International Guild of Knot Tyers Forum

General => Practical Knots => Topic started by: xarax on September 15, 2010, 05:37:52 PM

Title: 4 ends bend. (Another interesting knot problem)
Post by: xarax on September 15, 2010, 05:37:52 PM
0
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: Sweeney on September 15, 2010, 05:46:55 PM
How about the Zeppeilin Bend? The Zeppelin Loop has the strain on 3 different ends and is structurally similar to the bend so placing a strain on all or any of the 4 ends should be reasonably secure (and the 4 ends are at right angles which may help).

Barry
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: roo on September 15, 2010, 06:00:17 PM
  Tie two ropes with a bend that can wthstand loading/pulling from any two of the four ends.

   ( The first thing that comes to mind is two intelocked cow or clove hitches. There should be many simple, but more secure solutions, I guess.)
  
What's your application?

This may work:
http://notableknotindex.webs.com/midspan.html

The problem with a Zeppelin Bend is that pulling the free ends hard can cause a jam.  In the loop, one of the free ends of the Zeppelin Bend structure only sees 50% of the parent line tension.
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: roo on September 15, 2010, 07:04:27 PM
There are the so called bungee/"occy" straps that are very popular - and very useful in a variety of applications, indeed. I have thought of a knot capable to handle similar situations, where one wishes to wrap one or more objects from many sides, or to connect many points together, not knowing in advance which of all those rope ends will remain loaded and tensioned. ("Occys" are named from the 8 octopuses ends that some of them have. We better solve the problem in the simpler 4 ends case first... :))
 
I'm not sure I understand your application.  Could you explain it another way, or can you find pictures?

If it's a matter of connecting many bungees to coincide at one point, one could work with the ends of the bungees.  Start with a Zeppelin Loop at the end of one bungee.  Then, link into the starter loop with other bungees using their own Zeppelin Loop.
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: knot4u on September 15, 2010, 08:53:43 PM
If I just need the knot to work here and the only problem with the Zeppelin Bend is risk of jamming, then I'd be inclined to use the Zeppelin Bend.

Other options, how about using connections that aren't technically bends?  For example:

-two Butterfly Loops interlocked
-two Zeppelin Loops interlocked
-two Adustable Grip Hitches interlocked like on this page: http://www.layhands.com/Knots/Knots_Bends.htm

If it's a matter of connecting many bungees to coincide at one point, one could work with the ends of the bungees.  Start with a Zeppelin Loop at the end of one bungee.  Then, link into the starter loop with other bungees using their own Zeppelin Loop.

Same concept I think.
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: roo on September 15, 2010, 10:02:25 PM
The standing ends and the tails of all those loops and hitches are very different in their walk inside the knot body, that is, if one loads the standing end and the tail of one knot, and the loop itself is not loaded -as it might happen in our unknown 4 ends bend - the knot will not work properly, or will jam.  

I don't think the tails or free ends of any of the loops will ever be loaded.
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: knot4u on September 15, 2010, 10:07:29 PM
-two Butterfly Loops interlocked
-two Zeppelin Loops interlocked
-two Adustable Grip Hitches interlocked

The standing ends and the tails of all those loops and hitches are very different in their walk inside the knot body, that is, if one loads the standing end and the tail of one knot, and the loop itself is not loaded -as it might happen in our unknown 4 ends bend - the knot will not work properly, or will jam.  

I don't think that problem would happen with interlocked Butterfly Loops.  In fact, one Butterfly Loop performs better if both the standing end and working end are loaded.  Meanwhile, a Butterfly Loop will perform well enough if only one end is loaded, as long as the the knot is dressed tightly.

I think there are many solutions to your problem in the original post.  If you add the requirement of "elegance", then we need to think harder.
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: Dan_Lehman on September 16, 2010, 05:45:50 AM
   Tie two ropes with a bend that can wthstand loading/pulling from any two of the four ends.

   ( The first thing that comes to mind is two intelocked cow or clove hitches. There should be many simple, but more secure solutions, I guess.)
   

I'm going to assume one unstated condition : that the knot be stable
under any given sequence of loading -- e.g., end-1 versus -2
and then end-3 vs end-1 and then ... should not result in failure,
which is defined as coming untied (but irrespective of jamming).

The KISS principle works here, and the winner is the venerable
Overhand knot -- which becomes, upon the various loadings,
(i) an Overhand (w/another rope engaged),
(ii) a "strong-form" Water knot,
(iii) a "weak-form" Water knot, and
(iv) an Offset Water knot .

.:. / QED.

To just take the iv-th structure, one can tie (in the bight!)
the bights of the two ropes into an Offset Water knot and
have pretty much like loading behavior across all pairings.
(This implements Xarax's "quadrapus" ideal.)

.:. / QED'


It's funny to see the same ol' favorites trotted out as though they
were panaceas.  Rather than Rosendahl's bend, one should first think
of the more symmetric False Zeppelin, which though a bit unstable
in holding form in some orientations, does seem to hold well enough,
and gives just a trio of effective orientaitons:

[ naming ::  Rope-Numeric -: ends "1" & "2" ':- ;  Rope-Alpha -: ends "a" & "b" ]

1 v 2 = a v b
1 v a = 2 v b
1 v b = 2 v a

 - - - - - - -

Quote
The [XYZed] Loop has the strain on 3 different ends and is structurally similar to the bend so placing a strain on all or any of the 4 ends should be reasonably secure ...

This logic is --as my alteration indicates-- apparently applicable to all
eye-knot/ends-joint cognates; it has some obvious, serious failings
if taken e.g. for the Bowline & Sheet Bend.  And these failings suggest
that the Mid-span Sheet Bend be kept out of the game: recall the
issue of ring-loading and the Bowline (becoming in effect the
inferior Lapp knot orientation).

Quote
... two Butterfly loops interlocked ...

(NB, Xarax: SParts & tails are rather indistinguishable here!)

Okay, let's take this solution further, w/Occam's Razor trimming:
cut the two eyes; fuse the cut ends appropriately to constitute
the challenge case of two pieces of rope  (well, yes, one can do
this with a single piece and let the close-ends/bight side pretend
to be separate) -- in short, dispense with the two eyes, for they're
unneeded impediments to knot/material efficiency.  We might
call the result "Back-2-Back Butterflies" (or something more
amorous, but I leave that to Xarax's & Derek's imaginations  ;D ).
Recall what is written recently in a nearby thread:  the Butterfly
transforms into itself in swapping SParts for tails, by flyping!

--dl*
====
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: roo on September 16, 2010, 04:10:58 PM
  And these failings suggest
that the Mid-span Sheet Bend be kept out of the game: recall the
issue of ring-loading and the Bowline (becoming in effect the
inferior Lapp knot orientation).
Wrong.  The Midspan Sheet Bend never assumes a Lapp Knot orientation since the leftover loop is not used.
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: roo on September 16, 2010, 04:18:14 PM

The KISS principle works here, and the winner is the venerable
Overhand knot

Since the original poster has now clarified that he has bungee in mind, I'm wondering if you've tried this with bungee, either with four or eight "arms".
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: Rrok007 on September 16, 2010, 05:46:07 PM
Here's a thought....

And I know this probably goes a little outside of the "simple" field...

Why not use either a partially or completely tied star knot?

The Star knot cause be tied using virtually any number of ropes. I've laid two ropes across each other to make a four-point star knot just as easily as I've used 8 strands to make an eight-point star knot. It would fit your needs regardless of needing a quad or an octo, and an octo can even work like a quad if needed.

The knot itself has enough of a body to form a grip point if one was called for.

I would think (I'm sure Dan can call me out if I'm wrong ;-P) that as long as the body was dressed up tight enough, you wouldn't need to tie a full and proper star knot. I would think just the first few steps (hitching around the adjacent rope, crowning back the opposite direction, then tucking back down through in the original direction, would be enough if it's tighten up right.


Another thought is the Prosperity knot, which, when tied with two parallel ropes, forms an open-ended Ocean Braid Mat.
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: knot4u on September 16, 2010, 07:25:45 PM
Roo, your idea of the midspan sheet bend has characteristics of simplicity and non-jamming.  Also, it seems to solve the original problem.  I've tested it, and it seems to work well enough.  To address security concerns, I propose a Double Midspan Sheet Bend.  I've tested that too.  It's quite secure and non-jamming.

In contrast, your pic in Reply #8 ventures off from the problem at hand.  The original post would like to load any of the four ends of only 2 ropes.  Your pic shows 4 ropes, which is  fine as long as we can load any of the 8 ends.
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: roo on September 16, 2010, 10:24:55 PM
Roo, your idea of the midspan sheet bend has characteristics of simplicity and non-jamming.  Also, it seems to solve the original problem.  I've tested it, and it seems to work well enough.  To address security concerns, I propose a Double Midspan Sheet Bend.  I've tested that too.  It's quite secure and non-jamming.

In contrast, your pic in Reply #8 ventures off from the problem at hand.  The original post would like to load any of the four ends of only 2 ropes.  Your pic shows 4 ropes, which is  fine as long as we can load any of the 8 ends.
Since the topic has shifted from rope to bungee (and eventually eight arms), I had to adapt my response.  Most knots just aren't reliable with slick sheathed bungee.  So I looked to a more brute force solution of Zeppelin Loops in a hub system.  To most people, a reliable solution will win out over the number of lines used.  Further, the longer the lines, the more attractive is a solution that only uses the ends of the line if a purely on-the-bight solution isn't practical.

Another advantage is that you don't have to reassess the security or stability of your knot every time lines are added (as the knot grows into a planet). :D

Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: DerekSmith on September 16, 2010, 10:57:02 PM
Well, I just tied the Myrtle (ABOK #1445) mid way in two lengths of bungee cord - dressed and set it and it turns out to be a stunning little knot for this application and possibly the simplest solution going.

Derek
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: roo on September 16, 2010, 11:06:23 PM
Well, I just tied the Myrtle (ABOK #1445) mid way in two lengths of bungee cord - dressed and set it and it turns out to be a stunning little knot for this application and possibly the simplest solution going.
Ashley calls it the worst single Carrick bend, and in my bungee, it rolled freely.   At least the overhand knot solution just flipped a few times before jamming.

P.S.  Did you decide to call the ABoK #1445 the "Myrtle".  If so, why?
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: Dan_Lehman on September 17, 2010, 07:15:23 AM
Since the topic has shifted from rope to bungee (and eventually eight arms), I had to adapt my response.

You're reading too quickly: the problem was stated in the OP and hasn't shifted,
except in your view of it.  (Xarax explicitly states, e.g., that we are not concerned
with elasticity.)

 And these failings suggest
that the Mid-span Sheet Bend be kept out of the game: recall the
issue of ring-loading and the Bowline (becoming in effect the
inferior Lapp knot orientation).
Wrong.  The Midspan Sheet Bend never assumes a Lapp Knot orientation since the leftover loop is not used.

Again, you're hasty in reading : I responded to the general case of taking
an eye knot into this situation, and so remarked at how this can lead to
trouble, as with the Bowline .  Now, while it's true that the MSSBend
here (but not in the 2-loops-knot & one cut thread) doesn't present
the orientation called "ring-loading", it does present another circumstance
that Bowline challengers like to put forward -- i.e., that of pulling on
the tail to supposedly create a "slip-knot" ... .  I.e., pulling the bight/U-part
apart : and I don't favor using it where that can happen (be so loaded).
YMMV on results.


As for Have I tried the Overhand (both) in shock cord? , no, I hadn't
(for, again, that wasn't specific to the challenge).  But I have done so now,
w/o problem.  MY cord is that from the sea, and so has lost slickness --
enough so that the Overhands can be tightened and hold; in some sense,
then, I convert the elastic material in the knot to relative inelasticity.
(slightly diff. dia.s, too:  I'll guess 1/4" & 3/16" ?  --that sort of difference)

And I've just now tried to bump the structure with one more "line" (I just
folded one cord into a bight, then laid the other parallel and tied the Oh.:
still fine.

AND, I tried Derek's Disaster :  what you said!  (fails readily (yet is a PITA to untie!))
(Derek, how did you get past the obvious, here, to the keyboard??!)


--dl*
====
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: DerekSmith on September 17, 2010, 11:23:17 AM

Ashley calls it the worst single Carrick bend, and in my bungee, it rolled freely.   At least the overhand knot solution just flipped a few times before jamming.

P.S.  Did you decide to call the ABoK #1445 the "Myrtle".  If so, why?

Likewise, in my bungee it rolled freely, until I set the knot by heavily (fully) extending the bungee on all four cords.  This preshrank the cords and the knot and placed it under 'elastic tension'.  This stopped any hint of rolling within the forces of the bungee cord's normal elasticity and locked up the knot so effectively I had to use as fid to prise it open.

The Myrtle is a single loop knot and was named by Dave Root as he found it holding up a Myrtle tree.  I cited #1445 to clarify the knot structure much as one might cite the sheetbend structure to describe a bowline to someone who was unaware of it (some chance).

'Why?'[was it called Myrtle] - why not?  It is as easy to remember as it is to tie - cast a Constrictor, feed the end into the constrictor, pull the constrictor and let it unfold, taking the end around with it and 'pop' - a Myrtle loop...  or in this case a four legged bungee 'quadrapus'.  But yes, I take your point, it is a rather new name for a knot that has been around for some time, but then we are used to giving new names to things in order to make them more memorable, and in some cases even 'Nicknames' to further improve memorability within small groups.

Personally, I think that when we get too precious about names then we start to squish some of the fun out of knotting - you only have to look at some of the wonderfully descriptive names that JD~TIAT is bringing to the game - although I am not a great exponent of rope 'knitting' I have to credit him with some lovely designs and highly memorable descriptive new names.

Derek
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: Dan_Lehman on September 17, 2010, 04:24:41 PM
Ashley calls it the worst single Carrick bend, and in my bungee, it rolled freely.   At least the overhand knot solution just flipped a few times before jamming.

Likewise, in my bungee it rolled freely, until I set the knot by heavily (fully) extending the bungee on all four cords.  This preshrank the cords and the knot and placed it under 'elastic tension'.  This stopped any hint of rolling within the forces of the bungee cord's normal elasticity and locked up the knot so effectively I had to use as fid to prise it open.

Well, okay, that's understandable, but little less worrisome : can one
trust the very-tight knot to remain so over time (where the constant
tension is asking for the relief of some loosening) ?  The Overhand
also wants tightening, but I don't think my tightening went to the
extremes that the *Single Carrick* construction requires.  Now,
also over time, the material I think loses some of the slickness that
makes knotting challenging (as perhaps is indicated by the apparent
trouble Roo had w/Overhand (rolling) and lack of that I had (using
some well non-new material).

--dl*
====
Title: Re: 4 ends bend. (Another interesting knot problem)
Post by: Rrok007 on September 17, 2010, 07:36:09 PM


Thank you Prok007,

It's Rrok007, no "P". Which is not a major issue, just making sure.

 I have to say that I disagree with your view that a Star knot solution goes outside the field of required simplicity...
   Simplicity is not judged by the mere number of initial elements, but of the symmetry of the transformations that produce the final, completed form. Simplicity is not a matter of economy in the senses, but in the mind. The Star knot is a very simple knot, one of the simpler knots I know, and just because it is so simple, it is also so beautiful. :)
   Could you please tell us the kind of hitches you would use to tie each individual leg ? Is it possible to present us with a picture of a well dressed a 4 legs star knot? Does a Star knot retain its pure form regardless of which and how many rays/legs are loaded?
   One advantage of your proposed solution is that it can be generalized in any number of legs. One disadvantage is that the individual hitches are not working together, in a central knot nub, so we might say that some rope element is wasted and some nipping force is used not as effectively as it could.

I guess when I think of simplicity, I mean the nuber of overall steps. Figuring 3-5 steps at most for the entire knot. I do agree that, once you know how it's done, the Star knot is quite rather a simple knot to tie. It's the repition all the way around the knot that makes it less than simple in my mind. I was also keeping in mind that it is generally considered a decorative knot, and as such may not initially thought of to have much practical value. For me, I think it's one of my favorite knots to tie.

As for load bearing, I have never really tried it as a means of supporting weight and can't really say without taking the time to test it out. I know that, depending on how complete you tie it, the legs will be coming from either the outer edge of the knot, or from the center of it. I will say that I have a few small knots that I have, from time to time, pulled on the legs and not had the body deform much. The "petals" if you will of the star knot may flex slightly, but not much, if at all. In fact, the tugging on the legs only helps to tighten the knot and keep it firm. I would imagine in your purposes, and weight loaded onto the legs would have a similar effect, essentially making the knot even more secure. I suppose it's possible for the knot to collapse if enough weight were loaded onto it, but I think it would have to be a disproportionate amount, for example, using the vacumn cause by high altitude cabin pressure loss to clean a carpet, when a simple handy vac would have done the job. Indeed, if one is using bungie cords, I would suspect there would be almost as much chance of the legs, or the supportng structure, breaking as there would be the knot collapsing. IE, if I used a star knot with long legs, then drilled a hole in my ceiling to feed the legs down through, and then hung a lead-lined corpse (don't ask) from the legs, I think it would be just as likely that the ceiling would come apart before the knot did.

Another advantage is that the number of legs in the knot can be doubled, simply by tying it close to the center point of each leg. IE, two ropes can make a four-point star; four ropes can make either a four-point, or an eight-point, star knot.