Windlass won't feed splice, tips?

[lancelot9898]

I'm using an eye splice on the chain/rope interface. While this makes for a larger splice and does not help in getting the splice over the gypsy, I think is provides better strength vs the common rope chain splice. I'm able to get the entire 5/8 inch 3 strand through the chain length without undoing the strands. After getting the rope through the chain, I then take apart the strands and start weaving. I am using 3/8 inch BBB chain which has a large enough end length to allow the 5/8 inch line to pass. FWIW

[waterman46]

I saw the diagram in the beginning of this thread where only two strands are run through the chain link. Did I understand this correctly? Why wouldn't three strands be better than two? The third strand can be run the other way through the chain link then the three strands are woven back into the rope just like a standard end splice.

[estarzinger]

^^ Using 2 strands is 'better' primarily because it makes a smaller more flexible splice that feeds thru rope/chain gypsy's better.

Per my testing, both the '2 strand' and '3 strand' splices, if properly done, are full strength. Brion Toss says in his testing the 2 strand was stronger than the 3 strand - but I believe my testing is better

you will ask how can 2 strands be as stronger (or even stronger than) 3 strand?

The answer is a combination of two factors:
(1) loop splices are 'double strength', because (for instance) the '2 strand splice' has 4 legs (strands) holding the load (2 coming and two going - and yes that doubles the strength). Now the base rope has only 3 legs (strands) holding the load - so the splice is at 'full strength' (eg the rope will break at the base of the splice taper not at the splice (and same with the '3 strand splice' which has 6 legs).

This is complicated by
(2) Bend radius - ropes are weakened by bend radius. By how much depends on the relative diameter of the bend and the 'effective bent rope' size. So, for a start here the tests assume a typical chain to rope size ratio - the strength of the splice would be lower if you used really really tiny diameter rod in your chain, and in that case neither of these splices would be full strength. But for normal chain/rope combinations that is not a big factor. The difference between the 2 strand and 3 strand is subtle, and is contained in the word "effective bent rope size". The 'effective; size is NOT the rope size, but the diameter of how it is as it bends around the link. The 2 strand splice has a bit more room to spread out and the 'effective' size is actually rather less than a strand diameter - close to a filament diameter, while the 3 strand is jammed in there and the effective size is close to strand size. This means that the 3 strand is hurt more by bend radius than the 2 strand.

That's probably complex to process in words - but be assured good tests suggest the 2 strand is just fine.

I personally used the 3 strand splice, because it worked on my gypsy and it was easier to make. If I had more skill with the 2 strand I would have used it, but I generally sucked at making it.

[thinwater]

Originally Posted by lancelot9898

I'm using an eye splice on the chain/rope interface. While this makes for a larger splice and does not help in getting the splice over the gypsy, I think is provides better strength vs the common rope chain splice. I'm able to get the entire 5/8 inch 3 strand through the chain length without undoing the strands. After getting the rope through the chain, I then take apart the strands and start weaving. I am using 3/8 inch BBB chain which has a large enough end length to allow the 5/8 inch line to pass. FWIW

This is a common misunderstanding. However, there is no testing that suggests it is true.

1. The number of strands going around the link is the same: three.
2. The bend radius of the strands is better when the line is unlaid that when it is whole. The only way to achieve this with a whole rope is to use a thimble and a shackle.
3. The number of strands approaching the last link is the same in both cases: three. In fact, four have proven to be enough.
4. Chafe is reduced, since the link splice stops motion at the link.

I do not believe there is any controversy in this. Yes, it seems counter intuitive, but not after you break down the engineering and do the tests. It is simply a better way, and no more complicated.


You are free to do what you like, but it is worth clarifying that the an eye splice with no thimble is actually slightly weaker because of the horrible bend radius.

[lancelot9898]

Good discussion about the bend radius and etc. Maybe Mainesail can do some testing if there are no tests available between 1. an eye splice with thimble 2. an eye splice without thimble 3. two strand chain splice 4. three strand chain splice ....and denote where failure is occurring.

[UNCIVILIZED]

Originally Posted by lancelot9898

Good discussion about the bend radius and etc. Maybe Mainesail can do some testing if there are no tests available between 1. an eye splice with thimble 2. an eye splice without thimble 3. two strand chain splice 4. three strand chain splice ....and denote where failure is occurring.

I think that Evans has done/has access to loads of data on this kinds of stuff. Some of it's been posted before regarding bend radii & Spectra/Dyneema.

[estarzinger]

^^ Yea, I and Brian Toss and ThinWater have done such testing. There has also been testing done outside the 'yachting' world - in commercial marine and heavy lifting and military.

There are small variations in the test results - due (I believe) to different chain configurations and different skill with the splices, but overall the big results are identical.

#1 done correctly all of them are 100% strength or nearabouts. Brian is a bit of an outlier as he believes the 3 strand splice(s) is only 85% (ish) strength and the others are 100%.
#2 Failure with a thimble is at the base of the splice taper (or sometimes with a perfect taper, just randomly somewhere in the rope). Failure without a thimble (all three styles of splice) is usually at the taper base also, but also sometimes at the bend radius. I believe this has to do with how the individual splices were made (even tension and spreading the bend radius out well).

The general conclusions among the 'experts' is that: (a) if you can, use a thimble and shackle, but that will (usually) not run thru a windless, so (b) otherwise there is not a lot between these options, with some people favoring one and some favoring another.

Mainesail does terrific work, and has the equipment to do this sort of thing, but I think generally he has not really taken on rope work as an area.

[dakno]

Video of anchor splice . www.animatedknots.com/chainsplice/#ScrollPoint.