NASA can't open its asteroid capsule | Hacker News Reader

NASA can't open its asteroid capsule

tortoisemedia.com

122 pointsadolph72 comments

Original source from NASA:

https://blogs.nasa.gov/osiris-rex/2023/10/20/nasas-osiris-re...

Much better, and without the hyperbolic headline.

but no photograph. the photo here is most valuable. I find the electrical receptacles with the yellow extension cord plugged in and the cheap shelving unit with almost nothing on it that is pretty much the identical shelves I use in my basement to store groceries to be the most fascinating aspect of it. seems like there'd be a more...I dont know, "special" kind of room to unpack rock samples that cost $1.16 billion to acquire

That’s deliberate - NASA goes out of its way to use ordinary consumer goods when possible.

You can imagine how much money you could spend on a special room for billion dollar rocks! Surely 150M is nothing!

The actual problem statement: “we need a safe room for some rocks” only requires a couple plastic bins and some tape.

I’m always so pleased by how hacky NASA stuff looks!

Brings back memories for me. I used to work on particle physics experiments.

The "science groupies" would be amazed at what happens behind the scenes. There were a couple of deliberately broken vacuum cleaner parts that happened to play a surprisingly important role in certain neutrino parameter measurements....

I would definitely like to hear more. It's very HN worthy story.

I considered linking to that one instead but it hides the lead. My son and I were checking in on the latest from the mission because everything had gone quiet since the initial JSC announcement. Then one of the space YouTubers mentioned the screw problem as an aside and I found the linked article from a source of unknown reputation that was backed up by other sources.

So if it is judged click-baity blogspam, I plead guilty and I'd do it all over again to read the comments about stainless steel galling/spalling.

Saying can’t for a problem with a short term solution is the worst kind of clickbait. “This might delay the process by a few weeks.” Is really just an abundance of caution situation.

Worse, "we're choosing to delay to minimize chances of contamination" vs "cannot".

Ooh - let me guess... It's a stainless screw in a stainless thread and has suffered spalling?

Spalling is a really wierd effect that only happens with screws and threads made of stainless steel. You can tighten a screw with your fingers (no tools, just finger tightened), and then try to undo it and it'll suddenly be stuck. You try with tools, and it's still stuck. You use a lot of force, and it snaps off!

It happens because stainless steel can act a little like animal fur - rub in one direction and it slides easily, while when you change direction the fur meshes and refuses to move at all.

Doesn't happen reliably either - you can have 100 identical screws in 100 identical holes, and perhaps 5% will get stuck.

Normally the fix is to put vaseline or some other lubricant on the screws (before it gets stuck - after it's stuck you're sol), but I figure the NASA team doesn't want that lubricant contaminating their sample.

I like this forum because this kind of explanations. Thank you

Is there a way to anticipate the “direction” of the “grain” of stainless steel?

I'd be a bit surprised if they have two moving parts, made of the same metal, in contact with each other: https://en.wikipedia.org/wiki/Cold_welding

I'd really love that, perhaps, perhaps, we get an update to this comment -- “Hey guys, NASA just contacted me.” :-)

From what I can see the TAGSAM head did use stainless steel screws, but the majority of the structure is aluminium [1]. So its possible there is a galvanic corrosion issue that caused the screws to bind, but I would be a little surprised at that based on how the TAGSAM has been handled throughout its life. Still does feel like the most likely issue though.

Regarding lubrication on screws in space, it is done quite regularly with a variety of lubricants (as well as thread lockers) [2]. I can imagine that none were used in a system that is meant to be kept as free of contaminants as the TAGSAM is though. Hard to know with any amount of confidence.

From experience with spacecraft integration, when the article mentions that they didn't have a suitable tool for removal of two screws, my head immediately jumps to an access issue. Its amazing the number of times I have heard 'it fits in the CAD' from the designers, when you have no access at all for getting a screwdriver to the screw, let alone a torque wrench. So it could be as simple as that.

[1] - https://link.springer.com/article/10.1007/s11214-018-0521-6 [2] - https://ecss.nl/hbstms/ecss-e-hb-32-23a-threaded-fasteners-h...

There's a lot of push in the defense world to get to so-called "Digital Twin", where cabling, conduit, maintainability and other late-stage product work is done purely in CAD and in simulation. I'm very open to the idea . . however . . when I get a pile of geometry with no fastener material data, no soft bodies of any kind (tape, string, wire, cloth, rope, etc), broke fillets . . and then someone tells me "there you go, you can figure out the maintenance breakdown now. DIGITAL TWIN!" . . I'm not filled with confidence.

Someone, somewhere, needs to spec out the minimum required for each stage of this kind of work . . they might find that doing things like wire channels in new product isn't fantastically easier than doing it in a prototype. You can do it in CAD< but you still have to get the design right. Tool clearance doesn't magically appear in the right place by screaming DIGITAL TWIN.

This is all aerospace. In another industry, with less demanding performance requirements, the easy solution to all problems is always "increase the tolerances". Then you can jam anything in there.

Another problem: logistics systems want to know the sparing level for what's being supported, not what's being designed. This, incredibly, isn't addressed so far as I can see; adopters are just sort of pretending that bleeding edge is the same as the as-built, supported configuration. When I bring it up, the general response is that LSA can be batch-create from the design, totally ignoring the fact that a lot of changes come the other way, from field modifications. The solution for this - for NGAD and Raider - is to just not do logistics as it's been practiced for the last forty years. Spoiler: this is the correct answer. 30 year sustainment programs are hilarious fictions that do nothing but generate high margin work for primes.

"DIGITAL TWIN!!" would be a lot less of a punchline if I understood the prototype on my desk well enough to predict its behavior. And I can look straight at that one. The idea that we can understand real hardware in unusual conditions well enough to make a simulation/Digital Twin the One True Reference is simply hilarious.

Though as you point out, it sure does generate a lot of high margin work and great ad copy.

I'll concede that when you write off maintainability down to the nuts-and-bolts-level, that does make things a whole lot easier. But I don't think that's digital twin; that's just tighter tolerance, and aaalllll the things that ride along with that.

But doing something like scheduled maintenance just from the Twin?! Ehhhhhh sheesh, I dunno. I'm not one of the smart guys, but you would need to combine the solid-model FM sort of simulation with the more fluxy CFD models, and then have them interact. I realize computers are super powerful these days, but that is a HELL of a lot to ask from a simulation running on commodity equipment with commodity software. I've seen single-assembly CFD choke when it had to extend to just a single fuel system's materials data - and that was nothing compared to doing that for an entire airframe. Or even a reasonable chunk of an airframe.

I'm at peace leaving this stuff to the guys who know what they're doing, but I don't see a lot of those people actually doing the decision making. It's finbros all the way down. I'm afraid what I'll be ending up with, at the end of the day, is the Reification Fallacy made flesh: a box of parts that's supposedly a flyable airframe. "Just write a preflight checklist for it!" they might say. "Copy and paste something or whatever."

Why knows, they're RIFing more or less everyone with technical knowledge this week, so it probably won't be my problem for long.

The Springer link is to the original TAGSAM paper--very interesting. I found it while doing a screw count of the below sampler head--there are 21 phillips-like head and 6 with allen head. The NASA press release [1] says "two of the 35 fasteners on the TAGSAM head could not be removed with the current tools approved for use." That doesn't match, which lead me to the Bierhaus et al. at Springer [2] which shows how they put the sampler head screws down into the dirt. The diagrams from Bierhaus and Walsh et al. [3] are both interesting.

0. https://www.nasa.gov/wp-content/uploads/2023/10/nelson-tagsa...

1. https://blogs.nasa.gov/osiris-rex/2023/10/20/nasas-osiris-re...

2. https://link.springer.com/article/10.1007/s11214-018-0521-6

3. https://link.springer.com/article/10.1007/s11214-022-00887-2

Generally in aluminum parts in aerospace, locking stainless steel threads inserts (“helicoils”) are added for pullout strength, repairability, reducing risk of damage when fastening multiple times and to meet the requirement (which almost always exists) for secondary back-off prevention (primary is usually controlled preload). The locking feature is 1-2 threads in the insert that are deformed to provide a decent amount of friction even if preload is lost. But for a special case like this it’s possible a unique design was used; I would be interested to read about it.

From the images, it’s not clear which fasteners are removed but if the issue is not access it could be a stuck fastener for whatever reason and maybe a rounded head. I wouldn’t be surprised if nothing is damaged but the procedure was stopped due to higher than expected torque, and they are taking it slow to decide how to proceed.

I hadbt consudered a stuck locking helicoil in fairness. And seeing the pictures of the fasteners it does seem unlikely to be an access issue, so you are probably right about the torque.

> procedure was stopped due to higher than expected torque

I should have done this. Just broke a brand new 10-32 tap!

I work in the industry and I’ve never heard of spalling with relation to stainless steel fasteners. The effect that happens with stainless steel fasteners is galling, when there’s enough friction in 2 mating fasteners of similar material to break through the surface oxide layers and the mating threads weld together.

Lubrication helps but in aerospace there’s typically requirements to use dissimilar materials for screws and female threads. The dissimilar material can be a coating but I think they’d probably avoid unnecessary materials in this case and use (for example) titanium screws and inconel threaded inserts. Any mechanical engineer at NASA and its vendors knows this and it’s covered in detail in the NASA fastener handbook.

> “…when there’s enough friction in 2 mating fasteners of similar material…and the mating threads weld together.”

I’m not in industry, but i wrench my motorcycle. I’m learning so much here.

I’m usually worried about “galvanic corrosion” when aluminum and stainless steel corrode and making removing the steel screws all but impossible.

(At the risk of leading you down a rabbit hole): consider that most torque specs are for dry torque, but in many instances, anti-corrosion measures and anti-seize compounds are quite fine to use, but the torque values have to be adjusted waaaay lower to avoid over-torquing.

Consider, by way of example, the oil pan bleed screw. It and its threads are well lubed, and while granted, it is not holding much of a clamping load, it is still subject to heat cycles, vibration, etc, and yet it stays happily in.

(Not an Mechanical Engineer, just a fellow occasional wrench-spinner who has battled rust) When a clamping load is involved, it starts to matter if there is an expectation of vibration, at which point something like Nord-Lock is called for.

I am sure my much more knowledgeable peers here will point out the glaring mistakes and omissions I have made, causing learning to have occurred. :)

This is something I’ve seen several times. often in aerospace the threads are supposed to be lubricated but if you get lubricant under the head you can get way more preload than you expect. Same for changing the washer material for a thermal or other reason and not realizing it has much lower friction.

For unique applications in aerospace, it’s common to do testing to determine a non-standard torque-preload curve.

In some industries, preload is measured more directly using hollow bolts (so you can measure the before and after length therefore determine preload), ultrasonic measuring devices (same reason) or preload sensing/indicating washers (there are many types).

I think, as mentioned, the term is "galling."

This is kind of funny as the "normal" usage of the word (by which I mean non-engineer usage) would mean causing annoyance or resentment. That seems pretty apt for the current situation!

Maybe the atoms in the screw and the atoms in the capsule decided that they now form a single piece of metal.

Cool, never knew this, now all those home improvement projects where I got a screw stuck and had to drill it out make more sense.

You're unlikely to be using stainless steel screws very often in home DIY

Fancy mirrors and fixtures. You'd be surprised.

I worked with ultra high vacuum equipment and am pretty familiar with this.

The term I've heard is "galling."

UHV is not compatible with the use of greases either. We would use silver plated screws to avoid stuck fasteners, and to avoid galling for small and delicate fasteners inside of the chamber where silver couldn't be used as it might be a contamination source, Trichloroethylene would work as a lubricant that evaporated without residue.

The NASA fastener handbook specifically calls for using silver plated nuts on steel bolts for aerospace. Very cool.

>Silver plating is cost prohibitive for most fastener applications. The big exception is in the aerospace industry, where silver-plated nuts are used on stairdess steel bolts. The silver serves both as a corrosion deterrent and a dry lubricant. Silver plating can be used to 1600 “F, and thus it is a good high- temperature lubricant.

https://ntrs.nasa.gov/api/citations/19900009424/downloads/19...

I wonder how they designed it all so that it would be machined in earth's atmosphere at ambient temperature, made to open in a cold near-0K vacuum, collect electrostatically charged fine dust material of unknown chemistry, close up the container despite deformation from pressure and temperature differential since machining, reenter atmosphere at high temperature... then come back to earth... and then the bolts are supposed to not have seized or deformed or obstructed or be under excessive force from the pressure.

I'm sure they somehow designed around all this? Or am I over complicating the nature of the problem?

I read it as though they might have miscalculated which tools they would need to open it, and now that it’s partly opened inside the “anti-atmosphere box” they can’t easily add additional tools.

They’re going to need enclose that all in a bigger anti-atmosphere box.

I want to be the person in the monkey suit who said "We're going to need a bigger boat."

Does this count? https://www.youtube.com/watch?v=f0lpiXAHuyA

FTA:

> This might delay the process by a few weeks.

SO, kinda clickbaity?

It’s still interesting to get some insight into small technical problems

Unfortunately, the article doesn't really provide any of that. Frankly, is not really even an article.

Very much clickbait - it is worded as if NASA were incapable of removing a bunch of screws, but the problem is not that they can't open a container they built themselves, the problem is that they want to make sure that they won't accidentally contaminate the collected samples by using the wrong instruments. Thus only pre-approved tools must be used, and they're not the right ones for the job.

Surely, that can be called a hickup, or lack of proper planning, but it really isn't that much of an issue. That's why it will only delay the operation by a short amount of time.

My immediate thought when I read the headline is I can open that box - but I'll contaminate the contents in the process. Opening a sealed box is easy, it is when you want to preserve the box or the contents it gets hard.

Run the lid under some hot water

I look forward to the analysis to see what the root cause was and whether the new drive system was part of the problem, or if the threads were simply seized beyond the capability of any drive system.

If they extracted almost all of the fasteners, that seems like it would be the latter.

Do the screws have the latest firmware update?

On a serious note, when you think of what achievment this mission is, it's silly how this is the issue they are dealing with. Good they were able to get some of the sample out though and this will also help develop better solutions so they don't get screwed again :)

They should just give up and toss it out.

At least it's on Earth, from the title I thought a mission in space that landed on an asteroid was borked for collecting samples

Isn't this called cold welding in space?

ACR Torq-Set Drive System . . . is an offset cruciform drive system that allows more torque to be applied.

Oh my, Im looking forward to getting some of those. Never thought I would be so excited about not stripping out screws. I bet their next approved tool kit is going to come with a drill bit

interesting ... and tiring of this site ... this post just lost a point. Why? no response, no reasoning against wanting screws that perform well. Just a loss of points. FU y combinator ... i think ill spend more time posting elsewhere

From their website: "Are you interested in Licensing this technology...?"

It's bugs me they can get a patent on a shape with features that you or I could design in a couple nights. I thought the whole reason the world didn't already adopt better screw heads than blasted Phillips (like Robertson, or square bits for those who aren't familiar) is that licensing arrangements weren't economical.

I personally think that torx should be used more often(the star shape).

Agreed. Torx might not be ideal, but it is widespread and (relatively) cheap. And miles better than anything else that is widespread and cheap.

The local hardware stores' own brands offer a wide range of cheap, torx-compatible screws. I think this started not too long ago (but can't say for sure, probably 2019 when the latest torx patent expired?), and they still offer screws with the normal cross shape. Both are of the same, decent quality.

I only buy the torx/star variants, since they're just superior.

Regular people cannot handle using torx.

Philips is nice because (untrained) people naturally will over-tighten the fuck out of every screw they encounter. Philips cams out and prevents them from going too far overboard. In fact its literal design goal was to make a screw that is hard to over-tighten.

I know if we used torx on some of our products, we'd spent 10000% more time on removing snapped screw bodies.

It's me, I'm untrained people.

Is there a decent convention for how much a screw should be tightened, or is it fully case-by-case?

There are torque specs and torque wrenches to solve this problem in manufacturing.

For day to day, 95% of the time you can tighten a screw gently until it stops, its head fully against the surface, then just "snug it up" a bit to seat it.

"Snug it up" is of course vague, so refer back to line 1.

Square Robertson screws are my go to. My father had a wood shop and used them for production as they were nearly impossible to strip and the screw holds tight into the slightly tapered driver. So I grew up using them for various projects as we had boxes of them. You can easily one hand a 3-4 inch #10 deck screw. Those crappy Torx things they sell don't compare.

If we are playing pin the plan on the happenstance I guess the profit from the patent could pay for the sales work necessary to turn a random shape into a standard anyone uses.

You or I could not "design it in a couple nights". For a one-off part with no particular requirements, sure, although it's a lot of work just to manufacture and test once for a critical application. And there is no point to design such a thing except for a critical application with some unique requirements. But to make a new standard, mass-produced fastener to replace an existing one and make sure it's manufacturable, make sure its actually better than something like Torx and is worth making for your niche customers, make sure you've got the dimensions correct for all different sizes, then make prototypes and test them it could easily be person-years of work.

That being said, pozi-driv and JIS "+" style screws are far superior to philips and readily available. Even in Europe half the time you see a "+" it's philips and it's kind of impossible now have everyone know the difference and to deprecate philips even though they do usually have markings.[1]

[1]https://rtstools.com/jis-vs-phillips-screwdrivers-and-where-...

Boy, those look expensive to make. Not gonna use those to build a deck.

Agreed. I'd love to see the manufacturing process for those heads. I can't imagine such fine detail being forged. Maybe it's some type of multi-step blind broaching process. Like you said, expensive.

Most screw heads are formed in a cold header. If you've never seen one running, you'd think it was magic.

Wire goes in one side, and using a surprisingly simple and old (WWII era) machine made with some heavy castings, plain bearings, oily camshafts, and the magic of tool steel, the wire is incrementally upset, extruded, formed, punched, and thread rolled. Screws roll out the other side at an astonishing rate - hundreds per minute.

I can't share video of my customer's machine running, but here's one from Youtube:

https://youtu.be/WoN2KKfbzLA?si=iN42q6n1wBkoJ-gS&t=187

I imagine that many screws on spacecraft use CNC-cut machine screw threads, and have significant QC measures in place, but I'd be unsurprised if they had cold-formed Torx or hex heads.

See also MorTorq, also by Phillips, which can work in fasteners with low-profile heads.

Nah, they need Rockwell Automation's retro encabulator. The inverse reactive current it produces (designed for use in unilateral phase detractors) should also be capable of automatically synchronising cardinal gram meters. My guess would be that the gram meters have become unsynchronised.

https://www.youtube.com/watch?v=RXJKdh1KZ0w&ab_channel=rlcar...

Soak it in WD40?

"Complex organic molecules found in samples returned from asteroid Bennu! Spectral analysis looks a lot like AeroKroil."

Better headline: “NASA can’t access asteroid sealed in alien force field.”

Reply on news.ycombinator.com