Posted: Sat Mar 31, 2012 5:21 pm Post subject: Thread engineering 201
Binding pullout is under discussion and foam core skis are taking a beating again. I'm always unsure of the engineering implied in some analyses, vs how much the explanation is--however correct--ultimately subjective & intuitive.
With permission, a fundamental engineering question. Some years ago, while sharing my vast binding expertise gleaned at Ttalk, while skiing along with a friend who (was) a "nuts & bolts" mechanical engineer at Lawrence Livermore, he shocked me. He said in engineering threads, you assume all the load will be borne by one thread. Some time ago I emailed him to confirm my memory, and expand a little:
Quote:
Yes, you are correct. When one looks at thread strength it is assumed that one thread carries all the load. I really can't remember the details behind this, but I would guess that it is assumed that there is no way to make the threads so exact that all can be equally engaged simultaneously. The only way to get all threads to physically touch is for there to be deformation and that brings in a whole lot of other issues. That said, you don't want to design a nut with one thread.
I know what you're thinking. He goes on to say:
Quote:
There are some standards. Normally one wants to engage a length of threads equal in length to the diameter of the threads. For example, a standard 1/4-20 nut is 7/32" thick. Yes, it is less than 1/4" but the diameter they are using is the root of the threads (or something close to that). A thin nut would be 5/32" thick for 1/4-20.
The point here is, why would you care what the core of a ski is if you have a binding plate that holds? Presumably the densolite core of Atomics could hold so little the thin delrin plate must be good enough. If your plate is metal and glued to the ski, is it doing the whole job? I've no doubt maple Scotty Bobs can hold a screw well but I wonder how much solidity these pithier woods--paulownia, poplar--cores provide? I know epoxied fiberglass alone can do the job, at least well enough to hold 3-pins to some glassed, cardboard cored Fischer XC skis I once had.
My other engineering curioisty is, if it is the whole screw you're using, via slip-friction or plasticity, wouldn't that inevitably cause a little volcano and introduce serious flex stress on the screw?
Joined: 12 Jun 2006 Posts: 2971 Location: the cascade foothills
Posted: Sat Mar 31, 2012 7:06 pm Post subject: Re: Thread engineering 201
Whiteout wrote:
The point here is, why would you care what the core of a ski is if you have a binding plate that holds? Presumably the densolite core of Atomics could hold so little the thin delrin plate must be good enough. If your plate is metal and glued to the ski, is it doing the whole job? I've no doubt maple Scotty Bobs can hold a screw well but I wonder how much solidity these pithier woods--paulownia, poplar--cores provide?
IF the binding was "riding" on the heads of the screws alone, it would fail pretty quickly. What the screw actually does is create a wide area of surface contact under pressure between the binding base plate and the ski's surface.... This area under pressure gives the binding a greater ability to resist the torsion appied to it by the skier. Once there is some play between the binding and the ski surface, it's just a matter of time before that torsional pressure ovalizes the screw holes in the metal top sheet enough to allow the binding rip out to easily occur....
This is one of the reasons that I paint the top of the ski with a thin coat of epoxy when I mount a binding. The "bedding" of the base plate in the liquid epoxy creates a negative image of the binding base plate shape so the skier's force can not move the binding laterally once the epoxy cures,...... THUS saving the integrity of the hole in the ski's top sheet metal..... AT that point, the only way the mount can fail is generating enough force to rip out a 100% intact screw to ski connection.... which is the rarest occurance.. I have mounted many BD01's and never had a person I mounted them for report a single rip out, (but I make furniture for a living,... so I am pretty meticulous and accurate... which helps...
You owe me a beer.... send this to your engineer friend.... and see what he thinks... _________________ the fall line is your friend.... resistance is futile
Joined: 09 Apr 2007 Posts: 1479 Location: Valmorel, France
Posted: Sat Mar 31, 2012 7:19 pm Post subject: Re: Thread engineering 201
Whiteout wrote:
Binding pullout is under discussion and foam core skis are taking a beating again. I'm always unsure of the engineering implied in some analyses, vs how much the explanation is--however correct--ultimately subjective & intuitive.
With permission, a fundamental engineering question. Some years ago, while sharing my vast binding expertise gleaned at Ttalk, while skiing along with a friend who (was) a "nuts & bolts" mechanical engineer at Lawrence Livermore, he shocked me. He said in engineering threads, you assume all the load will be borne by one thread. Some time ago I emailed him to confirm my memory, and expand a little:
Quote:
Yes, you are correct. When one looks at thread strength it is assumed that one thread carries all the load. I really can't remember the details behind this, but I would guess that it is assumed that there is no way to make the threads so exact that all can be equally engaged simultaneously. The only way to get all threads to physically touch is for there to be deformation and that brings in a whole lot of other issues. That said, you don't want to design a nut with one thread.
I know what you're thinking. He goes on to say:
Quote:
There are some standards. Normally one wants to engage a length of threads equal in length to the diameter of the threads. For example, a standard 1/4-20 nut is 7/32" thick. Yes, it is less than 1/4" but the diameter they are using is the root of the threads (or something close to that). A thin nut would be 5/32" thick for 1/4-20.
The point here is, why would you care what the core of a ski is if you have a binding plate that holds? Presumably the densolite core of Atomics could hold so little the thin delrin plate must be good enough. If your plate is metal and glued to the ski, is it doing the whole job? I've no doubt maple Scotty Bobs can hold a screw well but I wonder how much solidity these pithier woods--paulownia, poplar--cores provide? I know epoxied fiberglass alone can do the job, at least well enough to hold 3-pins to some glassed, cardboard cored Fischer XC skis I once had.
My other engineering curioisty is, if it is the whole screw you're using, via slip-friction or plasticity, wouldn't that inevitably cause a little volcano and introduce serious flex stress on the screw?
Saturday curious.
Your friend refers to nuts&bolts, so his thinking is true for a metal/metal thread. In the case of skis, the screw creates its own thread in the wood (or foam), and the wood is rather soft and flexible compared to the metal screw. So each thread is engaged. _________________ Dré dans l'pentu !!
"Why can't people here be a bit more serious and why are so many posts utterly bs?" (Telesaint)
Posted: Sat Mar 31, 2012 8:30 pm Post subject: Re: Thread engineering 201
tele.skier wrote:
IF the binding was "riding" on the heads of the screws alone, it would fail pretty quickly. What the screw actually does is create a wide area of surface contact under pressure between the binding base plate and the ski's surface.... This area under pressure gives the binding a greater ability to resist the torsion appied to it by the skier. Once there is some play between the binding and the ski surface, it's just a matter of time before that torsional pressure ovalizes the screw holes in the metal top sheet enough to allow the binding rip out to easily occur....
This is one of the reasons that I paint the top of the ski with a thin coat of epoxy when I mount a binding. The "bedding" of the base plate in the liquid epoxy creates a negative image of the binding base plate shape so the skier's force can not move the binding laterally once the epoxy cures
Well this is certainly my feeling about bindings---critical to get full, contact, under slight compression, of binding shim to topsheet. But the screw's role is unchanged. Meet that shim in compression and not yield. And the question remains, do you need 6 threads to accomplish this or will one do? For sure, with a wood mounting block and fiber deformation, likely many threads are contacting at the same time. But if you have a metal plate--or hard plastic with metallic properties--you DO have metal on metal. And you could engineer it so all you needed was, say, two threads. And you wouldn't care if the core was balsa.
Side note tele.skier (I'll bring a six) moved HHeads to new skis last weekend. They were on DIMyself quiver killer shims with nuts. I had "embedded" them to the skis not with epoxy, but silicone caulk. Way softer, but...they were incredibly hard to remove. Had to carefully pry them off with a chisel. No way could they exert any sideways--or upward--motion on their holding wood screws. In past think I've done what you do: clamp slightly, snug screws, back off, clamp firmly, tighten screws.
Joined: 23 Dec 2010 Posts: 412 Location: in a subaru down by the dry wash
Posted: Sat Mar 31, 2012 9:01 pm Post subject: Re: Thread engineering 201
Whiteout wrote:
Well this is certainly my feeling about bindings---critical to get full, contact, under slight compression, of binding shim to topsheet. But the screw's role is unchanged. Meet that shim in compression and not yield. And the question remains, do you need 6 threads to accomplish this or will one do? For sure, with a wood mounting block and fiber deformation, likely many threads are contacting at the same time. But if you have a metal plate--or hard plastic with metallic properties--you DO have metal on metal. And you could engineer it so all you needed was, say, two threads. And you wouldn't care if the core was balsa.
Binding screws are basically #12 sheet metal screws - greater thread depth than say wood screws, designed originally for screwing into a thin piece with perhaps only one thread engaged. I believe that much of the holding power comes from the thin reinforcement sheet in the binding area.
But I also suspect you should think of the threads in the core material as adding preload to keep the screw tight. It's probably like the nylon in a nylock nut. The nylon doesn't add absolute strength to the metal nut; it keeps preload on the fastener so it doesn't loosen from vibration or load/unload cycles.
Bindings, especially tele bindings, see a lot of load cycles. It's likely that binding failure is preceded by a gradual loosening of the screw - if small motions start to wallow out the core material and the screw rocks in the hole, it's then much more likely to rip through the binding mounting sheet. This is conjecture - I've not ripped a binding out or inspected a tear-out. Also, it's warm here and there is no snow, blargh.
Posted: Sat Mar 31, 2012 10:16 pm Post subject: Re: Thread engineering 201
polemonium wrote:
Bindings, especially tele bindings, see a lot of load cycles. It's likely that binding failure is preceded by a gradual loosening of the screw - if small motions start to wallow out the core material and the screw rocks in the hole, it's then much more likely to rip through the binding mounting sheet. This is conjecture - I've not ripped a binding out or inspected a tear-out. Also, it's warm here and there is no snow, blargh.
This is an important point, and can't be stressed too much (no pun intended) in this recurring thread.
This is also why the role of epoxy is sometimes misunderstood. The epoxy helps prevent rotation / gradual loosening of the screw; in and of itself, it is not 'strengthening' or increasing the direct pull-out failure force.
And since there may be some Dynafit users reading this, I'll throw in my usual comment about the Dynafit engineers being smart enough to thread the baseplate of the toe...this means that each of the 4 screws is held perpendicular to, and by, the baseplate, which means the constant cyclic forces in plane with the topsheet are less likely to begin to work each screw hole transversely.
So ignore the small group of maggots who tell you to drill out those pesky threads on the dynafit baseplate.
On a related note: since it can be difficult sometimes to get each screw properly epoxied to the hole, I occasionally had rotte cobra baseplates start to work the screws a bit loose. Usually just cleaning the threads & re-epoxying fixed this problem, but I also found that double-side carpet tape in between the metal mount plate and the ski topsheet worked really well at resisting those transverse, repeated forces.
I've not had to use the carpet tape trick on any NTN baseplates; I suspect that's because the roller system/round toecup on NTN reduces those twisting forces. _________________ "Now is not the time for sound-bites" - David Cameron
But I also suspect you should think of the threads in the core material as adding preload to keep the screw tight. It's probably like the nylon in a nylock nut. The nylon doesn't add absolute strength to the metal nut; it keeps preload on the fastener so it doesn't loosen from vibration or load/unload cycles.
Nice! I can feel that. Bet there is another term for "preload,"--tension?--but like the idea of elastic resistance.
Jim wrote:
Just remember - a screw is just a simple lever.
I thought it was an inclined plane. I know, only going in. Once there, it's lever properties enhance the benefit of having a good grip at the long end of the lever, that part in the ski-core. Its uplift resistance may be immaterial, but once embedded, it provides great resistance to movement of the short end by skiing forces. And well embedded, it turns the fulcrum from just a couple of threads to the length of screw below the topsheet. That's what this has me thinking.
So, the straight pull-out may be just one thread, but the mount-killer is ovalling vibration and torsion forces perpendicular to the screw. The the nylock stop-nut comparison makes me think that the return-to-shape properties of the delrin mounting plate on an Atomic ski makes it even better than a metal mounting plate. (paren--and it's not like densolite is styrofoam...think you shape it with a file). A lot of bindings stay put screwed only to the plastic shim, the shim being screwed to the ski.
nils wrote:
This is also why the role of epoxy is sometimes misunderstood. The epoxy helps prevent rotation / gradual loosening of the screw; in and of itself, it is not 'strengthening' or increasing the direct pull-out failure force.
Been on board with this for a while, but apply an equally effective substitute to epoxy (for a good reason.) No substitute with enlarged hole. Was successfully using silicone caulk but have now stopped: doesn't really stick to wood well and leaves a residue in the hole hard to clean when you need to switch to epoxy. Now use Titebond III wood glue: waterproof, dries hard but elastic w/o shrinkage, sticks to anything, vice versa.
I haven't quit silicone caulk to "embed" the shim to the ski--the carpet tape application. It is like wetting a suction cup before sticking it to glass. Tape would be handier in a lot of cases--had forgotten that.
Re the abundance of mounting talk but scarcity of calculation of the "2X enough" line and the boundary of "works for me" overkill. Fun quote in a story on false alarm that neutrinos had travelled faster than light and Einstein been disproven. Astrophysicist Arthur Eddington: "No experiment should be believed until it has been confirmed by theory."
Joined: 20 Oct 2005 Posts: 17735 Location: following Diogenes, but the ba$tard threw away the lamp so I'm just stumbling along in the dark!
Posted: Sun Apr 01, 2012 5:42 pm Post subject:
Jim wrote:
Just remember - a screw is just a simple lever.
Bingo!
One that is anchored in a hole subject to ovalization in the context of a seriously flexed ski!
While helicoils and other metal receivers for said screws/bolts might not ovalize, the hole they are in certainly will change shape as the ski bends. If the bonding agent can, while being stretched along one axis and pinched on the other, maintain good grip on the ski...no problem!
I like that double sided tape concept! It would put some of the stress on the skin of the topsheet...but there there is the rigidity of the binding base plate to consider...it that could flex with the ski, problem solved....except for metal fatigue issues of said plate.
Might be nice to see if a system plate for mounting tele bindings would reduce or eliminate the problem altogether.
Quote:
Fun quote in a story on false alarm that neutrinos had travelled faster than light and Einstein been disproven. Astrophysicist Arthur Eddington: "No experiment should be believed until it has been confirmed by theory."
Logic, who needs it _________________ "Moderate is not the new Low" - Chris Joosen, USFS Lead Snow Ranger (Tuckerman Ravine, White Mountains National Forest)
Joined: 21 Sep 2007 Posts: 17707 Location: EL/R -6.12, SL/A -8.15 in NW VT and slightly south of the Poutine Curtain
Posted: Sun Apr 01, 2012 7:16 pm Post subject:
thread engineering is tricky business. The true art of the craft is when one is able to pull off a blatant troll and nobody figures it out. Oh, wrong kind of engineering, wrong kind of thread. darn. _________________
Now use Titebond III wood glue: waterproof, dries hard but elastic w/o shrinkage, sticks to anything, vice versa
TitebondIII doesn't stick to anything to anything - it's a wood glue and is a pretty crappy bonder of metal to wood. Epoxy or if you want to spend the $ 3m5200 are much better bonders of wood to metal. TIII just does an ok job of keeping out water; it's not meant for below the waterline use.
the original atomic mounting plates weren't delrin I believe; something else. The aftermarket ones are sometimes delrin.
the original atomic mounting plates weren't delrin I believe; something else.
Right:
Quote:
-------- Original Message --------
Subject: Mounting
Date: Tue, 28 Mar 2006 09:03:41 -0500
From: Rick Halling <RJHalling@atomic-usa.com>
To: <charleshwhite@mindspring.com>
CC: Fezz Ali <AliF@atomic-usa.com>
The Kongur has a Rilsan plate in the binding area that the alpine skis
do not have for reinforcement.
Same difference I bet.
ghostofcarl wrote:
TitebondIII doesn't stick to anything to anything - it's a wood glue and is a pretty crappy bonder of metal to wood.
Wrong:
Here's a jambsaw I made to cut dormer siding butting up against roof shingles. The black arrow is epoxy, but the entire wood-to-metal contact IS Titebond III--red arrow. Took a lot of use, still holding. AND, when I break the bond, I expect it to break clean. Just like I want. Just like I want for my binding screw.
ghostofcarl wrote:
TIII just does an ok job of keeping out water; it's not meant for below the waterline use.
Neither is plain marine epoxy...without additives. Check West Marine. My binding is above waterline--no hydrostatic pressure to speak of. TIII--like one thread--is ok enough, AND meets the hard requirement of being field-repairable 4 days from any shop.
Hey, all we do here is all hilarious babble. Until there is an eyeball in your vomit. Or it's YOUR binding broken 4 days back, no hot screwdriver in sight.
Last edited by Whiteout on Tue Apr 03, 2012 2:43 pm; edited 1 time in total
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
All of the comments above are owned by the
poster, telemarktips.com is not responsible in any way for the
content. The views expressed by the posters are not necessarily
those of Tt.com, its management or owners. Ski safe, be happy,
rip it up, smile on your brother and sister!
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
I know, only going in. Once there, it's lever properties enhance the benefit of having a good grip at the long end of the lever, that part in the ski-core. Its uplift resistance may be immaterial, but once embedded, it provides great resistance to movement of the short end by skiing forces. And well embedded, it turns the fulcrum from just a couple of threads to the length of screw below the topsheet. That's what this has me thinking.
