Important Double Sheet Bend query

Practical Knots
41

It is ages since I analysed anything statistically (30 odd years) so I am extremely rusty, but number of trials required also depends on spread of results. Any statisticians here able to lend me a hand? There is nothing commercial in any of this. I do not even have a blog, do not participate in Facebook, do not have a website. Results will be freely shared. I will report any conclusions on the cruising forum as well.

The problem on board is that apart from equipment, I am extremely limited by materials. I currently have old jib sheets and halyards I can use - precious as testing on old line is, I feel, vital. Old line has been stretched, UV exposed and salt impregnated. Unlike climbers who would replace line frequently after big loads, sailors continue using line for years. Tests in new line are great (and I will do these too), but unfortunately don’t replicate ‘real world’ performance. Simply storing large quantities of line in different diameters is difficult.

The line (and particularly the knot) shouldn’t be reused for trials as the trauma to the line may affect the subsequent result. Over a metre of line (x2) is used each time. So imagine how much line I am going to need if 25 trials are done per knot (per diameter, as thicker line performs very differently: the thicker the line, the bigger the gap when it bends). Just way beyond my scope.

I think perhaps all I can do is get a rough idea what knots look promising and need to be investigated further.

42

And the spread of the results depends on the number of trials ! :slight_smile: The more tests you perform and the more results you acquire, the spread gets bigger. What will tell you the required size of sample you need , is the required reliability : of course, we do not need 5 sigma in knotting, as in Particle physics ! :slight_smile: ( = one chance in two million, to be wrong ! )

Now you know one of the many reasons I prefer OPT ! :slight_smile: ( Other People s Tests ).

No, that is irrelevant : the relative geometry of all the elements of a knot does not change with size.
What does not scale uniformly, is the size of the knot ( which changes ) in relation to the size of the individual fibres ( which remains the same ).

A “rough idea” is a much much much better thing than “no idea whatsoever” ( = the black box we were talking about the other day…). Go on !

43

No, it is not irrelevant at all. It is critical. Wider diameter lines do not bend as easily, they become stiffer as the diameter increases. Imagine a 10 cm thick line. It can simply not be bent back well on itself - there is too much difference in the distance the outer part of the bend needs to travel compared to the inner (the inner portion can simply not compress enough to accommodate this). Let you mind go further and consider a metre wide line. How well will this bend? Then consider a thread. A very sharp turn is possible.

One unusual characteristic of unsheathed Dyneema is that it can be bent back at acute angles. Hugely in its favour. Imagine how much worse it would perform if this were not the case.

Thanks for the encouragement. The work I do now will just be ground work. It will perhaps give me something interesting to work on when age or ill health make me a ‘landsman’ again.

44

It is ! :slight_smile:

“Stiffer”, is a measure of the force you need to bend them. The wider the diameter, the bigger the size, the greater the force you need to apply, to bend a rope. I claim that, all other things remaining the same ( the braiding of the individual fibres, for example ), the size and the stiffness scale up uniformly.

You are still talking about the force required to bend a rope. This force is relative to the size of the rope, of course. The difference in the length of the paths of the inner and outer threads scale up in exactly the same way as the size of the rope.

Now you shift the goalpost again ! :slight_smile: You mention a property of the material, not of the geometry…Dyneema does the same thing in small sizes and in large sizes, so the size of the line is irrelevant. I made a comment on your post, where you said that :

Actually, I believe that, on a boat, it may be easier to perform tests of knots than on land - where you have so many other much more seductive/important things to distract your attention and consume your time ! :slight_smile: All you need is a long ram hydraulic bottle jack ( I do not think that the mechanical winches are suitable for strength tests…), your old lines, or old lines of other boats you will encounter in your journey, ( which may be weaker, but, if they are not worn locally, they will generally behave like just ropes of a smaller size…), and patience ! :slight_smile:

45

Just coming back to this issue, I tested the single lapp bend using the same single line yesterday. I literally burst out laughing at the ease in which it slipped on the winch (I never imagined knot testing could be so funny)- it slipped like a hot knife through butter. I think has I tested this in thin line (still keeping the same ratio of 10:16 for the line diameter) the results would have been very different.

46

Well, let us agree to disagree then.

And how exactly would I secure a hydraulic jack? Or have the power to run it? Decks are aluminium. Winches are reinforced with extra aluminium underneath. And how well would a hydraulic jack survive in a wet salty environment, where decks regularly have salt water streaming over them? And if it was dismantled after each test, where would I secure this heavy item safely so that it wasn’t a flying object the next time I hit bad weather? I live in a space that is probably not much bigger than most people’s kitchens :).

47

Why ? If the material and the construction of the ropes were the same, the geometry does not change ( apart from the metric properties, which scale up or down ), so the results would have been the same.
However, this may not happen, because, as the size of the rope increases, usually the pattern of the braiding of the rope in the core and at the sheath changes ( so, the strength of the core and the and the friction at the surface of the sheath changes ), and the very composition of the materials used for the rope ( the ratios between the quantities of the usually more-than-one kinds of fibres, for example ) changes, too - for various reasons, which we can only speculate : rope construction is a black art and a classified science ! :slight_smile: So, it is a wise thing, indeed, to do the tests of knots directly on knots tied on ropes of the most common sizes, in the most critical applications ( as rescue and climbing, for example ). Otherwise, the extrapolation of the results of tests on knots tied on ropes of smaller sizes to knots tied on ropes of bigger sizes would be debatable, indeed.

48

You do not have to “secure” it somewhere ! :slight_smile:
A hydraulic bottle ( forget the “jack” word, which conveys the impression that it should be placed vertically, on a floor ), can be “free floating” : you can just wrap the rope around it, so the forces would be compressible, and would be applied only on the body of the jack, not on the boat. You will only need to attach two pulleys in each of its two ends, of a diameter that will allow the rope ( and the bend on it ) to turn around, without contacting the cylinder.
With hydraulic jacks, you do not need “power” - just a manual pump, which, along with a good breakfast, can offer you all the power you will need ! :slight_smile:

49

I used this way to test some HMPE slings I had made, just wrapping them over the jack. None of them broke, neither did they slip, and when I struck them with a metal object they gave a high-pitched “ping” telling me, that the tension was indeed very high.

The jack i used had a safety valve, so that at about 3.5 “tonnes” the fluid recirculated.

I was unable to break polyester or polyamide this way, as it elongated under tension more than the jack could move. However I succeeded in jamming a few knots severely.

Back to the subject of the sheet bends, I too have found that properly used, the single is often less prone to fail than the double. However, for a long time I haven’t used any of them, as very seldom a bend is needed, and at those occasions, the Carrick Bend is faster to tie, more secure, and easier to open after heavy load. This is also the main reason why I wouldn’t use the Zeppelin Bend, as it is not so swiftly tied. It is however amply secure. So my solution was an alternative, not trying to improve the sheet bends. Otoh, as a becket hitch, I use any of them, single or double, depending on thickness difference.

One important advantage of the Carrick Bend is that it is as easily tied in ropes of large diameter, and that it needs not, or rather should not, be worked tight. It takes the correct form inevitably under load if left to its own devices. So any “proper dressing” can just be forgotten. It does dress properly all by itself. A boon is that it never jams.

I use my own method of tying the Carrick Bend, reminiscent of the Weaver’s method of tying the Sheet Bend:
http://web.comhem.se/~u77479609/Carrick_Bend.pdf

http://www.youtube.com/watch?v=wTmwqYQI93Y

Like any choreography, it takes a bit of training to get the movements into the memory of the muscles, but once learned, just like cycling or swimming, it isn’t forgotten.

50

If a single-stage hydraulic cylinder is not long enough ( nylon, for example, is a very stretchy material, and some more complex knots may consume some unpredictable amount of ropelength before they reach their final, most compact forms ), one may use a two- or even a multi-stage telescopic cylinder. Due to their construction ( extended areas of contact between the pistons, more seals, etc ), such cylinders are not so robust as the single-stage ones. However, even taking into account that the force required to break a, say, 1/2 inch line of the sling wrapped around the two pulleys is more than twice the MBS of the single line ( the mechanical advantage of such a simple machine works now as a dis-advantage, and the pulleys themselves, as Evans had noticed, by their friction, “protect” the line a little bit ), it is still well within the range a quite cheap telescopic hydraulic cylinder may reach.
Now, if one wishes to dispense with such a telescopic cylinder able to deliver twice the MBS of the rope, he may use a less powerful, smaller pull-back ram cylinder instead - but then he will need two firm anchor points to attach the one end of this cylinder and the one end of the rope. On board of a small sailing ship, he will have to use an anchor s really looong shank ( cruising ships do not carry such long anchors nowadays …), the boom ( and, perhaps, the boom vang itself, if it is strong enough, attached along the boom ), the lower part of a keel-stepped mast, or the reinforced part of bilge near the keel for this. I would nt submit any other part of the ship s structure ( the deck, for example…) to such heavy, repeated loadings.

51

Xarax, thanks for putting thought into your suggestions, but they are totally impractical for a yacht on the water. Even at anchor, swell can rock a boat from side to side violently. If another vessel passes and the wake hits beam on (even tankers or cruise ships in the far distance), unsecured objects can go flying unexpectedly.

Some of the points you suggest for securing the line during testing would be pitifully inadequate, and for other more secure points I doubt if any boat owner would be willing to deliberately subject them to loads like this.

By the way, I can’t imagine how you could use secure an anchor adequately to use its shank (and what would be holding you stationary if it was up on board and not on the seabed :)), and I’m not sure what you mean by cruising ships not carrying anchors with long shanks - ours is nearly a metre long :).

But your suggestions may be useful for anyone contemplating setting up some kind of system at home.

Performance is king when it comes to practical knots. Factors such as time taken tying, ability to do this single handed or in the dark and ability to undo the knot after load has been applied may be important, but holding ability is often paramount when it comes to assessing performance.

Load tests are really the only way of assessing this.

52

I have heard this several times now. May I ask how you tie the double (#1434 or #488)?

Have you used the Carrick and Zeppelin for lines of different diameter? I haven’t been game.

Yes, I agree the Carrick goes into muscle memory easily.

53

Which points do you mean ? Those I mentioned are adequately strong, in any boat large enough for a couple to live in, I guess. I was replying on you comment about the “aluminium deck”. You do not have to attach your test rig on the deck at all.

It depends on what do you mean by “cruising ships”. You had described a boat with a living space “smaller than a kitchen”, and that made me imagine a small boat, not a large kitchen ! :slight_smile: :slight_smile: :slight_smile:
So, you live on a big enough boat ( larger than 60-65 ft LOA, I suppose ? ), on board of which I am pretty sure you can secure a simple ad hoc test rig like the one I suggest - even with some difficulty. You can possibly use this or any other long and strong enough shank as part of this rig, indeed, and attach the line at its one end and the body of a short pull-back cylinder on the other. Or attach your cylinder along the boom, or along the lower part of the mast. On such a boat, the boom and the mast will not suffer from the loads we are talking about - they are designed to withstand much greater strains. ( I used to sail on a Sun Odyssey 52 ft boat, so I do have some idea of the dimensions and the allowable loads of those parts I suggested you attach your test rig on ). Anyway, to prove that you are worth your salt, and truly “seaworthy”, find a way to perform load tests of ropes and knots on board of your boat - it would be an interesting challenge per se ! :slight_smile:

54

I stand uneasily on both sides of this debate! :-\ :-\

On the one hand, the argument(s) that X. advances
seems perfectly understandable. OTOH, I have SEEN
the odd cases --frequently, at one site (trawlers)–
of capsized bowlines in mooring lines, and yet that
knot elsewhere, in other lines much thinner, doesn’t
do this (or do it so seemingly readily)?!? And whereas
I had thought to explain this by Seaworthy’s assertion
of greater stiffness in the larger ropes, in fact most of
those ropes seem surprisingly suppler --more so, e.g.,
than low-elongation caving ropes (e.g., BlueWater II)
and rockclimbing ropes. (We can note that Mark Gomers’s
article on bowlines shows a BW II knot that has gone
some way torwards capsizing, but still hasn’t.)

So, I think that this is a puzzling question.
(As for HMPE’s flexibility (at least in some forms),
that I think partly aids its slipping.)

–dl*

55

Always 1434.

I think the spilling problems arise from the different diameters of the lines, that the single is never used for such large differences as the double sheet bend.

I have done security trials with both the Carrick Bend and the Zeppelin with different diameters, to see whether they can be used for such bending, and the Carrick Bend works well when the difference is not very large. So for two lines of say 3/4" and 1/2" a Carrick bend is secure, while it cannot be trusted for joining a 5 mm line to a 12 mm line. The Zeppelin is quite another thing. To join different diameters, just an extra turn with the smaller line makes it secure.

There is an earlier thread on this topic: The asymmetric Zeppelin Bend

In fact, it can even be used for a ridiculous difference in sizes between the ropes, as can be seen in post #13 in that thread.

http://igkt.net/sm/index.php?action=dlattach;topic=2159.0;attach=2280;image

56

If you are ever playing again with securing lines of different diameter could I get you to try #488 that I have always used? I have only recently become aware that #488 will not tighten into #1434, as I had always assumed it would want to. From test results I have performed #488 does seem to slip considerably less (not always indicative of the final strength of a knot, but it is a good start).

I think I have a good contender then for tying lines of significantly different diameter. The Zeppelin with an extra final turn seems to be cropping up again and again as an option. This is really where I (and probably thousands of other sailors) flounder when tying lines. I think none of us feel confident with the double sheet bend for this purpose, although it is a great quick option when similar lines are being joined.

PS Good photo!

57

Hi Seaworthy!
You ask if maybe the old salts had some preference as to which of the double sheet bends they prefer. I think they did, and they seem to agree with your opinion. I looked it up in Sam Svensson’s “Handbok i sj?mansarbete” (title means handbook of sailor’s work) where he puts the two double sheetbends in pictures next to each other for easy comparison, and Svensson says that ABOK #488 (the one made with one roundturn followed by one turn tucked under) is superior to ABOK #1434 (the one with both turns tucked under) for the following reasons: It is easier to tie it; it is is not as bulky (it is more elegant); it cannot jam; and it is always easy to untie. However, Sam Svensson says, the ABOK 1434 is preferable when the difference in size is very large; then you can make more turns (3-5) of the thinner line around the thicker bight (here he apparently means turns that are tucked under the incoming standing part; I like to think of them as “tucks”, even if thay involve roundturns as well). Svensson doesn’t discuss how secure they are, so I guess they both hold well enough in hemp or manilla. Svensson does not refer to ABOK numbers, but I added them above, for clarity.

These two knots are, as you have noticed, so different that it is impossible to transform one into the other by any kind of rearrangement or “dressing” of the parts inside the knot without untying it first. They should have different names since they really are different knots. Maybe #488 could be called “singly-tucked double sheetbend” and the other “twice-tucked double sheetbend” or something along those lines. In that way, when someone talks about the double sheetbend, we can ask them to specify if they mean the single-tucked or twice-tucked double sheetbend.

58

I think that I agree - but then, by the same token, should the 2 pairs of the different “Eskimo” bowlines have different names ?

59

I don’t know. What are Eskimo bowlines?

60

There are 4 variations of what most people call " the Eskimo bowline ", which are, more or less, functionally ( as rope mechanisms ), geometrically and topologically different from each other.
http://igkt.net/sm/index.php?topic=3233.msg23797#msg23797
http://igkt.net/sm/index.php?topic=5015.msg33060#msg33060