of Dudley Dooright-Where the lead dog is the only one that gets a change of scenery.
Posts: 2,132
Likes: 0
Received 0 Likes on 0 Posts
Question Cigarettes and other things.

I was not implying that the SRB were end burners. I stated that smaller airborne rockets were end burners.

The “O” rings were never designed to contain a flame front. If the temperature were that high the”O” rings would fail even if they were not compromised.

I can’t say that the SRB had controlled cut-off panels but most rockets that must cut off at a specific point have blowout panels at the pointy end. When these panels are opened the internal pressure is vented and the panels may also act as retro rockets both slowing the rocket motor and at the same time deflecting the motor away from the main stage.

Lu Zuckerman is offline  
Old 5th Feb 2003, 14:16
  #9 (permalink)  
 
Join Date: Jul 2000
Location: Everywhere
Posts: 783
Likes: 0
Received 0 Likes on 0 Posts
Lu,

to have a fire you had to have fuel, Oxygen and a fire source. Where was the fuel that would result in such a violent explosion?
Well the thing is, if you take Hydrogen as a gas and light it, it will go bang.

I know you'll correct me if I've overlooked something, and in my naivety as a young student (!) would like to know!! but, Oxygen behaves in some similar ways to Hydrogen. Hydrogen burns in air because there is oxygen present. The hydrogen is the fuel. Liquid oxygen is inflammable and behaves as a fuel source, try burning it, well.. ok I wouldn't reccomend that but still.

I guess I'm trying to say that with the presence of oxygen in the air and a fire source, liquid oxygen will burn. As the energy of the oxygen increases (in micro seconds) it returns to a gaseous state where it expands very rapidly producing the explosion.

Any chemists set me right!!

Andy.
The African Dude is offline  
Old 5th Feb 2003, 14:29
  #10 (permalink)  

Victim of Blackmailing Scouser
 
Join Date: Aug 2000
Location: Zürich, Switzerland (But a Brit)
Age: 59
Posts: 113
Likes: 0
Received 0 Likes on 0 Posts
fish

Lu,

Before I proceed, I need to state that I think the Challenger accident bears no relation to the dreadful Columbia accident. The NASA management, culture and working practices are very different 17 years on.

I've read your original post three times, and I still don't see what you're trying to say here. Maybe I'm thick, but could you spell it out for me? Words of two syllables or less would be a great help.

The Rogers Commission report is a very famous document indeed. The report summary devotes about a page to the actual reason(s) for the failure, and about twenty other pages to major deficiencies in the NASA management and quality culture at that time. Two pages were devoted to political pressures from Washington!

Basically, what actually brought down the Challenger was almost irrelevent: one of the many zillions of components was bound to fail. Fate happened to select the SRB O-ring on that sad occasion. What really brought the Challenger down was sloppiness, a badly compromised safety culture, and endless political interference by Washington.

Read the report, and you'll see I'm right.

This time around we seem to be faced with some unresolvable design compromises, rather than any actual sloppiness in NASA processes and procedures.

Accusations that the Shuttles are 'too old' are laughable. The damn things are pretty much brand new every time they launch! No, it's just that previously acceptable risks have become unacceptable over the last two decades.

Safety and routineness have been grossly oversold.

TW

Last edited by Tricky Woo; 5th Feb 2003 at 14:42.
Tricky Woo is offline  
Old 5th Feb 2003, 14:42
  #11 (permalink)  
 
Join Date: Feb 2003
Location: London
Posts: 1
Likes: 0
Received 0 Likes on 0 Posts
A couple of thoughts. First, as to how the flame front got to the O-ring. As you say, the O-rings sealed the segments of the casing together. Each segment had its solid, rubbery propellant cast into it separately. Therefore, on a level with the O-rings there was a discontinuation of the propellant, an inevitable result of the segmented structure.
When the propellant was ignited the pressure build-up caused the individual segment cases to bulge out into a barrel shape. This would then allow the flame front to travel between the segments of propellant. As long as the O-rings maintained the seal then there wouldn't be an escape route for this flame front and so the pressure would be maintained. If the sealing putty and O-rings were burnt away, though - as happened to Challenger - then the situation is much worse.

As for how the explosion occured, the view is that the flame leakage impinged on one of the lower struts that attaches the rocket to the fuel tank. The strut eventually failed, which caused the rocket to rotate around the remaining top strut and puncture the Hydrogen tank. Mix that with the leaking Oxygen, plus the flame from the still-burning rockets, and you get the big bang.
matthewbell is offline  
Old 5th Feb 2003, 15:02
  #12 (permalink)  
PPRuNe Enigma
 
Join Date: Feb 2001
Location: Scotland
Posts: 427
Likes: 0
Received 0 Likes on 0 Posts
That's about right, matthew - the strut failed allowing the top of the booster to puncture the main tank.

Once this had happened it's largely irrelevant what the Oxygen and the Hydrogen did. As Feynman pointed out, arguing about what happened AFTER the original failure is a bit like going over the pieces of a train wreck working out which car broke apart first, instead of concentrating on what brought it off the rails in the first place.

The root cause here was the clevis joint rotation and subsequent 'O' ring erosion. The design required the 'O' ring to expand rapidly to fill a widening gap in the clevis joint. This is NOT the way that 'O' rings are supposed to work so the design was really doomed from the beginning.

The fact that 'O' ring erosion on previous flights didn't bring them down was simply random chance and good luck, not confirmation that it was safe for this to happen.

I still can't help feeling that this sounds ominously similar to the discussion that debris had damaged tiles on previous flights. The tiles were never intended to be bashed off by bits of debris - the fact that we got away with it on previous occasions is not a good enough reason to declare that it was OK for this to happen and safe to continue flying.
Grainger is offline  
Old 5th Feb 2003, 15:10
  #13 (permalink)  

Self Loathing Froggy
 
Join Date: Jun 2002
Location: elsewhere
Age: 18
Posts: 546
Likes: 0
Received 0 Likes on 0 Posts
African Dude

Liquid Oxygen does NOT burn, but, if given a suitable source of heat (droplets falling in air can get it), it will vaporize. The density of liquid Oxygen at boiling point at ambient pressure (Temp=90K, or -183°C or -297 F) is slightly below the density of liquid water, whereas gaseous oxygen is slightly heavier than air. This gives us an expansion factor of the order of 1000 at ambient temperature and pressure. At 500 hPa, this factor is 2000. Liquid hydrogen is found at even lower temperatures (BP =20K or -253°C or -423F), and is much less dense (leading to those huge tanks, where liquid oxygen takes only a small volume), but the order of magnitude would be in the same range (too lazy to do the math)

Both liquid oxygen and liquid hydrogen tanks are pressurized, both for structural and thermodynamic reason, about which I will not elaborate. If the tanks get punctured or break apart, without combustion taking place, you will very quickly get a cloud of liquid droplets vaporizing and expanding very quickly, but this not enough to generate an explosion in the open.

Combustion of hydrogen with oxygen occur only in the gaseous phase, but as soon as it has started, the heat of combustion is sufficient to vaporize droplets. In that case, the mixture is very likely to explode (inflamation of hydrogen in air at ambient pressure is between 4% and 75% volume fraction, this range widens when you increase oxygen concentration, but I do not have exact figures at hand).

Lu

If combustion takes place (be it in air or oxygen-enriched air), it will produce huge amounts of heat and water vapor, which means that the water will not necessarily condense down to solid state. Beacuse of said heat, the flow in the cloud will be turbulent which will give you a cumlus-like cloud, even if you have ice forming. Moreover, in the case of an explosion, the flame is in the center and the cloud expands radially, hiding the flame from sight.

Last edited by Bre901; 5th Feb 2003 at 15:54.
Bre901 is offline  
Old 5th Feb 2003, 15:26
  #14 (permalink)  
 
Join Date: Feb 2003
Location: In a stable
Posts: 1
Likes: 0
Received 0 Likes on 0 Posts
The sequence of events with the strut being weakened which then allowed the booster to puncture the H2 tank is pretty much the was I remember the report reading.

It is also worth remembering that statistically, the shuttle was likely to fail on something like 1 in 100 missions. Wasn't the recent one something like #107? I am not sure how much of the shuttle is 'new' each time it flies, but for sure it is pretty much gutted and rebuilt with broken bits being replaced. When there is talk that the shuttles are old and dated, its more a case that if the project was done now, the result would have been very different.

As TrickyWoo has said,'previously acceptable risks have become unacceptable over the last two decades'
RaGaRoO is offline  
Old 5th Feb 2003, 15:33
  #15 (permalink)  
 
Join Date: Jun 2001
Location: UTC +8
Posts: 2,626
Likes: 0
Received 0 Likes on 0 Posts
Thanks for the Physics 101 refresher class.
GlueBall is offline  
Old 5th Feb 2003, 18:08
  #16 (permalink)  
 
Join Date: Feb 2001
Location: US
Posts: 604
Likes: 0
Received 0 Likes on 0 Posts
If the tanks get punctured or break apart, without combustion taking place, you will very quickly get a cloud of liquid droplets vaporizing and expanding very quickly, but this not enough to generate an explosion in the open.
If the tanks get punctured, could that be enough to cause the tanks themselves to fail structurally?
OFBSLF is offline  
Old 5th Feb 2003, 18:43
  #17 (permalink)  

Plastic PPRuNer
 
Join Date: Sep 2000
Location: Cape Town
Posts: 1,898
Received 0 Likes on 0 Posts
Other people also seem to have read "Challenger - A Major Malfunction" (1987) by Malcolm McConnell. This is an excellent book which contains a wealth of technical detail about the incident and the mechanics of the O-rings and their failure modes. Also a lot of stuff about Thiokol and the political background - I humbly suggest that you read it Lu.

For reasons not unconnected with my work I am also interested in the chain of events leading up to nasty accidents and the links are always much the same. Just as in the Turkish Airlines 981 DC-10 disaster there had been many indications for a long time that something was seriously amiss with a criticality 1 component. In both cases "fixes" were designed and applied in the hope that the patch would render the problem less critical, but quite a few people remained seriously concerned. The financial and time implications of a full redesign and retrofit were so great that this was judged impractical.

O-ring erosion (which wasn't supposed to occur AT ALL) occurred on virtually all launches after the lightweight SRB casing was introduced. It was known to be worse in for launches in cold weather when the cold-stiffened chromate putty failed to conduct the pressure of the ignition transient and correctly seat the O-rings. These problems became significantly worse after a preflight pressure test (which was supposed to check O-ring integity) was introduced. Quite a few people were extremely worried (e.g. Roger Boisjoly) and wrote scads of memos. Still, nothing blew up and O-ring erosion came to be accepted as almost normal. People consoled themselves that a new joint design was on the way. You know the rest.

Right from the start there were considerable worries about tile integity and bonding and these have grumbled on in the background since the first flights. Tile damage on launch wasn't supposed to occur. It did, and insulation loss from the big tank was known to be a primary culprit. I think a lot of people were pretty worried and held their breath for each re-entry. Still, nothing blew up and tile loss/damage can to be judged an acceptable risk.

To quote the late, great Richard Feynman (of the Rodgers Comission): "I read all of these reviews and they agonise whether they can go even though they have some blow-by in the seal or they had a cracked blade in the pump of one of the engines, whether they can go next time or this time and they decide yes. Then it flies and nothing happens"

"Then it is suggested, therefore, that the risk is no longer so high. For the next flight we can lower our standards a little bit because we got away with it this time....It is a kind of Russian roulette."

When you win, it just until the next time.
And when you lose, you lose bad.....
Mac the Knife is offline  
Old 5th Feb 2003, 19:47
  #18 (permalink)  

Self Loathing Froggy
 
Join Date: Jun 2002
Location: elsewhere
Age: 18
Posts: 546
Likes: 0
Received 0 Likes on 0 Posts
OFBSLF
If the tanks get punctured, could that be enough to cause the tanks themselves to fail structurally?
The answer is very probalbly yes as they are thin walled and are designed to withstand structural loads only when pressurized (just look at a plastic soda bottle just before first opening, and just after).

This being said, I would support Grainger when he says :
Once this had happened it's largely irrelevant what the Oxygen and the Hydrogen did
Nevertheless, the answer is quite straightforward : if they were allowed to meet, with the slightest source of energy available (may I remember that we have some rocket engines running at full throttle just below), there is only one answer : BOOM. Period.

Even if the tank is not punctured, if the booster rotates, they are doomed, because the controls will probably not be able to prevent the shuttle to veer off trajectory, and I'm quite sure the maximum structurally admissible angle of attack is quite low. Anyone got figures ?
Bre901 is offline  
Old 5th Feb 2003, 23:52
  #19 (permalink)  
Cunning Artificer
 
Join Date: Jun 2001
Location: The spiritual home of DeHavilland
Age: 76
Posts: 3,127
Likes: 0
Received 0 Likes on 0 Posts
Exclamation

Yes Bre901, its very easy to get a LOX explosion. We used LOX on military transports (I assume they still do, too - the good old VC10 remains in service) and during hot weather it was customary to use the stuff to fast chill a Coke. You had to be careful to keep the Coke cans clean, though. One day one of our engine fitters didn't realise he'd transferred a film of grease from his greasy hands onto his Coke can and as soon as the vented LOX vapour hit it - BANG! Most impressive! Without any source of ignition too! I believe that can of Coke is still in orbit.

Still, it was a most impressive demonstration of two important safety points.

1. The guys who write the safety precautions know what they are talking about.
2. Complacency is the greatest danger faced by engineers.

Both these safety points are relevant to the Challenger accident. (Columbia too, I think)

**************************
Through difficulties to the cinema
Blacksheep is offline  
Old 6th Feb 2003, 12:07
  #20 (permalink)  
 
Join Date: Jun 2002
Location: London
Posts: 8
Likes: 0
Received 0 Likes on 0 Posts
For anybody who hasn't read it, Feynman's appendix to the report on the Challenger disaster is online at http://www.scs-intl.com/online/

Graham
Zipperhead is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.