Hacker News Reader

As a cyclist who does my own maintenance as a hobby, I am pretty willing to pay retail-type prices on exotic hardware that requires less effort on my part and is more reliable on the road.

Not the biggest market, sure, but it's a market where you could very plausibly sell a single fastener for ten-plus dollars with the right marketing.

Also a cyclist, biggest issue I have is getting things to the correct torque as carbon fibre can be a bit of a bugger, I can't recall anything coming lose on my bike in a long time though.

I doubt I'd buy a 10 dollar but either but if you marketed it as aero I know some who would ;)

i suppose if they market it right, such a boutique style use-case is viable. But i doubt it'd be all _that_ profitable...i m going to guess it's going to be a family style business at best.

> However as some have noted, cost will keep it from being adopted. Let's face it, fastening is the last thing considered in a project and the last thing bought. Even for large projects it's at most a few percent of budget and people are desperate to limit it's cost as much as possible.

This is incredibly true IME. When I was working at Xerox, they were in the process of switching to vastly inferior fasteners just to save a couple bucks per $500K machine.

I agree it's a little specialized and likely quite pricy, but as a sometime roboticist this kind of innovation makes me hopeful. Fasteners and connectors are unsolved problems, and robotics is plagued with such issues. Even an expensive solution for a few critical points would be valuable. Loc-tite works OK most of the time, but the ideal is a nut that actually stays tight all by itself.

A sincere question from another part-time (hobbyist) roboticist: what problems do you have with loosening that nylock[1] nuts don't solve?

[1] https://www.google.com/search?q=nylock+nut&espv=2&source=lnm...

Likely high temperatures and/or chemical resistance, maybe very high tightening torques or force, but possibly others, too. Mostly more specialized fastening, I would guess.

But think about the bolts and nuts that hold the exhaust system to a car engine. Temperatures are going to be very high and nylon wouldn't withstand those temperatures. Things like that.

Lock washers?

According to a NASA fastener design manual[1](page 9) split-ring style lock washers are useless for locking purposes. They say that star washers do proved some additional friction/locking action, but at the cost of damage to the surface(the friction comes from digging into the surface).

[1] http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/1990000...

Thanks. That's great to know. And lots of other useful information.

Fretting. Under high lateral force and vibration, I get fretting and I hate it.

I imagine aerospace may be an interesting application as well. Pretty darn hard to just "go tighten it" when it is on its way to mars!

I know you're probably joking but there are time-tested solutions to that problem.

e.g. http://en.wikipedia.org/wiki/Safety_wire

truly there's no need for a product like this unless the stress on the bolt, nut and what's being fastened have to be manage just perfectly.

Engine blocks.

Engine blocks aren't all that crazy. Most of the time (at least these days) they're made out of aluminum. That means that the stress on the bolt is limited more by the aluminum threads that the bolt mates with rather than by the bolt itself.

Further engine blocks are pretty cheap, all things considered. Turbines have tighter engineering constraints and wider price constraints so I would look there before looking at reciprocating ICEs.

So why are blown head gaskets such a problem?

Blown head gaskets have a lot less to do with bolt torque and a lot more to do with a lot of stuff all going on at roughly the same place. If they didn't separate combustion from coolant and both of those from oil, they would fail a lot less.

Yes the bolts must be torqued properly but that has more to do with ensuring that even pressure is applied across a huge gasket face than anything to do with the bolts self-loosening. If the heads on an engine were steel and much thicker you could do the same job with a couple of bolts instead of a dozen.

If an auto manufacturer was enterprising they could probably move some of the oil passages away from the coolant and combustion (which by their nature need to be close to one another) and they could make cavities for o-rings and seal the oil from the coolant a lot more effectively. But the failure rate for head gaskets is fairly low already and that would be adding a fair amount of cost and fiddly-ness to a not-that-big problem. When I say "fairly low" what I mean is that it might comprise 10% of engine failures but engine failures are already pretty rare, so 10% of 1% isn't all that big of a deal. Those are made-up numbers, btw, so don't crucify me if they're wrong.

"Yes the bolts must be torqued properly but that has more to do with ensuring that even pressure is applied across a huge gasket face than anything to do with the bolts self-loosening."

Ensuring that even pressure continues to be applied cross a huge gasket face is somewhat helped if the nuts do not become loose.

Agreed, but performing head gasket maintenance is a rare occurrence and loosening of the head gasket bolts is even rarer. Most of the time that head gasket failures happen is due to process problems on the part of the gasket manufacturer. The auto manufacturers have figured out how to ensure that they're torqued properly at the factory.

http://olybrake.com/pdf/fel_pro_torque_specs_guide.pdf

Most of the plain old torque specs in that brochure are at least 60 ft-lbs and there are some upwards of 100ft-lbs. That is no joke. If the bolt is a grade 5, 5/8" nominal diameter tightened to 57 ft-lbs that's over 4 tons of clamp force. At 82 ft-lbs it's just shy of 6 tons. At 113 ft-lbs it's over 7 tons. That's PER BOLT.

https://www.fastenal.com/content/feds/pdf/Torque-Tension%20C...

Applying 60 or 80 or 100 ft-lbs of torque to a bolt will cause huge amounts of clamp force which highly motivates the bolt not to wiggle loose. A lot of the problems where bolts (or nuts) self-loosen in high vibration environments is due to a torque spec for a sub-maximal amount of clamp force. If you're making machine tools and you need to tension ball-screw bearings (an application mentioned in the video) you can't torque the shit out of things because of the bearing pre-load needing to be a substantial but not insane level. A few hundred pounds of preload (a reasonable range for CNC spindles or ball-screws) can be achieved with only a few foot-pounds of torque; it's exactly what the machine needs but not enough for the bolt to keep itself from working loose.

Engine heads and head gaskets are complicated because you have to apply a lot of clamp force very evenly over a large area. The bolts are more than good enough for this, I've got several friends who are mechanics and none of them has heard of a head bolt that's loosened itself. The problem is that manufacturers often use "inferior" (less expensive) grades of head gaskets because the failure rate is already quite good.

EDIT: I don't disagree that needing to maintain even pressure is necessary (it is), nor do I disagree that nuts which can't wiggle loose doesn't help mitigate the problem. What I'm trying to say is that the failures which do happen are caused by the weakest link, and that in many cases the weakest link these days isn't the bolts or the heads or the block or whatever, but the gasket itself.

This could mean a lot for suspension components like spindle nuts where torque is so critical. Not only would that potentially mean no more cotter pins and staking the nut to the spindle, the preload on a wheel bearing would be more precise potentially leading to an increased MTTF just by using a different fastener.

You're correct in that for applications of limited complexity it's not likely cost effective. However we don't know what the premium is over traditional fasteners. It's also probable that in an assembly with many fasteners it may be cost advantageous to just use a product that solves the problem, rather than spend time and money on man-hours and possible production delays.

I vaguely remember seeing a TV show on amusement parks and thought that mechanics had to check and tighten every every bolt on a roller coaster rather frequently. I wonder if this type of nut would result in any labor savings for them?

Probably not, as they still might need to be checked periodically...

At least, that's what i want them to do :P

I don't know a lot about mechanical engineering, but at least in many aviation applications, the issue of nuts coming loose is resolved with a split pin that is bent into a locking shape after the nut has been tightened.

http://www.technologystudent.com/joints/comnut1.htm

Of course, this adds some extra overhead when disassembling the part, but the nut won't come loose unless there's a different problem (e.g. corrosion).

could the force that is trying to loosen the nut bend the split pin? would the self-tightening bolt perform better in this situation under the same condition?

What would be a miracle nut? Is it just about cost?

"Almost all bolts in the world are under-tightened."

You learn something new everyday.

I don't wanna alarm ya, but scientists say forty percent of America's pictures... are hanging crooked.

So are you saying that most people would think its nuts to spend so much on a fastener?

Lesson learned. Never make bad engineering puns lest you feel the wrath of the downvote onslaught.

So you are also saying as a former mechanical engineer there are no applications were tighter nuts are needed?

Well personally I know a few scenarios were the nut has keep steady when its hard to do. But for that case I prefer two nuts stacked on top of each other.

They're called jam nuts and they're a nightmare in production. These days I think they're only used where the application isn't critical, the labor is skilled and underpaid, and you're fresh out of lock washers and nylocks.

I haven't seen jam nuts used outside of construction in a long time.