Author Topic: Alternatives to the triple fisherman s bend.  (Read 9412 times)

xarax

  • Sr. Member
  • *****
  • Posts: 2781
Alternatives to the triple fisherman s bend.
« on: October 29, 2011, 05:16:06 PM »
  Cords and ropes made from UHMWPE fibres (1), ( Dyneema, Spectra and the like), are very slippery, and the manufacturers that produce them - with no exception - advice the use of the triple fisherman s bend. Of course, any sufficiently convoluted bend would do the job, but such bends are often quite bulky, difficult to tie and dress correctly, and unknown !
   I have played with some easy-to-tie bends, that are based upon the retracing of simple "marginal"  knots ( simple single-line links ). I have not tested those bends with ropes made from UHMWPE fibres, so I can not tell if they offer any improvement over the "standard" triple fisherman s bend - but I hope that somebody out there will, and inform us about his findings.
   In fact, all we have to do is to chose a single-strand stopper, lanyard or button knot, and to retrace it with the second line. Some simple examples of such bends, with their corresponding marginal knots, are shown in the attached pictures.

1.  http://en.wikipedia.org/wiki/Ultra-high-molecular-weight_polyethylene
« Last Edit: October 29, 2011, 09:52:36 PM by xarax »
This is not a knot.

xarax

  • Sr. Member
  • *****
  • Posts: 2781
Re: Alternatives to the triple fisherman s bend.
« Reply #1 on: October 29, 2011, 05:17:36 PM »
2.
This is not a knot.

xarax

  • Sr. Member
  • *****
  • Posts: 2781
Re: Alternatives to the triple fisherman s bend.
« Reply #2 on: October 29, 2011, 05:18:52 PM »
3.
This is not a knot.

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 3754
Re: Alternatives to the triple fisherman s bend.
« Reply #3 on: November 01, 2011, 07:32:03 PM »
  Cords and ropes made from UHMWPE fibres (1), ( Dyneema, Spectra and the like), are very slippery, and the manufacturers that produce them - with no exception - advise the use of the triple fisherman's bend [aka double grapevine -dl*].

One should be specific, here.  I presume that the cordage at issue
is the narrow segment of rockclimbing / caving / <etc.> kernmantle
rope users, with the typical (if not universal) material being a 5.5mm
cord with HMPE fibres only in the core.  (Actually, there are some thicker,
"main" ropes used for canyoneering that have, IIRC, hi-mod fibres in
both the mantle & the kern (Technora, also to HMPE (Dyneema/Spectra));
but such ropes are not typically knotted together, in any way.)
This is in contrast to ropes made entirely of the fibres, such as can
be found in yachting and some other, commercial applications
(where knots are simply advised against).

What can happen in these ropes with the slippery (and no-stretch/static)
core fibres is that the core slides within the sheath, leaving load-bearing
to the latter, and weakening the knot, thus.  On rupture of the sheath,
it has been seen that the core just then pulls out of the knot,
like snake shedding its skin. (!)


Quote
Of course, any sufficiently convoluted bend would do the job,

If the job is merely staying tied, yes, this is likely.
But typically such cords are tied end-2-end to form slings,
and the knot can see some incidental knocking/sliding about,
where abrasion resistance is wanted.  Also, these knots are
pretty much considered permanent, and the knot should be
secure when slack, even jamming.  Not every convolution
of cordage will meet these criteria.

Quote
I have played with some easy-to-tie bends, ...
 I can not tell if they offer any improvement over the "standard" double grapevine bend
...

I can tell you that none of these presented end-2-end knots
(and note that each image shows TWO possible such knots,
as you do not specify which ends are to be loaded (well, make
that THREE, if we include offset  loading!) )
satisfactorily yields compactness and abrasion resistance
like the dbl.grapevine --and likely they're weaker.

It is odd to see you present a traced strangle knot ,
as that knot is the component in the grapevine bend which
is the end-2-end joint presumed dubious in this material!
With the slightly broader curves of the loaded strands of this
traced knot, I'd expect the security to be less, not more.

An end-2-end joint I'd like to see tried is some version of
the blood knot.


--dl*
====


xarax

  • Sr. Member
  • *****
  • Posts: 2781
Re: Alternatives to the triple fisherman s bend.
« Reply #4 on: November 01, 2011, 10:12:55 PM »
   Thank you Dan Lehman,
   The difference you mention, between all-UHMWPE ropes, from the one hand, and ropes with a UHMWPE core inside a sheath weaved around it, -made by a less slippery material-, from the other, is great indeed. I can not even imagine which one of the two types would slip more or less inside a particular knot, so I can not tell what we should seek for each case : more turns, tighter turns, alternating left and right turns...
 
...but such ropes are not typically knotted together, in any way.
...(where knots are simply advised against).


  That does not mean we should not find out and know which knots would hold if tied with those materials, and which not - even out of a pure "academic" interest.

..note that each image shows TWO possible such knots,
as you do not specify which ends are to be loaded (well, make that THREE, if we include offset  loading!) )


  Due to high symmetry, thiis is not true for some of the knots presented here...Anyway, we should test all the possible loadings, because we can not tell, in advance, which one would be more and which would be less secure than the other, can we ?
 


This is not a knot.

xarax

  • Sr. Member
  • *****
  • Posts: 2781
Re: Alternatives to the triple fisherman s bend.
« Reply #5 on: November 02, 2011, 03:42:33 PM »
   I can tell you that none of these presented end-2-endknots...
satisfactorily yields... abrasion resistance like the dbl.grapevine --and likely they're weaker.

   How ? I mean, how do you know it ? I could possibly claim the exact opposite - and just add a "likely"  in my claim, 2-B-sure... :), - but I could not prove  my claim, could I ? We need tests, and to get tests, we need people to make those tests !  :)
   If you are going to test those knots with UHMWPE cords and ropes, you could possibly also test the simple "extension" of the double eight bend shown in the attached picture - just in case it would be sufficiently convoluted for this purpose.
This is not a knot.

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 3754
Re: Alternatives to the triple fisherman s bend.
« Reply #6 on: November 03, 2011, 07:54:04 PM »
..note that each image shows TWO possible such knots,
as you do not specify which ends are to be loaded (well, make that THREE, if we include offset  loading!) )


  Due to high symmetry, this is not true for some of the knots presented here ...

Wrong : it is true for exactly ALL of the R1..R4 knots presented above!
(Note exact relations, e.g., of twin parts per entry/exit point.)

--dl*
====

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 3754
Re: Alternatives to the triple fisherman s bend.
« Reply #7 on: November 03, 2011, 08:00:43 PM »
   I can tell you that none of these presented end-2-endknots...
satisfactorily yields... abrasion resistance like the dbl.grapevine --and likely they're weaker.

   How ? I mean, how do you know it ? I could possibly claim the exact opposite - and just add a "likely"  in my claim, 2-B-sure... :), - but I could not prove  my claim, could I ? We need tests, and to get tests, we need people to make those tests !  :)

One can tell this simply by examining the shape of the knots.
The "barrel" shape of the grapevine series is obviously robust
in resisting abrasion vs. any strand as the strands are tightly
compacted together at nicely rounded, equal surfaces; only
some pulling-into (stopper-like) situation is going to get some
direct impact against the highly loaded SPart(s).  In all of the
above cases, there are shapes more irregular, more bulging
in one way or another (with the possible exception of the
traced strangle knot).


--dl*
====

xarax

  • Sr. Member
  • *****
  • Posts: 2781
Re: Alternatives to the triple fisherman s bend.
« Reply #8 on: November 04, 2011, 12:07:22 AM »
... the strands are tightly compacted together at nicely rounded, equal surfaces; only some pulling-into (stopper-like) situation is going to get some
direct impact against the highly loaded SPart(s).

  Thanks. My approach was more naive, perhaps because I have not the knotting experience that would allow me to attempt such "theoretical" predictions. I thought that, to tie a secure bend for very slippery materials, we can just make one step upward in complexity, - in relation to some ordinary simple practical knots -, and keep retucking the resulting bends, until we reach a sufficiently secure one. How will we know that we have already reached such a bend ? By trial and error, by testing the thing using UHMWPE ropes. And how do we stay within the limits of easy-to-tie practical knots ? By tying bends made by retracing relatively simple, symmetric knots.
   So, if those bends are not more secure than the triple fisherman knot, we simply should go to the next level - and then to the next, etc.

Wrong : it is true for exactly ALL of the R1..R4 knots presented above!

Correct : although the differences of the "two" R1 or R4 bends are slight, strictly speaking, they exist. In all those bends, I guess that the more interesting variations are the "non-offset" ones where, when they enter the knot s nub, the standing parts are closer to its longitudinal axis.
This is not a knot.

xarax

  • Sr. Member
  • *****
  • Posts: 2781
Re: Alternatives to the triple fisherman s bend.
« Reply #9 on: March 03, 2014, 05:46:54 PM »
   Although one might argue that the two parts that compose the Eternity knot (1), or the similar ABoK#583, are not interwoven sufficiently within each other, and so they do not form a tight whole, the retraced knot will not suffer from this : in each link, the one part is pulled and the other is pushed towards the centre of the knot, by the pulling and by the pushing action of its own or of the other link s Standing part, respectably.
   I do not see many differences, regarding the general pattern, between this knot and the other knots presented in this thread, which are based on well known knots. Moreover, the more sharp U-turns the Standing parts of the two links follow may be considered an advantage, in comparison to the other retraced stoppers, where the Standing parts trace wider, smoother curves. I have came to believe that this - the many alternating, meandric U-turns inside the knot s nub, the many 'inner collars" - might be beneficial on the working of the knots when they are tied on Dyneema, but this is only a conjecture - only detailed tests can tell if this is a valid theory, or just another arm-chair knot tyer s gut-feeling that turned to be wrong. Knots are complex machines - we can not analyse them as much as we would had wished, and that is why we can seldom predict their behaviour...

1. http://igkt.net/sm/index.php?topic=4807.0
« Last Edit: March 16, 2014, 07:08:35 AM by xarax »
This is not a knot.

shadowjack

  • New Member
  • *
  • Posts: 6
Re: Alternatives to the triple fisherman s bend.
« Reply #10 on: May 12, 2014, 06:53:26 PM »
I especially like extended fig. 8 - I think it will have good lead (compared to grapevine). I do not have any UHMWP cordage to test it - it would be interesting.