Non slipping bend in Dyneema

[estar]

I have a bend that does hold - it's a modification of the blood knot (pictures here:http://www.bethandevans.com/pdf/bend.pdf).  It is a bit more complex than is ideal, but it is "low profile", and is the only bend I have tested that does not slip.

It is 100% in 'loop configuration' (as I was reporting all the other bend numbers - not sure if I made that clear), so 50% as a straight knot (that's two pulls). I think 50% is roughly as good as it gets in dyneema.
 
There are two slightly different ways to tuck the tails back - one way they slip at about the same loads as the triple fisherman, and the other way they hold, and unfortunately I am not sure which way I photographed.  I will have to do some checking tomorrow.

Also, two Estar's might be quicker and easier to tie than this, and we know they don't slip and are strong and low profile

[allene]

Estar just indicated that he has been testing loops so it seems like my knot is not 68% but rather 34% as it is only taking half the force because of the loop.  The blood knot above is 100% or 50% on the knot.   I have asked for clarification but wanted to alert any readers here that the strength has most likely been overstated. 

You should get some of this line and try and tie a knot in it and see if it slips.  I think it would be useful for you to understand how slippery it is.  I will send you some if you like.

Allen

[allene]

  What would you suggest we test next?

   I have already shown the Strangle bend ( http://igkt.net/sm/index.php?topic=4756.msg30732#msg30732 ) - so we go one only step beyond the Fisherman s knots. I try to compare smaller apples to larger apples -  this is why I had made this suggestion. Next, one more step further from this, where the wraps are interweaved a little more, is the also very simple and easy to tie Trefoil bend, shown at the attached pictures. Test them, and then we will see. I will not present you some really big guns we have  - we will use them later. The true reason is that I try to figure out the Achilles" tendon of this material... By comparing results of tests on similar bends, I believe will gradually be able to understand more things.

   There are members of this Forum who have used Dyneema ropes, and know some things about knots tied on this material. Personally, the only thing I know for sure is that the recommendations for the use of the triple Fisherman s knot are quite common - I believed until now that this bend was extensively tested on exactly the material you cite. I have even tried to guess some more convoluted bends that could serve as substitutes of the triple fisherman s knot - and, to read my hand-weaving, vague arguments ( in the absence of a test rig, I could nt do anything else...), and get an idea of some big guns, see :

http://igkt.net/sm/index.php?topic=3668

I have completed some testing using a 500 pound test version of this line called "Lash It"  I am able to make loops with that so that I can pull hard enough to cause the knot I submitted to slip. When I used figure of eight loops to terminate my knot, the figure of eight knots would break before my knot slipped.  The funny thing is that when it slipped, a puff of smoke came out of the knot!.  I then made a loop with the trefoil knot and interlaced with a loop with my knot.  Kind of a tractor pull to find the strongest knot.  The trefoil slipped without any noticeable slippage on my knot.

Allen

[Dan_Lehman]

Estar tested the above knot and in his test, unlike mine, it slipped ...

Before my sleepy eyes fail fully,
IIRC of EStar's on-line documentation,
his supposed zeppelin bend was NOT that,
{{edit to correct: EStar has it right, at www.bethandevans.com/load.htm
   so, "sleepy eyes" was seeing things!
   But awake eyes now sees there the "tails opposite" version
   of the sheet bend, which is generally regarded
   as the inferior version --tails should be on same side.
   So, its slippage in this version is less a surprise.
   Also, I'd never orient the butterfly like that;
   a better orientation should see higher values.}}
but a sadly too common misrepresentation
of it.  The initial structures of the two ends
should form what can be seen as "p & d"
not "p & b" --the draw on the loaded ends
will rotate the nipping turns in the same
direction, not opposite.

NB: the butterfly eye knot is asymmetric,
so testing of that should indicate which end is being
loaded; and the dressing of the knot as I see it
most often is not what I'd expect to get more
strength (but I must admit that the good strength
it sometimes gets, thus, doesn't leave much room
for improvement!).  In forming the eye knot by
what has been jocularly called "the twirly flop"
method, there is a natural tendency by torsion
for the eye legs to cross in their exiting
of the knot, not be pressed against each other
in a plane including the axis of tension.
(In introducing this knot to climbers back in 1928,
Wright & Magowan specifically pointed out this
crossing as desired; I don't recall that they had
a testing basis for so desiring, but maybe just
a sort of "go with the flow" pointer, to say "yes,
it should go like this".)

As for slippage of the double grapevine (a name
I favor for it better matches to parts of the knot
--the "double", i.e.), Xarax's alarm should be ameliorated
by realization that those who make the recommendation
for its use do so on the basis of their testing the
material in which it's recommended to be tied
--i.p., HMPE-cored, polyester-sheathed small cord.
And they have break tests of that, not slippage.
YMMV, again.
(I sure would like to see video of it slipping,
though, as seen for the dbl.bowline "#1"!)


Thanks much for all the testing & exploration!
Cheers,
--dl*
====

[Dan_Lehman]

I have a bend that does hold - it's a modification of the blood knot
(pictures here:http://www.bethandevans.com/pdf/bend.pdf).

NB: what you show (as do many others, I know)
is not a blood knot, really, unless you get
those turns to capsize into wraps around straight
S.Parts (i.e., the loaded ends)!  This is sadly something
lost to the knot-parroters who've seen such images
in fishing-knots books and taken them into rope
and ... the transformation might not occur.
It's a distinction of what Barnes called "in-coil"
& "out-coil"; in nylon fishing line, upon setting,
the result is the same --in-coil.  (And, surprisingly,
there is enough force transmitted into the wraps
to leave indentation impressions on the S.Parts,
with break coming at the center of the knot
where the tails caused a deflection --and not
at the hard (1-diameter) turns of the S.Parts
where they turn back & wrap!)

I suggest that you endeavor to get this arrangement
in the HMPE line, perhaps must adding a 2nd tucking
of the tails, for security?

You've essentially tied extended anchor/fisherman bends
of each end around the other.

It is 100% in 'loop configuration' ..., so 50% as a straight knot (that's two pulls).

I take it that you are making a statement of
equality here, not reporting separate test results?
Beware this apparent (logical!) equality, as it might
be more the case that the naive adding of the 100%
side and the whatever-% knotted side is more to
the truth of closed-loop sling strength, practically
--the point being that with knot compression that
side will lengthen and lighten its load, and equalization
around the pins or whatever is pulling the sling apart
might be inefficient and so indeed the unknotted side
could shoulder a bigger portion of the load, and both
sides reach their resp. maximums, say 100% + 60% !
(I believe I've read reports where the break came at
the pins, not at the knot, too.)

I think 50% is roughly as good as it gets in dyneema.

If not better than it gets!   

--dl*
====

[allene]

I can  only test one thing against another so that is what I do.  I am using smaller line now but not mixing results, all results I give of one vs another are in the same size line.  I am not picking line for any other reason than this is what I can test.  That said, I tested my knot against the four knots here http://igkt.net/sm/index.php?topic=3668 and mine is in the middle.  The first two knots slipped first but in the second two, mine slipped.  I did not test the last two against each other as I find them just too hard to tie to be practical on a boat.

PS.  This is a reply to post #29.  Sorry, I am not familiar with how your forum works.

[allene]

   Estar, when, in the files I had send to you, you find a bend that slips below 50% of the MBE, just throw it out of the window ! 

Just a note that Estar is testing loops (that is what we call them) so when he says it slips at 38% of line strength, the knot is slipping at 19% of line strength.  This is slippery stuff.

[estar]

I am a bit less enthused about the modified blood knot this morning - I have tested 12 pulls, and it slips (at about the triple fisherman slip load) in about 1/3 of them (that may or may not be due to the capsize issue mentioned above).  Practically speaking it will be easier and more secure (and equally compact) to tie two EStar loops, so for now that is the best solution I have found.

As to those suggesting we look at 'simpler' bends . . . I would love to find one that works . . . but I have now tested quite a few and just can't see how we are going to find one "simpler' than (for example) the triple fisherman that will work.  The EStar is the "simplest" knot I have found that holds securely in this stuff.  It is my benchmark for now.

The EStar tests at 50%.  The only thing I have tested higher is the Polamar at 54% (Just barely statistically significantly stronger). But you cannot (easily) tie the Polamar as a bend.

For those questioning the triple fisherman slipping.  I have a series of photos below. #1 is no load, #2 is 1300lbs of load after it is all tightened up and the tails start slipping (I put some new black marks on the tails at this point to watch them being sucked into the knot), and #3 has one of the tails almost into the knot.  It does this every-time, consistently.

[allene]

the Strangle bend - shown in the attached pictures - , because the wraps of the one link squeeze the ones of the other. Could you, please, try to pull this bend - because I am dying out of curiosity ! Of course, I suppose you would set and pre-tighen the bends a little bid, before the final loading.

The strangle bend slipped against my knot.

[allene]

Diamond Bend

Diamond bend slips against my bend.  It also took me 15 minutes to figure out how to tie it.  I must say that this class of knots, at least with just trying to tie it by following these pictures, could never be tied on a moving boat at sea.  Perhaps there is an easier way but as they slip against a very easy to tie knot there really isn't a point.  That, of course, does not apply to the two knots that did not slip against my knot so if there is an easy way to tie those two I would love to know what it is.

Allen

[roo]

^^ it is unclear what solution is being suggested here

It's in the linked post:

http://igkt.net/sm/index.php?topic=2013.msg14229#msg14229

Sandra  made a diagram of the tying procedure for this extended Stevedore bend (if you wish to call it that) later in the thread. 

Strangely, it can be untied when needed in rope.  Again, simple but obese.

[allene]

  Those knots are complex in their form, but simple in their tying - and, as symmetric, instantly inspected. The idea was to re-trace the path of a symmetric stopper. Just tie the stopper, as the one link of the bend, on the end of the one line, and retrace its path, without any further thinking, with the end of the other the opposite way - that is, just follow what the first line had followed as it closed in itself.

That is how I tied them.  I call them retrace knots, like a water  knot.  It is probably harder with the line I am using which is very small.

I still find this category of knot difficult to tie.  The benchmark is a bowline, which I just tied one handed behind my back without looking.  The bend I proposed takes two hands so it isn't as easy as a bowline.  But  you need to be able to tie  a knot in your shower standing on one of those exercise balls with the water on of course :-)

Allen

[allene]

^^ re-tucked water knot slipped against my knot.

[Dan_Lehman]

I am a bit less enthused about the modified blood knot this morning
--I have tested 12 pulls, and it slips (at about the triple fisherman slip load)
in about 1/3 of them
(that may or may not be due to the capsize issue mentioned above).

What I mentioned about "capsizing" / transforming
cannot occur in your knot with its tucked tails;
rather, it should occur --a design goal, i.e.--
for the blood knot, and hence my suggestion
for revision (tie "in-coil" (working end wrapping back
over the overlapped S.Parts to the center tucking
point), and make an extra tuck in hopes of locking).

For those questioning the triple fisherman slipping.

NB : what you show here is the triple grapevine hence
quadruple fisherman's knot !  (And we can see
why I favor the former name --it matches the visible
wraps (triple ; a (single) fisherman's knot has none,
not one).  The pre-loaded knot looks as though it could
be better tightened?  Still, the slippage is so eye-opening,
knots-thinking-shattering!!  Thanks.

Note that the tails of these knots could be taken
through their opposite halves; I've no idea if this
would enhance security.  It certainly would complicate
tying the knot, as one cannot form one half and then
the other, as this novel tail-tucking needs part of the
"other half" to be formed, into which to tuck.


Let me suggest a relatively "simple" knot --the base
is the overhand : Ashley's #1452, sometimes presented
as Ashley's bend.  I'll ask that it be tried, as is,
and then with the simple securing measure of putting
in a 2nd course of tails tucking --which bulks its central
turning-around mass to four diameters, and puts
double collars against the SParts' entry.
((Sorry, don't have a good image for that in 100kb
or less, so let's look at the similar Ashley's #1408.))

I can't imagine this slipping (to become untied ; slipping
to generate heat on rapid loading and ... , maybe);
but then I couldn't imagine a multiple grapevine
slipping, either (or the rope flowing out of a dbl.bwl
from collapsing the eye!).

From the original knot shown in the attached photo,
just take each tail and *trace* the finish of the
opposite tail --e.g., the white tail will turn clockwise
and lie atop the yellow tail in making the 2nd tuck;
the yellow will reciprocate, running counterclockwise
behind the white tail's initial tuck & exit.  (And this
extension --the 2nd tucking-- can be tweaked to
try to work out the best version, if it looks fruitful.)


--dl*
====

[allene]

   Thank you. You keep testing everything against your knot, and not against each other, and doing so you deprive us from valuable information

I tested the two knots that made mine slip against each other.  I made a loop of each knot and larks head knotted the two loops together.  It broke at the larks head.  I was not going to re-tie those things so did not report it.

I tested the strangle against whatever knot you asked me to.  One of the knots suddenly exploded and the test piece fell.  I am reasonably sure the strangle survived but not 100% so again I did  not report this.

These knots are pretty hard to tie and not suitable for sailing imho so my motivation to do multiple tests when I screw up is limited.

Allen