Possibly Original Thought About Chain Catenary - or - The Myth of the Bar Tight Chain

[thinwater]

^^ I like the approach!

[Dockhead]

Originally Posted by Kelkara

I still think I'm calculating it right.

This graph shows the same 100m of 3.3Kg/m chain in 30m of water (3.3:1 scope) with a horizontal pull of 500, 1000, 2000, 4000 and 8000 Kg horizontal force (there will also be 330Kg vertical force). The chain will take the shape of the curve between the dots. 510Kg is the point at which the last link lifts off the bottom and starts to pull up on the anchor, at which point there is only 1.5m of "slack" left in the catenary for snubbing.

Also because I was interested. for the same 100m of chain I plotted the angle of pull at the anchor after the last link starts to lift for 3:1, 4:1, 5:1 and 7:1 scope.

I also plotted how straight the chain is in the same situation ... This seems to be predominately controlled by the force, and not the scope ... these graphs are only for once the anchor starts lifting, and there is essentially no "snub" left in the catenary.

What a great thread this has turned out to be! There are more interesting offshoots from this idea than I imagined.


I think this is the data I wanted, and if it's correct, then it sort of disproves my idea, for these parameters (I suppose that for 200 meters of 15mm chain, what I was guessing will be true).


Your data does show that catenary is still working a lot, at 2 tonnes of force, reducing the pulling angle on the anchor from 19.5 degrees to 13 degrees, which is what you would get with scope of 1:4.5 with rope rode or mixed rode with small amount of chain -- still a decently effective scope for a good and well set anchor.


I started this to try to explain the observed behavior of my own ground tackle, and so this fact alone - assuming your calculations are right, and I have no reason to doubt them -- would seem to more or less do that. ABYC table suggests that I would see about 1.5 tonnes of load in 42 knot of wind, so it is easy to see that according to your data I would be reasonably secure at 3:1 in something more than 42 knots of wind if the anchor is well set.


The snubbing value of the remaining catenary is harder to evaluate. I would have expected a much greater sag than you have calculated, but it may be non-obvious how a small amount of sag of heavy chain works to absorb energy. Your last table seems to indicate, surprisingly, that scope does not much effect the snubbing effect of catenary, and that the curve starts to fall over for all scopes at around 1.5 tonnes of force, so, 42 knots of wind. Does that fallover point correspond to where energy stops being absorbed effectively? It's not completely obvious, but I guess that is probably the case, even though we seem to be talking about only centimeters of sag. So I guess it is right that I rig a snubber when the wind gets up, even in deep water and with all the chain out.



As Thinwater said, we also have to be careful not to look at this as a static system -- there are many dynamic things going on. But I think what we are looking for here is whether there is any energy absorbing capacity left at 42 knots of wind -- looks like it is just petering out by that point, and is gone by 2 tonnes, which is the WLL of the chain. So a snubber is definitely needed.


The same graph would explain another thing, namely why in 20 knots of wind on this boat I have no need of a snubber at all, with the catenary able to absorb huge amounts of energy, whereas with my previous boat and much lighter chain, I would smash the bow roller to pieces if I failed to rig a snubber by 20 knots of wind.




Very helpful and very interesting!! Great work!!

[conachair]

Originally Posted by Dockhead

But I think what we are looking for here is whether there is any energy absorbing capacity left at 42 knots of wind

If the anchor stays stuck then where else will the energy of a moving boat go to stop it moving?

Might make more sense to think of how far it takes for the energy of the moving boat to get transferred and the boat to stop, double the distance, half the force.

[Moo]

Originally Posted by Dockhead

What a great thread this has turned out to be! There are more interesting offshoots from this idea than I imagined.


I think this is the data I wanted, and if it's correct, then it sort of disproves my idea, for these parameters (I suppose that for 200 meters of 15mm chain, what I was guessing will be true).


Your data does show that catenary is still working a lot, at 2 tonnes of force, reducing the pulling angle on the anchor from 19.5 degrees to 13 degrees, which is what you would get with scope of 1:4.5 with rope rode or mixed rode with small amount of chain -- still a decently effective scope for a good and well set anchor.


I started this to try to explain the observed behavior of my own ground tackle, and so this fact alone - assuming your calculations are right, and I have no reason to doubt them -- would seem to more or less do that. ABYC table suggests that I would see about 1.5 tonnes of load in 42 knot of wind, so it is easy to see that according to your data I would be reasonably secure at 3:1 in something more than 42 knots of wind if the anchor is well set.


The snubbing value of the remaining catenary is harder to evaluate. I would have expected a much greater sag than you have calculated, but it may be non-obvious how a small amount of sag of heavy chain works to absorb energy. Your last table seems to indicate, surprisingly, that scope does not much effect the snubbing effect of catenary, and that the curve starts to fall over for all scopes at around 1.5 tonnes of force, so, 42 knots of wind. Does that fallover point correspond to where energy stops being absorbed effectively? It's not completely obvious, but I guess that is probably the case, even though we seem to be talking about only centimeters of sag. So I guess it is right that I rig a snubber when the wind gets up, even in deep water and with all the chain out.



As Thinwater said, we also have to be careful not to look at this as a static system -- there are many dynamic things going on. But I think what we are looking for here is whether there is any energy absorbing capacity left at 42 knots of wind -- looks like it is just petering out by that point, and is gone by 2 tonnes, which is the WLL of the chain. So a snubber is definitely needed.


The same graph would explain another thing, namely why in 20 knots of wind on this boat I have no need of a snubber at all, with the catenary able to absorb huge amounts of energy, whereas with my previous boat and much lighter chain, I would smash the bow roller to pieces if I failed to rig a snubber by 20 knots of wind.




Very helpful and very interesting!! Great work!!

Interesting thread. All I can say is that when anchoring in high winds at a point somewhere between 40 and 50 knots the chain appears to become bar-tight whether it is 6mm chain on a 30 footer or 18" links on a ship.

It looks and feels like a bar. On a yacht you can actually feel the boat move back and forth as the waves pass under as the geometry of the straight line describes its arc.

[Bruce K]

Sort of off topic, but how about a Dyneema rode with a sinking surgical type rubber core? This would give strength, stretch (Dyneema would compress the rubber as it is pulled tight), lightness, and should not float if rubber is dense enough. Is this a silly idea?

[wsmurdoch]

Alain's site looks into some of this both static and dynamic. His spreadsheets are back now. Tuning an Anchor Rode

One thing missing is the dampening effect of nylon.

Bill

[valhalla360]

"Bar Tight" is kind of like "Hull Speed" In reality, the definition doesn't allow for calculating an exact value.

There is always some cantenary no matter how tight you make it but at some point, it's small enough to be irrelevant just like at some point it's silly to try and make a displacement boat go faster by throwing more HP at it but it will go a little faster for every extra HP.

The link below is to a spreadsheet I developed many years ago for estimating cantenary for traffic signals. With some updates to the assumptions, it would be easy enough to calculate anchor chain cantenary. You could even add a kellet if you like to see the effect (put it in as a custom signal head of whatever weight you like at whatever point you like). Some thoughts on making it work:
- Set the "POCH" height to zero for the anchor end and anchor roller height for the boat end above bottom.
- Update the weight per ft based on the chain weight...but discount it for buoyancy provided by the water (it you want to be really accurate, figure out the point where it meets the water and use different values in and out of the water).
- It does model it as 1ft links, which is probably close enough but you could reconfigure it to be based on the length of chain per link (which is less than the length of a link since they overlap).

https://www.michigan.gov/mdot/0,4616...628--F,00.html


PS: If you really want to get fancy, you could use the 3way tie off part to estimate for multiple anchors.

[wsmurdoch]

Originally Posted by Paul L

What load cell are you using? How is it rigged, to the chain, to the snubber?

I bought this 0-2000 lbf dynamometer for a bargain price on ebay a few years ago. I have found it useful for quantifying anchor loads, measuring the boat's reverse thrust, measuring genoa sheet loads, and other things. I've loaned it to others to do the same. It has been a useful toy.

Unlike Thinwater's load cells, it does not give an electronic and remote reading, and it is not waterproof.



Bill

[groundtackle]

For added entertainment we could introduce this factor;

5.3.3 Elongation
5.3.3.1 All chain which is heat treated must be in the quenched and tempered condition
before elongation is measured.
5.3.3.2 The elongation test specimen shall consist of a length from the lot containing at least
the number of links in the Table in Section 5.3.1.
5.3.3.3 A positive load not exceeding 10% of the proof test shall be applied for determining
the original gauge length (LO).
5.3.3.4 The elongation shall be based on the total extension at fracture. This is expressed
as a percentage of the change in length (∆L) divided by the original gauge length
(LO). The elongation may be determined by the equation below or by autographic
recorder or side scale.
Elongation (%) = {∆L / LO}x 100
Where
∆L = The test specimen final length at fracture - the test specimen original gauge
length (LO).
LO = The original gauge length (the sum of the inside lengths of the test chain links
not counting the fixture links, or as determined in 5.3.3.3).
5.3.4 Elongation shall not be less than 20% for Grades 80 and 100; 15% for Grades 30, 43, 70, and
Stainless; and 10% for Machine, Coil, and Passing Link chain

Chris

[s/v Jedi]

Was the weight reduction for being immersed in water being used for the weight of the chain?

[MartinR]

Originally Posted by Dockhead

This is the curve calculated by Peter Smith, for 48 meters of 12mm chain, under 2 tonnes of force (about 50 knots of wind on a certain size boat) and in only 8 meters of water:


Attachment 187771


That's the "bar tight effect", which I think clearly exists under these conditions.



But what does this curve look like with 100 meters of chain and 30 meters of depth? The sag goes up with the square of the length of the cable, all other things being equal, so length changes a lot.

There is plenty of useful catenary left in this graph. Effecting the angle of attack by the anchor a few degrees and giving some damping.

To get all the sag out of a chain you need need infinite force. Hard to apply in practice

[thinwater]

ABYC table 1 does NOT related to wind load, and only to rode tension in very specific circumstances (shallow water, long fetch, all chain, and no snubber). It is worst case.



Your measured tension will be 2-5 times less, depending on the set-up and situation. Many testers have reported this.

[thinwater]

Originally Posted by Bruce K

Sort of off topic, but how about a Dyneema rode with a sinking surgical type rubber core? This would give strength, stretch (Dyneema would compress the rubber as it is pulled tight), lightness, and should not float if rubber is dense enough. Is this a silly idea?

No advantage over polyester. In side-to-side chafe testing on rock, single braid Dyneema is about the same as polyester. Probably not much more stretch.



By the way, some mooring systems (Stormsoft) take a related approach.

[thinwater]

Originally Posted by wsmurdoch

I bought this 0-2000 lbf dynamometer for a bargain price on ebay a few years ago. I have found it useful for quantifying anchor loads, measuring the boat's reverse thrust, measuring genoa sheet loads, and other things. I've loaned it to others to do the same. It has been a useful toy.

Unlike Thinwater's load cells, it does not give an electronic and remote reading, and it is not waterproof.



Bill

My first one was just like that, and rigged much like that. Very useful. In truth, there is so much fluctuation you don't need to fret over 5-digit accuracy!


The other thing that is interesting to test, is the effect of sailing-at-anchor on the load. With some boats, it can double when the boat gets somewhat broadside.


I like the bar-tight vs. hull speed analogy. Both suggest an important principle, but neither is a wall. For example, bar-tight is real for a cat in shallow water (boat can move fast and really snatch hard... one more reason to use a nylon bridle), and hull speed is less abrupt in a cat.

[conachair]

Originally Posted by s/v Jedi

Was the weight reduction for being immersed in water being used for the weight of the chain?

On my link, yes, factored in. Something like 0.78 for "generic" seawater from memory.