Double collar bowlines based on the Clove hitch or the Girth hitch.

[X1]

   The Clove hitch and the Girth hitch can serve as nipping structures of double nipping loop bowlines - and, with the addition of a second collar, they can be transformed into the Double collar Water bowline, or the Double collar Girth hitch bowline ( the "Mirrored" bowline ). I believe that these very secure bowlines can replace the retraced fig.8 loop, in climbing, SAR and other life-threatening situations.
   However, there might be a fly in this ointment : Mark Gommers, in his " Analysis of Bowlines" (1), mentions that " Under load, the 2 (nipping) loops will separate...". (*). Now, this might not be immediately dangerous, but it is not something that one would have wished / ordered... The "walking" of the "lower" nipping loop away from its twin and towards the tip of the bight can drag the "lower" collar along with it, and, as a consequence, the knot will become more elongated and less compact that it already is !
   On the contrary, using a "closed", self-constricting double nipping structure, like the reversed double coil or the twisted Pretzel knot, the two nipping loops will always remain in contact, because their two ends will always pull and push the one loop towards the other. ( See the attached pictures). ( The same can be said for a Constrictor or a reversed Constrictor double collar bowline. )
   I have no experience of those double nipping loop / double collar eyeknots under extreme conditions - load, humidity, vibrations, etc., so I can not really tell if this theoretical danger I describe will ever materialize... The reader who has tested those knots under such conditions is kindly requested to enlighten us on this issue.
   
(*) This is a remark on the Clove Hitch -based Water bowline - but I can not see why the situation would be different on the Girth Hitch -based bowline....   
1.   http://www.paci.com.au/downloads_public/knots/Bowlines_Analysis.pdf

[Dan_Lehman]

   The Clove hitch and the Girth hitch can serve as nipping structures of double nipping loop bowlines
--and, with the addition of a second collar, they can be transformed into the Double collar Water bowline,
or the Double collar Girth hitch bowline ( the "Mirrored" bowline ).
I believe that these very secure bowlines can replace the retraced fig.8 loop,
in climbing, SAR and other life-threatening situations.

They have shown --via a single test, only, admittedly (so "they"
is only implied by the particular one (mirrored bowline)--
ability to hold securely under load in slick, HMPE 12-strand rope.
Such use might only come in infrequent situations --the norm
being not to knot HMPE rope--, but then it will be helpful!

However, there might be a fly in this ointment : Mark Gommers,
in his " Analysis of Bowlines" (1), mentions that " Under load,
the 2 (nipping) loops will separate...". (*). Now, this might not be
immediately dangerous, but it is not something that one would
have wished / ordered...   The "walking" of the "lower" nipping loop
away from its twin and towards the tip of the bight[eye]
can drag the "lower" collar along with it, and, as a consequence,
the knot will become more elongated and less compact that it already is !

I think that this is incorrect analysis/projection : of course these
turNips tend to separate, as they're loaded respectively from
SPart & eye-leg by 100% & 50% against their joining connection
which awaits the transmission of force through the two "hitch
components" (as Derek might point out    ).  But note that this
is a sort of leaning apart more than full movement of these turns.

Further, re the "dragging of the tail," realize that the SPart's
turNip has the greater force and hence greater nip, which
should impede the eye-side's turNip from taking parts
from it (but perhaps some vagaries of parts positioning has
some ability to help/hinder?).  Much of any "movement" will be
just the draw of the SPart & eye leg  tightening their turNips,
the flow of material of the turn, not the turn itself vis-a-vis
other parts.   (Bouncing 50# weights on 8mm kernmantle
tied with both of these knots --1 @either end-- didn't show
to me any feared movement.  This is surely much less force
than was applied by Mark.  Using a 5:1 pulley and body mass,
I see nothing happening to the just-bounce-loaded knots.)

On the contrary, using a "closed", self-constricting double nipping structure,
like the reversed double coil or the twisted Pretzel knot, the two nipping loops
will always remain in contact, because their two ends will always pull and push
the one loop towards the other. ( See the attached pictures). ( The same can
be said for a Constrictor or a reversed Constrictor double collar bowline. )

The simple double-turn lacks the security aspect of the
challenged forms --shown both by the Toss video of the
dbl.bwl collapsing the eye under high load, and also being
less secure when slack (the rockclimber's concern).


--dl*
====

[X1]

.. hold securely under load in slick, HMPE 12-strand rope.

  How much load ? Was it, say, 50% of the rope s ultimum strength ? Was it static or dynamic ? There are many things one has to take into account, beyond the well-known fact, that a single test can reveal but a miracle ! 

of course these turNips tend to separate, as they're loaded respectively from SPart & eye-leg by 100% & 50% against their joining connection

 (I call this "joining connection" a "bridge").
 So, you say that the situation regarding this "leaning apart" of the two nipping loops at the Girth hitch and at the Clove hitch, is the same  ? There should be some difference, as the geometry is very different, but I can not isolate and analyse it... I can only point out that the curves at the two ends of the "bridge" / "joining connection" are sharper in the case of the Girth hitch, because there the "bridge" itself runs parallel to the axis of the knot - while the "bridge" at the Clove hitch is inclined (see the attached picture). However, I do not understand how this can influence the "difference" that, although I can not see, it should be there nevertheless...
  It is not only the difference in the loading ( 100% and 50% ) that tend to pull the two loops apart... As I had tried to explain at (1), the nipping loop of the common bowline does not "walk" ( "walk", not only "lean" ) towards the tip of the eye ( the "eye" is the whole shape of the "bight"/"loop"...) mainly because of the friction forces between the rim of the nipping loop and the penetrating legs of the collar. As this friction is "proportional" to the gripping power of the nipping loop, the friction forces between a heavily loaded, "strong" nipping loop and the segments it encircles are greater than those of a "weaker" one - and the "lower" nipping loop, tensioned by 50% only of the load, is a weaker one, indeed. So the chances it will "walk" towards the tip of the eye are greater.

re the "dragging of the tail," realize that the SPart's turNip has the greater force and hence greater nip, which should impede the eye-side's turNip from taking parts from it

  I suppose that the "first" / "higher" / stronger nipping loop will, more or less, remain in its initial place, because it is further immobilized by the pull of the "higher" / strong collar  - but the "second" / "lower" / weaker one will be the one which will start "walking" towards the tip of the eye, dragging the weak collar along with it, and pulling the tail downwards. I just can not see how the "higher" loop, being firmly attached to the "bridge" and the legs of the collars, will impede the "walking" of the "lower" one...
   The 50 whatever weight is too light for a 8 mm climbing rope... We want to use those safe bowlines in extreme situations, as replacements / substitutes of the retraced fig. 8 knot. So, I believe that any tests should reach bouncing dynamic loadings close to the ultimum strengths of the ropes on which they are tied.

   The simple double-turn lacks the security aspect of the challenged forms -- shown both by the Toss video of the dbl.bwl collapsing the eye under high load, and also being less secure when slack (the rockclimber's concern).

   Yes, I have also sensed the same weakness - and I can not explain it ! Perhaps it is due to the friction between the two adjacent coils of the double nipping loop, or the sharp curves at the standing part s and eye leg s of the standing part first curves.

I have seen that the simple two-coils nipping "tube", can not constrict the tail in a sufficiently secure way. Two nipping loops, in contact or/and aligned to each other, form a straight "tube", that helps rather than blocks the slippage of the segments that penetrate them ! I do not know why this happens, and I believe we should test this before we can jump into premature conclusions... As we can not PULL the tail - even a not-very-efficient nipping structure, when it is accompanied by the collar(s), is very resilient - we can PUSH it, and try to feel the resistance its constriction by the double nipping structure imposes on this push.

1. http://igkt.net/sm/index.php?topic=3233.msg23449#msg23449
    http://igkt.net/sm/index.php?topic=3233.msg23450#msg23450
    http://igkt.net/sm/index.php?topic=3233.msg23452#msg23452
   
    I know that most people would be surprised by a "single turn" component of the bowline walking along the standing part, but this is exactly what would have happened had it be no friction between the turn s rim, the wheel, and the pair of the collar s legs, the wheel s axle. If you glue the two legs together, and pass them through a ball bearing, and then roll the turn around this ball bearing, as soon as you load the bight the one leg will wind around the rim of the bearing while the other will unwind - and the turn, as a whole, will move towards the direction the collar pushes it...

2. http://igkt.net/sm/index.php?topic=4306.msg26857#msg26857 

[X1]

   I might have conveyed the wrong message here : I do not say that the possible split, the possible spatial separation of the two nipping loops relatively to each other, is "dangerous" to the integrity of those most secure "double / double" bowlines.  I say that IF the "lower"/ weaker nipping loop would crawl away from the "higher" one, dragging the "lower"/weaker collar and the tail along with it, the knot would become more elongated / less compact than it is already - and this, in the eyes of a conservative re knots rescue worker or climber, for example, can become a cause of some concern, disquiet, and anxiety - and those feelings are dangerous, indeed, in situations demanding the maximum of attention be focused on the purpose / target, not on the means / tools. We have to admit that the retraced fig. 8 knot eyeknot offers a "look" and a sense of high security, and any replacement / substitute of it should provide the same, at least, psychological easiness - otherwise it will not be adopted.
   So, I have thought that the double nipping structures where the two parts are pulled / pushed by their ends upon each other, would be preferable. However, there is a caveat in those knots, too. A portion of the tensile forces converging into the nipping structure by both ends is consumed, and "wasted", within the structure itself, and so its gripping potential on the penetrating collar legs may be jeopardised. That is the most simple explanation I can think of for the weakness of the reversed double coil nipping structure. The two nipping loops are clinched tightly, but the legs of the collar slip through them without much resistance.
   The "twisted" Pretzel bowline, and the "reversed" Constrictor bowline, might suffer from the similar problem : The rims of their nipping loops are embracing each other, and I fear that at this area of mutual contact they will "upload" a great deal of their potential as gripping structures. The tensile forces would be "wasted" as friction forces within the two nipping loops, and so they will not be utilized and transformed, in the maximum possible degree, to friction forces between the inner surface of their rims and the penetrating collar legs. I have tied the double collar un-twisted Pretzel bowline, but, for some reasons, I did not like it very much (1). That leaves the double collar Constrictor - which would be a subject of another thread.
   I will use the opportunity here to restate my present beliefs, for what they are worth : The most secure post-eye-eyeknot I know, and the only I would recommend as a replacement to the fig.8 eyeknot, is the Teedledee "bowline". A very interesting, but very "wide" - so not very stable and compact - knot I have not evaluated yet in all loading circumstances, is the Constrictor x2 "bowline". The most secure and beautiful double / double "Eskimo""bowline" is the Girth hitch - based one.
   Last, but not least, I mention the double collar Helical loops, in general, and the Helical Constrictor or Strangle loops, in particular.( See the attached pictures). In those tangles, the Standing part is not crossing itself, so it dissipates the tensile forces gradually, along an extended area. Its curvature is wide, and constant throughout its path around the encircled ( better, enhelix-ed  ) collar structure. We may add or subtract helical turns, and so control the amount of "shock absorption" we wish it to present during sudden, abrupt pulls.
    Perhaps we were enslaved, by the King of Knots, the bowline, for too long all those thousands of years... and we were seduced and/or forced to live with one or two nipping lops around our neck. Perhaps it is time to get rid of this necktie, and wear some new garment accessory - a scarf / Helical loop, for example. 

   "Ancient Rome is one of the first origins of the scarf, where it was not used to keep warm, but to keep clean. It was called the sudarium, which translates from Latin to English as "sweat cloth", and was used to wipe the sweat from the neck and face in hot weather. They were originally worn by men around their neck or tied to their belt. Soon women started using the scarves, which were made of cloth and not made of wool, pashmina, or silk, and ever since the scarf has been fashionable among women.
   Historians believe that during the reign of the Chinese Emperor Cheng, scarves made of cloth were used to identify officers or the rank of Chinese warriors.
   In later times scarves were also worn by soldiers of all ranks in Croatia around the 17th century. The only difference in the soldiers' scarves that designated a difference in rank was that the officers had silk scarves whilst the other ranks were issued with cotton scarves. The men's scarves were sometimes referred to as "cravats" (from the French cravate, meaning "Croat"), and were the precursor of the necktie..."

1. http://igkt.net/sm/index.php?topic=4306.0

[X1]

   Just an example out of the many possible Helical Loops...( See  also: (0), (1), (2), (3))
   ( I have not yet decided which is the "best" core knot - it may be a fig.8, a fig. 9, a strangle, or even a simple S shaped segment formed on the returning eye leg. Also, the number of helical wraps can vary, according to the use, and/or the material. Also - and this is a very complex issue -, I do not know how tight should this core knot be... People are used to pull the tails of the bowlines, so they get the tightest possible collars, but this is probably detrimental to the overall strength of the knot -  because a tight collar would force the standing part to follow a tight first curve - not a good thing.)

0. http://igkt.net/sm/index.php?topic=3020.msg21688#msg21688
1. http://igkt.net/sm/index.php?topic=3020.msg22085#msg22085
2. http://igkt.net/sm/index.php?topic=3020.msg22086#msg22086
3. http://igkt.net/sm/index.php?topic=3020.msg23685#msg23685

[Dan_Lehman]

[I should have remarked that I think my prior post was "2323"
--twice "23 Skiddoo" (some ol' American saying)     ]

.. hold securely under load in slick, HMPE 12-strand rope.

  How much load ? Was it, say, 50% of the rope s ultimum strength ?
 Was it static or dynamic ? There are many things one has to take into account, beyond the well-known fact, that a single test can reveal but a miracle ! 

Of course not 50% --that just isn't seen in HMPE !
(not in eyeknots, anyway)  And slow-pull loading,
as you ought to have seen --do WATCH that video,
and the one of the dbl.bwl slipping .
Now, re the slipping knot, that I believe couldn't
slip fast enough in a dynamic load to spill before breaking
(or holding a lower load); maybe the movement would
contribute to breakage (heat).

As I had tried to explain at (1), the nipping loop of the common bowline
does not "walk" ( "walk", not only "lean" ) towards the tip of the eye

And I challenge that there IS any such walking.
(I understood your assertion, but doubt it; I put what
Gommer's observed in a different light --"leaning" (and
[perhaps to some initial extent, walking-adjusting
to a high-force set, but not continual adjustment.)

( the "eye" is the whole shape of the "bight"/"loop"...)

No, not to me : "eye" has less geometric and more functional
significance --it is that loopish part that DOES ...  (and I don't
care to scrutinize whether its legs cross ("->loop") or not ("bight")!
And esp. with the knots at issue, we have already a plethora
of "bights" (in the sense so unfitting for "in the bight" use,
where thus it might be best to say "without ends" --such
dramatic geometric difference!).

As this friction is "proportional" to the gripping power of the nipping loop, the friction forces between a heavily loaded, "strong" nipping loop and the segments it encircles are greater than those of a "weaker" one - and the "lower" nipping loop, tensioned by 50% only of the load, is a weaker one, indeed. So the chances it will "walk" towards the tip of the eye are greater.

re the "dragging of the tail," realize that the SPart's turNip has the greater force and hence greater nip, which should impede the eye-side's turNip from taking parts from it

  I suppose that the "first" / "higher" / stronger nipping loop will, more or less, remain in its initial place, because it is further immobilized by the pull of the "higher" / strong collar  - but the "second" / "lower" / weaker one will be the one which will start "walking" towards the tip of the eye, dragging the weak collar along with it, and pulling the tail downwards. I just can not see how the "higher" loop, being firmly attached to the "bridge" and the legs of the collars, will impede the "walking" of the "lower" one...

You have just concurred in the reasoning that the
SPart's loop will have a stronger grip,
so why don't you concur in my conclusion that IT
will better HOLD parts than will any weaker loop
pull them away?  In this (challenged!) tug-of-war,
the stronger grip should win ; the weaker turNip
will walk only *lightly*, if at all --not carrying baggage.

Btw, how can this weaker turn walk away when it
is joined by the bridge?  It would have to give back
material into the bridge (just as a wave through
water doesn't transport the actual water)!

The 50 whatever weight is too light for a 8 mm climbing rope... We want to use those safe bowlines in extreme situations, as replacements / substitutes of the retraced fig. 8 knot. So, I believe that any tests should reach bouncing dynamic loadings close to the ultimum strengths of the ropes on which they are tied.

One doesn't use cordage near its believed ultimate strength!
And, FYI, dangerous, ratcheting slippage in cyclical loading
of the offset water knot in tape was observed NOT at
high loads, but at low loads --exactly those that might
be generated on abseil, where the knot is in one side of two
of an anchoring sling and getting only half of the roughly
man-x-2 forces.  So, the low-load bouncing was looking
to any such hint of movement where some recoil of the
material might be easier.

But, yes, there are other loadings, and so I did the pulley
loading for that, which was a look but hardly conclusive.
--to check a mere conjecture, though, coming w/o observation
beyond that of one that can be otherwise explained.

   The simple double-turn lacks the security aspect of the challenged forms -- shown both by the Toss video of the dbl.bwl collapsing the eye under high load, and also being less secure when slack (the rockclimber's concern).

   Yes, I have also sensed the same weakness - and I can not explain it !
Perhaps it is due to the friction between the two adjacent coils of the
double nipping loop, or the sharp curves at the standing part s and eye
leg s of the standing part first curves.

I can explain it : material feeds more easily into the
easy turns, whereas in the hitch-bases (clove/cow)
there are sharper turns, and material rubs shoulders
with what should suffice frictionally to impede easy
movement.  The double turn at least *allocates* an
absolute feed over two vs one turn, but that's only
a reduction of degree, and maybe in time (more time
allows more material, to then achieve equal looseness).
The hitch bases seem to better impede such loosening,
and there are ways to stuff more material among these
parts for increasing the simple frictions.
In the bowline, the eye-side of the turNip can be
readily available to help loosen this turn; but in the
hitch-based knots, it flows into the "bridge" as you say,
and so doesn't easily abet the SPart's loosening.


--dl*
====

[X1]

Of course not 50% --that just isn't seen in HMPE ! (not in eyeknots, anyway)  And slow-pull loading, as you ought to have seen --do WATCH that video,

  At least 50%, of course ! Otherwise "security" would only lie in the hands of the weaker... In order to be able to speak about security, we SHOULD test knots at 50% of the ultimum rope s strength, under static as well as dynamic loading - otherwise we better buy some glue... It has not be seen, because almost NOTHING has been seen - and HMPE ropes are expensive, and they require heavy, expensive instruments.
   ( I was not replying about this video, of course - which is one of the first things I had watched since I came to this Forum. This is what I plan to do ; Built a proper laboratory, and test all the existing practical knots, tied on every available material...  )

re the slipping knot, that I believe couldn't slip fast enough in a dynamic load to spill before breaking (or holding a lower load);     maybe the movement would contribute to breakage (heat).

  I do not know what the various standards / requirements on rope safety tell about this... Personally, I would say that, if a knot remains in one piece, i.e., does not slip and/or does not break after a first extreme pull ( a fall, for example), it should be considered as a safe knot. And by this "extreme" pull, I believe we should mean a loading corresponding to a 50%, at least, of the rope s ultimum strength, static and/or dynamic. A rule of thumb, but a reasonable one...

As I had tried to explain at (1), the nipping loop of the common bowline does not "walk" ( "walk", not only "lean" ) towards the tip of the eye

  And I challenge that there IS any such walking.

   I was referring to the common bowline, and to the theoretical possibility of such a walking, had the friction forces between the rim of the nipping loop and the two legs of the collar were absent. Read the whole three (long) posts !
   I do not doubt that there would be no such walking in the double Clove hitch or Girth hitch bowlines. I was just expressing my concerns about it, because I do not have the means to test this hypothesis. I would nt bet in more than, say, a 3.25% odds that there would be such a phenomenon. 

( the "eye" is the whole shape of the "bight"/"loop"...)

  No, not to me : "eye" has less geometric and more functional significance

   An eye is an eye is an eye : Look what an eye is, at Google images !  We could well have named this shape "a fish" - and then, it would be easy to denote the tip of the eye as the head of the fish, and the eyeknot s nub as the tail of the fish...

You have just concurred in the reasoning that the SPart's loop will have a stronger grip,
so why don't you concur in my conclusion that IT will better HOLD parts than will any weaker loop pull them away?

I did / I do !

In this (challenged!) tug-of-war, the stronger grip should win ; the weaker turNip will walk only *lightly*, if at all --not carrying baggage.

  The stronger grip / "higher" nipping loop, just because it will better HOLD parts than the weaker / "lower" nipping loop, it will bite harder the penetrating two legs of the collar(s) and the "bridge", and so, relatively to them, it will probably remain in place - it will not be translated at all. If somebody is going to "walk", it will probably be the weaker / "lower" grip / nipping loop, because it can not hang itself from those three lines as effectively as the stronger grip /"lower" nipping loop. When the load will be pulling its leg, it will run the danger to "walk" downwards, the gravity road. That is what Gommers is saying, I believe.

Btw, how can this weaker turn walk away when it is joined by the bridge?  It would have to give back material into the bridge ..!

   NOW, at last, you have seen what you should have seen just right at the beginning ! That is the crux of the matter.
   THAT is the meaning of "walking" ! It would give as much material as it would take. In a sense, it would slip and rotate at the same time. I have explained it in the post you had not read... It is like a nipping loop tied around a rod ( a pencil, at that post... ), with its two limbs at the two ends of this rod. If the rod is revolving around its axis, the nipping loop will be transported along its length. If the friction forces wold not be strong enough ( and, in this weaker / "lower" nipping loop, they might not be strong enough, indeed ), the rod = eye leg would not have to rotate, because the weaker / "lower" nipping loop itself would be able to slip on its surface. So, it can well be transported till the tip of the eye / head of the fish, without any movement of the 'higher" end of the "bridge", that would have been immobilized by the stronger / "higher" grip /  nipping loop.
   If you will understand what I mean, I am sure you would be able to defend it much better / with a stranger language, than I do ! 

  One doesn't use cordage near its believed ultimate strength!

  I do not understand either the exclamation mark, or the reason for this ! He rather should - taking safety precautions, of course. He should remain far away from the tensioned cordage, but not SO far away from its ultimate strength !
 
 

  And, FYI, dangerous, ratcheting slippage in cyclical loading of the offset water knot in tape was observed NOT at high loads, but at low loads --exactly those that might be generated on abseil, where the knot is in one side of two of an anchoring sling and getting only half of the roughly man-x-2 forces.  So, the low-load bouncing was looking to any such hint of movement where some recoil of the material might be easier.

   Very good explanation. The absence of full loading/squeezing of the one or both links might result in a loose nub, which can let the material slip through - if I understood it. Believe it or not, I have never thought of such a case ! My understanding was that, as the loading will progress, the knot will be tightened by itself and become more compact. So, if it would hold at the stage of maximum loading, it would have hold all the way down the stages of lighter loadings...and that would have meant that if we test the slippage of a knot and find it OK at maximum loading, we should not bother to test it at lighter loadings...

--to check a mere conjecture, though, coming w/o observation beyond that of one that can be otherwise explained.

   My conjecture was based on Mark Gommers report, at fig. 10 - 11 - although I have seen such a behaviour by myself.
   The simple fact is that, if something can go wrong, it will - so, if we do not have an explanation for the opposite ( which we do not...), we must suppose that the feared effect, however improbable, is, in fact, possible. I do not like fear, and I am not whistling in the dark. However, I appreciate that you "have understood [my] assertion", before you remembered that you should rename it into a "mere conjecture", "coming without observation"... )[Next stop : "Baloney"]

   I have also sensed the same weakness - and I can not explain it !
   Perhaps it is due to the friction between the two adjacent coils of the double nipping loop, or the sharp curves at the standing part s and eye leg s of the standing part first curves.

material feeds more easily into the easy turns, whereas in the hitch-bases (clove/cow) there are sharper turns,
The double turn at least *allocates* an absolute feed over two vs one turn,

   The turns of the double reversed coils nipping loop are at least as sharp, if not sharper than the curves in the hitch-bases nipping loops. And if the same absolute feed would be "allocated" over two turns, rather than one, that means that each one of the two turns of the double rev. coils nipping structure would consume half of it, so each turn would be loosened less than if the same amount of material was consumed by two turns. I can not follow your reasoning ! See the attached pictures, for the sharp curves of the weaker structure...

[Dan_Lehman]

Of course not 50% --that just isn't seen in HMPE ! (not in eyeknots, anyway)  And slow-pull loading, as you ought to have seen --do WATCH that video,

  At least 50%, of course ! Otherwise "security" would only lie in the hands of the weaker... In order to be able to speak about security, we SHOULD test knots at 50% of the ultimum rope s strength, under static as well as dynamic loading - otherwise we better buy some glue... It has not be seen, because almost NOTHING has been seen - and HMPE ropes are expensive, and they require heavy, expensive instruments.
   ( I was not replying about this video, of course - which is one of the first things I had watched since I came to this Forum. This is what I plan to do ; Built a proper laboratory, and test all the existing practical knots, tied on every available material...  )

You miss the point : HMPE ropes (ditto other "hi-modulus" ropes)
don't yield such strengths as (even) 50% when knotted.
(My strongest eye knot was 42%.)


--dl*
====

[X1]

You miss the point : HMPE ropes (ditto other "hi-modulus" ropes) don't yield such strengths as (even) 50% when knotted.
(My strongest eye knot was 42%.)

   Thank you, I was not aware of this. Perhaps we should discover stronger eyeknots than yours  , or accept an exception for knots tied on this material. Taking into account that it suffers from fatigue, too, we should lower the bar to 33 % ? ? 

[DDK]

   The simple double-turn lacks the security aspect of the challenged forms -- shown both by the Toss video of the dbl.bwl collapsing the eye under high load, and also being less secure when slack (the rockclimber's concern).

Yes, I have also sensed the same weakness - and I can not explain it ! Perhaps it is due to the friction between the two adjacent coils of the double nipping loop, or the sharp curves at the standing part s and eye leg s of the standing part first curves.

I can explain it : material feeds more easily into the easy turns, whereas in the hitch-bases (clove/cow) there are sharper turns, and material rubs shoulders with what should suffice frictionally to impede easy movement.  The double turn at least *allocates* an absolute feed over two vs one turn, but that's only a reduction of degree, and maybe in time (more time allows more material, to then achieve equal looseness). The hitch bases seem to better impede such loosening, and there are ways to stuff more material among these parts for increasing the simple frictions. In the bowline, the eye-side of the turNip can be readily available to help loosen this turn; but in the hitch-based knots, it flows into the "bridge" as you say, and so doesn't easily abet the SPart's loosening.

I was warned via automatic notice that this thread is 120 days old and that I should consider starting a new thread.  To prevent repeating what has already been discussed, I have ignored this advice.

I am not sure that I would explain the observed performance of the nipping structure in the same way as dl.  I am not disagreeing with the mechanics of the process he describes, but, questioning whether it is the primary mechanism at work.

As background, I have found, as I am sure others have, that by increasing the double bowline to a triple bowline, one finds the middle collar (nipping turn) in the triple bowline can be difficult to tighten in some ropes and, after loading, may not even contact most of what it is suppose to be nipping.  Clearly, this is not an increase of 50% more "nipping" power as compared to the double bowline. This certainly brings into question the analysis of Gommers who implies a doubling of the surface area for a double bowline which one might conclude should lead to a doubling of the nipping power of a single bowline.

The relative angle of attack between the material being nipped and the nipping turn is irrelevant for the middle collar (nipping turn) in the triple bowline, at least with the materials I have examined.  What I feel is not being satisfactorily addressed is the importance of the tension within the nipping turn.  This tension (referred to as "hoop" stress by mechanical engineers) generates a normal (meaning perpendicular) force, ie. pressure, on the material within the hoop (turn) and, of course, is what produces the frictional forces on the materials being nipped.

Anything that reduces or otherwise impedes this tension within the nipping turn will reduce the desired frictional forces applied (all other things being equal).  The angle of the nipping turn to the applied tensions is one of the important factors.  Consider this thought experiment.  Wrap a rope around a frictionless tube and apply tension (to both ends of the rope) perpendicular to the long axis of the tube.  In this case, the rope may incidentally touch itself, but, effectively does not interfere with the tension within the turn.  The tension within the turn, for this case, is everywhere the same as (100% of) the applied tension.  As one rotates the tube (edit: rotate long axis relative to direction of tension) and the rope forms the crossing point, the rope interferes with and reduces the tension within the turn.  The sharper the turn at the crossing point, the lower the tension within the turn, the lower the pressure applied and the lower the frictional forces gripping the material within the nip.  One can take this to an extreme.  Consider the hoop stress in the loop of an overhand loop knot as one pulls on the ends - yes, it is zero.  The knot has completely impeded the transfer of the tension to the loop.

In the double (triple, etc) bowline, the nipping turns can interfere within one another reducing their tension and their ability to grip. In this respect the term "double bowline" is a misnomer and would be more accurately applied to the water bowline whose nipping loops are more independent.  Additionally, I consider the "leaning" of the nipping turn in the water bowline to be a positive.

DDK

[Dan_Lehman]

Firstly, yes, you should make good judgement in reply
to an "old" thread --after all, it IS what you want to
address, so why not ... ?
 

Secondly, reading your text is tricky because you write
"middle collar" and there is no plurality of collars in the
"triple bowline" to which I think you intend to refer
--i.e., that with one more wrap of the nipping turn/coil,
and not the tie-a-bowline-with-a-bight which yields a
trio of eyes and two (the bight-legs') collars (still, though,
no "middle" one to that!)

Finally, you seem to be addressing a different aspect
from what I was discussing --dynamic/loaded security
vs. (my point) slack-security.  I was pointing to why
I think that the one structure won't so readily loosen
as the other, when the line is not in tension --this is
a concern for some rope users, i.p. rockclimbers.
You are looking at possible generated pressure within
the nipping loop/coil, which is something that comes
only with loading.


--dl*
====

[DDK]

dl - Yes, the "middle collar" I spoke of was as you have guessed - the middle nipping turn of the triple bowline.  Sorry for the confusion.

I did take your analysis to be of the loaded structures, that is, the "rubs shoulders" to be of nipped line through the misaligned/"leaning" nipping loops.  I see now your reference was in regard to the sharpness (none for double bowline) of the connection between adjoining nipping turns and the ease with which tension/slackness would be easily tranferred or not. Thanks for the help!

DDK

[SS369]

When I first read this thread I didn't think about how the naming of it seemed a little misleading. When I think of "collar" in regards to the bowline I think of the part that wraps the standing part at its entry. That's its collar right?

So I tied a bowline with a clove hitched collar and it is a fairly simple (to remember and build) loop knot. Its attributes are pretty positive in all respects.

Of course one could add a clove or girth based nipping structure.

SS

[Sweeney]

So I tied a bowline with a clove hitched collar and it is a fairly simple (to remember and build) loop knot. Its attributes are pretty positive in all respects.

This is interesting so I tried it but realised that if after tying the clove hitch and then leaving the tail outside of the nipping loop it locks tight. This does seem to be a testament to the security of the tail if it is tucked as usual.

[Luca]

Hi SS,

I'm asking you as a newbie because I continue to be a newbie as an experimenter of the behavior of knots: you happened to compare this Bowline with the Braided Bowline ? Which do you prefer?

                                                                                                                      Bye!