Catenary - would you buy krypton piano wire?

[AntarcticPilot]

I thought the catenary debate arose because it is now fairly widely accepted that under the tensions experienced in relatively gentle tide/sea states, an inelastic chain cable is drawn taught between the boat and the hook...so the idea that its weight helps prevent snubbing, ceases to apply.

In fact with no elasticity at all, it's more likely to break out the hook, than a rope rode...so sections of stretchy line need adding to the chain, in effect to replicate the catenary, which alone isn't up to the job.

My point is that it isn't a matter of opinion, but (relatively) straightforward analysis that many people on this forum could handle easily.
One point that people miss is that a chain ALWAYS describes a catenary curve. It may be a very flat catenary, but it is never dead straight. Further, the maximum curvature of the catenary is at neither the surface nor the anchor, and is therefore not observable. Whether the catenary is sufficient to allow damping of shock forces is another matter, but it is something susceptible to relatively straightforward mathematical analysis. There are several people on the forum who could carry out such an analysis; not me, I'm afraid - although I knew the maths existed, it would require too much revision of the relevant techniques for me. But even with my rusty skills, I can follow the maths in the Wikipedia article! But I can think of several forumites who could carry out the analysis, if they wished. I could - but it would take me too long to recover the skills I once had!

 

[Greenheart]

I'd trust your view (and your maths) far sooner than my own, A.P. :encouragement:

But whatever the absolutes and theoreticals, metal doesn't stretch like nylon...so it's not hard to see why the question has arisen.

When I buy a cruiser disgust::tranquillity::sleeping I expect it will be small enough and lowly enough not to have a windlass, and not to cope well with a hundredweight of chain near the bow...

...so I'm keen to believe that a few feet of chain between the multiplait warp and anchor, (which must be a pretty good, proven, modern hook) will leave all-chain rodes as a memory of something the old-timers relied upon.

My (hopeful) vision of a well-spread pair of such anchors - preventing 'roaming' and to allow somewhat for failure, strikes me as better than taking on the heavyweight cost, ongoing maintenance and physical burden of chain.

 

[GHA]

My point is that it isn't a matter of opinion, but (relatively) straightforward analysis that many people on this forum could handle easily.
One point that people miss is that a chain ALWAYS describes a catenary curve. It may be a very flat catenary, but it is never dead straight. Further, the maximum curvature of the catenary is at neither the surface nor the anchor, and is therefore not observable. Whether the catenary is sufficient to allow damping of shock forces is another matter, but it is something susceptible to relatively straightforward mathematical analysis. There are several people on the forum who could carry out such an analysis; not me, I'm afraid - although I knew the maths existed, it would require too much revision of the relevant techniques for me. But even with my rusty skills, I can follow the maths in the Wikipedia article! But I can think of several forumites who could carry out the analysis, if they wished. I could - but it would take me too long to recover the skills I once had!

Some graphical displays of other people's equations.
http://www.moondogmoving.co.uk/catenary.html
Just of when the last link lifts. Worth bearing in mind,mwhen that happens the catenary can still absorb more energy and go flatter but will do so over very short distances, so when you have a load of wind and a boat sheering about the forces go through the roof as the chain tries to stop the boat moving in such a short distance . Snubbers help by making that distance longer.

 

[thinwater]

... Whether the catenary is sufficient to allow damping of shock forces is another matter, but it is something susceptible to relatively straightforward mathematical analysis. There are several people on the forum who could carry out such an analysis; not me, I'm afraid - although I knew the maths existed, it would require too much revision of the relevant techniques for me.

I went through that exercise for an article a few months ago. As a general rule, if the water is less than 20 feet deep, the point at which the catenary can no longer absorb much energy (less than 2' of stretch) happens before the chain lifts off the bottom. There are quite a few variables, but the point is that these two things generally happen at about the same time; chain lifts and chain no longer has enough stretch remaining to matter. I think that qualitatively makes sense. The more chain, the more wind both require.

 

[GHA]

As a general rule, if the water is less than 20 feet deep, the point at which the catenary can no longer absorb much energy (less than 2' of stretch) happens before the chain lifts off the bottom. .

To be pedantic,not sure that's quite correct physics?, if the boat is moving the energy has to go somewhere for it to stop. Big enough anchor well dug in the catenary can still absorb the energy , it will just do it over a very short distance with a huge final force. Energy =force x distance. Though in our boat situations it's probably more likely that something else will give first, like the anchor dragging.

 

[lw395]

When you most need damping and snubbing, the bow of the boat will generally be moving from side to side somewhat. Maybe up and down too.
The rode won't be straight, it will be moving in the water. The mass and viscous drag of the rode moving provide springing and absorb energy. The heavier the chain, the more springing and slower response. The fatter the chain, more damping.
When the boat is not dead still in a steady wind, on a flat bit of sea, the maths get very hard. Also small changes in conditions give you very different answers.

If there's no motion and the rode is at a constant catenary, there is no problem, so long as the anchor holds. There is no damping to be done, no kinetic energy to be absorbed.

If you ever get in a situation where the boat is veering over an acre of ocean, is more or less mass in the rode going to make it better or worse?
I reckon it could go either way, you might just be changing the resonant frequency.

 

[thinwater]

To be pedantic,not sure that's quite correct physics?, if the boat is moving the energy has to go somewhere for it to stop. Big enough anchor well dug in the catenary can still absorb the energy , it will just do it over a very short distance with a huge final force. Energy =force x distance. Though in our boat situations it's probably more likely that something else will give first, like the anchor dragging.

Yes, you being pedantic and you know it. The first law of thermodynamics endures. The readers either already know this or don't care.

 

[thinwater]

When you most need damping and snubbing, the bow of the boat will generally be moving from side to side somewhat. Maybe up and down too.
The rode won't be straight, it will be moving in the water. The mass and viscous drag of the rode moving provide springing and absorb energy. The heavier the chain, the more springing and slower response. The fatter the chain, more damping.
When the boat is not dead still in a steady wind, on a flat bit of sea, the maths get very hard. Also small changes in conditions give you very different answers.

If there's no motion and the rode is at a constant catenary, there is no problem, so long as the anchor holds. There is no damping to be done, no kinetic energy to be absorbed.

If you ever get in a situation where the boat is veering over an acre of ocean, is more or less mass in the rode going to make it better or worse?
I reckon it could go either way, you might just be changing the resonant frequency.

Please sketch out the maths for us in simple terms. Drag of the chain through the water (<0.5- 1 knots, only the part near the boat?) vs. the wind force on the beam of the boat (>45 knots?) at an angle (15-25 degrees?), for example. Try two chain sizes.

Interesting, but we need numbers to go with the words. The alternative is testing. I have done some of this (variable chain size) and the shipping industry has analyzed this to death. I've read some wonderful studies by marine engineers that included both theory and testing. For example, placing a drogue on the rode would add even more side resistance, and yet has been shown to be relatively ineffective. But I would be glad to revisit this, particularly for fiber rodes. So far, all of my testing suggests that underbody, windage, and rigging (in that order) are what matter, and that rode type is a rounding error by comparison, once the wind gets up. Get the dinghy off the foredeck. Add a dodger. Lift the rudder if you can.

"At-Anchor Wind-Induced Yaw Instability of a Monohull"
Jonathan Soja, Donald L. Blount and Associates, Chesapeake, Virginia, USA
Jeffrey Bowles, P.E., Donald L. Blount and Associates, Chesapeake, Virginia, USA

 

[GHA]

Yes, you being pedantic and you know it. The first law of thermodynamics endures. The readers either already know this or don't care.

Seems the right thing to do to try and spread a little knowledge even if you disagree. Most people get no further than energy transfer which is only part of the story, the rate of energy transfer is much more important as far as force goes.

and you know it.

Not the case at all, even if you want to accuse me off having a dig for the sake of it.

First law of forums endures, point of some inconsistencies in a post and high probability of a grumpy reply.

 

[thinwater]

... First law of forums endures, point of some inconsistencies in a post and high probability of a grumpy reply.

Now how could I possibly disagree with that axium!

 

[Neeves]

As AP says to argue about catenary is hardly productive - its maths and well defined. As LW395 says, a number of times, its more complicated than that because the catenary is constantly changing. When the yacht is yawing in 30 knots of wind, or is hit by a bullet, and reaches the end of its tether all catenary advantages have gone and we rely, in the absence of anything else, on the anchor absorbing the energy of the moving yacht. Everyone has ignored that Thinwater pointed out that steel is elastic and particularly HT steel and it has an ability absorb a more significant amount of energy than realised. Many of us now use sensible snubbers and the snubbers and the inherent elasticity of the chain mean the anchor no longer need take all the impact of the snatch load (in fact does not take a snatch load at all). There will be a different between a snatch load with an 8mm chain compared to a 6mm chain - but the differences are not relevant (in my opinion) as 8mm is bad enough to need a remedial solution - and the same, or a similar solution applies to 6mm. There maybe a difference in the hydrodynamic effect of larger chain compared with smaller chain - but at 30 knots its academic as that snatch load at 8mm, with the larger chain - still needs that remedial method - a snubber (or an endless supply of chain and lots of room). Snatch loading does vary in the same anchorage, big heavy displacement yachts are less skittish and if the anchorage is shallow - there was not much catenary in the first place - so there is no one size fits all. I tend to look at worst case scenarios.

My conclusion is that larger chain is significantly overrated, it has been well sold, but technology has moved on and we can now enjoy lighter chain to no ill effects to snatch loading - the answer is snubbers or bridle.

From my own experience I think we have more to learn about using snubbers and bridles and my original ideas of the 10m snubber/bridle may not be the ideal and longer, much longer might have significant benefits. Noelex has made the point that monohulls cannot use bridles effectively as they have insufficient beam and the bridle angle is 'too' narrow. I think this simplistic as its is quite possible to have a 30m bridle with a bridle angle of 45 degrees, or 90 degrees. Its not rocket science. But it needs to be tested - as there may be other variables that have not been considered.

The unknown, or the unknown that has been raised, is wear resistance of the chain - and whereas Noelex reports are sobering - I'd like to see some similar evidence from the 'other' manufacturer as I don't find, given the reputation of the supplier, that the report are typical (of downsizing chain). Noelex reports underlines, to me, that consumers should be more attentive and demanding of galvanising quality and not leave that quality to chance. As Noelex correctly underlines, that has been known for decades - galvanising determines chain life. Tests I have conducted illustrate that Hot Dipped Galvanising quality varies and the coating thickness also varies - even from one manufacturer.

The benefits of the lighter chain are more obvious and have simply not been mentioned, less weight in the bow locker, less volume needed to store, lower electrical drain for the windlass, potential for a smaller windlass. If you hand retrieve - you back is very grateful.

The unknown is the abrasion/corrosion issue - I will be getting it in hand. But given the number of variables - not an easy nut to crack.

Monohull bridles of 20m and 45 degree bridle angle.

Attach bridle to bow cleats, run to turning block on stern, back to fairlead to common chain hook a few metres outboard of the bow. Play around to get 45 degrees, or 90 degrees. But you have an effective bridle of 45 degrees or 90 degrees and its deck length (lots of elasticity), say 20m. You need to watch for chafe points, you might want tiny turning blocks at the fairlead - its simply a modification of what I have been using (to give a potential 30m bridle).

Jonathan

 

[lw395]

....
Monohull bridles of 20m and 45 degree bridle angle.

Attach bridle to bow cleats, run to turning block on stern, back to fairlead to common chain hook a few metres outboard of the bow. Play around to get 45 degrees, or 90 degrees. But you have an effective bridle of 45 degrees or 90 degrees and its deck length (lots of elasticity), say 20m. You need to watch for chafe points, you might want tiny turning blocks at the fairlead - its simply a modification of what I have been using (to give a potential 30m bridle).

Jonathan

Does this do any good?
I get the idea of using a bridle on a cat, it puts the pivot point of the rode a good distance ahead of the boat.
With a monohull, it's hard to move the pivot point much forward of the bow, unless the boat is very beamy at the (not very) pointy end?

It seems much simpler to have the snubber in front of the boat where there's nothing for it to chafe on?

 

[Neeves]

Does this do any good?
I get the idea of using a bridle on a cat, it puts the pivot point of the rode a good distance ahead of the boat.
With a monohull, it's hard to move the pivot point much forward of the bow, unless the boat is very beamy at the (not very) pointy end?

It seems much simpler to have the snubber in front of the boat where there's nothing for it to chafe on?

A snubber is almost defnitely going to chafe somewhere, better it chafes somewhere useful than simply meeting convention.

Many catamarans have overly short bridles the pivot point, or chain hook not far forward of the bow, but they do have an obtuse bridle angle.

Horses for courses - just because some have pointy boats does not mean everyone does. Commonly the mantra is - 'snubbers for monohulls, bridles for catamarans' - its not one size fits all. Try it and see before condemning it - and when you have tried it and it does not work, let us know.

Jonathan

 

[Neeves]

I thought I might summarise the results of an abrasion test that was reported in Practical Sailor some years ago.

First the tensile strength.

As seems to be common with G30, G40 and G43 chain strengths were well over specification. Strength of Maggi G70 was excellent almost meeting a G70 spec but strengths of Peerless and Campbell G70 were lower, consistent with their galvanising a Q&T G70 chain. I deduce that possibly Maggi start with a stronger chain, possibly a G100 - this would be consistent with the strengths they achieve.

The test was conducted by attaching one end of similar chain lengths, about 300m, to a 1500mm piece of rebar. Supporting the rebar at each end and hanging of the transoms of our cat such that the chains were just touching the seabed at high tide, 2m, the bar, and chains then moved as the yacht moved on its swing mooring. The chains were periodically check but the length of time for the test was 70 days. The test was conducted twice with overlapping qualities used as standards. Also to check for any preferential wear one sample was attached in the middle and at one end. When we used the cat the chains were retrieved and secured on the trampoline.

The seabed is silica sand, from Sydneys sandstone bed rock.

Initially the chains were weighted - but this was found unnecessary as at 70 days wear was visually obvious - the raw steel could be seen.

This is an accelerated test, as it turns out, as some chain was worn bare - but I have never heard of a chain being worn bare so quickly in real life. So like a salt spray test this differentiates but does not say anything is bad - just some are better.

Most of the chains were 5/16th" or 8mm - so there was no conscious attempt to look at the effects of chain size.

All the chain was supplied by the manufacturers - except for CMP and PWBs samples - which were bought from a local chandler. Anyone who knows of some of my previous publication will know that CMP will not extend any largesse to me.

Campbell, part of Apex Tools in America, was undoubtedly the best and most consistent. Their G70, G30 and G43 were the best - along with Maggi's G40. Peerless G70, G30, CMP G30 and were 'middle of the road' (with Peerless G70 borderline between middle of the road and excellent) and PWB G30 (Grade L), Maggi's G70 and Peerless G43 had the most abrasion. In the second test Maggi's G40 and Campbell's G43 were again excellent performers.

I don't draw too many conclusions for the test - for conclusions you would need to test, say, every 6 months. I did find it interesting the variation within Peerless' samples and Maggi's samples. I would have expected stronger consistency - as was evident in Campbell's samples. There is no evidence that there is any difficulty in galvanising G70, Campbell quality was excellent and Peerless not far behind.

Galvanising thickness measured varied between 50 microns for Maggi's G70 to 180 microns for Maggi's G40.

In with these were 2 Armorgalv samples from America a 82 micron coated 3/8th" chain G43 (the only 3/8th" tested) was slightly worse than the 80 micron G80 5/16th". But the difference was small.

The results of the lower abrasion resistance of Maggi's G70 is consistent with Noelex suggestion - but the idea that all G70 performs the same way - these is no evidence to support the suggestion. There is evidence to suggest that galvanising quality is variable - and buyer would be wise to look at Vyv's website which gives one method to check for quality.

There was evidence of corrosion on some samples of chain, primarily those that abraded most quickly. The corrosion was at the crowns. There was no attempt to measure this corrosion (I did not know how to!)

One piece of chain, cheap, simply described as 'chain' of Chinese origin, but short link 8mm, bought from a local chandler had appalling strength and had virtually all the gal worn off. I only hope purchasers use it to chain up their dog - it would have no place on a yacht (nor in a chandler).

Jonathan

 

[thinwater]

Galvanizing is not just a coating, like paint. It protects even after it has a few gaps by creating a protective current. This is vital, or every chain would fail in a year due to minor nicks. Instead, we observe that chain doesn't corrode much until a significant portion of the galvanizing is gone. Of course, this only makes testing more difficult to accelerate, since the link-to-link wear/corrosion pattern is the concern.

 

[Neeves]

Developing a test whose primary focus is the link to link abrasion is taxing my imagination. It is easy to simulate abrasion on links, there are a number of methods - but link to link abrasion and subsequent corrosion is difficult. I can measure overall abrasion, by simply weighting before and after. You can visually look at the areas of worn chain - they are massive worn areas (like half the link bare of gal), less massive (bare patches showing through) or none. But measuring that tiny metal to metal contact at the link - I am so far bereft of ideas. I have a micrometer - but it has flat anvil edges which will not measure within the crown. I have callipers - but I don't think them sufficiently accurate.

Furthermore my own idea is that as abrasion is a function of pressure, the pressure of the item subject to abrasion, then a large heavy chain is likely to have similar abrasion as a small light chain - so abrasion tests as I conducted will abrade a 6mm chain similarly to a 10mm or 12mm chain - the difference will be coating thickness or strength of coating - as percent of the original. So if a 6mm chain has a 70 micron gal coating and a 12mm chain has a 70mm gal coating I'd expect to see the same area pf bare steel. Furthermore this weight of bare steel will totally mask the point to point abrasion.

In fact because the smaller link is light and would spend more time off the seabed than a big heavy chain I would not be surprised if in real life the gal of the light chain (with the same gal thickness as the heavy chain) does not lost longer than the same gal on the 12mm chain. On top of this the proposed gal on the light chain is harder than that on conventional chain and the steel also more abrasion resistant.

Similarly with crown to crown abrasion - if the light chain is subject to tension but is off the seabed it will have less sand at the crown to crown interface - so again less abrasion. When the tension is low there will be less abrasion, less pressure - so in light winds abrasion effect will be less. Abrasion will be at its worst in strong wind, more tension (more pressure to accelerate abrasion) but for the lighter chain, less sand at crown interface - because it is washed out.

It is unrealistic to suggest 2 rodes - the expense is huge (you would need 2 different gypsies, two new rodes). And if you have 2 rodes - it would be difficult to produce the same conditions on one chain as the other.

Any ideas on how to test the abrasion of 2 differently sized chains with a focus on the abrasion of link to link coating?

Jonathan

 

[thinwater]

First, I think wet/dry cycles will help. This accelerates corrosion, because the deposits become VERY saline as they dry. This is also realistic, represening chains that are stored in lockers much of the time.

Second, we need a variety of loads, from light to heavy.

I'm thinking I would insert sections in docklines sort of like some folks use rubber snubbers. The load would vary with the weather and wind direction. Choose lines that dunk in the water occasionally. I'm thinking these would be a supplemental pair of spring lines, with nylon lines backing them up.

Or could you splice them into your mooring chain, with the main chain still there as back-up? Or in the pendant, same idea?

This will accelerate things. But it will still take years. Measuring with calipers should work (+/- 0.001 inch), the trick being to record the EXACT locations so that they can be accurately compared.

But get started! I think this is worth doing. Of course, product changes over the years....

 

[Neeves]

First, I think wet/dry cycles will help. This accelerates corrosion, because the deposits become VERY saline as they dry. This is also realistic, represening chains that are stored in lockers much of the time.

Second, we need a variety of loads, from light to heavy.

I'm thinking I would insert sections in docklines sort of like some folks use rubber snubbers. The load would vary with the weather and wind direction. Choose lines that dunk in the water occasionally. I'm thinking these would be a supplemental pair of spring lines, with nylon lines backing them up.

Or could you splice them into your mooring chain, with the main chain still there as back-up? Or in the pendant, same idea?

This will accelerate things. But it will still take years. Measuring with calipers should work (+/- 0.001 inch), the trick being to record the EXACT locations so that they can be accurately compared.

But get started! I think this is worth doing. Of course, product changes over the years....

Our mooring blocks weight 1.5t - it would cost me a few hundred $ to lift and attach as suggested and then a few more hundred $ to lift to check, and/or complete the test.. But, advantageously, the chain would then be in the seabed, as the blocks sink with time in the sand. If I splice into the lines at sea level - I'd have no sand - abrasion would then be link to link (but there would only be link to link abrasion). I could do this easily - shackle 2 short lengths of chain together and splice into an 'extra' mooring line (so it takes the tension) where it touches the water, turn the 2 lengths end for end to ensure they both have similar environment.

I'd thought of a rat tailed file in a drill with the link hung on the file and a weight on the link. The problem I forsee is - its going to get hot! But I have a number of such files, of varying diameter and some bigger ones look the same grade of file as the smaller ones. Ideally I'd use a, say 10mm file in a 12mm link and a 4mm file in a 6mm link - but I'm not sure the files are the same grade, they just look 'similar'. Then abrade for a fixed time allocation.

I could use the same small file for both links - but not sure that that would be acceptable.

I could abrade for, however long 15 minutes, 30 minutes, (suck it and see) dump in a little seawater in a bucket for a week (in the sun), restart.

I was thinking initially of a 2.5kg weight on both links (as I have 2 x 2.5kg weights). I could have the 2 links abraded together side by side if I use a smaller file.

Jonathan

 

[thinwater]

I think I was not clear. Attach two prusik slings to the mooring chain or pendant and put a length of chain between them, so that they carry the tension, but without disturbing the original chain. A lashing could be used to transfer the tension. A little diving, but no hauling or re-rigging.

 

[Neeves]

The mooring block is 10m deep, all the chain is on the seabed, its the 'snubber' and is a combination of 2" mining chain, 5m and a sweep chain of a further 5m of 3/4" chain. Neither chain really lend themselves to joining to 6mm chain, easily. The pendant 1.5", spliced to a swivel on the 3/4" chain, floats, and does not touch the seabed with a bridle spliced into the 'top' at sea level.

Lashing might not last - the swivel lasts about 3 years if left unattended.

Jonathan