PDA

View Full Version : Challenger saveable just in time?



wd40
2016-Mar-25, 02:41 AM
In OTL the Challenger explosion occurred 73 seconds after launch.

What if the puffs of dark smoke from the SRB had been spotted and the hazard it indicated diagnosed at launch, with the order to abort given at 40 seconds, and the order to jettison both boosters at 50 seconds, could the crew be saved?

Was dumping the SRBs at that low altitude an available option and was it survivable?

Jens
2016-Mar-25, 04:00 AM
No, it was not an option. I think that for a number of reasons, you could not separate the orbiter from still-burning SRBs or even from the external fuel tank.

schlaugh
2016-Mar-25, 04:17 AM
Right, once the SRBs lit, and until they burned out, there were no abort modes.

It was suggested that the first flight of Columbia be used to test the Return to Launch Site abort mode but John Young vetoed the idea. Colorfully, IIRC.

ETA: https://en.wikipedia.org/wiki/Space_Shuttle_abort_modes?wprov=sfii1

wd40
2016-Mar-26, 06:11 PM
In theory, what would be the effect on the Shuttle's ascent if both boosters were dropped after 50 seconds?

Solfe
2016-Mar-27, 06:34 AM
The shuttle won't handled the stress of that, once the SRB's light, they can't turn off. At 50 seconds, they are still firing and would out accelerate the shuttle. I guess they would smash into parts of the shuttle and you would lose the vehicle.

If they vanished by magic, the shuttle wouldn't have the ability to climb too much further.

Edit - My evil, disaster minded brain just offered up another problem. The ET is full of fuel. It would slosh and that would do something to the shuttle. The ET separation isn't supposed to happen while accelerating. It would likely become very unpredictable if you magically lost the SRBs, dropped or not.

cjameshuff
2016-Mar-27, 11:42 AM
The shuttle won't handled the stress of that, once the SRB's light, they can't turn off. At 50 seconds, they are still firing and would out accelerate the shuttle. I guess they would smash into parts of the shuttle and you would lose the vehicle.

If they vanished by magic, the shuttle wouldn't have the ability to climb too much further.

Edit - My evil, disaster minded brain just offered up another problem. The ET is full of fuel. It would slosh and that would do something to the shuttle. The ET separation isn't supposed to happen while accelerating. It would likely become very unpredictable if you magically lost the SRBs, dropped or not.

It's supposed to happen under acceleration (they can't stop and restart the main engines), but with much less in the tanks. The vehicle is too heavy to fly without the boosters early on, so if the SRBs disappeared early on, the shuttle would just hit the ground in a fireball.

On the first launch, overpressure from the SRBs caused severe damage to the tiles and drove control surfaces at the back of the orbiter beyond their mechanical limits. If they had known how bad it was during ascent, they would have ejected once they reached a safe altitude. Direct exposure to the full exhaust plume of a departing SRB would probably just shred the vehicle.

Solfe
2016-Mar-27, 12:57 PM
Yes, you are right. It is accelerating, but it is past the peak. I think the power falls off by a large amount before separation.

wd40
2016-Mar-27, 01:32 PM
If Challenger's SRB retaining strut had burnt through before the explosion and it flew off without damaging the Shuttle, could it have reached an abort mode with the remaining SRB? Could the computer still maintain flight control on such an unbalanced vehicle? Was there any NASA scenario envisaging only one SRB working?

Noclevername
2016-Mar-27, 02:13 PM
If Challenger's SRB retaining strut had burnt through before the explosion and it flew off without damaging the Shuttle, could it have reached an abort mode with the remaining SRB? Could the computer still maintain flight control on such an unbalanced vehicle? Was there any NASA scenario envisaging only one SRB working?

No, no, and no.

swampyankee
2016-Mar-27, 02:34 PM
There was, during the Shuttle system's preliminary design phase, a plan to use a vertical stack, but it was deemed impractical, leading to the configuration actually used. I suspect that, had there been no DoD involvement, the Shuttle would have been significantly smaller and lighter, and a vertical stack may have been practical. This would have eliminated the failure mode that destroyed the Challenger and probably would have eliminated the failure mode that destroyed the Columbia.

The Shuttle's design was forced by a large number of compromises, as is the case with all complex systems. You can see examples of this if you look at any of the sites involving design of homebuilt aircraft: mission drives the design, but even there, where the chief designer and end user are frequently the same person, there's back-and-forth in the design process and compromises in the end design. I've asked permission to post a thread where exactly that happens, and I'll edit this post to add it if I get that permission: I believe a number of posters here have absolutely no idea what goes on in the design process.

LotusExcelle
2016-Mar-27, 02:40 PM
I could be wrong but if I recall.. the crew was safe until water impact. For me this changes the question to: If the crew section had a BRS of some sort - would the outcome have been different?

schlaugh
2016-Mar-27, 03:09 PM
BRS = Ballistic Recovery System? Like a scaled-up version of what was done in the F-111 (http://www.ejectionsite.com/texans/f111d_34rt.jpg)? A BRS would add a tremendous amount of weight to the final design of the orbiter and would also need the ability to sense when a breakup was taking place; although that part is probably feasible. But even so, then what? The BRS would have to eject the entire crew compartment - or maybe the entire front section of the orbiter - and you risk ejecting the crew into the flames of the SRBs or even the still-running main engines. Unless you have an emergency escape system on the nose of the orbiter...and here we go again.

swampyankee
2016-Mar-27, 03:15 PM
BRS = Ballistic Recovery System a scaled up version of what was done in the F-111 (http://www.ejectionsite.com/texans/f111d_34rt.jpg)? A BRS would add a tremendous amount of weight to the final design of the orbiter and would also need the ability to sense when a breakup was taking place; although that part is probably feasible. But even so, then what? The BRS would have to eject the entire crew compartment - or maybe the entire front section of the orbiter - and you risk ejecting the crew into the flames of the SRBs or even the still-running main engines. Unless you have an emergency escape system on nose of the orbiter...

The Challenger disaster was a failure brought about by excessive confidence in the system and by managerial arrogance; the latter would probably have removed any kind of BRS long before the Challenger disaster occurred.

cjameshuff
2016-Mar-27, 03:49 PM
If Challenger's SRB retaining strut had burnt through before the explosion and it flew off without damaging the Shuttle, could it have reached an abort mode with the remaining SRB? Could the computer still maintain flight control on such an unbalanced vehicle? Was there any NASA scenario envisaging only one SRB working?

It really is as simple as schlaugh said: there was no abort until the SRBs burned out. If the vehicle couldn't reach that point intact enough to perform an abort, it was lost. It would have been destroyed by the exhaust or by mechanical or aerodynamic stresses it wasn't designed for.

Ara Pacis
2016-Mar-27, 05:18 PM
BRS = Ballistic Recovery System? Like a scaled-up version of what was done in the F-111 (http://www.ejectionsite.com/texans/f111d_34rt.jpg)? A BRS would add a tremendous amount of weight to the final design of the orbiter and would also need the ability to sense when a breakup was taking place; although that part is probably feasible. But even so, then what? The BRS would have to eject the entire crew compartment - or maybe the entire front section of the orbiter - and you risk ejecting the crew into the flames of the SRBs or even the still-running main engines. Unless you have an emergency escape system on the nose of the orbiter...and here we go again.

I was thinking of something like this too. It may not have been feasible in the past with all the control panels, but with newer fly-by wire and thin displays, maybe everything could be crammed into a small capsule sized ejection pod. There might be on large round one that holds two decks worth of crew seats, or several pods at different crew stations, each one holding a couple seats. Of course, ejection risks hitting the vertical stabilizer, so maybe that should be ejectable as well.

Ara Pacis
2016-Mar-27, 05:22 PM
There was, during the Shuttle system's preliminary design phase, a plan to use a vertical stack, but it was deemed impractical, leading to the configuration actually used. I suspect that, had there been no DoD involvement, the Shuttle would have been significantly smaller and lighter, and a vertical stack may have been practical. This would have eliminated the failure mode that destroyed the Challenger and probably would have eliminated the failure mode that destroyed the Columbia.

The Shuttle's design was forced by a large number of compromises, as is the case with all complex systems. You can see examples of this if you look at any of the sites involving design of homebuilt aircraft: mission drives the design, but even there, where the chief designer and end user are frequently the same person, there's back-and-forth in the design process and compromises in the end design. I've asked permission to post a thread where exactly that happens, and I'll edit this post to add it if I get that permission: I believe a number of posters here have absolutely no idea what goes on in the design process.

I tried to design an aerospace taxi in a thread on this site. I ran into that issue immediately.

swampyankee
2016-Mar-27, 06:03 PM
I tried to design an aerospace taxi in a thread on this site. I ran into that issue immediately.

I think you see that whenever somebody starts posting about how cheap it should be to design a man-rated launcher.

Ara Pacis
2016-Mar-27, 06:38 PM
I think you see that whenever somebody starts posting about how cheap it should be to design a man-rated launcher.

I still don't know exactly what it takes to man-rate a booster. But, some people want to reduce or eliminate regulations, so maybe they want or expect the parameters to change.

joema
2016-Mar-28, 10:38 AM
...I suspect that, had there been no DoD involvement, the Shuttle would have been significantly smaller and lighter, and a vertical stack may have been practical. This would have eliminated the failure mode that destroyed the Challenger and probably would have eliminated the failure mode that destroyed the Columbia...The Shuttle's design was forced by a large number of compromises, as is the case with all complex systems.

The basic design parameters were not dictated by DoD -- these were NASA decisions. This includes wing type, wing size, cross range capability, payload bay size and parallel stage design. One of the shuttle's main jobs -- which profoundly affect its design -- was assembling and supporting the space station. Funding limitations precluded developing both shuttle and station concurrently, but the shuttle's design had to support this. You can't build a space station with a significantly smaller, lighter orbiter. Suggestions like maintaining the Saturn V to loft the station then servicing it with a smaller shuttler are NOT an alternate shuttle design, they are an alternate reality.

In the CAIB hearings Bob Thompson (Space Shuttle Program Manager from 1970 to 1981) said: "NASA did not put cross range in the vehicle because the Air Force forced us to. NASA put cross range in the vehicle because we thought that was the right way to build the vehicle and it just happened to give the Air Force some capability they wanted. But we wanted it for abort capability during the launch and we wanted to start flying the vehicle right at entry. We didn't want to keep the thing above stall all the way down to landing area and then flip it around. So the myth that the Air Force made us do something we didn't want to do is absolutely a myth." http://caib1.nasa.gov/events/public_hearings/20030423/transcript_am.html

The payload bay size decision was discussed in Thompson oral history interview for NASA:

http://www.jsc.nasa.gov/history/oral_histories/ThompsonRF/ThompsonRF_10-3-00.htm

"We went back and forth, and we finally settled on a fifteen-foot diameter, sixty-foot-long module. So that set the size of the payload bay."

"Now, you can read in the history of the thing that people were looking at ten-foot diameter, thirty-five-foot-long payloads. We looked at different-sized payloads, but we never were very serious about anything other than the fifteen-foot diameter, sixty-foot long, because modular space station, if you get the modules too small, aren't very practical."

Re parallel staging, it is not clear in the Challenger disaster whether serial staging would have saved it. An SRB burn through would probably propagate until the SRB itself failed or exceeded steering authority, at which the vehicle would have broken up.

Re Columbia, a top-mounted orbiter would reduce -- not eliminate -- the risk. The shuttle hit birds several times during ascent, just good luck not at a bad place or velocity. Any exposed TPS is at risk -- the error was not checking it on orbit as a normal procedure. Putting the orbiter on top introduces other risks and problems. An orbiter on top puts heavy weight with lots of wing area up there which creates bending moments, esp critical for a winged vehicle. Cross-winds during ascent become harder to handle -- it's a kite on the end of a stick.

While on the pad, wind shear force on entire stack becomes more severe. Payload access is another big problem: with orbiter on top it's harder to reach. Yet another problem is orbiter access for on-pad maintenance. On top of a tall stack it's harder to reach, which then requires a much more expensive taller tower or a trip back to the VAB, which itself entails risk.

All these items were carefully weighed and considered during the design. The orbiter wasn't hung on the side as some kind of undesired compromise -- after carefully considering all pros and cons NASA felt it was the best overall design at that time. Some of these decision factors were discussed at length in the 2005 MIT video lecture series 16.885J, which is available on line.

Nicolas
2016-Mar-28, 12:01 PM
Hermes tried to place a small Orbiter on top of an Ariane 5. Apart from the money/politics issue that killed the project, there were also major Technical issues in having even a small Orbiter on top of your launcher. I'm not saying it's impossible, but certainly not easy. And Hermes was way smaller than te STS Orbiter.

Nicolas
2016-Mar-28, 12:02 PM
By the way, my computer likes to give some words like Orbiter and Original a capital letter. Just ignore them until I can fix the issue.

swampyankee
2016-Mar-30, 11:12 PM
The basic design parameters were not dictated by DoD -- these were NASA decisions. This includes wing type, wing size, cross range capability, payload bay size and parallel stage design. One of the shuttle's main jobs -- which profoundly affect its design -- was assembling and supporting the space station. Funding limitations precluded developing both shuttle and station concurrently, but the shuttle's design had to support this. You can't build a space station with a significantly smaller, lighter orbiter. Suggestions like maintaining the Saturn V to loft the station then servicing it with a smaller shuttler are NOT an alternate shuttle design, they are an alternate reality.

In the CAIB hearings Bob Thompson (Space Shuttle Program Manager from 1970 to 1981) said: "NASA did not put cross range in the vehicle because the Air Force forced us to. NASA put cross range in the vehicle because we thought that was the right way to build the vehicle and it just happened to give the Air Force some capability they wanted. But we wanted it for abort capability during the launch and we wanted to start flying the vehicle right at entry. We didn't want to keep the thing above stall all the way down to landing area and then flip it around. So the myth that the Air Force made us do something we didn't want to do is absolutely a myth." http://caib1.nasa.gov/events/public_hearings/20030423/transcript_am.html

The payload bay size decision was discussed in Thompson oral history interview for NASA:

http://www.jsc.nasa.gov/history/oral_histories/ThompsonRF/ThompsonRF_10-3-00.htm

"We went back and forth, and we finally settled on a fifteen-foot diameter, sixty-foot-long module. So that set the size of the payload bay."

"Now, you can read in the history of the thing that people were looking at ten-foot diameter, thirty-five-foot-long payloads. We looked at different-sized payloads, but we never were very serious about anything other than the fifteen-foot diameter, sixty-foot long, because modular space station, if you get the modules too small, aren't very practical."

Re parallel staging, it is not clear in the Challenger disaster whether serial staging would have saved it. An SRB burn through would probably propagate until the SRB itself failed or exceeded steering authority, at which the vehicle would have broken up.

Re Columbia, a top-mounted orbiter would reduce -- not eliminate -- the risk. The shuttle hit birds several times during ascent, just good luck not at a bad place or velocity. Any exposed TPS is at risk -- the error was not checking it on orbit as a normal procedure. Putting the orbiter on top introduces other risks and problems. An orbiter on top puts heavy weight with lots of wing area up there which creates bending moments, esp critical for a winged vehicle. Cross-winds during ascent become harder to handle -- it's a kite on the end of a stick.

While on the pad, wind shear force on entire stack becomes more severe. Payload access is another big problem: with orbiter on top it's harder to reach. Yet another problem is orbiter access for on-pad maintenance. On top of a tall stack it's harder to reach, which then requires a much more expensive taller tower or a trip back to the VAB, which itself entails risk.

All these items were carefully weighed and considered during the design. The orbiter wasn't hung on the side as some kind of undesired compromise -- after carefully considering all pros and cons NASA felt it was the best overall design at that time. Some of these decision factors were discussed at length in the 2005 MIT video lecture series 16.885J, which is available on line.

According to one of the participants, related here, the Air Force's cross-range requirement eliminated whole swathes of lower risk, lower cost alternatives. http://ocw.mit.edu/courses/aeronautics-and-astronautics/16-885j-aircraft-systems-engineering-fall-2005/video-lectures/lecture-2/

joema
2016-Mar-31, 01:16 AM
According to one ofthe participants, related here, the Air Force's cross-range requirement eliminated whole swathes of lower risk, lower cost alternatives. http://ocw.mit.edu/courses/aeronautics-and-astronautics/16-885j-aircraft-systems-engineering-fall-2005/video-lectures/lecture-2/

There is nothing in that video which says that. Early in the program there were *many* early alternate designs considered such as Faget's straight wing orbiter and the Chrysler SSTO "capsule". In the MIT video these were mentioned by Aaron Cohen who reported to Robert F. Thompson, overall shuttle program manager (whose interview and testimony I referenced). Just because an early "paper" design was sketched and superficially looked interesting doesn't mean that was borne out by further study. A good example is Faget's straight wing orbiter.

Max Faget personally liked his straight wing design. However this would not necessarily produce a hugely lighter, cheaper shuttle. As you can see here, one estimated dry orbiter weight with a straight wing was 85,500 kg (188,500 lbs), which was *heavier* than the delta-wing shuttle actually built (165,000 lbs): http://www.astronautix.com/lvs/shulelcr.htm

The issue is *not* how the shuttle could designed could be tweaked and might be a little better. The issue is how to use early 1970's technology and the available development budget and make it *much* better. Nobody has ever authoritatively demonstrated what that unchosen miracle design was.

E.g, it's true the Air Force wanted a larger cross range than a few of NASA's early shuttle concepts, but upon further study NASA also felt this was needed. This is discussed in the book "Facing the Heat Barrier: A History of Hypersonics", available on line in PDF format (links below):

"Charles Donlan, acting director of the shuttle program office at Headquarters...wrote that high cross range was 'fundamental to the operation of the orbiter.' It would enhance its maneuverability, greatly broadening the opportunities to abort a mission and perhaps save the lives of astronauts. High cross range would also provide more frequent opportunities to return to Kennedy Space Center in the course of a normal mission....Delta wings also held advantages that were entirely separate from cross range. A delta orbiter would be stable in flight from hypersonic to subsonic speeds, throughout a wide range of nose-high attitudes. The aerodynamic flow over such an orbiter would be smooth and predictable, thereby permitting accurate forecasts of heating during re-entry and giving confidence in the design of the shuttle's thermal protection. In addition, the delta vehicle would experience relatively low temperatures of 600 to 800 F over its sides and upper surfaces. By contrast, straight-wing configurations produced complicated hypersonic flow fields, with high local temperatures and severe temperature changes on the wing, body, and tail. Temperatures on the sides of the fuselage would run from 900 to 1,300 F, making the design and analysis of thermal protection more complex. During transition from supersonic to subsonic speeds, the straight-wing orbiter would experience unsteady flow and buffeting, making it harder to fly. This combination of aerodynamic and operational advantages led Donlan to favor the delta for reasons that were entirely separate from those of the Air Force."

Faget himself came to realize this and his later orbiter design proposals used delta wings.

Part 1: history.nasa.gov/sp4232-part1.pdf
Part 2: history.nasa.gov/sp4232-part2.pdf
Part 3: history.nasa.gov/sp4232-part3.pdf


Various other design elements were considered, which superficially might seem better, cheaper or more reliable. However when NASA actually examined these more closely, they were not. E.g, metallic TPS rather than silica tiles. People often think that was better and the silica tiles were a mistake. Yet metallic TPS upon further study was considered very dangerous because the slightest scratch of the coating would cause oxidation and a burnthrough. Faget himself was afraid of metallic TPS.

There is nothing in the MIT videos that argues there was some hugely better design available, or the shuttle design was severely flawed due to compromise.

bknight
2016-Mar-31, 11:58 AM
I believe the ceramic tiles where the way to go as they absorbed heat quite reliably, the only drawback was the adhesives and continual replacements.

CJSF
2016-Mar-31, 02:29 PM
Years ago, when I studied the Challenger accident, I came to the conclusion that the Nixon (and subsequent) administration's capping and cutting the STS R&D budget was a serious factor in many of the design decisions. I don't know how well that bears up as the years have gone by and more information has come to light (nor how the materials joema references address it).

CJSF

joema
2016-Mar-31, 04:16 PM
Years ago, when I studied the Challenger accident, I came to the conclusion that the Nixon (and subsequent) administration's capping and cutting the STS R&D budget was a serious factor in many of the design decisions. I don't know how well that bears up as the years have gone by and more information has come to light...

It was obviously a factor, but it is unclear whether more money would have helped to a major degree. Robert F. Thompson (head of the shuttle program) addressed this in his CAIB testimony. He said even had the money been available and they tried to build a fully-reusable TSTO design, it would have probably been more expensive to fly or failed outright due to technical complexity. His opinion was the technology didn't exist at that time -- even given increased funding. The winged fully-reusable manned booster discussed at that time would have been gigantic, and development cost and risk typically scales roughly with gross takeoff weight at a given design complexity.

Transcript: http://caib1.nasa.gov/events/public_hearings/20030423/transcript_am.html

CJSF
2016-Mar-31, 06:46 PM
Well, if you want to get an immediate answer, the largely political decision to go with segmented boosters transported across the country instead of unsegmented on-site production was a more pressing cause. No segments, no O-rings. Sure, there were other likely failure modes (the loss of tiles almost being one of them), but the actual Challenger accident could not have happened in that case. But we're drifting off the OP, and frankly, this accident still digs up traumatic memories for me, so maybe it's better to leave it for another time.

CJSF

joema
2016-Mar-31, 09:19 PM
...the largely political decision to go with segmented boosters transported across the country instead of unsegmented on-site production was a more pressing cause. No segments, no O-rings. Sure, there were other likely failure modes (the loss of tiles almost being one of them), but the actual Challenger accident could not have happened in that case....

This is sometimes discussed, however it is not clear the segmented design *itself* was the cause, nor that a monolithic design would have been free of problems. Aerojet proposed a monolithic SRB, however they (nor anyone else) had any production experience making one. By contrast Thiokol and Hercules had lots of experience making large segmented solid rockets. This was an issue since the shuttle SRB would be the largest solid rocket ever flown by far, and there was some concern about the manufacturing safety and quality control when making a single gigantic solid fuel casting.

A monolithic, non-segmented rocket can easily fail and explode, as is shown in this Minuteman test: https://www.youtube.com/watch?v=rRERvYr5G9A and also in this Navy SM-2 failure: https://news.usni.org/2015/07/27/navy-restricts-use-of-a-number-of-sm-2-missiles-following-uss-the-sullivans-launch-failure

Although the Challenger SRB failure was a segment field joint, it almost failed previously on a nozzle joint. The Aerojet monolithic SRB would still have had nozzle joints, and still been subject to failure there.

The problem with the SRB was not the segmented design but the field joint design was suboptimal, combined with NASA's poor operating procedures. In the Challenger case the overall Thiokol Director of the SRB project (Allan J. McDonald) was at KSC to sign off on the launch, but he refused because he felt it was unsafe. His exact words were: "I wouldn't want to be the person that has to stand in front of a Board of Inquiry to explain why we launched outside of the qualification of the solid rocket motor."

Despite him being the *Director* of the SRB program, he was overruled and they launched anyway. Decision making and management practices like that will eventually doom any vehicle -- no matter what the design. You can't design around that.

Afterward then the SRB was redesigned, it was vastly safer, and used a much better field joint. In fact during tests they drilled holes in the O-rings and imposed all kinds of adverse measures, yet it worked well. A segmented SRB design will still be used on the Space Launch System.

For anyone interested, by far the most detailed account of this ever written is "Truth, Lies, and O-Rings: Inside the Space Shuttle Challenger Disaster": http://amzn.com/B00B91LIVO

Re jettisoning the SRBs, this wasn't possible while they were firing. The firing SRB exerts tremendous force on the struts toward the direction of flight. The struts are designed to absorb that and only yield laterally (perpendicular to flight direction) when thrust is near zero: http://history.nasa.gov/rogersrep/v3o241a.jpg

Initially they planned to have thrust termination ports in the SRB nose, which would mostly neutralize the thrust even though the SRB was still firing, then jettison them in a contingency. However detailed mechanical analysis indicated this would require beefing up the vehicle structure by 20,000 lbs, which would cost nearly 1/3 the payload. That is the difference between sketching something that looks plausible vs doing the engineering analysis to determine what's actually required. There is a lesson there.

Ara Pacis
2016-Apr-01, 04:35 PM
Could there have been a way to safely dump liquid hydrogen? Also, is there some sort of chemical that could poison the SRB combustion reaction chemically or physically (like cryo nitrogen)? What about the flame out ports they use on solid rocket ICBMs?

swampyankee
2016-Apr-01, 06:51 PM
Could there have been a way to safely dump liquid hydrogen? Also, is there some sort of chemical that could poison the SRB combustion reaction chemically or physically (like cryo nitrogen)? What about the flame out ports they use on solid rocket ICBMs?

For the first? Possibly, but it's hard to imagine a dumping process that would get rid of LH2 faster than the engines burning it would. Poisoning SRB combustion? That would be difficult, if for no other reason than there is a lot of complicated fluid mechanics and chemistry going on in there.

joema
2016-Apr-01, 07:45 PM
Could there have been a way to safely dump liquid hydrogen? Also, is there some sort of chemical that could poison the SRB combustion reaction chemically or physically (like cryo nitrogen)? What about the flame out ports they use on solid rocket ICBMs?

There is no physical way they could dump LH2 fast enough to make a difference in an abort situation. Re poisoning SRB combustion, you may be thinking about liquid injection thrust vectoring. In this scheme a liquid (often freon) is injected on one side of the nozzle which creates an asymmetric thrust and steers the missile. It can be very effective, as seen in this photo of a Nike Sprint ABM: http://www.postwarv2.com/sprint/photos/js640_andrey_sprint_02.jpg

However there is no way I've seen to quench a large burning solid rocket. By definition it is providing its own fuel and oxidizer.

You are correct the Minuteman 3rd stage used thrust termination ports. However I think this was not to snuff out combustion but to momentarily produce a forward-firing negative thrust which allowed the post-boost vehicle to cleanly separate at a specific delta V.

Early during shuttle development a similar scheme was considered for the SRB. However -- unlike the Minuteman 3rd stage -- the SRB was gigantic and the orbiter much more fragile. Using ICBM-like thrust termination ports resulted in so much force the orbiter and ET would have required 20,000 lb of structural reinforcement, which was about 1/3 of the total payload.

VQkr
2016-Apr-02, 04:19 AM
Could there have been a way to safely dump liquid hydrogen? Also, is there some sort of chemical that could poison the SRB combustion reaction chemically or physically (like cryo nitrogen)? What about the flame out ports they use on solid rocket ICBMs?

Sure, detach from the ET. Liquid fueled engines can be stopped without exhausting the fuel (the SSMEs were always turned off, not run to depletion). If risk aversion necessitated the need to shut off the 1st stage boosters early during an abort, the solution would have been to switch to a liquid-fuel booster. This is what the Buran system used.

swampyankee
2016-Apr-02, 10:51 AM
Sure, detach from the ET. Liquid fueled engines can be stopped without exhausting the fuel (the SSMEs were always turned off, not run to depletion). If risk aversion necessitated the need to shut off the 1st stage boosters early during an abort, the solution would have been to switch to a liquid-fuel booster. This is what the Buran system used.

I think that liquid fueled boosters were one of the options examined early in the design process.

joema
2016-Apr-02, 12:53 PM
...Liquid fueled engines can be stopped without exhausting the fuel (the SSMEs were always turned off, not run to depletion). If risk aversion necessitated the need to shut off the 1st stage boosters early during an abort, the solution would have been to switch to a liquid-fuel booster. This is what the Buran system used.

Liquid engines do not solve all problems and they introduce their own risks. On STS-51F, the crew and vehicle were nearly lost when a spurious SSME shutdown happened and a 2nd engine almost failed. Ironically the solution in this case was inhibit redline limits, running the engines in open loop mode which would prevent another sensor-based shutdown but allow them to blow up.

The most critical phase is early in the ascent. The ability to shutdown a liquid fueled engine is often useless. Had the Saturn V lost a *single* engine during the first approx. 15 sec of flight it would have fallen back into the launch pad.

Likewise had Buran lost a single liquid-fueled booster engine, the opposite engine would be cut to maintain control, which would have dropped thrust:weight ratio below 1, and it also would fall back on the pad and be destroyed.

This actually happened with the N-1 launch vehicle: https://www.youtube.com/watch?v=m79UO4HOQmc

The RD-170 engines used in Buran/Energia are similar to the RD-180 which almost doomed this launch when it also shut down prematurely: http://spaceflight101.com/cygnus-oa6/by-the-numbers-how-close-atlas-v-came-to-failure-in-this-weeks-cygnus-launch/

And this liquid-fueled NK-33 engine simply exploded in flight: http://spaceflightnow.com/wp-content/uploads/2014/10/20141028-Antares-Explosion-for-story.jpg

Liquid fueled heavy lift boosters typically require many engines, each having highly stressed turbopumps. The more engines, the more probability of a spurious shutdown or an uncontained failure. Due to thrust:weight limitations, an otherwise benign shutdown can destroy the entire vehicle.

So the "safety" feature of shutting down a liquid fueled engine is often illusory. If it made that big a difference in reliability they wouldn't use the same segmented solid design on the upcoming Space Launch System.

bknight
2016-Apr-02, 08:09 PM
Liquid engines do not solve all problems and they introduce their own risks. On STS-51F, the crew and vehicle were nearly lost when a spurious SSME shutdown happened and a 2nd engine almost failed. ...
My bolding. From what I've read the crew was in no imminent danger as the second temperature anomaly occurred 8 minutes into the flight, might have been a one orbit, or they may have enough velocity to limp into an even lower orbit than accomplished. The mission might have been lost in that case.

joema
2016-Apr-02, 10:55 PM
The SSME shutdown happened at MET 05:49, which was after negative return and only 35 sec after press to ATO. This means it came within 35 sec of forcing a two-engine TAL abort with a 2nd engine about to fail. Before STS-51L (Challenger), they did not have bailout and ocean ditching was unsurvivable. IOW from 05:49 to 07:00, both surviving SSMEs had to work perfectly or they were dead, yet a 2nd engine indicated problems.

This is why engine sensing limits were inhibited -- they could not survive another shutdown -- whether it blew up or shut down would have been loss of crew and vehicle.

After 07:00 they reached single-engine TAL at which they could survive a 2nd SSME shutdown, thus they re-enabled redline limits. However when the 2nd SSME later showed more problems they risked another shutdown, so again inhibited redline limits all the way to orbit. They guessed it was more likely an instrumentation problem, and an abort to orbit was safer than a TAL abort, so they gambled on inhibiting limits.

It turns out both SSME problems were spurious instrumentation not real issues. This are some of the downsides to hyper-stressed liquid engines which are spring-loaded to shut down to avoid exploding. Each fuel turbopump is the size of a trash can, spins at 34,000 rpm and produces 63,000 shaft horsepower, with a discharge pressure of 6,000 psi. The previously-posted photo of the Antares failure was a turbopump explosion.

Jens
2016-Apr-03, 08:38 AM
There seems to be a bit of confusion in this thread. The original question was about Challenger specifically, not whether it could have been designed differently. I think the answer (no) has been fairly clearly stated.

CJSF
2016-Apr-03, 04:08 PM
Aw, I think it's alright that we drifted a little into the overall design and heritage of the STS, but yeah, the OP has been answered.

CJSF

publiusr
2016-Apr-03, 08:20 PM
The pre-Buran concept called OK-92 might have been able to do the trick
http://cosmoquest.org/forum/showthread.php?12512-Cassini-and-Saturn-s-moons&p=2347843#post2347843
http://www.buran.ru/htm/ok-92.htm

According to the book--the OK-92 had a large central solid for escape purposes--but the translation seems to imply a liquid.

Had Challenger been made like OK-92--with this F-111 escape cabin:
https://s-media-cache-ak0.pinimg.com/736x/94/f9/22/94f922e53b7641635e4f4d570c57dc32.jpg

The crew would have lived. Energiya used liquid strap-ons anyway--and with engines under the ET--there is no oxygen ramp to shed foam on Columbia either.

Some other ideas
http://www.wired.com/2014/05/a-relocated-relocatable-main-engine-cluster-for-the-space-shuttle-1975/
https://www.aiaa.org/uploadedFiles/About-AIAA/History_and_Heritage/Final_Space_Shuttle_Launches/ShuttleVariationsFinalAIAA.pdf

joema
2016-Apr-04, 03:05 PM
...Had Challenger been made like OK-92--with this F-111 escape cabin...The crew would have lived...

A cabin escape system was considered early in the design phase, but the experience with similar F-111 and B-1A systems was not good.

The one time it was used on a B-1A, one crew was killed and the other severely injured because of malfunctions in the complex system. Similarly many crews were killed when using the F-111 cabin escape system:
http://www.ejection-history.org.uk/Aircraft_by_Type/F_111/F-111.htm

It is easy to draw a picture of a cabin escape system. Making that actually work over a wide operational envelope is another thing entirely. A cabin escape system requires many different pyrotechnics, separators, sequencers, aero-stabilization under extreme conditions, air bags, etc. Yet it's used when the surrounding vehicle is torquing, disintegrating, maybe even exploding. All those complex separation systems would have to work in that environment. They add additional risk of fire, explosion or spurious activation. Because of the weight penalty and lack of confidence in the complex cabin escape system, these were dropped from later versions of the B-1 and F-111.

Even something that looks simple -- like adding thrust termination ports on the SRB -- has major repurcussions when it comes time to actually designing it and making it work. See below quote from History of the Space Shuttle, by T.A. Heppenheimer: http://amzn.com/B00K9MIZRO

"Studies showed that thrust termination would impose a heavy jolt upon the shuttle in flight. To prevent this sudden mechanical shock from producing structural damage, the orbiter needed more strength— and this extra robustness threatened to cost 19,600 pounds in additional weight.

VQkr
2016-Apr-05, 10:40 PM
If it made that big a difference in reliability they wouldn't use the same segmented solid design on the upcoming Space Launch System.

I was under the impression that that was largely a politically-motivated decision, not a safety or engineering driven one. Yes, there are still failure modes with liquid boosters.

joema
2016-Apr-06, 12:12 AM
I was under the impression that that was largely a politically-motivated decision, not a safety or engineering driven one...

If use of SRBs in the Space Launch System was mainly politically motivated, why did the independently-developed DIRECT launch vehicle also use those? https://en.wikipedia.org/wiki/DIRECT

It is more likely the primary determining factors were safety, having a known heavily-tested technology, and constraining initial development costs for a vehicle with a limited flight rate.

For later versions of the SLS, NASA studied using strap-on liquid-propellant boosters. I think the current plan is to use advanced solid boosters (not liquid) as a follow on to the five-segment shuttle-derived SRBs, but maybe it's still being studied.

Besides the high development and manufacturing costs, plus risk from a shutdown, another big problem with large liquid strap-on boosters is inability to control thrust. Unlike SRBs which have a tailored thrust profile: https://en.wikipedia.org/wiki/Space_Shuttle_Solid_Rocket_Booster#/media/File:Srbthrust2.svg

Liquid strap-on boosters are typically fixed thrust which tends to cause over-acceleration and high g forces. One solution is using throttleable engines which add yet more cost, complexity and risk of failure. One of the frequently-practiced shuttle contingencies was a stuck throttle on an SSME.

VQkr
2016-Apr-07, 01:44 AM
If use of SRBs in the Space Launch System was mainly politically motivated, why did the independently-developed DIRECT launch vehicle also use those? https://en.wikipedia.org/wiki/DIRECT

Because the entire premise of DIRECT was to re-use shuttle hardware.

Ara Pacis
2016-Apr-07, 06:29 PM
You are correct the Minuteman 3rd stage used thrust termination ports. However I think this was not to snuff out combustion but to momentarily produce a forward-firing negative thrust which allowed the post-boost vehicle to cleanly separate at a specific delta V.

Early during shuttle development a similar scheme was considered for the SRB. However -- unlike the Minuteman 3rd stage -- the SRB was gigantic and the orbiter much more fragile. Using ICBM-like thrust termination ports resulted in so much force the orbiter and ET would have required 20,000 lb of structural reinforcement, which was about 1/3 of the total payload.

What was the force from? Was it the struts between the SRB and the ET holding onto the SRBs forcing the Orbiter to pull them along? Or was it from the Orbiter suddenly having to carry the ET all by itself with more fuel than normal at SRB separation?

If the former, do the SRBs need explosive bolts or something or would a nested-cup design suffice to hold them together but allow them to separate cleanly in the event of an abort using thrust termination ports?

joema
2016-Apr-07, 07:15 PM
What was the force from? Was it the struts between the SRB and the ET holding onto the SRBs forcing the Orbiter to pull them along? Or was it from the Orbiter suddenly having to carry the ET all by itself with more fuel than normal at SRB separation?

If the former, do the SRBs need explosive bolts or something or would a nested-cup design suffice to hold them together but allow them to separate cleanly in the event of an abort using thrust termination ports?

I don't know. They initially thought it would work and maybe did back of envelope calculations which confirmed this, else they would never have proposed thrust termination. They probably later did more detailed simulation and mechanical analysis which indicated the forces were too high without major structural reinforcement.

Story Musgrave described the orbiter as a "butterfly bolted to a bullet". This conveys how delicate it was if subjected to forces outside the load paths it was designed for. However this is not unique to the shuttle. The tyranny of the rocket equation requires that all launch vehicles be extremely light. They cannot be designed to take lots of extra loads in unexpected directions "just in case" -- otherwise they can't lift the required payload. The Atlas booster that launched the Mercury astronauts was a thin metal balloon supported by internal gas pressure. If laid horizontally empty it would collapse if the internal pressure leaked out.

Before the early 1990s the struts holding the shuttle to the ET could not withstand a double SSME engine failure, even if both SRBs worked perfectly. After Challenger STS-51L in 1986, many upgrades were done, which continued through the 1990s. These enabled many more abort options including (theoretically) a triple SSME failure immediately after liftoff. I don't know the mass penalty for the reinforcement but it was obviously not severe, as the shuttle did not take a big payload hit afterward. See attached charts.

Unlike SRB ignition which which has a rapid but controlled thrust ramp up, SRB thrust termination would happen suddenly at full thrust. I believe even for SRB ignition the maximum permissible left/right time differential is 100 milliseconds -- more than that would have destroyed the structure. Thrust termination would necessarily be more abrupt and less controlled, and probably subject the ET to significant loads.

As you said, if thrust termination happened early in the flight, aero forces on the SRBs and ET would also place large loads shuttle/ET struts and SRB struts. These SRB strut loads would be in the reverse direction of the normal load path.

Hornblower
2016-Apr-08, 01:31 AM
I would say it is important to remember that the shuttle bundle at liftoff and early ascent is fragile compared to a warhead being delivered by an axially symmetrical Minuteman missile. With the latter it is no big deal if the thrust termination measures with a solid fuel rocket mean cutting off with a bang.

publiusr
2016-Apr-09, 06:34 PM
It is easy to draw a picture of a cabin escape system. Making that actually work over a wide operational envelope is another thing entirely.

That's true--that's why both it and liquid fuel boosters should work together.

In Challenger--the crew cabin came away anyway--I'm just asking for a chute for it.

The modularity from an Americanized Energiya/Buran was really what I wanted all along before settling for Ares V/SLS.

Despite the risk--I like parallel staging/ side payload mount in that very wide aerobrake disks and hypersonic boilerplates can be launched instead of the orbiter or payload pods.

This way, large hypersonic craft can be launched on pop-up trjectories, blow the Energiya/SLS type core--and glide back for full scale tests--not top mount where pitch loads and bending moments play havoc.

With engines on the ET (Energiya) the orbiter can be simplified.

In fact--I can imagine a payload free orbiter made a little more strongly--passengers only--looking a bit like this:
https://en.wikipedia.org/wiki/Airbus_Defence_and_Space_Spaceplane

Combined with this:
https://en.wikipedia.org/wiki/Airbus_Defence_and_Space_Spaceplane

Launch heads up away from foam--no SRBs.

This may be all we have left to look for:
https://en.wikipedia.org/wiki/SpaceLiner

Pure rocket--no airbreathing complexity.

John Mendenhall
2016-Apr-14, 12:13 AM
Hmm. Seems to be a lot of disagrrement about the development history of th Space Shuttle.

Here's a lighter piece of non-history: our two boys were both under six years old for the early flights, which they followed avidly. When I got home, I got daily reports on what the 'Space Shovel' had done that day. Too fitting to correct!

swampyankee
2016-Apr-15, 03:49 PM
Certainly, retrospect (and even contemporary thinking) shows that there were problems with both the conceptual and detail design of the SLS, but a more immediate problem was that NASA's leadership overestimated their technical competence and underestimated the risk: the Challenger disaster was not an unpredictable part failure, but was the result of ignoring a known issue with a critical part. There were certainly alternative elastomeric and non-elastomeric sealing systems that had the potential to replace the compound used for the SRB O-rings; there does not seem to have been any post-Challenger investigation into alternatives. Now, I've worked in aerospace, and it is not unknown for aircraft systems, especially helicopters, to have components where failure is likely to be fatal: Sikorsky changed from built-up sheet metal spars for its helicopters' rotors to extruded tubular spars, as the latter could have a real-time crack detection system installed (the blades were pressurized; a crack would cause the spar to lose pressure and trigger a warning. The pilot would then have a reasonable period -- several hours -- to land before the rotor blade fell off). In general, manufacturers find ways to either correct or ameliorate this sort of problem; with the SLS the amelioration seems to have been a change in operational parameters: don't launch when it's too cold. Columbia showed another failure mode, which, bluntly, should have been known: foreign object damage is a well-known and well-studied problem in aerospace, and NASA had known that ice and/or foam was being shed from the external tank. Birds have caused the loss of many military and commercial aircraft (including nearly killing a DC-10 full of people, because GE hadn't bothered to actually test their CF-6 engines), one airline stopped using 737s because the engines had a habit of vacuuming gravel up from a runway, a Concorde was lost because of debris left on the runway, and it's a real design concern for seaplanes and for aircraft with pusher propellers. Where the ice shed from airframes goes is also a concern (I've done some of those analyses), as that ice slows down rapidly and can damage parts of the airframe where it hits.


(eta)

Don't tell me that this would not happen with a private contractor. First, the DC-10 and the Concorde were being operated in revenue service, not government service, and second, I can tell you some horror stories about companies knowingly using counterfeit parts.

joema
2016-Apr-16, 12:30 AM
...There were certainly alternative elastomeric and non-elastomeric sealing systems that had the potential to replace the compound used for the SRB O-rings; there does not seem to have been any post-Challenger investigation into alternatives.

After Challenger, many different O-ring sealing systems were evaluated, including non-elastomeric metal-to-metal sealing rings. They were not only analyzed, but actually test fired in full size motors. The improved triple-sealed J-joint design using viton rubber O-rings was selected on the basis of those tests as the best performing, most reliable option. The final tests included full duration test firings with severely degraded O-rings that had been intentionally damaged. They were totally successful. This technical history is documented in tremendous detail in Allan J. McDonald's book "Truth, Lies, and O-Rings: Inside the Space Shuttle Challenger Disaster": http://amzn.com/B00B91LIVO



Columbia showed another failure mode, which, bluntly, should have been known: foreign object damage is a well-known and well-studied problem in aerospace, and NASA had known that ice and/or foam was being shed from the external tank...

This was well known. The shuttle "red book" design specification stated the maximum TPS allowable impact was 0.006 foot pounds, which is tiny. The vehicle was being operated far outside the design spec. It was not a case of not understanding FOD. Rather it was a case of willfully operating far outside the specification. The TPS was generally very resilient and over designed for a single mission. Over time this resiliency led to a false confidence and a twisted view that if the vehicle survived x amount of TPS damage it must be safe to operate in that zone. As Richard Feynman termed, this was "normalization of deviance".

VQkr
2016-Apr-16, 01:37 AM
This was well known. The shuttle "red book" design specification stated the maximum TPS allowable impact was 0.006 foot pounds, which is tiny. The vehicle was being operated far outside the design spec. It was not a case of not understanding FOD. Rather it was a case of willfully operating far outside the specification. The TPS was generally very resilient and over designed for a single mission. Over time this resiliency led to a false confidence and a twisted view that if the vehicle survived x amount of TPS damage it must be safe to operate in that zone. As Richard Feynman termed, this was "normalization of deviance".

What would the practical method have been to bring impacts on the TPS back into design spec? Heavier external tank insulation system?

swampyankee
2016-Apr-16, 03:43 AM
After Challenger, many different O-ring sealing systems were evaluated, including non-elastomeric metal-to-metal sealing rings. They were not only analyzed, but actually test fired in full size motors. The improved triple-sealed J-joint design using viton rubber O-rings was selected on the basis of those tests as the best performing, most reliable option. The final tests included full duration test firings with severely degraded O-rings that had been intentionally damaged. They were totally successful. This technical history is documented in tremendous detail in Allan J. McDonald's book "Truth, Lies, and O-Rings: Inside the Space Shuttle Challenger Disaster": http://amzn.com/B00B91LIVO


I didn't follow NASA very well at the time, so I was not aware evaluated alternatives.




This was well known. The shuttle "red book" design specification stated the maximum TPS allowable impact was 0.006 foot pounds, which is tiny. The vehicle was being operated far outside the design spec. It was not a case of not understanding FOD. Rather it was a case of willfully operating far outside the specification. The TPS was generally very resilient and over designed for a single mission. Over time this resiliency led to a false confidence and a twisted view that if the vehicle survived x amount of TPS damage it must be safe to operate in that zone. As Richard Feynman termed, this was "normalization of deviance".

Allowable impact of 0.006 ft-lbf is insanely low: it's something like a 1 grain object traveling at 50 ft/sec. Was that allowable based on no damage to the TPS or was it based on an amount of damage that was low enough not to compromise the TPS's effectiveness?

The report I remember seeing about the Columbia was that some of the people involved did not believe the foam insulation could damage the tile; if that was an accurate description of people's beliefs, there was a severe lack of understanding. Later, an engineer finally got around to shooting a piece of foam at a leading edge at something like 300 ft/sec; not only did the chunk of foam damage the leading edge TPS, it damaged the metal structure behind it.

joema
2016-Apr-16, 11:51 AM
What would the practical method have been to bring impacts on the TPS back into design spec? Heavier external tank insulation system?

There was no simple solution for this. They knew the TPS would likely sustain some damage. Originally they planned to fly a tile repair kit and a maneuvering backpack on all shuttle missions, including the very first one STS-1. Martin Marietta developed the backpack and tile repair kit. Attached is a photo of astronaut Anna Fisher testing it in 1980, and overview diagrams of the system. It was determined flying close to the underside might risk damaging other tiles during the repair effort, so they decided not to fly it.

joema
2016-Apr-16, 12:17 PM
...Allowable impact of 0.006 ft-lbf is insanely low: it's something like a 1 grain object traveling at 50 ft/sec. Was that allowable based on no damage to the TPS or was it based on an amount of damage that was low enough not to compromise the TPS's effectiveness?

That was just the early design spec, which indicated the original designers knew the tiles were fragile and from a safety standpoint they wanted essentially zero impacts. However from another perspective the tiles were very robust. The TPS was designed to protect the vehicle for 100 flights -- not one flight like an Apollo heat shield.

That excess margin meant the vehicle could (and often did) safely reenter with many tiles entirely missing. Large sections of many tiles could be gouged to a thin layer, yet still provide adequate protection for a safe landing. More maintenance would be required to refurbish the vehicle, but it was not a flight safety issue. So the tiles that cover most of the orbiter were superficially fragile, yet far over-designed from a thermal protection standpoint. Instead of addressing the root problem, over time NASA came to rely on this margin which was *not* there to compensate for damage but a side effect of the multi-mission durability.

The RCC panels which caused the Columbia disaster were totally different, but were viewed as tough and able to shrug off considerable impacts. I don't know what data supported this or whether it was a subjective belief. I don't recollect the CAIB report probing for past test data during TPS development which supported this.

There were two separate TPS damage prediction software programs, one was "Crater" used for the silica tiles, the other for an RCC panels. During the Columbia mission Crater indicated the possibility of severe tile damage but this was discounted because the program generally over-predicted damage. Of course if you're going to discount the results why even run it?

The other damage prediction program for the RCC tiles was never well described in the CAIB report. This was a significant shortcoming since RCC panel failure was the actual cause of the disaster.


...some of the people involved did not believe the foam insulation could damage the tile; if that was an accurate description of people's beliefs, there was a severe lack of understanding.... Yes, whether the RCC panels or tiles, many at NASA likened the risk to an empty styrofoam cooler blowing off a pickup truck bed and striking the following car. That rarely does any damage.

However the formula for kinetic energy is KE=1/2mv^2, which shows impact force goes up as the *square* of the velocity. A light object can do damage if traveling fast. Anyone who has been struck with a aggressively-hit ping-pong ball knows that. A true "rocket scientist" should understand basic math and be guided by that, not by what happened on his last trip to the beach.

publiusr
2016-Apr-16, 08:07 PM
Venture Star type heatshields were going to be metal right? That'd been perfect for an Americanized Buran replacement.

joema
2016-Apr-16, 08:40 PM
Venture Star was to use metallic tiles, usually described as safer but that depends on the exact composition and characteristics of the tiles. Metallic tiles were considered for the shuttle but they were very susceptible to oxidation, protected only by a thin coating. They were overall considered less safe than the silica tiles actually used. I suspect the Venture Star tiles were better but I don't remember the details.

Venture Star had very low mass for its surface area, which decreased the reentry heating problem. That and 25 years more technical progress may have enabled a different type of metallic tile which wasn't available when the shuttle was designed in the early 1970s.

Jens
2016-Apr-19, 02:40 AM
a Concorde was lost because of debris left on the runway,

I know this is a bit of a nitpick, but my understanding is that the Concorde was not solely lost because of the debris. Apparently it was heavily overloaded, had a bad center of gravity, might have had problems with the landing gear (which was repaired a few days before the crash), and the crew shut down one of the engines due to an alarm when the safety procedures say that you shouldn't shut down an engine until reaching 400 feet and achieving stable flight. The captain pulled the plane off the ground with insufficient speed because it was veering off the runway. Initially the French investigators seems to have wanted to pin the blame on the other plane (Continental?).

joema
2016-Apr-19, 04:13 PM
I know this is a bit of a nitpick, but my understanding is that the Concorde was not solely lost because of the debris. Apparently it was heavily overloaded, had a bad center of gravity, might have had problems with the landing gear (which was repaired a few days before the crash), and the crew shut down one of the engines due to an alarm when the safety procedures say that you shouldn't shut down an engine until reaching 400 feet and achieving stable flight. The captain pulled the plane off the ground with insufficient speed because it was veering off the runway. Initially the French investigators seems to have wanted to pin the blame on the other plane (Continental?).

That is misleading or incorrect. The accident was caused by several factors:

(1) The metal debris was struck at about 175 knots, which is well beyond the V1 accelerate/stop speed of 150 knots. There was no way the plane could stop so was committed to taking off. Investigators simulated what would have happened had the pilot tried to abort the takeoff, and concluded it would have resulted in a high-speed runway overrun and a catastrophe no different from the crash.

(2) To simplify and lighten the aircraft, it was designed without flaps or slats which necessitated a very high takeoff speed. E.g, Vr (rotate speed) was 198 knots (228 mph), roughly the space shuttle landing speed. By contrast a Boeing 777 has a typical Vr speed of 160 knots.

(3) The landing gear and tires were known to be susceptible to failure. 57 prior Concorde flights had experienced burst tires. Of the 57 prior incidents, 19 were caused by foreign objects and 37 times the tire just failed. Twelve of those tire failures produced debris that damaged the plane, and six times the fuel tanks had been punctured from tire or related debris.

(4) Previously when tires exploded it would sometimes tear off the metal water spray deflector behind each tire, and that debris would strike the aircraft. British Concordes were modified with a retention system to prevent this but French Concordes were not. It is unknown if that modification would have lessened the damage on flight 4590.

(5) When the tire exploded it caused a fire alarm in engine #2 which was already shutting down by itself. After #2 rolled back to essentially idle power by itself, the pilot called for fire suppression on #2, then shut it down manually. The other three engines automatically went to 104% emergency power based on the control system. However engine #1 was also damaged by the event so was slower to respond. This created a thrust asymmetry during takeoff that was pushing the aircraft off the runway, so the captain rotated early to avoid running off the edge. Engine #1 partially recovered to about 70% then after takeoff it also rolled back to idle by itself.

(6) The severe thrust asymmetry in flight was pushing the aircraft out of control, so the captain commanded #3 and #4 back to idle. This was a final desperate attempt to try and crash straight ahead instead of being totally out of control.

Other conclusions of the investigation

- Aircraft weight exceeded max specified by 1 ton, however this had negligible effect on takeoff performance.
- Metallic debris strip was from a DC-10 which had been improperly maintained.
- Crew tried to retract landing gear which would have greatly decreased drag. However gear was stuck down due to damage from the tire explosion.
- Even had all engines worked perfectly and stayed at full power, the severe structural damage and fire would have led to loss of the aircraft.

The complete investigation report is here: https://www.bea.aero/docspa/2000/f-sc000725a/pdf/f-sc000725a.pdf

joema
2016-Apr-20, 09:42 PM
....Was dumping the SRBs at that low altitude an available option and was it survivable?

See attached photo from the new book "Into the Black: The Extraordinary Untold Story of the First Flight of the Space Shuttle Columbia and the Astronauts Who Flew Her", by Rowland White. The caption reads: "'No stone unturned.' While the astronauts practiced abort procedures in the sims, NASA used wind tunnels to test whether in an emergency it might be possible to separate the orbiter from the stack at hypersonic speeds while the SRBs were still burning. They thought not."

swampyankee
2016-Apr-20, 11:43 PM
Joema, as an aside, tailless deltas, like the Concorde, usually don't have flaps, as trailing edge flaps cause a nose down pitching moment, which can only be counteracted by deflecting the elevons upwards, which decambers the wing and reduces the lift coefficient. On a highly swept delta, like the Concorde, there's a controlled separation from the wing's leading edge, which leads to vortex lift. See here (http://soliton.ae.gatech.edu/labs/windtunl/classes/unstaero/vortflo1/vortflo1.html), here (http://www.concordesst.com/history/eh2.html), and, more generally, here (http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19710012498.pdf), so leading edge flaps may be superfluous.

joema
2016-Apr-21, 03:33 PM
Joema, as an aside, tailless deltas, like the Concorde, usually don't have flaps, as trailing edge flaps cause a nose down pitching moment, which can only be counteracted by deflecting the elevons upwards, which decambers the wing and reduces the lift coefficient. On a highly swept delta, like the Concorde, there's a controlled separation from the wing's leading edge, which leads to vortex lift. See here (http://soliton.ae.gatech.edu/labs/windtunl/classes/unstaero/vortflo1/vortflo1.html), here (http://www.concordesst.com/history/eh2.html), and, more generally, here (http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19710012498.pdf), so leading edge flaps may be superfluous.

You are generally correct, however first some background:

Delta wing aircraft with tails often have flaps, such as the A-4 Skyhawk: http://www.fiddlersgreen.net/aircraft/McDonnell-Skyhawk/IMAGES/landing-A4-Skyhawk.jpg
And Gloster Meteor: https://upload.wikimedia.org/wikipedia/commons/a/a9/Gloster_Javelin_FAW.1_XA563_FAR_10.09.55_edited-2.jpg

While it is generally true that tailless deltas do not have flaps, there are exceptions such as the TU-144 which effectively had flaps via symmetrical eleveon displacement, as can be seen in this landing photo: http://www.luchtzak.be/pictures/albums/userpics/10239/normal_tu144newyork4.JPG

However this did not actually provide a lower landing speed than Concorde since the wing design was less refined. But designers often consider adding lift-increasing devices to delta winged aircraft. The planned follow-on Corcorde "B" had leading-edge slats and a lower landing speed: http://www.concordesst.com/graphics/baircraft.jpg

The Boeing Sonic Cruiser would have used trailing-edge flaps on its delta wing: http://vignette2.wikia.nocookie.net/aircraft/images/d/d4/Boeing-Sonic-Cruiser-Concept.jpg/revision/latest?cb=20131022195750

Boeing's delta-winged 2707 SST also used leading-edge slats and trailing-edge flaperons (see attached).

The tailless delta-wing space shuttle orbiter had a large adjustable body flap, however this was not used for landing: http://science.ksc.nasa.gov/shuttle/technology/images/aft_fuselage_2.jpg

The Beech Starship was technically a tailless delta wing, although it did have "tip sails". It used trailing-edge flaps which can be seen here: https://upload.wikimedia.org/wikipedia/commons/6/69/Beechcraft_Starship_fly-by.jpg

The Mig-29 uses leading-edge "Krueger flaps", but these are nonetheless classified as flaps: http://i234.photobucket.com/albums/ee39/desaix_photos/kub2.jpg Use of Krueger flaps on various pure delta-winged aircraft has been studied, and I saw a reference the F-16XL used these but am not sure.