Worrisome engineering patches

Space Shuttle Discovery being transported to the launch pad. Photo credit: Larry Tanner, USA

From Slashdot: BJ_Covert_Action writes:

“NASA engineers have finally discovered the root cause of the cracks that have been found on space shuttle Discovery’s main external tank. The main tank, one of the ‘Super Lightweight Tank’ models developed by Lockheed-Martin, employs an aluminum-lithium alloy developed by Lockheed-Martin specifically for this application.

The new alloy is used in various structural stringers throughout the SLWT design. Unfortunately, the batch of this alloy used in the tank that is currently mated with the Discovery shuttle appears to be of low quality. The alloy used in the stringers has a ‘mottled’ appearance, compared to the nominal appearance typically used in the main tank stringers (see picture in article).

This appearance is indicative of a fracture threshold that is significantly lower than typical. NASA has determined, through testing, that this low grade alloy has only 65% of the fracture strength of the nominal alloy typically used.

NASA engineers have devised a potential fix to the problem that they are currently testing to ensure the repair will cause no unintended consequences. NASA plans to have the Discovery shuttle ready to launch again by February 24th, 2011.”

===========================================================

I’m reminded of another ship with low grade metal. Let’s all hope the solution they devise is sound, otherwise it may be a terrible ending for the shuttle program.

Advertisements

71 thoughts on “Worrisome engineering patches

  1. I am hoping that this wasn’t a cost saving, outsourcing issue. The understanding of contamination wasn’t understood as well for a previous ship with brittle metal. There is no excuse today for this.
    It is interesting that such parts made it into the system in a program that has been running for as long as the shuttle program.
    John Kehr
    The Inconvenient Skeptic

  2. If quality control and basic batch testing are out the window, why not just make it out of pot metal? I could mention other issues with and uses for lithium, but I am SO not going there.

  3. That we’re still using the outdated shuttle system instead of being able to take scheduled recreational flights to Mars is the true scandal here. Pathetic.

  4. Anthony:
    The problems with the NASA space shuttle are a tragedy. Despite that, some may be amused by the following pertinent facts.
    The Roman military set standards for sizes of things: standard nails, boots, etc.. One standard was the axle length of carts, and it derived from the separation distance of the rods or ropes that connected a cart to each side of a horse.
    Throughout the following millennia this standard axle length continued in use for horse-drawn carts. And railroads existed for frequently used tracks mostly associated with mines and quarries, so the railroad tracks had a standard separation determined by the axle length.
    The standard rail separation continued in use when the steam age arrived (although Brunel attempted to get it increased). Railroad bridges and tunnels were constructed with according size.
    The parts of NASA space shuttle were commissioned from organisations around the US. These parts had to be transported to Florida for assembly, and the larger ones were transported by rail.
    So, each major component of the space shuttle was designed such that it could be carried on a railroad truck and could pass through railroad bridges and tunnels.
    Is it any wonder that the space shuttle has had problems when all its major parts are defined by the size of a horse’s rear end?
    Richard
    REPLY: While that story seems plausible, it has been shown to be false.
    http://www.snopes.com/history/american/gauge.asp
    -Anthony

  5. I think this is very serious. A structural failure in the tank would almost certainly bring the vehicle down, resulting in the loss of the crew (as happened with Challenger, but for a different reason). Also, I understand that Lockheed have stopped building the tanks in the factory New Orleans.
    Far better, in my opinion, to end the shuttle program now rather than take any undue risks. I think that the shuttle is a magnificent work of engineering, but it has already killed 14 people. There are other ways to finish the space station build.

  6. Engineering design usually has a fairly large safety factor built into material specifications. On such a weight-critical application, this may have been pared to the minimum. Material not up to spec in those circumstances is a big issue. If the designers and test engineers have accurately determined the strength reduction, they should be able to compensate with additional stringers, at the expense of payload.
    A nervous time for the mission astronauts, who don’t need their stress compounding with press speculation.

  7. The entire premise of the Shuttle was flawed.
    Why try to make a boxcar both reusable and safe enough for passengers?
    Why not make a smaller, reusable glider for passengers and a separate launch vehicle for cargo? That way, you could focus all the safety issues on a much smaller vehicle. Costs would have been much lower and it would have been safer for the passengers.

  8. Fantastic and they found this problem just in time for the very last (maybe) space shuttle flight ever. After that, what are we left with? An incredibly expensive orbiting piece of space junk that the US can’t even get to on its own any more. It would be cheaper just to fly it into the ocean and use the budget savings to do some real work on climate science – but no, all that friction on re-entry would contribute to global warming and some piece of it might hit a penguin who wasn’t able swim fast enough to get out of the way because of the tag on his wing.

  9. NASA does politics, not engineering. They wouldn’t have changed the insulation on the tank for greenie reasons and lost a shuttle if they did engineering.

  10. OK, so I had a cup of coffee (with sarcchrin) and looked at the article again. The REALLY distressing thing is that the difference between the good stuff and the bad stuff is about as obvious as looking at your feet to see if you have shoes on.
    So the manufacturer made the alloy, but nobody tested it or even looked at it, the fabricator made the parts, but again nobody QC’d it or even looked at it and asked why it looked funny; then the NASA fuel tank contractor received it and did not QC it or even look at it and the assemblers built the tank (not their first one) without noticing that it looked a little strange. Millions of dollars later, somebody finally got down to LOOKING AT IT and guess what? It looked funny and didn’t work as specified.
    Sort of makes you wonder about the other zillions of parts that make up the Shuttle. Maybe it is a very good thing that it is being retired.

    • This really is a scandal.
      All engineers, everywhere and of whatever persuasion, should call for a thorough and detailed public inquiry into what went wrong.
      If we don’t it will happen again and again.

  11. This is reminiscent of the awful tragedy of the beautiful but fatally flawed British commercial airliner, the De Haviland ‘Comet’, that was designed and built before much was known about the long-term structural integrity of various lightweight metals.
    I once worked with a former US Navy flyer who had gone into the military directly from college. His recounting of the safety speech his first Navy flight instructor made to the new trainees has stuck in my mind for over forty years;
    “Never forget that anything you fly that was commissioned for the US Armed Forces was built by the lowest tenderer and is maintained by high school dropouts!”

  12. Ground it, and end the Shuttle Program. 2 disasters too many, no need to roll the dice for a 3rd.
    The last Shuttle crew will be disappointed, but very much alive. They can be the museum curators for the exhibit that NASA should open…. feauturing a real live surviving Shuttle.

  13. No worries here. All those Shuttle folks have jobs as long as that bird is sitting on the ground. No need to hurry. None at all. In fact, if this gets fixed, we can probably find something else that’s a problem.
    This is better than the Post Office, huh?

  14. Is has been a long time since I’ve used my Material Science degree, but I can assure you this falls into the “not good” catagory.
    But, this thing *should* have been so overengineered, 65% should still get it done comfortably.

  15. Was it here that I read (or perhaps a book on the subject) that the Apollo system had over 6 million parts and that NASA safety requirements allowed it to fly as long as they were confident that there would be less than a 1% failure rate. That would still leave about 60,000 parts that could fail. Pretty brave those astronauts!

  16. Anyone who is into aircraft development might remember that lithium-aluminum alloys were in the initial design of the Boeing 777. Boeing suffered major delays when cracking issues emerged in the lithium-aluminum structures. I am sure NASA was aware of this issue and I am also sure that metals companies made improvments in both the alloys and their processing methods. I am sure that the parts that NASA sourced were well quality controlled but is their an aging issue with this material. Metals with small atomic numbers and low melting points might see atoms quickly migrate which changes the microstructure of the metal and makes them age faster than expected.

  17. Vendor QC for the loss. Those stringers weren’t QC’d at Michoud Assembly Facility. And they were sitting on the shelf for a couple of years before being assembled on the stringer panels about 2-3 years ago. Your toes would curl if you knew some of the covered up quality defects and union worker sabotage that went on at that Lockheed run facility.

  18. Months of delay w/this launch — but I agree w/this considering there’s an engineering problem. Then why didn’t they consider delaying the launch of Challenger in 1982 in freezing weather (& after warning of danger from engineers) for just a few days?

  19. I recall an issue with the foam on the shuttle some time ago that had to do with some kind of change due to an environmental regulation. I wonder if this could be the result of a similar thing?

  20. Many years ago in a materials class, the professor mentioned about “Liberty Ships”, and how a few broke up in the N. Atlantic. Seems they also had a problem with the North Atlantic cold, causing the steel to change from ductile to brittle.

  21. There is hardly anything that comes out of NASA that I accept without thinking about potential political angles.

  22. I think you are rather unfair to the Titanic. It was actually an outstandingly safe ship.
    Remember that it was:
    – run at full speed (>20 knots) through a known iceberg area
    – when a collision was inevitable, maneuvred in a way that was guaranteed to maximise damage
    Even then it floated for four hours after the collision, allowing ample time for evacuation. However this, in line what had happened before, was handled with outstanding incompetence, causing many lifeboats to be launched not even half-full. Moreover most of these boats deliberately refrained from picking up survivors after the ship had sunk.
    The Titanic disaster was caused by a combination of regulatory failure (specifying much smaller lifeboat capacity than needed) and an almost incredible lack of professionalism by the captain and crew.

  23. I saw some Titanic show on the History Channel or something where they investigated the hull remains and concluded that manually formed cast-iron rivets in the lower bow of the ship (where the riveting machine couldn’t fit) failed, allowing the metal plates that run horizontally along the hull to buckle. The rest of the rivets were steel and machine-formed, and much stronger.

  24. This subject of failure analysis of aluminum alloys is one that I once could claim some expertise. Base on what they have found and reported, I think this is a classic case of intergranular stress-corrosion cracking resulting from improper heat treatment. Ordinary tensile testing for quality control won’t show this. I doubt the metal orginally had the mottled appearance because that would have been a sign that there was something wrong with the heat treatment. They seem to have found a solution and didn’t blame CO2.

  25. Appears to be possible surface corrosion which high strength aluminum alloys tend to do unless surface treated.
    There was also some dissension on the Challenger disaster:
    http://www.miaminewtimes.com/1990-08-22/news/bent-out-of-shape/
    In addition, NASA management had predicted a mission failure rate of about 1:100,000, their engineers about 1:100 and empirical experience being about 1:35. The mishap appeared to fit the empirical data the best since the mishap occurred on the 31st flight (counting test flights) for this general design (not specific vehicle). The Columbia fared somewhat better, lasting until the 113th mission. How many would decline to fly knowing the actual odds?
    Because of the predicted low failure rates, there also was never any rescue systems developed in case of problems while in orbit. I suspect that this is the real reason that NASA declined offered telescope examination of the Columbia tiles, as there was nothing they could do anyway except let everyone know in advance how doomed they were.
    NASA management had also ignored engineering warnings about the use of 100% oxygen atmospheres in the Apollo, resulting in the Apollo-1 loss of crew.
    Can’t wait to see if the space vehicles being developed for public use do any better.

  26. Burma shave signs from a fighter pilots lounge: If you taxi without a care you may be buying a one way fare. Best they caught the problem before shuttle takeoff.

  27. Obama is yanking money from space programs and we go back 100 years to rail. Railroads don’t have this problem. Trains don’t go where we want to go. So it means they want us not to have cars but save energy.

  28. Back in my flight instructor days, I had the dubious opportunity to use a new type of basic trainer. This aircraft made by a major manufacturer, was a two seat 112 hp
    putt-putt that was supposed to give a “high performance” feeling to the flight training experience. It did. truly nasty stall/spin characteristics that rivaled that of a jet
    fighter (notably the F-104) AND it required a level of skill to recognize the immediate
    need to keep the heck out of a spin! On top of this, they were having ah, quality control
    issues at the factory. Little things, like the rudder would jam hard over as the airplane
    entered the spin and -stay there- enhancing the spin experience.Then, the wing spar
    carry through it had several joints and bolts holding the wing on, except when they weren’t all put on at the factory. I had a demo flight with the dealer that consisted
    of spins, (no locking rudder-found that out later) and wingovers, chandelles , etc.
    then several hard landings (as part of the demo ) to show how rugged the aircaft
    was for light training. We taxi up to the hangar, and the chief mechanic walked and said:”Heard from the factory gotta pull the inspection panel.” Three of the four bolts
    that held the wing on were missing. We’d already paid for this airplane. Which would’ve been a great wind tee (direction pointer) and darn little else. took about
    10 years to engineer that thing to where it is safe. The shuttle I feel is not safe..
    I want to ride my atomic airliner to the Space Port-to go to the Moon then Mars!

  29. I’m at a loss here to understand how this could happen without gross corruption in the process.
    I’ve manufactured flight grade parts, both for aircraft and space vehicles. The pedigree for the metals (“Certs”, as they’re commonly called), the manufacturing processes themselves, almost to the chain of custody, it all is tracked and documented. Inspectors will come on site and check your documentation, and they’ll check your inspection tools for calibration. You guys think science is rigorous??
    The quality of the parts should never be in question. The ONLY questions that remain are:
    1. The inherent design, and
    2. The integrity of the manufacturing chain.
    #1…Well, we get what we get. But #2? A failure there points to corruption, either for political or financial gain. In the case of the former, it’s usually tied to the latter.
    So I’m puzzled here because I know the process, and I don’t understand how such an obvious flaw could slip by the cadre of inspectors. There’s got to be that faint scent of cover-up at Canaveral.

  30. Now that they have discovered the problem. They must inspect all critical components for any sign of cracks. The probability of fatigue cracking failure at launch (from high vibrational stress at the bottom of the crack) is too high. Pilots and mechanics inspect for cracks. I know because I was a navy pilot and did research at the Naval Air Materials Center specifically with aluminum alloys. http://www.stormingmedia.us/68/6848/0684884.html

  31. “Then why didn’t they consider delaying the launch of Challenger in 1982 in freezing weather (& after warning of danger from engineers) for just a few days?”
    There’s no schedule pressure anymore. NASA knows there’s only a few missions left and no-one really cares about them other than NASA workers and astronauts… delaying the launch just means they have jobs for longer.
    And that’s probably the root cause of this problem. When people discover they’ll be losing their jobs, the usual reaction is for the better ones to leave so the project is left in the hands of the few, if any, who are totally dedicated and the ones who can’t get a job elsewhere.
    In the Challenger case I seem to remember they did look at a graph of O-ring leaks vs temperatures, but it only showed the temperature range where there were problems so there was no obvious correlation. Combine that with the pressure to launch ASAP and ‘go fever’ pushing people to allow a flight in conditions where things had gone wrong before but no-one had died yet and that explains pretty much everything.

  32. >>The Titanic disaster was caused by …. an almost incredible
    >>lack of professionalism by the captain and crew.
    You mean that they did as they were told by management. A bit like the Challenger disaster, where management said ‘launch no matter how cold it is’. So were the Challenger crew culpable, or NASA management? Were the Titanic crew culpable, or the shipping-line management?
    .

  33. I wonder how much of the fact that once they engineers who built the tank are done they are out of a job. How much effort goes into work when you have been given your two week notice. I hope not but it could be there. Why go over the pieces without the the normal microscopic check when it worked before and it is the last one to make.

  34. Richard S Courtney, re railroad gauges.
    New Zealand railroads have an exceptionally narrow guage, designed to facilitate getting trains around bends in mountainous terrain. It tends to limit speed somewhat, as locos will ‘rock’ themselves right off the rails if speeds are too high. If you are interested, look up NZ’s ‘Raurimu Spiral’, which winds up something like 3000 feet from it’s start and is an interesting daylight journey. This solved a huge construction problem – the originator was an engineering apprentice!

  35. http://www.itl.nist.gov/div898/handbook/pmc/section2/pmc231.htm
    http://www.samplingplans.com/aqlprimer.htm
    http://www.siliconfareast.com/ltpd_aql2.htm
    Re: NASA and allowing 1% defective – never happened – see above.
    At least for semiconductor components consumer product manufacturers required an AQL of .065% for eg transistors, way back in the 1960s. MIL specs were tighter. By 1995 Bosch required AQL less than 3 ppm for complex integrated circuits for use in ABS assemblies. I have no NASA experience, but would expect inferred AQLs in the few ppm range. However for the shuttle program no one made more than a few thousand of anything for them, so for sure 100 % testing and inspection was employed with large guard bands and zero defect requirements. The issue then becomes MTBF. Its conceivable but unlikely, that they never expected the shuttles to fly this long so clould have been light on MTBF requirements.

  36. Douglas DC, that’s either the Grumman AA1A Yankee or the Piper Tomahawk.
    My wife and I had the misfortune to do our power flight training in a Tomahawk. Both of us survived. I had the advantage of over 2000 hours in gliders.

  37. But, this thing *should* have been so overengineered, 65% should still get it done comfortably.
    Safety margins for flight capable vehicles are in the 10% range. Safety is maintained by continual inspection and appropriate repairs. So if the max rated G load is 3 Gs the aircraft will be designed to about 3.3 Gs. Now some of the safety margin is in the G load. But still.
    For aircraft everything structural is tested to destruction. To make sure it meets the design criteria. Everything is weighed. And documented.
    We have a rule in aerospace that “no aircraft flies until the weight of the paperwork equals the weight of the aircraft.”
    A lot of someones didn’t do their job on this one.

  38. Gerry says:
    January 15, 2011 at 2:19 pm
    Nothing like strapping yourself onto four million pounds of explosives built by the LOW BIDDER…

    Kinda like using the Met Office’s supercomputers to predict climate change, eh…

  39. Those stringers were formed and then painted with a green zinc chromate at an outside vendor to prevent corrosion. No one would know the aluminum was aging or defective visually. I don’t recall that area of the tank having any sort of x-ray inspection either. The intertank, where these stringers are used, is a completely riveted structure. Lockheed had a lot of trouble with the aluminum-lithium alloy when I was there. Its very hard to weld without defect, and the super lightweight tank is what it is because of its extensive use of aluminum-lithium alloy.

  40. Anthony:
    Thankyou for your link to
    http://www.snopes.com/history/american/gauge.asp
    However, I fail to see where it “disproves” the story. In fact, it seems to confirm it, although it does say that if the Confederacy had won the Americam civil war then a different railroad gauge would have been adopted in the US.
    Perhaps you would be so kind as to explain the how the story “has been shown to be false”. I have no real interest in the matter, but I would like to know as a matter of curiosity.
    With thanks in anticipation.
    Richard

  41. George Steiner says:
    January 15, 2011 at 3:51 pm
    “There is no engineering patch for inferior material.”
    I agree 100%, but they will go ahead anyway.
    I had to think about the “patch work” performed on Concord that eventually caused the Paris crash when a chunk of tire separated and penetrated the left wing section puncturing the fuel tank.
    Fixation to a specific problem resulted in the application of heavier reinforced tires.
    No engineer had thought about the effect of heavier chunks of rubber hitting the wing surface if such a tire blows.
    The risks of patchwork performed on aircraft (or spacecraft) in this case often have consequences that effect the entire design.
    In the case of the main tank there could be an increase in weight resulting in a reduction of the payload. Patchwork performed on the structural parts of the tank could effect the way stress is transfered to the outer skin.
    The worst scenario is a crack resulting in fuel loss during take off.
    The entire project is under pressure and that is another factor that slipped into Murphy’s Law in more than one occasion.
    This is still an old school space project where relative high risk factors are standard
    and accepted.
    In the case of Concord however the 9/11 attack and the following slump in air travel grounded the supersonic flyers for good.
    New aerodynamic developments however have solved the problem of the sonic boom and when the current surge of green madness no longer obstructs progress a new generation of supersonic airliners will emerge.
    http://en.wikipedia.org/wiki/Air_France_Flight_4590

  42. ***
    tty says:
    January 15, 2011 at 7:53 am
    I think you are rather unfair to the Titanic. It was actually an outstandingly safe ship.
    Even then it floated for four hours after the collision, allowing ample time for evacuation. However this, in line what had happened before, was handled with outstanding incompetence, causing many lifeboats to be launched not even half-full. Moreover most of these boats deliberately refrained from picking up survivors after the ship had sunk.
    The Titanic disaster was caused by a combination of regulatory failure (specifying much smaller lifeboat capacity than needed) and an almost incredible lack of professionalism by the captain and crew.
    ****
    I think you are insulting the professionalism of the ships officers and crew who were doing their job while knowing they were going to die.
    The lifeboats were launched on a timetable, by a small crew, one after the other. The problem was that few passengers were willing to get in the initial boats, they knew the sea was going to be cold and rough in a small boat while the Titanic seemed safe. They even launched a set of (unauthorized by UK Board of Trade) canvas fold up lifeboats that people were supposed to get in while in the water (dumb design for winter North Atlantic).
    There is a very good reason the lifeboats moved away from the ship (falsely explained to passengers that they thought there was another ship over there). A lifeboat is easily swamped by people trying to get aboard from freezing waters. There had been a spectacular foulup on another sinking with lifeboats sunk by passengers trying to get aboard). So the safe proceedure was to move away and let the people who jumped into the water freeze to death. When most were dead it would be safe to return to rescue a few hearty survivors. The time it took to die was very short, and most were not functional within just a couple minutes of water exposure.
    Lifeboat rules can be very cruel and inhumane, but they are designed for the preservation of the maximum number of lives in a very dangerous situation. There is a reason the lifeboat captain is armed and has authority to use his weapon.

  43. Mike Borgelt says:
    January 15, 2011 at 1:12 pm
    “Douglas DC, that’s either the Grumman AA1A Yankee or the Piper Tomahawk.
    My wife and I had the misfortune to do our power flight training in a Tomahawk. Both of us survived. I had the advantage of over 2000 hours in gliders.”
    Sorry to hear that-I had instruction time in both and they both were marginal as basic trainers. This is an example of poorly thought out design and perpetrated on the
    general public. I am a fan of the old school-Piper Cub, Taylorcraft, Aeronca/Champion
    Cessna 150/152,etc. Do like the Diamonds, however and some of the light sport aircraft that are now coming out. Designs whether spacecraft or light planes,
    should not worry the occupants whether or not that the craft is actively trying to
    kill them.
    “Dave Bowman: Open the pod bay doors, HAL.
    HAL: I’m sorry, Dave. I’m afraid I can’t do that.”

  44. Fred H. Haynie says: “…Base(d) on what they have found and reported, I think this is a classic case of intergranular stress-corrosion cracking resulting from improper heat treatment….”
    I’m wondering what the tank was exposed to, internally and externally. I’m reminded of the hydrostatic testing of a 6Al-4V titanium alloy space vehicle tank. After the test had been conducted with methanol as the test fluid, it was discovered that methanol was corrosive to that alloy. Who’da thunk?
    http://oai.dtic.mil/oai/oai?verb=getRecord&metadataPrefix=html&identifier=AD0649187

  45. “real work on climate science “
    How absurd. Dump huge amounts more money to closer study random movements of huge winds in the atmosphere? Looking for a gubmint grant, are you? The Earth rotates on its axis, air has mass, inerta and viscosity. Mountains/mountain chains bull their way through the atmosphere, and the result is huge vortices of air, creating high velocity east to west or west to east moving winds, randomly varying northward and southward.
    With knowledge in the fundamentals of both chemistry AND physics, statics/dynamics, thermodynamics and electrical and hydro/aero dynamics, not all that difficult to comprehend the processes, but completely impossible to model, high speed computors or nay.
    Best spend money to figure out how to survive the awesome powers of natural forces . . . or find a gullible multiple billionaire to sponser you. The US taxpayer has plumb run out of money.

  46. So how many men and women have died in the far simpler task of learning how to get relatively safe airplanes to ply the skies? The airplane didn’t go quickly from the Wright Brothers contraption to SST, and many experimental airplanes crashed and burned in the process.
    Also, this is done by government contract, with schedules and spending budgets. And a fickle public and politicians to please.
    It takes men and women of great courage to pilot and crew these no more than huge possible bombs, and the crews know their lives are at great risk.
    I consider it to be a near miracle that the space program has worked out so well, and salute the engineers and technicians who made it happen, and the courageous men and women who knowingly risked their lives in a quest for knowledge.
    But then, I am a mere civil engineer, who has managed large highway projects, both as a government-paid engineer, and a contract-to-government engineering project manager. Long since retired now.

  47. This is what happens with “lowest bidder” contracting where no weight is given to quality of work.

  48. George Steiner,
    There is no engineering patch for inferior material.
    Quite so!
    There is no engineering patch for inferior thinking either. A hodge is a hodge. I have never made an engineerig change without unintended side effects. Do the job right and get the right material.

  49. jorgekafkazar said on Worrisome engineering patches
    January 15, 2011 at 6:43 pm
    My guess is that it is the atmosphere at the launch pad. They are near the ocean and salt spray travels inland further than we realize. It doesn’t take much salt to increase the conductivity of condensed moisture and stimulate corrosion (especially if the rivets are cathodic to the stringers.

  50. “another ship with low grade metal”.
    An interesting and perhaps better example, is TankerSchenctedy.jpg. This should take you to the Wikipedia page which shows the tanker Schenectedy which was an early all-welded ship.
    About 10 years before the war Griffith, who was an aircraft engineer, showed that if a crack developed in a brittle material it would continue across the material until it reached the edge. It was quickly found that a crack could be stopped by a properly drilled hole in it’s line of travel. The hole had the same effect as an edge.
    Schenectedy must have developed such a crack, but whether or not the crew noticed it, or whether or not they knew how to stop it, I don’t know, but anyway, it is a matter of history that the ship broke, in a spectacular manner.
    But Titanic was not an all welded ship. In fact I imagine that nowhere on that ship was there anything welded. The Titanic was an all riveted construction, made of steel plates. The plates could have been about 2 feet by 6 feet, perhaps and assembled with the long edge longitudinally. The entire ship, hull, bottom, inner bottom, longitudinal bulkheads, transverse bulkheads, decks, were all assembled from those huge all-rivetted steel plates. An enormous tonnage of rivets. Such a design meant that any crack would be stopped at the affected plate’s edge. Obviously then an excessive crack would be transferred from the originating plate, now failed, to the two adjacent plates, but now there would be two plates resisting the stress. And so on.
    The materials would certainly allow for low temperatures, and the design would allow for the routine bending expected from alternatively being supported amidships on a wave, then supported at both ends only, which is the routine life of a ship. In fact the worst bending conditions imagine for a large ship would have been experienced during launching.
    So perhaps the failure was not as simple and clear cut as the professors claim.
    Nothing to do with your interesting piece, but perhaps worth noting.
    Note also, Titanic was illustrated by a lurid sketch, whereas Schenectedy had a real, live photo.

  51. Perhaps NASA wants the manned U.S. space program to go out with a bang?
    It wouldn’t be the first time “end users” were sacrificed in order to secure funding for a new iteration of a defective product. Why, it’s the entire basis for the business plan of Microsoft, Inc.

  52. DJ Meredith says: January 15, 2011 at 9:21 am
    “I’m at a loss here to understand how this could happen without gross corruption in the process.
    I’ve manufactured flight grade parts, both for aircraft and space vehicles. The pedigree for the metals (“Certs”, as they’re commonly called), the manufacturing processes themselves, almost to the chain of custody, it all is tracked and documented. Inspectors will come on site and check your documentation, and they’ll check your inspection tools for calibration. You guys think science is rigorous??”
    The explanation is simple. That alloy is a very low production item, and some of the fabrication techniques are unique to the Shuttle. It does not have the decades of practice by hundreds of users to uncover design faults as other alloys do. If there is a design flaw, all those certifications do is ensure that every one will have the same flaws as the first one. If you want flight safety, there is no substitute for lots of flight test, of which the Shuttle had very little. Unfortunately, that’s true of all manned space flight vehicles so far. It’s hard to get 1000 test flights on a vehicle that flies 4 times a year (Shuttle orbiters) or thrown away after every flight (all other NASA vehicles to date).

  53. The aluminum-lithium alloy usage was the result of a project to take weight out of the main tank to allow for a larger payload. Talking to some of the locals here who work at Kennedy Space Center it seems that it might have been overdone. The foam shedding was a result of getting away from a freon foam ‘expander’ to go with something a bit more ecologically sound. There had been very little or no shedding before the change, but we are all familiar with the after.
    NASA seems to be at a point that no matter how much money is dumped into their hopper they always seem to come up short in their space exploration activities. I personally have given up hope that they can continue to lead the way. Rather, they seem to be on a path to follow, at great expense, companies like Space-X and Bigelow.
    As far as the quality of the metal used in the problem tank, earlier comments here have not been the first I’ve heard about union activities and worker lack of focus. It is the Government, isn’t it? When I was in the Air Force the motto of the day was “Good enough for Government work.”

  54. I need to clarify my NASA comments somewhat. The people who actually work on the shuttle are highly skilled, carefully selected people. You see license plate frames with the comment “Doing daily what others only dream of.” They love their work.

  55. Anthony, thanks for debunking that railroad gauge story, which is a version of an old story that people adapt to their particular hobby horse.
    Bridges and tunnels are not designed to the track gauge, but to the width and height of the body of the cars which always overhang the track – and probably some have been built wider and higher to accommodate even larger loads such as airplane fuselages and double-stack container railcars or automobile-carrying railcars.
    As have highway overpasses (one was raised a few years ago on I-405 in the Totem Lake area east of Seattle WA, another on I-5 in Burlington WA may have been after a trucker tried to demolish it a few years ago by ignoring height information – and on the TCH east of Vancouver BC you can see scars on the lower overpasses).
    Tallbloke
    Aerospace structure is weight-critical.
    Boeing had difficulty with stringer ends on the 787 wing structure in the wing-body, though the pieces were of composite material and IIRC the defect was an analysis problem revealed by static tests.
    The disturbing problem here is a large shortfall in materials quality, as marcoinpanama suggests not noticed by anyone. Boeing has had comparable anomalies in 787 sub- assemblies that puzzle me as to why none of the supposedly skilled aviation people noticed them before final assembly.
    Mark G, I think you’ll find on researching that the trend of o-ring erosion on the space shuttle was clear enough to be very concerned about risk. How clear may depend on how the data was plotted, but it should have been carefully examined given the downside risk. The big problem was political in Morton-Thiokol management – they took a vote among executives, only one of whom had technical experience. That’s “consenus”, the scam pulled by climate alarmists.
    NASA too had lost visibility of risk because problems for which a fix was scheduled were no longer highlighted, despite the FMEA’s effect of dual o-ring failure: “loss of mission, loss of crew”, which tragically came to pass.
    M Simon, fact is that there is a 50% overload factor in normal aircraft design – the difference between “limit” and “ultimate” cases. Everything else is either component tests or calculated. 10% sounds like a rule of thumb for thorough analysis or test cases as a margin to cover remaining imprecision in that analysis (which should have already used minimum spec part size and material strength values). We could talk about how loads are calculated of course, which is perhaps what you refer to as “G” factors. And fact is that only critical structure, like the wing, is actually tested to the required load, often to failure.

Comments are closed.