Analysis of Bowlines paper uploaded for review and comment (PACI website)

[alpineer]

Hello Mark,

Here are images of the Tresse Bowline which you requested. If you prefer a white background, I can do that. If you require full size images, that can be arranged.
I'm sending tech info via PM.

alpineer

[alpineer]

And the Tresse Bowline on a Bight.

[Stagehand]

Hello Mark, Thank you for your work in industry and in the Guild.  Thanks for the opportunity to contribute.  I would like to limit my remarks on the Analysis of Bowlines to the section on the Carrick Loop, ABoK #1033.  I question the suggestion that the Carrick Loop, ABoK #1033, is based on the Carrick Bend, ABoK #1439.  The Carrick Loop, when loosely tied, can be pushed by the fingers into the outline of a carrick just as any partial carrick may take up the carrick resemblance.  This carrick in the Carrick Loop lacks the complete interweaving of the Carrick Mat or the Carrick Bend.  There are a number of such partial carrick knots.  The Reef Knot, the Thief Knot, the Granny Knot, the Sheet Bend, and the Bowline are a few notable examples.  Especially for the demanding purposes in your view, a more clear basis could be reserved.  As much as ABoK #1439 represents  a specific carrick, a true carrick, then it can be seen as the basis of the Knife Lanyard Knot, ABoK #787, and the Chinese Button Knot, ABoK #599.  A fixed loop with a complete, oblique carrick could also be similarly judged.

[Dan_Lehman]

I question the suggestion that the Carrick Loop, ABoK #1033,
is based on the Carrick Bend, ABoK #1439.

Agreed : this should be called the "carrick bowline",
leaving the name unfortunately chosen by Ashley
for a certain other eyeknot.

--dl*
====

[knoeud]

It has been a long time since the last update! Any news on this paper?

Also, I don't understand why Mark doesn't recommend the EBDB Bowline in this paper! I think EBDB is a very recommendable knot since:

*   It has exactly the same 3 rope system as in EBSB+Y
*   It has the two nipping turns which is historically known to make the knot much stronger and has always been recommended over the single bowline
*   It is much easier and faster to tie compared to the EBSB+Y (hard to pass the tail in both collar and binding loop in practice!)

Dan Lehman, any suggestion on comparing these two knots for climbing tie-in?

[Dan_Lehman]

... the EBDB Bowline

Whoa, that's one "B" Byond balance
(redundancy).   

* It has the two nipping turns which is historically
 known to make the knot much stronger ...

Testing is, for what it's worth (not so much),
not quite this clear --and to a much less than
"much" degree.  The real appeal is its better
security vs. loosening (which itself might be
of a degree too small for full comfort --hence
my extension).

It is much easier and faster to tie compared to the EBSB+Y
(hard to pass the tail in both collar and binding loop in practice!

But you must understand that Mark is deeply
in luv with the "Y" bit --I think he probably Ys
his shoelaces!   

Dan Lehman, any suggestion on comparing these two knots for climbing tie-in?

I'd go with the EBDB (albeit partly from pride,
not the best guide).  But I've seen that eyeknot
loosen, in some things, and more recently
taken an interest in the more complex, repetitive
knots such as the mirrored bowline --which can
be quite loose in setting but I think will just not
loosen further very easily.  Also, the essentially
double-eye knots --where there is a 2nd passing
of rope through the harness-- provide some good
assurance against (at least) unnoticed untying,
as there'd be a long tail flopping about.

Might be good to have more- & less-rope-efficient
tie-in methods just to put the knotting at two
different places along the rope.  Otherwise, there
might be a desire for a knot that works better in
the stiff, inflexible rope one can sometimes encounter
--and where tightening the EBDB isn't so easy.

BTW, here's one of my latest ideas:
take the tail of a fig.8 eyeknot --that knot
sometimes required by climbing gyms to be used--
and put in a bowline *guard* knot to it,
reeving the latter's tail through the former
and back through the central nip of the Bwl
(for that 3dia bulk).  I think that the fig.8 will be
less prone to jam, thus (esp. w/the tail reeving),
and the bowline can evaporate so far as safety
is concerned, leaving the 8 standing guard as req'd.


--dl*
====

[enhaut]

BTW, here's one of my latest ideas:
take the tail of a fig.8 eyeknot --that knot
sometimes required by climbing gyms to be used--
and put in a bowline *guard* knot to it,
reeving the latter's tail through the former
and back through the central nip of the Bwl
(for that 3dia bulk).

I would like to see this in a picture if it's possible.
ths

[roo]

[...] I think EBDB is a very recommendable knot since:

*   It has exactly the same 3 rope system as in EBSB+Y

We really need to get away from appearance-based evaluations.  I just did a security test with the belly of an "End Bound Double Bowline" linked to a Water Bowline in some bluewater II rope.  It didn't take much shaking for the bound end to spring out and become a regular, mediocre, old double bowline which then proceeded to become entirely untied.

The simpler, more compact Water Bowline stayed in form for the duration of the test. 

[Sweeney]

I was at the National Waterways Museum today and not being busy I did some rough and ready testing of the slack security of a few bowlines. The rope was 10.5mm 'semi-static' climbing rope which is stiff and not something I would normally use. In each case I tried a vigorous shaking of the knot for 30 seconds and the results were:

Ordinary bowline (#1010) - fell apart before the 30 seconds were up.

Water bowline, girth hitch bowline, double bowline, end bound double bowline, Lees locked bowline :  all of these loosened considerably though in each case when force was applied (only a pull using hand and foot) they all gripped and none collapsed. However I would be reluctant to trust any of them in this rope unless under constant strain.

Scott's lock : this was a surprise in that it was tightened to the same degree as the others but showed no sign at all of loosening even after a second bout of shaking.

This was an idle moment test born out curiousity rather than serious study so please treat the results with scepticism as I know you all will - and bear in mind that all I was testing was slack security not holding power under load.  Nonetheless Scott's lock is something I will look at using to see how it performs in everyday situations with 'ordinary' rope and cordage (I don't do climbing).

Barry

[xarax]

   I did some rough and ready testing of the slack security...
   I tried a vigorous shaking of the knot for 30 seconds...

   I believe we should establish a few guidelines for such tests, otherwise we would not be able to compare and evaluate any results we may acquire from them. Anybody would be able to "shake" an eyeknot differently, and "see" different things - because, as we all know, most knot tyers wish so badly to "see" only what they had already convinced themselves they "know" ... A future situation, where we would have a few circumstantial, differently designed and performed, and subjectively evaluated "tests", can be worse from the present situation, where we have no tests at all !   

  1. What was "vigorously shaken" ? The Standing End only, or the whole eye-knot, the eye included ?
  2. Which was the angle between the eyelegs during the shaking ?
  3. Was the eye free to move/rotate around the wrapped object, or not ?
  4. Was the eyeknot pre-tensioned at the start of the shaking, and, if yes, to what degree ? Any shaking, however vigorous and prolonged, will not be able to loosen a very tight knot, of course - so we should not only certify that the compared eye-knots were tightened to the "same degree", but we should also decide/define this degree.
  5. An important element that should be always decided/defined, too, is the length of the Tail. I have seen that, during "shaking", the motion of the Tail influences the behaviour of the whole nub : a longer Tail, which is shaken along the rest of the knot, can help the nub remain more compact, for a longer time.     
   
  Also, we should describe the "shaking", with more precise terms. For example, we can specify the exact location of the point of the line, and the path and frequency of its repetitive movement, which induce this "shaking". I have seen that a "shaking" of the eye-knot by the Standing End, which involves short instances of pulling of this End, has a very different effect on the loosening of the nub, as it was anticipated.
   
   Having said that, I wish to make clear that I am delighted those "tests" involved some newer eyeknots, as the Scott s and the Lee s locked bowline ( re-tucked Yosemite bowline ). Although my "tests" on the same knots had not the same results  , I am delighted when people decide to tie and try knots they were not aware of until recently. The history of knot tying had not started, and we should always remember that it had not ended, with the ABoK ! 

[roo]

Scott's lock : this was a surprise in that it was tightened to the same degree as the others but showed no sign at all of loosening even after a second bout of shaking.

This was an idle moment test born out curiousity rather than serious study so please treat the results with scepticism as I know you all will - and bear in mind that all I was testing was slack security not holding power under load.  Nonetheless Scott's lock is something I will look at using to see how it performs in everyday situations with 'ordinary' rope and cordage (I don't do climbing).

Barry

I do like the slack shaking security of this variant, but I have had instances where the knot form capsized under heavy load, and have discussed the issue with Scott on this forum:

https://igkt.net/sm/index.php?topic=5995.msg40270#msg40270

It should be noted that many loops are more or less secure depending on the direction the loop is hanging (e.g. if the belly of the  loop is above or below the knot body).

[SS369]

I do like the slack shaking security of this variant, but I have had instances where the knot form capsized under heavy load, and have discussed the issue with Scott on this forum.

Yes, we did discuss this variant, your claims, and to date I have not been able to duplicate your findings without leaving the knot poorly dressed and loose. As such it is no different than any other poorly tightened/dressed eye knot. A competent knot tyer/user knows to set their knots well.

I have recently pulled a tractor out of a ditch using two of these, one on each end and suspect that I was dangerously close to an unsafe working load (sometimes you use things you have the best you can.). I am not recommending this to anyone. But, the loops did the task and untied without the use of tools. No capsizing, no difficulty untying.

It should be noted that many loops are more or less secure depending on the direction the loop is hanging (e.g. if the belly of the  loop is above or below the knot body).

Want to qualify or expand on the quoted statement a little more?

SS

[roo]

I do like the slack shaking security of this variant, but I have had instances where the knot form capsized under heavy load, and have discussed the issue with Scott on this forum.

Yes, we did discuss this variant, your claims, and to date I have not been able to duplicate your findings without leaving the knot poorly dressed and loose.

The most convenient way to observe the nature of the Scott's Lock Bowline capsizing event is to tie the knot in bungee or shock cord.  Elasticity seems to play a role.

[Dan_Lehman]

[...] I think EBDB is a very recommendable knot since:

*   It has exactly the same 3 rope system as in EBSB+Y

We really need to get away from appearance-based evaluations.

Where that appearance suffices, it's fine;
I take it that the reference is to the 3 diameters
stuffed through the central nipping loop (and
suggested to be indicative of better strength).

I just did a security test with the belly of an "End Bound Double Bowline" linked to a Water Bowline in some Bluewater II rope.  It didn't take much shaking for the bound end to spring out and become a regular, mediocre, old double bowline which then proceeded to become entirely untied.

Of course, the obvious takeaway from this
report is that you made a good effort to put
the former knot at its loosest, or are entirely
incompetent at tying (dressing & setting) it!

But if we are using rope that is relatively stiff
and resistant to bending (and worse, springy),
it is critical for the EBDB to be set tightly and
that might be difficult; if loosened, its turns
around other turns can all just expand --which
in contrast the *bights* (by which I mean
U-shaped not O-shaped) want to open like
scissors and that can be easily impeded,
with good effect for arresting further loosening.

--dl*
====

[Dan_Lehman]

BTW, here's one of my latest ideas:
take the tail of a fig.8 eyeknot --that knot
sometimes required by climbing gyms to be used--
and put in a bowline *guard* knot to it,
reeving the latter's tail through the former
and back through the central nip of the Bwl
(for that 3dia bulk).

I would like to see this in a picture if it's possible.
ths

It's possible, but there are many ways,
but I'm unsettled on what's preferable
--doing what I'll describe again, and which
should need no graphic, seems overly bulky
(and needing some good guidance on how
much working end to have, in anticipation
of what one will need to do with it!).

Tie a bowline with amply large eye,
in which to then tie a fig.8 eyeknot
(with/in the former's eye), set close to
the bowline.  Now, tuck the bowline's
tail through the fig.8 --try different ways.

Now, the above tying method was given just
for simple illustration --to get one to the result--;
for tying-in, one cannot practically tie in-the-bight
but must reeve the tail of the fig.8.  Doing this,
though, with ample tail, enables one to then
put in a bowline to *guard* the fig.8
(since the former is "PET"=post-eye-tiable).

What I'm likely to prefer is some (novel) "quick8"
first-formed eyeknot, which (naturally, being "quick")
is more easily done, and then casting the guard knot
and then re-tucking the tail.  All of which is to result
with a bunch of knotting that, even loose-ish, should
prevent the tail from working out of.


--dl*
====