toad

Retrofit Composite Chainplates

Recommended Posts

Anyone have any advice or guides on how to do this? Looking at removing my glassed in SS ones to check for crevice corrosion and just considering replacing them with something newfangled. Have had an opinion that its better to build these into the layup rather than glue onto the top laminate.

Share this post


Link to post
Share on other sites

Unless it’s a very small boat they definitely need to be made in situ.  It is possible with the deck on but very fiddly 

Share this post


Link to post
Share on other sites

Well I sort of disagree that they have to be built into the laminate.

On my boat I made some very long (4'?) carbon chainplates on a bench. (actually since I needed 4 chainplates I made up a 12" wide laminate then cut it into 3" wide chainplates) Bonded them to the main mast beam but well after it was built. Then glassed over them.

They were for 5/16" 1x19.

If I was doing this I'd want to know that I could laminate carbon very effectively.

If I was doing this I might make a splash of the outside of the hull, make up a thinner glass test chainplate, and see how it fit to the inside laminate.

I'd also reinforce the area of the hull with more layers of glass.

Or are you bonding them to a transverse bulkhead?

Share this post


Link to post
Share on other sites

 I would make up a strip the same shape as the hullwith a glass tube on top for the pin that is tapered to nothing about 12" long. Round the corners on one side. On the bench Lay your carbon unis over the top of your section straight and at 5,10 and 15 degree angles about 6' long, so 3 each side then put a smear of thickened epoxy over the back side of the section, then slide the whole slippery mess through the deck from the bottom. Prefrebly vacuum bag it to the hull, or roll the air out by hand. More carbon or glass around the deck join and all good. This is how I did the bow chainplate on my boat and the 6 on my last boat, 34'. 

Share this post


Link to post
Share on other sites

sounds good, any photos? I cant quite picture how it looks

Share this post


Link to post
Share on other sites

It doesn't matter if you fabricate them on a bench and bond them into the hull. In fact you could vacbag and cure them out the boat. Whereas you only get to vacbag in the boat. The reality is you will put a ton of extra material regardless of how you do it. Some people do a combination. Make up your turnbuckle attachment on the bench then bond it to the hull. Then add on layers to distribute the forces further out into the hull and down to the mast step.

Share this post


Link to post
Share on other sites

Kurt Hughes has an interesting entry on his blob about how carbon chainplates are dumb because if you calculate the stress you find that 1.8 mm of carbon will do the job, but that would freak everybody out. So you make it super thick to make it "look right". And then you could do it all in uni E-glass because once you get it strong enough looking.

He's got a point but carbon uni has such a huge safety factor if you make it look right.

Share this post


Link to post
Share on other sites
6 minutes ago, Zonker said:

Kurt Hughes has an interesting entry on his blob about how carbon chainplates are dumb because if you calculate the stress you find that 1.8 mm of carbon will do the job, but that would freak everybody out. So you make it super thick to make it "look right". And then you could do it all in uni E-glass because once you get it strong enough looking.

He's got a point but carbon uni has such a huge safety factor if you make it look right.

That's pretty astute. We have 3mm SK99 chainplate ties on the Int 14, (Think an epoxy bound lashing) which is 1150kg breaking strength - per leg of lashing! He couldn't deal with just one wrap, so we went with 2. So 4 legs. Maybe stronger than the 1/4" stainless dog-bone...

4600kg breaking strength on a 150 pound boat....  and the damn chainplate is tiny and STILL looks undersized...  Pretty easy to make shit hella-strong without much cost of $s or weight with these materials...

 

 

Share this post


Link to post
Share on other sites
1 hour ago, Zonker said:

Kurt Hughes has an interesting entry on his blob about how carbon chainplates are dumb because if you calculate the stress you find that 1.8 mm of carbon will do the job, but that would freak everybody out. So you make it super thick to make it "look right". And then you could do it all in uni E-glass because once you get it strong enough looking.

He's got a point but carbon uni has such a huge safety factor if you make it look right.

Further to what Hughes said, I can't figure out why everyone is so hung up on carbon.  If strength rather than stiffness is the desired goal, UD E-glass is almost as strong as carbon. https://www.christinedemerchant.com/carbon-kevlar-glass-comparison.html One of the things working in favour of E-Glass is that it is a better match with epoxy than carbon - in the sense of matching elongation to failure.  Quoting from this article: "...In the end if Tensile Strength is your only concern, save your money and use e glass."

Share this post


Link to post
Share on other sites

The other thing to consider is that it is not a stand alone piece of kit. Its bonded directly to the hull sides. The load is spread across a large amount of an existing strong & stiff surface. Delam shouldn't be a concern because of  the bonding surface area. The part is in almost pure tension, therefore carbon is a waste of money.

I always liked using s-glass for these kind of structures or reinforcements. Unfortunately its been made obsolete by carbon and the availability and cost makes it a poor choice.

  • Like 1

Share this post


Link to post
Share on other sites

The chainplate is in tension but the bond to the hull or bulkhead is primarily in shear.  IIRC you use 15 MPa (roughly 2200 psi) as the shear strength for epoxy laminates.  So in imperial units, the bonding area should be at least equal to total shroud tensile strength x an appropriate SF / 2200. 

For example if total shroud tensile strength is 15,000 lb and a chosen SF of 3, then bonding area is 15000 x 3/2200 equals about 20 sq inches.  Not much really.  If wrapped around hull or bulkhead that is 10 sq inches per side.

Share this post


Link to post
Share on other sites

The first composite chainplates I did see early nineties on minis, it was done like hughes said, a few mm wraps over foam, scary looking, but it worked well.

Share this post


Link to post
Share on other sites

Don't forget he's bonding to existing e-glass laminate that was probably built with polyester resins. So be a bit conservative with the lap shear strength. So double it 40 sq in. About 2 sides x 2" x 10" long. Still plenty small isn't it :)

Share this post


Link to post
Share on other sites

Its surprising how little bond area need when you get someone to do the calculations. I remember when we were building the 40' race boat. We had the boys from NEB help us do the hull layup. We got to use their impregnator.  My friend who designed and built the boat is a composite/structural engineer. He calculated that including safety factor you need 2.5 inches of overlap between each roll of tri-ax when using Proset epoxy. The NEB guys had always done  6 inch laps. They built most of Star's and Stripes as well as a slew of other big name racing boats.

Imagine the weight of resin and glass of an extra 3.5 inches every 36" across the hull all the way along a 80' boat. Keep in mind the same rule applies every time you make a lap everywhere on the boat.

  • Like 1

Share this post


Link to post
Share on other sites
On 9/10/2018 at 6:40 PM, 12 metre said:

Further to what Hughes said, I can't figure out why everyone is so hung up on carbon.  If strength rather than stiffness is the desired goal, UD E-glass is almost as strong as carbon. https://www.christinedemerchant.com/carbon-kevlar-glass-comparison.html One of the things working in favour of E-Glass is that it is a better match with epoxy than carbon - in the sense of matching elongation to failure.  Quoting from this article: "...In the end if Tensile Strength is your only concern, save your money and use e glass."

This one has my brain spinning due to the inaccuracies in the statement!!!!!!

First of all, carbon and high grade epoxy have very similar elongation before failure, so that statement is pretty false. Its when you start trying to pair materials with subpar resins that this becomes an issue. Generally speaking polyester is crap (and in this particular case for a secondary bond SHOULD NOT BE USED). Good vinylesters and epoxies perform virtually identically; epoxy has the advantage of near infinite shelf life and the ability to cure in less than ideal conditions. Vinylester is less expensive and has the ability to cure at different rates depending on mix ratio so is a good choice in a production environment.

Second of all, GRP type composite structures are almost entirely deflection driven structures, not strength driven. As the article states, glass is strong enough in tension, but there is compression to deal with and carbon is much better there. Further, carbon on average has an elastic modulus that is 4x greater than glass. Effectively you need 4x less carbon than glass to fabricate a panel with the same deflection. In practice its not quite as good as epoxy makes up >40% of the laminate, so we generally say half, i.e 80gsm of carbon has equivalent stiffness to a layer of 160gsm glass when laminated. This stiffness issue is why you see modern race boats and even performance cruisers using a high amount of carbon in their structures. Often these are sized based on flexural requirements with the exception of say the beams on a performance cat which are typically designed in modern FEA programs and their structure iterated. Still, due to the principals of beam bending, a high modulus carbon beam will way 1/3rd less than the equivalent standard modulus carbon beam, on average, for the same strength and stiffness.

The trade off to carbon is all this stiffness means the structure can be a tad brittle. The other trade off when building high performance composite structures is you really want the fibers aligned with the loads; most folks quickly get into trouble as its not always easy to visualize the load path and often you have incorrect fiber orientation which quickly reduces the strength of the structure by an order of magnitude (check uni carbon in the 90 degree plane, its basically useless).

 

Share this post


Link to post
Share on other sites
On 9/10/2018 at 5:58 PM, Raz'r said:

That's pretty astute. We have 3mm SK99 chainplate ties on the Int 14, (Think an epoxy bound lashing) which is 1150kg breaking strength - per leg of lashing! He couldn't deal with just one wrap, so we went with 2. So 4 legs. Maybe stronger than the 1/4" stainless dog-bone...

4600kg breaking strength on a 150 pound boat....  and the damn chainplate is tiny and STILL looks undersized...  Pretty easy to make shit hella-strong without much cost of $s or weight with these materials...

 

 

Do you have a photo? I'm curious as this sounds like a good approach to chainplates on A class catamarans. Downside is when the SK99 chafes how do you replace the chainplate?

Share this post


Link to post
Share on other sites
1 hour ago, samc99us said:

Do you have a photo? I'm curious as this sounds like a good approach to chainplates on A class catamarans. Downside is when the SK99 chafes how do you replace the chainplate?

I can get one this weekend.

We waxed up the through-beam tubes so we shouldn't have any issue pulling out the sk99 if/when it chafes.

 

Top of beam: Ferrule

through beam: little carbon tubes just big enough for 2 passes of the sk99 on each side

bottom of beam, extra carbon laminate and a 1/4" stainless dog-bone

 

 

Share this post


Link to post
Share on other sites
1 hour ago, samc99us said:

This one has my brain spinning due to the inaccuracies in the statement!!!!!!

First of all, carbon and high grade epoxy have very similar elongation before failure, so that statement is pretty false. Its when you start trying to pair materials with subpar resins that this becomes an issue. Generally speaking polyester is crap (and in this particular case for a secondary bond SHOULD NOT BE USED). Good vinylesters and epoxies perform virtually identically; epoxy has the advantage of near infinite shelf life and the ability to cure in less than ideal conditions. Vinylester is less expensive and has the ability to cure at different rates depending on mix ratio so is a good choice in a production environment.

Second of all, GRP type composite structures are almost entirely deflection driven structures, not strength driven. As the article states, glass is strong enough in tension, but there is compression to deal with and carbon is much better there. Further, carbon on average has an elastic modulus that is 4x greater than glass. Effectively you need 4x less carbon than glass to fabricate a panel with the same deflection. In practice its not quite as good as epoxy makes up >40% of the laminate, so we generally say half, i.e 80gsm of carbon has equivalent stiffness to a layer of 160gsm glass when laminated. This stiffness issue is why you see modern race boats and even performance cruisers using a high amount of carbon in their structures. Often these are sized based on flexural requirements with the exception of say the beams on a performance cat which are typically designed in modern FEA programs and their structure iterated. Still, due to the principals of beam bending, a high modulus carbon beam will way 1/3rd less than the equivalent standard modulus carbon beam, on average, for the same strength and stiffness.

The trade off to carbon is all this stiffness means the structure can be a tad brittle. The other trade off when building high performance composite structures is you really want the fibers aligned with the loads; most folks quickly get into trouble as its not always easy to visualize the load path and often you have incorrect fiber orientation which quickly reduces the strength of the structure by an order of magnitude (check uni carbon in the 90 degree plane, its basically useless).

 

Referencing this: https://epoxyworks.com/index.php/if-you-were-just-more-flexible/

Typical strain to failure for CF ranges from 0.5% to 2.0%  High modulus CF being at the lower end.

Typical strain to failure for WEST System epoxy is 3.5% to 4.5%

Typical strain to failure for e-glass is 3.5% to 4.8%  

Agree that as long as the resin has higher strain to failure than the fibre, it is okay.  However, with epoxy/e-glass, the composite can strain more prior to failure, allowing the epoxy to take up more of the load.  Admittedly this is a small amount, but it does help.

I should have been more specific by saying " If tensile strength rather than stiffness is the desired goal, UD E-glass is almost as strong as carbon", but I thought one would have figured that out by my quote from the article.

Nowhere did I say e-glass was almost as good in flexure, but again I thought one would have figured that out when I said "...strength rather than stiffness" since flexure and stiffness go hand in hand.

My main point was not to rip on carbon but that some people spend a lot of money fabricating carbon fibre parts when e-glass may be just as suitable. 

Share this post


Link to post
Share on other sites

My composite/structural engineer friend is a big fan of glass. He adamantly believes that huge amounts of carbon fibre are wasted in many applications. Its well known that the tensile strengths of carbon & glass are effectively the same. It is well known that the compressive strength of glass is at least 2.5 times less than carbon. This would be on paper in the lab pulling/pushing unidirectional fibres.

Proper design can make the compressive strength nearly irrelevant. If you are making a composite I-beam the fibres on one side are always in tension the other side in compression. The tensile strengths are so much higher than the compressive strengths that the tensile is the limiting factor. Therefore you could use either carbon or glass. The grid in the bottom of the 40' race boat was made of glass, because it is essentially a bunch of I & C-beams bonded together. 

My father and friend are both engineers who worked at EB designing submarines. Jon was the project manager on the largest composite structure in the world back in the 2000's. Its the bow dome of a Seawolf class submarine. 16" of solid fibreglass. Its got to be transparent to the sonar.

I've learned a ton from them over the years. Its always a good idea to run some of these questions by someone who knows the math & physics. 

Share this post


Link to post
Share on other sites
15 hours ago, CaptainAhab said:

The tensile strengths are so much higher than the compressive strengths that the tensile is the limiting factor

I guess you made a typo, what you're saying is correct, but it should say that the tensile strength is lower

Share this post


Link to post
Share on other sites
On 9/17/2018 at 7:20 AM, CaptainAhab said:

Its surprising how little bond area need when you get someone to do the calculations. I remember when we were building the 40' race boat. We had the boys from NEB help us do the hull layup. We got to use their impregnator.  My friend who designed and built the boat is a composite/structural engineer. He calculated that including safety factor you need 2.5 inches of overlap between each roll of tri-ax when using Proset epoxy. The NEB guys had always done  6 inch laps. They built most of Star's and Stripes as well as a slew of other big name racing boats.

Imagine the weight of resin and glass of an extra 3.5 inches every 36" across the hull all the way along a 80' boat. Keep in mind the same rule applies every time you make a lap everywhere on the boat.

6 inch laps is a huge amount!!! what weight of material was being used? 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now