Help me understand how chainplate loads work?

[a64pilot]

OK just for reference, an aircraft .5' NAS bolt will shear 18.650 lbs, meaning of course it fails at that. Now that is a way stronger bolt than any SS bolt, but it is the data I have, so you can see that the bolts themselves should never, ever see a load that shears them, even if there were no friction from clamping force. Airplane world we discount the friction, but it is there, how much I have no idea.

On edit, I would assume that in any boat with good rigging and chain plates, likely the hull would fail first depending on method of attachment.
Surely no manufacturer thru bolts chainplates through a balsa core do they? surely its solid glass there?

[Rapanui]

You are transferring the load from a high modulus and strength material (shroud or stay) to a low modulus and lower strength hull material (grp etc). To do this you have to spread the load over a larger area to keep the stress down to allowable levels. To this end, clamping of the chainplate to provide static friction is essential. The more bolts you use the more even and stronger the total clamping force. If you loosen all the bolts to the chainplate and take all the load on the bolts in shear, the bolts would take the load no problem but the bolts would eventually cut into the glass and elongate the holes.
The other factor is stress concentration, any discontinuity in the structure will increase the strain and hence stress markedly over theoretical stress Force/area. Spreading the load over a wider area reduces both the theoretical stress and the stress concentration factor. Ideally a chainplate would taper down in thickness towards its edge to reduce stress concentration.

[ur2slo]

Originally Posted by Tessellate

Ok, so to sum up: the chainplate bolts take the load, as shear forces (perpendicular / 90 degrees). And the number of bolts in a chainplate are due to the medium they're in (wood core / FRP skin) rather than the strength of the bolts under shear forces.

And some excess redundancy is built in due to higher probability of the bolts rusting and difficulty of inspection (complete lack of inspection on most boats probably). Also thrubolts are relatively cheap, so why not throw in a few extra.

It does not seem to be the case that there are more bolts due to shearing forces being higher than inline forces - if that were true, then the shear forces on a clevis pin in the toggles above deck would have comparable force and be comparable to a single thrubolt.

I wish there were a rigging book that explained these forces (which I suspect are not much beyond high school physics, but it's been a long time since I took that). I suspect the only option would be to go to mechanical engineering textbooks.

All good design points to bring up, bud.

As far as sailboat design, I'm not, but knowledge is available to all and it transfer's.

Find a copy of an F.A.A Publication AC-43.13 Acceptable methods and Tech.

You seem able to WANT to learn and it has many structural aspects that transfer to your new life at sea........( and we won't hold being an engineer against you) lol


As always.............Listen to everyone, but follow no one.

Enjoy your endeavor.


Question Stupidty

[roverhi]

3 bolts holding the chainplates to bulkheads or knees have worked just dandy for almost 50 years on my 35' boat. Don't understand the fetish with swiss cheese many mulitple fasteners chain plate attachments.

Did replace the chainplates for fear of crevice corrosion not fastener fatigue. Found only a couple of places with minor pitting in the through deck area of the chain plates after 45 years.

[jheldatksuedu]

I am an engineer, taught Physics, ME and EE for 30 years at Kansas State, I also have sailed all kinds of sailboats from windsurfers to ocean circumnavigating ketches.

One bolt would probably hold the load, but if and when it failed it would be sudden, sure there will be some clamping friction, mostly just after installation, but eventually shear of the bolts will most likely transfer 90% of the load, with 9 bolts, when they work into position over time, probably a year or more, you can count on 6 or so of the bolts sharing the load.

Redundancy is important in long life of equipment, reduce it to a minimum in a racing atmosphere and gain a very slight advantage, but every advantage is important in competition.

[catsketcher]

Consider the other parts of the load path too,

Remember that the load has to be transferred nicely from the stay into the bulkhead/hull skin and out into the structure to rotate the whole boat as the wind pushes on the sails.

You could of course engineer one bolt and chainplate to big strong enough to take the whole load in shear. But where does the load go? Through the bolt and then what?

You have to engineer the whole join. A longer chainplate allows a stiff part, the chainplate, to spread its applied load over a larger area on the hulls skin and into internal bulkheads. This lessens the stress on the hull near the chainplate. With multiple fasteners in a material that stretches slightly each one of the fasteners will share the load and as such the stress in the internal structure is reduced.

This is not always the case with composite chainplates. You could also engineer the inside of the hull skin to take a shorter and fatter chainplate.

This chainplate design is the result of centuries of evolution of steel and wood joins. At the other end of the stay is the mast tang with exactly the same load. On an alloy mast with a high modulus (stiffness) and high mechanical properties you can use a bolt through a reinforced mast wall. On a timber mast it pays to use the idea of spreading loads because the mast material is less stiff and less able to cope with high stress. Because chainplates are bolted onto wood or glass hull sides you have to use the same engineering principles as you do at the top of the mast. Of course on a steel boat you can just do as you did up top (on an alloy mast) if you like.

cheers

Phil

[catsketcher]

Pic for comment above

[DumnMad]

Originally Posted by Rapanui

You are transferring the load from a high modulus and strength material (shroud or stay) to a low modulus and lower strength hull material (grp etc). To do this you have to spread the load over a larger area to keep the stress down to allowable levels. To this end, clamping of the chainplate to provide static friction is essential. The more bolts you use the more even and stronger the total clamping force. If you loosen all the bolts to the chainplate and take all the load on the bolts in shear, the bolts would take the load no problem but the bolts would eventually cut into the glass and elongate the holes.
The other factor is stress concentration, any discontinuity in the structure will increase the strain and hence stress markedly over theoretical stress Force/area. Spreading the load over a wider area reduces both the theoretical stress and the stress concentration factor. Ideally a chainplate would taper down in thickness towards its edge to reduce stress concentration.

Correct. Stumble is also correct.
This is commonly not well understood even among engineers.

Shear is carried by the clamping friction or the bolts the bearing on the edge of the holes through the hull material.
No slip in clamping friction, but with cyclic loading movement has to take place for the bolt to take its varying bearing load - hence clamping friction is the way it is designed.
If clamping fails then movement in the chainplate should be noticed and total failure avoided.

[Eigenvector]

Not to discount some of the previous...........let me expand a little more as an Engineer on this subject.

Anytime you have more that ONE fastener loaded in shear, you will have an indeterminate structure. Due to tollerances of the holes and fastener diameters, one fastener will always make contact first. Stress is a result of Strain (deformation) which is measured in micro inches that you can't see. As it loads up, one will ultimately carry all the load until the clearance is made up with one of the others. Normally by this time the load on the primary fastener has already exceeded the yield strength and plastic deformation has occured.

By designing shear carrying joints based on friction, this is eliminated. Of course there is a generous safety factor involved. Normally, the bolts/rivets/screws are sufficient to carry the load when loose, but damage will be done as the material deforms to pick up the load sharing between the fasteners.

Additionally in particular to fiberous materials like wood and frp's they are highly directionaly dependent with the lay of their fiber. In particular with glass and other modern fibers, they are strands within a matrix of epoxy. The stiffness mismatch between the fiber and resin are orders of magnitude different. As a result, the fibers take all the load and the binder is just to keep them contained as a unit. You don't want to compression load small fibers. This would occur if you had bolts imparting a compressive load load on the bore of the bolt holes.

By utilizing clamping friction as your primary load transfer mechanism, you have dispersed an otherwise concentrated load uniformly and nicely into the panel structure in a nice way without causing stress concentrations. Standard practice in bolted joint design.

I've never designed a sailboat, but a lot FRP stuff that went into the ocean. Hope this clarifies what's going on as to the load transfer.

[Stumble]

Eigen ,

Thanks for the explanation. I think I have mentioned this before but I am not an engineer though I have worked with them over the years.

Frankly I think any new boat that uses metal chainplates is missing out on a huge opportunity. Carbon or fiberglass chainplates woven into thehull laminate eliminate one of the most common causes of rig failure, and aren't that much more expensive to build or design. It really is a better mousetrap.

[ur2slo]

Originally Posted by Stumble

Eigen ,

Thanks for the explanation. I think I have mentioned this before but I am not an engineer though I have worked with them over the years.

Frankly I think any new boat that uses metal chainplates is missing out on a huge opportunity. Carbon or fiberglass chainplates woven into thehull laminate eliminate one of the most common causes of rig failure, and aren't that much more expensive to build or design. It really is a better mousetrap.

Lots of good brain food here.

Interesting thought on the use of woven fiber to alleviate stress concentration.

But as we all well know the KISS principle ( keep it simple stupid) has been well proven over the course of building sailing vessels.

Experience will tell, and you get experience from bad luck.

[rhubstuff]

Stumble is correct. A properly designed and installed chainplate resists loads by friction, not by shear on the bolts. The bolts squeeze the chainplate to the hull to create resistance to shearing force created when vertical tension is applied by the stays and shrouds. IOW, if the holes through the hull were slightly oversized and the bolts installed dead center in the holes, the bolts should not migrate to the tops of the holes under tension; they should stay centered. That is why more bolts are better; more bolts equals more friction between the hull and chainplate, which equals greater resistance to tension applied to the chainplate.

[Eigenvector]

Originally Posted by Stumble

Eigen ,

Thanks for the explanation. I think I have mentioned this before but I am not an engineer though I have worked with them over the years.

Frankly I think any new boat that uses metal chainplates is missing out on a huge opportunity. Carbon or fiberglass chainplates woven into thehull laminate eliminate one of the most common causes of rig failure, and aren't that much more expensive to build or design. It really is a better mousetrap.

I totally agree. However due to cost, it's really hard to pop a laid up structure out of the mold with these types of discontinuities. It can be done. The general rule I always quote to clients is: "Fast, Cheap, or Good. Pick Two." And when it involves FRP or injection molding, pick One.

[ur2slo]

Alrighty then, my brains churning.

if you all say the " friction coefficient" ? of the two surfaces interacting, ie; chain plate and mounting surfaces, is what supports the primary shear load interaction of the rigging attachment?


If that's' the case, then if I were to hypothetically remove the chain plate from the equation & attach directly to surface by bolting, what is taking the load?

I go back to my original injecture, that the bolts, no matter what, take the load.

I take it as the chain plate distributes said load across wider area.

So? Smarter minds than mine are at work here. I'm curious how you conclude your theory?

[DumnMad]

Originally Posted by ur2slo

Alrighty then, my brains churning.

if you all say the " friction coefficient" ? of the two surfaces interacting, ie; chain plate and mounting surfaces, is what supports the primary shear load interaction of the rigging attachment?


If that's' the case, then if I were to hypothetically remove the chain plate from the equation & attach directly to surface by bolting, what is taking the load?

I go back to my original injecture, that the bolts, no matter what, take the load.

I take it as the chain plate distributes said load across wider area.

So? Smarter minds than mine are at work here. I'm curious how you conclude your theory?

I'm afraid you are the one with the theory.
No theory to us - we know how friction grip bolting works. They transfer the load by shear friction. Used in all sorts of areas of the construction and manufacturing industries. Google "friction grip bolts."