Deck Cleats - Four or Two Bolts? SS or Aluminum?
- Thread starter Blitz
- Start date
- Status
Dave,
That was a minimum strength number. The engineer at West said he has seen bolts fail tapped into solid epoxy. But without testing he would not commit to more than 1200. He did tell me that a 1" hole in a 2X4 filled with thickened epoxy and a 5/16" bolt set into it held 2.5k..
We are not talking about threading cleat screws directly into the fiberglass we are talking about adding backing plates that will likely cause bolt failure before deck failure. We also need to remember what is in the decks. Nearly every production boat out there has plywood or balsa between the two very thin laminations, in many cases less than an 1/8" of glass on the top and bottom. If anything is going to fail it will be the core crushing well before the G-10. Of all the boats I have seen wash up on a beach in a storm I can't recall any of them being from a torn out cleat and the majority are using flimsy fender washers on a wood cored deck without even having the holes potted. The failure mode of the cleats are most often chafe or rode or pendant failure.
80% of the boats I have been on use dinky little fender washers and how often do we hear of cleats pulling through a dry deck in good condition?
If you can pull a fender washer through 3/8" G-10, before the deck or bolt fails, I will personally write you a check for $10.00..
Blitz,
G-10 is far stronger and far less prone to compression than any plywood but you pay for it. That being said plywood works well too and has for years provided it does not get wet. The only backing material I wont use is a solid wood as I have seen oak, mahogany and teak split along the grain. I had a PO installed teak backer on my own C-30 split..
Jibes—
The shear strength is 19000 psi, according to the spec sheet located HERE. Be aware this is a different manufacturer than the one I linked to previously, but both materials are made the same mil-spec standard, so should be reasonably similar in strength.
The shear strength is 19000 psi, according to the spec sheet located HERE. Be aware this is a different manufacturer than the one I linked to previously, but both materials are made the same mil-spec standard, so should be reasonably similar in strength.
This discussion is getting uglier by the minute. Every time you bring up a point you change the rules regarding what you are talking about. Who said anything about mounting a cleat using tapped holes in fiberglass? That would be a new definition of insanity. Casting a 5/16 bolt into liquid thickened epoxy in a 1 inch hole and then applying a tensile load to the bolt is a lot different than tapping a hole in a fiberglass part and then pulling on the screw. Kind of like mixing apples and pomegranates. What if the hole was 2 inches, what difference does the size of the hole make unless you are worried about the epoxy to parent part bond strength as opposed to the strength of the threads. In that case what is the epoxy bonded to as the strength will vary by how well it bonds. If you are bonding it to a hole with saran wrap for a liner I'll be a little worried.
My point was just don't let me catch you tapping holes in epoxy and expecting them to hold any kind of structural load. epoxy fully engaging the threads is a little different than a tapped hole where the interface is at the pitch diameter, the minor diameter of the hole is relieved for clearance with the root diameter of the bolt thread and the major diameter of the bolt is relieved so it is clearing the root of the female thread. The strength in a tapped hole in FRP is primarily just the strength of the plastic and in shear it is probably just a little better than peanut butter.
I am not in disagreement with you at all about the strength of the cleat assembly installation but if you engage in a design discussion be prepared to use facts and figures, not gut feel and opinion. I'll bet when you run the calculations on your finite element analysis software you will find the fender washer with no backing plate has a sufficient safety factor or we would in fact be seeing failures, but until you show me some numbers you can't prove the position. Adding a backing plate is cheap insurance and pretty much any structurally sound material will be fine (not wood). Backing plate or not you are still limited by the strength of the deck laminate and its' condition.
I expect to see a complete design analysis with assumptions for loading based on forces on the hull in a hurricane force wind and shock loading from waves looking at the cleat assembly as a complex loaded part and have it ready for my review in the morning. :>)
By the way I looked up the tensile strength of plywood and it was about 4496 PSI or 31MPa so Sailing Dog is vindicated that G10 is stronger than plywood. So are thousands of other materials. Personally if I had to make a backing plate I'd use AMS5670E in a forged plate configuration to make sure the grain orientation was optimized for the fatigue loading. If you want to put your boat at risk with a lesser design is a decision you will have to make.
My point was just don't let me catch you tapping holes in epoxy and expecting them to hold any kind of structural load. epoxy fully engaging the threads is a little different than a tapped hole where the interface is at the pitch diameter, the minor diameter of the hole is relieved for clearance with the root diameter of the bolt thread and the major diameter of the bolt is relieved so it is clearing the root of the female thread. The strength in a tapped hole in FRP is primarily just the strength of the plastic and in shear it is probably just a little better than peanut butter.
I am not in disagreement with you at all about the strength of the cleat assembly installation but if you engage in a design discussion be prepared to use facts and figures, not gut feel and opinion. I'll bet when you run the calculations on your finite element analysis software you will find the fender washer with no backing plate has a sufficient safety factor or we would in fact be seeing failures, but until you show me some numbers you can't prove the position. Adding a backing plate is cheap insurance and pretty much any structurally sound material will be fine (not wood). Backing plate or not you are still limited by the strength of the deck laminate and its' condition.
I expect to see a complete design analysis with assumptions for loading based on forces on the hull in a hurricane force wind and shock loading from waves looking at the cleat assembly as a complex loaded part and have it ready for my review in the morning. :>)
By the way I looked up the tensile strength of plywood and it was about 4496 PSI or 31MPa so Sailing Dog is vindicated that G10 is stronger than plywood. So are thousands of other materials. Personally if I had to make a backing plate I'd use AMS5670E in a forged plate configuration to make sure the grain orientation was optimized for the fatigue loading. If you want to put your boat at risk with a lesser design is a decision you will have to make.
AMS5670E - What is this? - Aerospace Material Specifications - not going to the moon, maybe just an afternoon sail though
OK. We're waiting. I've posted rough calculations for many things before, on forums and my blog; sometimes folks find errors and omissions, sometimes they don't. Go for it.
Nuff said. I have done that sort of calculations for cored hulls and you are generally correct. The wild card in any sort of general advise, of course, is that every deck is different.
I use 1/2" thick "Star Board" for backing plates, one piece, as large as possible to spread the load over a greater area.
http://www.kingstarboard.com/Pdf/PhysPropKingStarBoard12_01_2002.pdf
Bill
http://www.kingstarboard.com/Pdf/PhysPropKingStarBoard12_01_2002.pdf
Bill
Of course, the use of a backing plate often means that the cleat will hold even if the decks coring has rotted away... since the load is spread over a greater area...
As for using starboard for a backing plate... I wouldn't recommend it. Starboard is not designed or recommended for use in any structural setting. It is not stiff enough to really work well as a backing plate IMHO.
As for using starboard for a backing plate... I wouldn't recommend it. Starboard is not designed or recommended for use in any structural setting. It is not stiff enough to really work well as a backing plate IMHO.
True, but either will G10 Garolite, and for the price, I'd rather have the 3/8" G10, which is thinner, stronger and stiffer, and probably lighter than 3/4" starboard.
Starboard (HDPE) also creeps and sometimes cracks in cold weather or under impact loads. It is simply not the best choise and is not used by builders for that sort of purpose. It is for furnature, bearings, wear areas, and low-load spacer blocks. Things like that. I have used it and know its limitations.
The trouble with creap is it LOOKS like you have spread the load, but only the part right under the nut has any pressure on it. Fender washers can do this too, if overloaded; for cleat or high-load aplication I double them or fabricate a washer from something a little thicker.
FRP, SS and Al have better physical properties and are all used and suitable when properly enginereered. I have used all of these: Aluminum is easiest to work if, I can protect it from salt; SS is easy when fender washers are enough; and FRP is best for large plates (I get 1/8" - 3/8" pre-cast cut-off sheets free from a fabricator), but small SS washer are still needed under the nuts. IMHO.
One size seldom fits all.
Catsailor—
I still use fender washers with my fiberglass backing plates.... might not help... but can't hurt...doesn't really cost me anything to do so.
I still use fender washers with my fiberglass backing plates.... might not help... but can't hurt...doesn't really cost me anything to do so.
I had a stern cleat rip out because I didn't have my lines secured correctly, they slipped down the pylon and the 18" of tide with some rough weather removed it for me.
They were 6" aluminum, 2 hole cleats. I replaced them with 10" yacht style cleats ss, with 4 mounting holes. Backed by fender washers. I did properly pot the holes. My bow cleats are 10 yacht cleats ss, I backed those with 3/8" aluminum plate.
I like the bigger yacht cleats because they have so much room, it makes it easier to tie to.
They were 6" aluminum, 2 hole cleats. I replaced them with 10" yacht style cleats ss, with 4 mounting holes. Backed by fender washers. I did properly pot the holes. My bow cleats are 10 yacht cleats ss, I backed those with 3/8" aluminum plate.
I like the bigger yacht cleats because they have so much room, it makes it easier to tie to.
That is the cleat I used. And the begining of a repair where the other cleat vacted.
They are also called Herreshoff cleats.
They are also called Herreshoff cleats.
Attachments
This empirical data provides an abundance of information critical to our discussion. First let's assume this is a typical production boat cleat installation and that only fender washers were used. The load applied was similar to a prying motion from the description from Scott, the boat was being submerged on one side by pulling downward on the cleat from the side. This makes a strong case for the absolute need for a backing plate under your cleats. Note the line did not break, the deck failed. Increasing the area will drastically improve the situation as follows. Lets assume the cleat base is 2 inches by 4 inches x a 1 inch thick deck. The cross sectional area resisting pull out (assuming a backing plate that size) will be 12 square inches of shear area. SailingDog recommended in an earlier post use a factor of four times larger than the base. So if we make the backing plate 16 by 8 the shear area increases to 48 square inches. I happen to be looking at a ruler and a 16 x 8 backing plate is gigantic. I don't think I've ever seen a backing plate that big?
Perhaps this is why travel lifts use straps under the boat instead of just grabbing it by the cleats to pick it up? I wouldn't want to trust my boat in a breaking wave situation, anchored in the surf trying to save her where she is getting pounded on to not have the cleats rip out like this. Definitely need a stronger installation than what most builders are providing.
I'm surprised the piling didn't get pulled out of the mud before the deck ripped apart like this.
Perhaps this is why travel lifts use straps under the boat instead of just grabbing it by the cleats to pick it up? I wouldn't want to trust my boat in a breaking wave situation, anchored in the surf trying to save her where she is getting pounded on to not have the cleats rip out like this. Definitely need a stronger installation than what most builders are providing.
I'm surprised the piling didn't get pulled out of the mud before the deck ripped apart like this.
Four times the area would be 4" x 8" for the backing plate. You're going with 16 times the area, not 4... which is probably why you've never seen a backing plate that big... neither have I.
Also, pulling down on a cleat isn't helped by a backing plate, since you're forcing the cleat into the deck, not trying to pull the cleat's backing plate through the deck. Unless the backing plate is epoxied in place, it does nothing to resist forces pushing it down into the boat. If the deck core is rotted, then there is only a very thin layer of fiberglass preventing the cleat from pushing down into the hull.
Also, pulling down on a cleat isn't helped by a backing plate, since you're forcing the cleat into the deck, not trying to pull the cleat's backing plate through the deck. Unless the backing plate is epoxied in place, it does nothing to resist forces pushing it down into the boat. If the deck core is rotted, then there is only a very thin layer of fiberglass preventing the cleat from pushing down into the hull.
Design loads
OK, got all the theory about shear, threads, etc.
Do we know, or know how to calculate, the actual loads that a cleat will have to withstand? Probably pretty hard!
My thought is that I'd want to break my line before the deck or cleat tore out. Given that, add a safety factor of x.xx to feel comfortable about any assumptions you made and the go design the thing. From WM I got the following plated nylon data:
1/2" 7500 lb breaking strength
9/16" 9200 lb
5/8" 10400 lb
3/4" 15000 lb
since they specified breaking strength I probably don't need a huge safety factor and would guestimate 1.1 would be fine.
so my 5/8" anchor/drogue rode would part BEFORE the cleat got to 10400*1.1=11440 lb. so build the structure to handle that. Not much sense building it stronger than the rope it attaches to right?
50k psi materials will need 11440/50k=0.2288 square inches of area to withstand that
19k psi materials would need 0.602 square inches of area
I'm thinking that a washer would be fine provided the structure did not fail in another mode. The other mode being a rotten deck or thin FRG layup.
If someone has the strength of fiberglass I'll do the calc for the area you need to supply (via backing plate load spreader). Id assume that the core does not contribute to the holding in place and just increases the torque that the structure can handle.
Not sure why I'd need the cleat to be stronger than the line except to account for shock or fatigue loading. In which case I'd use 2.0 safety factor and go with 0.45 and 1.2 square inches respectively. Still pretty small IMHO
Other failure modes:
My engineer sense tells me that a deck cleat can fail in the following modes;
water runs out at the dock and the cleat is pulled up
water comes in at the dock and the cleat is pulled sideways and/or downwardish but mostly sideways
the wind / waves hit the boat and the cleat is pulled sideways
Cool! all three modes so we get to earn our keep.
Pulling up mode is either threads stripping, cleat bending, or deck failing in shear
Pulling down/sideways is the bolts bending by toque and or shearing or the deck failing in torque
Pulling sideways is the toe rail plus toe rail bolts plus the deck failing in tension or the bolts shearing off.
since we already know it does not take much surface area for a metal to develop these kinds of forces and I don’t know fiberglass's strength lets make some assumptions about the structure and see how highly stressed it would be and "get a feel" for it.
Assume the sideways load is max at 11440 lb and I provided ample metal to resist shearing off the bolts or bending them off. The failure mode is then pulling a (assuming here) rectangular shaped piece of deck sideways off the boat. If the deck core is 0.5" thick and the glass is 1/16" thick on both sides and the cleat is 4" long and 2" from the edge and there is no metal toe rail (glass takes all the load) we get:
length of the tear resisting area PSI on the glass
4+2+2" 2*(1/16*8)=1.0 11440/1.0=11440 psi
if it rips right out at the cleat and takes the deck to the toe rail with it, it could be made of wood veneer and still hold (wood typically is ~15k psi in tension). FTR I ignored that fact that the two 2" tears to the toe rail would be in shear but since the stress is pretty low I'm not real concerned.
I'm still thinking that washers are going to hold till the rope breaks
OK, got all the theory about shear, threads, etc.
Do we know, or know how to calculate, the actual loads that a cleat will have to withstand? Probably pretty hard!
My thought is that I'd want to break my line before the deck or cleat tore out. Given that, add a safety factor of x.xx to feel comfortable about any assumptions you made and the go design the thing. From WM I got the following plated nylon data:
1/2" 7500 lb breaking strength
9/16" 9200 lb
5/8" 10400 lb
3/4" 15000 lb
since they specified breaking strength I probably don't need a huge safety factor and would guestimate 1.1 would be fine.
so my 5/8" anchor/drogue rode would part BEFORE the cleat got to 10400*1.1=11440 lb. so build the structure to handle that. Not much sense building it stronger than the rope it attaches to right?
50k psi materials will need 11440/50k=0.2288 square inches of area to withstand that
19k psi materials would need 0.602 square inches of area
I'm thinking that a washer would be fine provided the structure did not fail in another mode. The other mode being a rotten deck or thin FRG layup.
If someone has the strength of fiberglass I'll do the calc for the area you need to supply (via backing plate load spreader). Id assume that the core does not contribute to the holding in place and just increases the torque that the structure can handle.
Not sure why I'd need the cleat to be stronger than the line except to account for shock or fatigue loading. In which case I'd use 2.0 safety factor and go with 0.45 and 1.2 square inches respectively. Still pretty small IMHO
Other failure modes:
My engineer sense tells me that a deck cleat can fail in the following modes;
water runs out at the dock and the cleat is pulled up
water comes in at the dock and the cleat is pulled sideways and/or downwardish but mostly sideways
the wind / waves hit the boat and the cleat is pulled sideways
Cool! all three modes so we get to earn our keep.
Pulling up mode is either threads stripping, cleat bending, or deck failing in shear
Pulling down/sideways is the bolts bending by toque and or shearing or the deck failing in torque
Pulling sideways is the toe rail plus toe rail bolts plus the deck failing in tension or the bolts shearing off.
since we already know it does not take much surface area for a metal to develop these kinds of forces and I don’t know fiberglass's strength lets make some assumptions about the structure and see how highly stressed it would be and "get a feel" for it.
Assume the sideways load is max at 11440 lb and I provided ample metal to resist shearing off the bolts or bending them off. The failure mode is then pulling a (assuming here) rectangular shaped piece of deck sideways off the boat. If the deck core is 0.5" thick and the glass is 1/16" thick on both sides and the cleat is 4" long and 2" from the edge and there is no metal toe rail (glass takes all the load) we get:
length of the tear resisting area PSI on the glass
4+2+2" 2*(1/16*8)=1.0 11440/1.0=11440 psi
if it rips right out at the cleat and takes the deck to the toe rail with it, it could be made of wood veneer and still hold (wood typically is ~15k psi in tension). FTR I ignored that fact that the two 2" tears to the toe rail would be in shear but since the stress is pretty low I'm not real concerned.
I'm still thinking that washers are going to hold till the rope breaks
Engineering 101
If the group is interested, I'd be willing to show how to determine the stresses on a generic structure. It is pretty easy math and all you really have to understand is how to come up with the area that has to resist the load and use some common sense about how the load is going to get from point A to point B.
If the group is interested, I'd be willing to show how to determine the stresses on a generic structure. It is pretty easy math and all you really have to understand is how to come up with the area that has to resist the load and use some common sense about how the load is going to get from point A to point B.
Mr. Roosa,
Thank you for the contribution to the thread. From my experience with bolts and threads and study of design characteristics I think a simple solution is to design the structure so it is stronger than the bolts. You mention the threads stripping as a failure mode. As long as about 4 threads are engaged the more likely failure mode to the bolt would be shear of the bolt or tension failure of the bolt, another possibility would be shear of the flat head from the bolt. With a fairly rigid washer of reasonable thickness the shear strength of the deck & core assembly for the slug that would be removed by the washer needs to be stronger than the bolt.
Mr. SailingDog,
Thank you for the correction about the area.
I wouldn't say the cleat is being pushed downward however. Rather it is being sideloaded with a force vector that has an angular component in the negative Y direction. This is putting a bending moment on the base of the cleat. I think Scott said it was a two bolt cleat so one side of the base is pushing down on the deck while the other side is pulling up. In short the screws are being bent so the nut is digging into one side of the washer while the other side is unloaded. This concentrated load, coupled with a possible wet core could explain the catastrophic failure. Using a four screw cleat as Scott did with his replacement provides a much larger base to resist this loading as the reaction moment on the base resisting the bending moment on the cleat will be much better. I am concerned as on my O'day my cleats are all of the two screw variety and possibly the hollow aluminum version that Maine Sail showed fractured. Pearson and O'Day were pretty close companions just before they both went backrupt. At one point Pearson owned O'Day, O'Day also ran Cal for a while and the last few years of the Cal brand looked just like an O'Day, but my year was still under the Bangor Punta banner along with Smith and Wesson. My stern cleats are arranged in the direction of loading however so that is better (lines apply loads in direction of the long axis of the cleat) for stress distribution.
Thank you for the contribution to the thread. From my experience with bolts and threads and study of design characteristics I think a simple solution is to design the structure so it is stronger than the bolts. You mention the threads stripping as a failure mode. As long as about 4 threads are engaged the more likely failure mode to the bolt would be shear of the bolt or tension failure of the bolt, another possibility would be shear of the flat head from the bolt. With a fairly rigid washer of reasonable thickness the shear strength of the deck & core assembly for the slug that would be removed by the washer needs to be stronger than the bolt.
Mr. SailingDog,
Thank you for the correction about the area.
I wouldn't say the cleat is being pushed downward however. Rather it is being sideloaded with a force vector that has an angular component in the negative Y direction. This is putting a bending moment on the base of the cleat. I think Scott said it was a two bolt cleat so one side of the base is pushing down on the deck while the other side is pulling up. In short the screws are being bent so the nut is digging into one side of the washer while the other side is unloaded. This concentrated load, coupled with a possible wet core could explain the catastrophic failure. Using a four screw cleat as Scott did with his replacement provides a much larger base to resist this loading as the reaction moment on the base resisting the bending moment on the cleat will be much better. I am concerned as on my O'day my cleats are all of the two screw variety and possibly the hollow aluminum version that Maine Sail showed fractured. Pearson and O'Day were pretty close companions just before they both went backrupt. At one point Pearson owned O'Day, O'Day also ran Cal for a while and the last few years of the Cal brand looked just like an O'Day, but my year was still under the Bangor Punta banner along with Smith and Wesson. My stern cleats are arranged in the direction of loading however so that is better (lines apply loads in direction of the long axis of the cleat) for stress distribution.
- Status