Shuttle Tank Foam FIX

[danwoodard]

The largest piece of foam that searated was inthe area of a hand repair, so a quality control or procedural problem rather than design limitation in the foam seems likely. However much one would like to reduce the problem to design specs and written procedures, all rockets have assembly procedures that require a lot of skill and craftsmanship that can't easily be learned, defined, or tested. <br /><br />Newer technolgy might be an answer. Just before it was completely cancelled the X-33 team had tested a new composite hydrogen tank that worked quite well.

 

[awuj2]

So how much that can be? <br />Shuttle can take 28803 kg to orbit.<br />the biggest ISS module to be launched (KIBO)<br />is circa 22000 kg<br />Antything we add to the Shuttle ET must be less <br />then 6800 kg.<br />Is that correct?

 

[grooble]

How about a chicken wire mesh?

 

[jaredgalen]

It was mentioned eariler that gas that gets trapped in the foam during application expands during ascent and <br />causes damage to the foam.<br /><br />What scenario would reduce this amount of trapped gas?<br />If it was reduced would it contribute significantly to the integrity of the ET foam?<br /><br />Perhaps applying the foam in a low pressure environment might help.<br />Not sure how plausible this would be in terms of facilities at the tank manufacturer.<br /><br />Edit: Although I guess that could affect the curingdrying of the foam.

 

[drwayne]

1207,<br /><br />What I was talking about is definitely covered under the disclaimer YMMV. <img src="/images/icons/smile.gif" /><br /><br />Wayne <div class="Discussion_UserSignature"> <p>"1) Give no quarter; 2) Take no prisoners; 3) Sink everything."  Admiral Jackie Fisher</p> </div>

 

[CalliArcale]

<blockquote><font class="small">In reply to:</font><hr /><p>"How about a chicken wire mesh? "<br /><br />Check out the link I posted. NASA's assesment is it would take years to certify and thus it is not an option.<p><hr /></p></p></blockquote><br /><br />I think a lot of people (generally not here, but often in the media) forget about that part of the problem. It sounds simple to just slap a bunch of netting over the ET, but if you just slap a fix onto it and go, you're abandoning every quality control and safety requirement in the book. There's a curious irony in that, because of course that's exactly the sort of process-related problem that led to the foam issues being so severely downplayed and insufficiently studied through the years.<br /><br />In a system this complex with so many safety-critical components, not mention a system as expensive as this, you have to make changes in a controlled fashion. I work in configuration management, which means I'm a process weenie. <img src="/images/icons/tongue.gif" /> But it also means I get a good look at how this sort of a process needs to operate. The principles apply everywhere, not just in the Shuttle program. It's good engineering discipline to make your changes in a controlled fashion. It mitigates risk, because it forces the changes to be made deliberately and with time to really explore the ramifications of them, and because it produces ample documentation to be able to trace problems resulting from the change and possibly delete the change from future builds. If this change process is properly implemented, it also forces you to justify your proposed changes and to test them and the overall system after they've been made.<br /><br />In general, the bigger the change, the more time, money, and risk will be involved in going through the proper change process. You'll need to do the research to justify the change and convince your change control board that it'll work, is desirable, is affordable, and won't screw something else <div class="Discussion_UserSignature"> <p> </p><p><font color="#666699"><em>"People assume that time is a strict progression of cause to effect, but actually from a non-linear, non-subjective viewpoint it's more like a big ball of wibbly wobbly . . . timey wimey . . . stuff."</em>  -- The Tenth Doctor, "Blink"</font></p> </div>

 

[anotheridiot]

shuttle guy wrote:<br />"The next shuttle is planning on taking 2 tons. " <br /><br />Where did you get that figure, that is not correct. They are taking up another MPLM full of supplies. <br /><br /><br /><br /><br /><br />This was right on the nasa website<br /><br />*Discovery will carry the Italian-built Multi-Purpose Logistics Module (MPLM) Leonardo, with more than two tons of equipment and supplies aboard. *<br /><br />

 

[drwayne]

Having worked SIV&V for 7 years, I have seen exactly what you mean Calli.<br /><br />I have seen test flights fly with known failure modes in the software because the SCCB determined that:<br />(1) The probability of occurance was very low<br />(2) The time needed to thoroughly test the fix was excessive/not available.<br /><br />To a degree, I think NASA has boxed themselves in here. By making all the comments about not flying until they have fixed the problem, they have rigged the risk/benefit analysis before the fix has even been identified. <br /><br />Wayne <div class="Discussion_UserSignature"> <p>"1) Give no quarter; 2) Take no prisoners; 3) Sink everything."  Admiral Jackie Fisher</p> </div>

 

[plasticman]

The foam as applied now is actually low pressure - ie atmospheric. The pressure drop is low from the initial to the last fill, since the dimension in this case is just the thickness.<br /><br />Filling closed vessels with long run lengths (ie imagine a temporary outer shell over the ET that was filled) has plusses and minuses-<br /><br />One BIG plus could be that one might design an "outer mold" that would reduce the hand placed pieces and cuts. (there might be reasons why all the parts that need covered can't be assembled then foamed..)<br /><br />Some MINUS factors of filling the shuttle with say a hard outer cover, " from the tank - an open space that is filled:<br />1) - High pressure drops along the flow lengths These usually need multiple "gates"<br />2) - Such long flowing leads to knit areas that can entrap gas/large voids<br /><br />... but there might be an interesting idea here -<br /><br />Place a thin skin of something tbd (Mylar, .010 AL foil..) on the inside of this "mold shell". <br /><br />Then after the inner filling, this smooth, solid, tough "skin" would be well bonded to the PU foam.<br /><br />Remove outer shell mold.<br /><br />ET tank has a smooth "thin" shell with the PU underneath.<br /><br />There are more problems and challenges to think through - but I am liking a vacuum bottle approach (ET shell, foam/vacuum/thermal break, thin outer skin).<br />

 

[dwilson]

Calli Posted..<br />"The principles apply everywhere, not just in the Shuttle program. It's good engineering discipline to make your changes in a controlled fashion. It mitigates risk, because it forces the changes to be made deliberately and with time to really explore the ramifications of them, and because it produces ample documentation to be able to trace problems resulting from the change and possibly delete the change from future builds. If this change process is properly implemented, it also forces you to justify your proposed changes and to test them and the overall system after they've been made."<br />_________________________________________<br /><br />Your analysis is very on point.. in the spirit of turning a problem over and over and inside-out to look at any solution and then deem them plausible or not you have struck on an overall overview that causes one to pause and step far back from the problem and look at the very large picture.. with a situation such as this, perhaps just squeezin all of the possible error in the application process of the current FOAMING process.. maybe the quality control there would find that the process should start and finish in large full steps.. and any deviation such as patches or encounters with unusual aspects of applying the foam should then result in a clean strip and reapply in one session or section sessions that can safely tie the joining of the sections, ie.. never spray the foam on completely dried foam since this would cause a foam on a dry foam surface that would not allow the 2 applications to to meld to one as it would sprayed at once in its entirety.. allowing the molecules to mix instead of "sit on top" of each other. <br /><br />One shot coverage.. and keep it flowing before it cools.. sounds like a mold that would go around it and then fill the space between the tank and mold.. that would maintain a more consitent material integrity throughout the process, creating a more proper single foam shell. The mold

 

[dwilson]

Hey Plasticman, <br />I swear I wrote my long diatribe including the mold idea without seeing your post first.. but perhaps that means that we are finding the most correct way as we come to similar conclusions from different trains of thought.. I credit Calli for providing the "funnel" for which I fell through to come to those conclusions for a possible "fix" utilizing a tighter control and application of the current foaming technique. It was strange to write so long on what I wrote and post it only to see you had arrived at the Mold cast idea a post before me.. "Great Minds think alike" if I may be so bold. This has been such a stimulating discussion, quite beyond my expectations when I started this thread.

 

[zenofjazz]

That was my thought, when this idea struck me. <br /><br />What other problems are there with the idea? As I understand it (used to work at a sub-contractor, worked on shuttle main engine controller, steerable nose wheel, and some payload related things on the orbiter bus) the insulation is to keep the cryo fuels cold, while waiting for launch. Launch to tank seperation is roughly 10 minutes.<br /><br />There are pressure vent valves on both the O2 and H2 tanks, that can be used to relieve ullage pressure, if it exceeds predetermined limits.<br /><br />So, what else am I missing, here?<br /><br />-Jazz

 

[zenofjazz]

Two ETs? SIde by Side? Front to back? Top and Bottom?<br /><br />For each of those choices, the questions come down to "how do you attach the SRBs? How do you release the empty?<br />how do you deal with the fact that the 2 tanks (because they are more sturdily built, and must more or less fit the existing form factor) are going to carry less fuel?

 

[bobw]

One potential problem I can think of -- bubbles in liquid can cause cavitation in the turbopumps. I would think that the inside of an uninsulated tank would look a lot like the bottom of a pot on the stove as you bring water to a boil; a layer of bubbles. <div class="Discussion_UserSignature"> </div>

 

[zenofjazz]

but both under gravity, and accelleration, the bubbles are going to tend to rise to the surface of the liquid fuel and oxidizer. How quickly would the H2 or O2 rise in temperature, over that 10 minute window (from engine start to Tank Separation)?<br />

 

[x24val]

Just fly baby! If you're only going to the station and you have a back-up ready on each flight...just fly!<br /><br />I'd put some pretty long odds on foam doing serious damage with only 20 or so missions left. If it does...ditch the bird.

 

[plasticman]

Sounds good to me DWilson <img src="/images/icons/smile.gif" /><br /><br />I can not imagineNASA/subs haven't thought of this - open mold casting is fairly well known - another term for large spaces filled slowly...

 

[dwilson]

I think we have done all we can do .. that is the best decision.. to improve the existing process.. now all we need is the right people to read it and 3m to insure a foam that will stay solvent long enuff to fill the mold.. since it is alot of foam.. for 1 pour. <br /><br />Look for my future thread on Retiring the Shuttle Fleet in space.. they would last longer without gravity tugging on them.. the uses are endless.. can you land a shuttle on the moon?<br /><br />..and why hasn't Hilton got money and living/space on ISS yet?<br /><br />

 

[danwoodard]

The orbiters have been hit over 10,000 times by foam over more the 20 years without significant damage. The spray-on foam is quite tough because of the outer skin or "rind" that developes when it is applied, which is about as tough as fiberglass anyway. <br /><br />The spray-on foam has never caused critical damage. The Columbia was hit by a glued-on block (the bipod ramp) that separated because the drying characteristics of the new bonding agent weren't adequately known. The large piece that separated due to lack of internal cohesion on STS-114 was a hand-applied repair in the also hand-applied PAL ramp. For the limited number of flights planned, the best solution is to correct these specific problems. There is little to be gained in changing the primary foam with chicken wire or anything else. Even at 1 MPH the impact of hundeds of pounds of intentionally separable foam on the orbiter would be considerably more likely to damage it, and heating the entire outside of the tank without insulation would cause a boiloff rate far in excess of any possible replenishing.<br /><br />The placement of a cyogenic tank ahead of a delicate payload was a design problem that is, like so many, obvious in hindsight and won't be repeated. For expendable rockets the problem is insignificant because the payload is ahead of the fuel tanks. The Delta IV originally had spray-on foam, but went to glue-on blocks for environmental reasons. I believe one block did fall off on the first launch, but there was nothing below it that was vulnerable to damage. Similarly the Saturn V accumulated about 2.5 tons of ice that fell off in huge blocks at launch, but did not damage anything vital.<br /><br />For the longer term I hope we someday get a chance to design a new fully reusable launch vehicle with all the knowledge that has been so painfully gained from the Shuttle. The CEV is only a partial solution; it will be safer but does not appear less expensive.

 

[zenofjazz]

reasons why this isn't going to be a problem...<br /><br />1) The shuttle drops the ET before it reaches orbit... (though it carries the take 90% to orbit!)<br /><br />2) once the tank is out of hte atmosphere, the majority of the pressures that work to tear the foam off are gone. So if the tank was being carried to orbit, the stresses that cause the foam to come off, wouldn't be affecting it.<br /><br />-Jazz<br />

 

[plasticman]

"The orbiters have been hit over 10,000 times by foam over more the 20 years without significant damage. The spray-on foam is quite tough because of the outer skin or "rind" that developes when it is applied, which is about as tough as fiberglass anyway. <br /><br />The spray-on foam has never caused critical damage. The Columbia was hit by a glued-on block (the bipod ramp) that separated because the drying characteristics of the new bonding agent weren't adequately known. The large piece that separated due to lack of internal cohesion on STS-114 was a hand-applied repair in the also hand-applied PAL ramp. For the limited number of flights planned, the best solution is to correct these specific problems. There is little to be gained in changing the primary foam "<br /><br /><br /><br /><br /><br />I really disagree. First, NASA used Boeing's CRATER impact program, which was is fundamentally flawed. It is a linear elastic impact modeler. In fact both the material and type of impact is with non-linear visco-elastic material (ie the foam) - and on in non-linear materals (the foam).<br /><br />This Boeing CRATER program is the one that that mislead the NASA team for months and led the administrator to use the incredibly technically wrong analogy of "styrofoam cooler" - Had the NASA and Boeing engineers involved, talked to experts in high-speed, ie high strain rate, low temperature polymeric impact - they most likely all (like myself) would have been agahst and tortured over the allowance of the "10,000 impacts that did not damage" -<br /><br />This was pure luck, but manipulated into an acceptable engineering risk by engineers with no relavent experience.<br /><br />Experienced people would have done the 500 mph foam low temperature impact tests in the 70's when that research was also known to the relavent community.<br /><br />The problem is - "foam" sounds so low tech and innocent - why worry (poor assumptions).<br /><br />The foam as a final surface must not ever fly again - wher

 

[awuj2]

Why only 2000 kg?<br />Can you elaborate it a bit?

 

[askold]

Instead of designing a foam that stays on better - how about a foam that shakes off at lift off. Something like packing peanuts glued together. Something that breaks up into small light pieces.<br /><br />You don't need the insulation after the shuttle launches - if it shakes itself apart in the first 1,000 feet of lift off - who needs it.<br /><br />I just wouldn't want to be the one who has to sweep up all the plastic peanuts ....

 

[CalliArcale]

Somebody suggested that earlier in this thread, and it could be a workable option. However, I think you would have to get the foam off a little quicker than that; I guess you'd have to do the impact studies and airflow studies to find out what the maximum airspeed is at which foam can separate without harming the vehicle.<br /><br />After that, I think the main engineering problem would be finding a way to break the foam off completely, consistently, and reliably without imperiling the vehicle. I don't think that's trivial, but I don't have enough background in materials science to even make a SWAG at what kind of timeframe we're looking at to explore that. My gut instinct tells me it'll be longer than the Shuttle program is running, so the fix would only be relevant for future vehicles, which in all likelihood won't be side-mounted and thus won't care about foam shedding anyway. <div class="Discussion_UserSignature"> <p> </p><p><font color="#666699"><em>"People assume that time is a strict progression of cause to effect, but actually from a non-linear, non-subjective viewpoint it's more like a big ball of wibbly wobbly . . . timey wimey . . . stuff."</em>  -- The Tenth Doctor, "Blink"</font></p> </div>

 

[drwayne]

Note that anything that comes off early in the flight would, it would seem, also be subject to coming off while the shuttle was being moved, on the pad etc just due to wind.<br /><br />By definition also it would seem that such insulation would have to be applied to the tank at the Cape.<br /><br />Nothing I have just said is a physical impossibility to deal with, however, it will take new procedures that will need toe be developed and refined and V and V'ed<br /><br />It is interesting how many ideas that fit under the category "Why don't they just ..." that take much more time to get going than a casual examination brings out.<br /><br />Wayne <div class="Discussion_UserSignature"> <p>"1) Give no quarter; 2) Take no prisoners; 3) Sink everything."  Admiral Jackie Fisher</p> </div>