Shock load on mooring/anchor chain & fittings

[MathiasW]

Interesting - thanks.

What's your definition of an excellent snubber for an 8m by 4m cat?

As Jonathan suggested, it is the snubber stretch that differentiates an excellent snubber from a poor one. For the maths, I do not care how long the snubber is. That has no effect on its ability to store energy. It is by how much it can stretch with a given load. Of course, the stretch should be not too much compared to the overall length of the snubber, so as not to overload and break it.

Sorry for the brevity of initial post.

 

[MathiasW]

I like the details. I have to ask is 12tons for a 10m boat common unless wooden or steel?

We are about 7.5tons with all the gin onboard.

The weight of the vessel only enters the calculation when using the maximal speed at anchor (caused by the gust or swell) to calculate the kinetic energy of the vessel that needs to be absorbed by the chain and the snubber. So, if your vessel is lighter, the results would correspond to a situation where this maximal speed is larger than the 0.6 kn I was assuming.

Whether such a weight is normal - I have seen very heavy monohulls. We are living on a multihull and there it is quite different, of course.

The values I chose for the table are all a bit at the edge, I admit, but this was to make the point very clear. In shallow water and with a lot of swell, your chain is not helping you at all, you do need snubbers. And even then relocating to deeper water (assuming the swell there is the same) will reduce the shock load at the anchor.

Cheers

Mathias

 

[thinwater]

It confuses me, I wish they'd stick to N and kN. I thought 10 to the 3,6,9 was the preferred system of units.

When in doubt, Google "convert x to y."

Centimeters, deciliters, decipascals, centigrade, and more come to mind.

 

[MathiasW]

I think you need to provide some information on vessel weight, in addition to physical size.

Mathias,

I think the question is valid - what makes a good snubber - I suspect the answer is in the data - you give the stretch - so what textile/rope/construction gives the stretch defined under the conditions you provide and then how can that be extrapolated for a specific yacht.


daN - 10 Newtons, or 10N

think of 1kg being 10N

so think of daN as being kg.

You will receive a much more erudite answer later, from those of a much richer scientific background . I just happened to be posting on Mathias excellent work.

I think once you get to 500kg you will have reached beyond crew tolerance levels, if you are married one of the crew being subjected to that sort of force will be checking the telephone number of the nearest divorce lawyer. If its a crummy anchor it will already have dragged. A I ton snatch load will bend some bow rollers and some (steel) chain hooks. If Mathias returns I suspect 0.6knots is an underestimate - but am happy to be contradicted.

Mathias - I am interested what length of snubber are you proposing for a 30m chain deployment if the snubber is stretching 1.6m? My guess is the snubber needs to be the length of the chain, approx.

To give some colour to the information - where are you now?

Take care, stay safe

Jonathan

Hi Jonathan,

Yes, your explanation is spot on and perfect!

I chose the numbers in an effort not to be unreasonable, but still to show the point: In shallow water the chain just fails to perform, and snubbers become absolutely vital. Even moving to deeper water is better than staying in shallow water and lots of swell. (assuming same swell there)

If a snubber is allowed to stretch by 20%, then a stretch of 1.6 metres implies the snubber has to be at least 5 x 1.6 metres = 8 metres long. But this would be at the edge. Twice that long would be my personal minimum.

As to our whereabouts - we are currently in Puerto Vallarta on the Pacific side of Mexico, but heading out today. All spare parts have arrived, the dental surgeries are completed, we are good to go!

Cheers

Mathias

PS: As to units, sorry, in Germany the sailing magazines all use daN, and so I had to adjust when submitting stuff there. My app also works in lbs/lbf.

 

[MathiasW]

The quoted displacement is a little under 3 tons, which doesn't seem much for a boat that is, by all accounts built like a brick outhouse.

What makes you think that? I worked with 10 metric tonnes for this table. But as said above, it does not really matter...

 

[Roberto]

You read that wrong. That is 230% of the strength of a single strand. 2.3 times the strength of the material, not 4 times.

What am I reading wrong? Starzinger talks about "underlying line strength"

cit.
I tested 10 soft shackles at an average 170% of the underlying line strength. The two keys to soft shackle strength are (a) ensuring the length/tension on the two lines in the shackle are exactly equal, and that the cover is milked down very hard and smoothly on the cover/bury portion of the shackle, and (b) reducing the loading on the diamond knot (which is where they always break). This can be accomplished by using longer soft shackles of lighter cord, and looping them between the load points several times (like a lashing) before closing the diamond/noose. Note: Soft Shackles 'theoretical' max strength is 4x the line used, because they are constructed with two strands in a loop (so four strands between the two load points). The diamond knot then reduces the strength by 50-60%, so you end up at about 170-200% of line strength.

 

[MathiasW]

It confuses me, I wish they'd stick to N and kN. I thought 10 to the 3,6,9 was the preferred system of units.

I think folks in Europe now use daN as it is in numerical numbers so close to what we used to use in Kg. (Which, by the way, has always been the wrong unit to use, as it is a unit of mass and not weight. Kp would be the correct unit to compare it with.) So, from the point of easing the transition from old to new units, it was felt that daN would be the best choice.

 

[roaringgirl]

What am I reading wrong? Starzinger talks about "underlying line strength"

cit.
I tested 10 soft shackles at an average 170% of the underlying line strength. The two keys to soft shackle strength are (a) ensuring the length/tension on the two lines in the shackle are exactly equal, and that the cover is milked down very hard and smoothly on the cover/bury portion of the shackle, and (b) reducing the loading on the diamond knot (which is where they always break). This can be accomplished by using longer soft shackles of lighter cord, and looping them between the load points several times (like a lashing) before closing the diamond/noose. Note: Soft Shackles 'theoretical' max strength is 4x the line used, because they are constructed with two strands in a loop (so four strands between the two load points). The diamond knot then reduces the strength by 50-60%, so you end up at about 170-200% of line strength.

I make my soft shackles as 2 strands with a button knot and a full-length bury inside each strand. These are supposedly stronger.

 

[thinwater]

What am I reading wrong? Starzinger talks about "underlying line strength"

cit.
I tested 10 soft shackles at an average 170% of the underlying line strength. The two keys to soft shackle strength are (a) ensuring the length/tension on the two lines in the shackle are exactly equal, and that the cover is milked down very hard and smoothly on the cover/bury portion of the shackle, and (b) reducing the loading on the diamond knot (which is where they always break). This can be accomplished by using longer soft shackles of lighter cord, and looping them between the load points several times (like a lashing) before closing the diamond/noose. Note: Soft Shackles 'theoretical' max strength is 4x the line used, because they are constructed with two strands in a loop (so four strands between the two load points). The diamond knot then reduces the strength by 50-60%, so you end up at about 170-200% of line strength.

Reread the last sentence. The diamond knot then reduces the strength by 50-60%, so you end up at about 170-200% of line strength.

There are 4 passes, so perfect would be 400% line strength. So a SS made with 5400-pound Amsteel is about 11,000 pounds, even though there are 4 passes. Roaring Girl used 400% in her OP. The button knot style can be as strong as 230% of 15 kN, or 34 kN. As I said, I wouldn't bet on more than 20 kN unless I had tested the knots I had personally tied.

As a point of comparison, Robline lists their 6mm (tied from 4mm) Dyneema SSs at 4400 pounds (19.5 kN), so I'm not far off, based on Robline's testing.

Finally, it is also quite easy to make SSs that are weaker than this, if the lines are not perfectly even. Also, consider the difference between minimum breaking strength and average breaking strength. MBS is the one that matters.

Just details.

 

[penberth3]

When in doubt, Google "convert x to y."

Centimeters, deciliters, decipascals, centigrade, and more come to mind.

Centigrade is a recognised unit. The other three you've quoted aren't, although schoolteachers seem to use centimetres for some reason.

 

[thinwater]

Centigrade is a recognised unit. The other three you've quoted aren't, although schoolteachers seem to use centimetres for some reason.

I'm not sure I've ever seen a ruler without cm. Google metric ruler. Centigrade is 100 division of an interval. Deciliters is odd, but not uncommon in formulations.

Don't forget decimeter. A liter is defined as one cubic decimeter. Volume could have been based on a cubic meter, for easier maths, but for most purposes it isn't. Of course, the gram was originally based on 1 cubic centimeter (that pesky odd unit) of water, rather than a millimeter or meter of water. Not convenient, I guess.

Centipoise. The most common SI unit of viscosity.

Few people would recognize slug as a standard unit, but it is recognized and required for even basic engineering math in traditional units.

They're just not recognized by everyone.

 

[roaringgirl]

I'm not sure I've ever seen a ruler without cm. Google metric ruler. Centigrade is 100 division of an interval. Deciliters is odd, but not uncommon in formulations.

Don't forget decimeter. A liter is defined as one cubic decimeter. Volume could have been based on a cubic meter, for easier maths, but for most purposes it isn't. Of course, the gram was originally based on 1 cubic centimeter (that pesky odd unit) of water, rather than a millimeter or meter of water. Not convenient, I guess.


Centipoise. The most common SI unit of viscosity.

Few people would recognize slug as a standard unit, but it is recognized and required for even basic engineering math in traditional units.

They're just not recognized by everyone.

Safest to stick to SI units for communication of formulae with strangers. I don't think I could have done my BEng in imperial units!

 

[MathiasW]

Safest to stick to SI units for communication of formulae with strangers. I don't think I could have done my BEng in imperial units!

Indeed! But daN, cm, etc are all derived SI units, just with a power of 10 prefactor. Given that our world stretches from well below 1 Angstrom to many light years, it just makes sense to use suitable pefactors to SI units... We'd go bananas with all the zeros before or after the decimal comma...

 

[thinwater]

Safest to stick to SI units for communication of formulae with strangers. I don't think I could have done my BEng in imperial units!

In the US the engineering tests typically alternate SI and imperial. Five problems with one system, then five with the other... and sometimes they mix them up.

When I have worked with multinationals, I always ask that they stick to SI and let me do the conversions. Chemicals should follow IUPAC name conventions, particularly abbreviations, which as hideous to translate. For sample HDPE is PEAD in Spain. Hard to look up.

 

[AntarcticPilot]

Centigrade is a recognised unit. The other three you've quoted aren't, although schoolteachers seem to use centimetres for some reason.

Well, I guess there are people of my generation who were brought up with the CGS metric units (Centimetre, Gram, Second) rather than the SI trio of Metre, Kilogram and Second. And millimetres are too small to be a convenient unit, and metres too big!

 

[roaringgirl]

Indeed! But daN, cm, etc are all derived SI units, just with a power of 10 prefactor. Given that our world stretches from well below 1 Angstrom to many light years, it just makes sense to use suitable pefactors to SI units... We'd go bananas with all the zeros before or after the decimal comma...

That's what scientific notation is for, eg: 1mm = 1x10^-3m

 

[thinwater]

That's what scientific notation is for, eg: 1mm = 1x10^-3m

So why didn't you use 4.0 x 10^-3 m Dyneema for those soft shackles?

Just playin!

[And for what it is worth, there is a space between the number and the units, other than %. Also, most of us get away with leaving off the degree mark, since couloms and farads are rare in our conversations.]

 

[Roberto]

I make my soft shackles as 2 strands with a button knot and a full-length bury inside each strand. These are supposedly stronger.

Hello,
yep I am trying that geometry too (if it's the Starzinger one, otherwise could you please provide a diagram for yours). Usually, getting a higher BS for soft shackles can easily be achieved by increasing the native dyneema 8strand diameter, soft shackles for chain snubbers have to fit inside the limited space between two chain links, which limits the finished soft shackle diameter so gives an advantage to splice geometries which maximize the breaking strength from a given starting rope diameter -->possibly the "Starzinger" one.
I have a limited interest in a "General Theory of Soft Shackles", just take this Starzinger geometry is the one possibly giving somewhat better results compared to others; from what he wrote, using a 4mm 8-strand braid as starting line to splice a SS, the resulting soft shackle shall roughly have 1.5-2x the breaking strength of the 4mm braid, quite possibly less if not done perfectly, maybe 2.3x if done the proper "Starzinger" way. I would not backwards infer any meaningful numerical value should they ever break.
FWIW, I am trying a few home made ones, end knot tied by winching it with a #50 winch. I applied a few different thickness coatings to try and increase resistance to chafe especially when threading the shackle into the chain link, which often results in filaments of bare 8strand rope being gripped by metal roughness. It's long and tedious as significant thickness can only be obtained with several coats, we'll see
Please give feedback with your findings if you like.
r.

 

[thinwater]

Hello,
yep I am trying that geometry too (if it's the Starzinger one, otherwise could you please provide a diagram for yours). Usually, getting a higher BS for soft shackles can easily be achieved by increasing the native dyneema 8strand diameter, soft shackles for chain snubbers have to fit inside the limited space between two chain links, which limits the finished soft shackle diameter so gives an advantage to splice geometries which maximize the breaking strength from a given starting rope diameter -->possibly the "Starzinger" one.
I have a limited interest in a "General Theory of Soft Shackles", just take this Starzinger geometry is the one possibly giving somewhat better results compared to others; from what he wrote, using a 4mm 8-strand braid as starting line to splice a SS, the resulting soft shackle shall roughly have 1.5-2x the breaking strength of the 4mm braid, quite possibly less if not done perfectly, maybe 2.3x if done the proper "Starzinger" way. I would not backwards infer any meaningful numerical value should they ever break.
FWIW, I am trying a few home made ones, end knot tied by winching it with a #50 winch. I applied a few different thickness coatings to try and increase resistance to chafe especially when threading the shackle into the chain link, which often results in filaments of bare 8strand rope being gripped by metal roughness. It's long and tedious as significant thickness can only be obtained with several coats, we'll see
Please give feedback with your findings if you like.
r.

What coatings?

I've tested several on polyester (Yale Maxijacket was best), and saw huge improvements, but not so much on Amsteel. I assumed this was because it comes with a coating, and because the coatings can't really bond to Dyneema. I gather they use specific polyurethanes on industrial Dyneema assemblies.

I would also think a heavy coating would gunk up the working of the SS. It looks like you are only coating the chain contact area, which makes sence.

I'm interested!

 

[Neeves]

Its taken me decades to get used to metric units - please don't use imperial!

I am sure soft shackles are marvellous for 8mm chain and above when attaching a snubber (see some qualification below) but when you get down to 6mm chain the aperture between the links is simply too small to easily and quickly thread a soft shackle through it - and the soft shackle itself looks like a toy.

If you check my article in YM and better Multihulls you will find we use a bridle and a bridle reduces yawing, because of the 2 snubbers forming a triangle. We have incorporated a vertical triangle, which then allows some management of swell. It becomes very messy to make a vertical triangle with a soft shackle..

Mathias shows that you need roughly' at least a deck length snubber, ours can be extended to 30m x 10mm. We found 12/13mm too beefy, not enough elasticity. If you have say a 15m snubber it extends a long way, in fact 15m, from the bow and if the wind drops and you are anchored in a 5m depth - then your snubber will lie on the seabed. Better you commence your snubber at the transom, then you have deck length down the deck, say 5m beyond the bow and unless you are anchored in less than 5m the snubber will never touch the seabed. Commencing a bridle, on a mono, at the transom allows the base of the triangle formed beyond the bow to be 'quite' wide - meaning you will reduce yawing (not as well as a multi - but better than a kick in the teeth). Finally if you commence at the transom your snubber can be as long as you want and if the wind picks up you can extend the snubber from the safety of the cockpit, in your diaphanous silk Jim jams if you want. Returning to soft shackles - I'm not sure of the longevity of soft shackles used on a 15m snubber constantly abraded on the seabed - personally I'd rather use steel (of some type). Of course you could use less than depth for the length of your snubber - but then it will not snub.

As pointed out the snubber, in order to snub, will not be as strong as the chain. It is not intended to replace the safety provided by the chain - the snubber is chosen for its snubbing characteristics. You cannot have strength and the required elasticity. Your snubber will one day - fail - you need a short snubber or chain lock to protect your windlass for the time when the snubber fails (and carry a spare snubber). Don't get excited about failure we have had 2 fail in 20 years.

If anyone wants to query any of this - feel free or send me an email address, PM, and I'll provide more detail.

As Mathias illustrates knowledge is of little value unless it is accessible to others. I'd agree with the data Mathias has calculated. With a short rode in shallow water and no chop subject to 35 knot winds we recorded snatch loads of 650kg. Add in some swell, with the bow rising and falling - I'd expect to pluck the anchor out. Add a snubber, a good snubber, and those loads drop to being of a size you can actually hold by hand, so around 70kgs. My data and that of Mathias will be 'different' but the same order of magnitude. Mathias has based his data on everything having different dimensions - and his data is exact and correct for the data he has used - you need to make some extrapolations for your specific circumstances. My measurements are actual for a specific catamaran in specific conditions - but I think underlines Mathias accurate calculations.

From memory Mathias has an app available for anyone who wants to plug in data for their specific yacht (I don't recall if he linked to it) - saves extrapolation and you will have specific data for your vessel.

If you can access, it may be behind a paywall for you, this was the original article on tension in the rode:

Anchor Testing and Rode Loads | Practical Sailor

Now - if someone wants to lend me a yachts 35' or bigger that I have access to thro the summer, I test during a developing Seabreeze, I can develop actual data for that yacht

Take care, stay safe.

Jonathan