International Guild of Knot Tyers Forum

General => Practical Knots => Topic started by: andy753421 on September 14, 2011, 09:56:32 AM

Title: Dressing a figure-eight loop?
Post by: andy753421 on September 14, 2011, 09:56:32 AM
I recently started rock climbing again and we were all taught how to tie a figure-eight loop (or whatever else you want to call it). We got the standard instructions on how to tie a figure-eight and then were told to follow the running end back through the knot.

It got me wondering though, I've been told how to tie a figure-eight loop many times but I've never been told how to properly dress it, if there is such a way. Just following through and pulling everything tight seems to lead to several rather ugly variants :)

I don't think there wouldn't be too much of an effect on the strength of the knot, but I could see some ways being easier to untie than others. That's just a guess though, has anyone looked into this before?
Title: Re: Dressing a figure-eight loop?
Post by: Hrungnir on September 14, 2011, 01:27:18 PM
The tying method is correct.

However, make sure  there are two strands beside each other at each part of the knot and that the strands are not crossing each other.

Correctly dressed knot:
(http://www.animatedknots.com/images/figure_8_loop_knot.jpg)
Title: Re: Dressing a figure-eight loop?
Post by: xarax on September 14, 2011, 01:56:48 PM
   See the symmetric variants at :
http://igkt.net/sm/index.php?topic=2198.msg15419#msg15419
   An "ugly" knot is not weaker than a pretty one   :), unless the "ugliness" is due to an asymmetry. In this case, and this case only, one can say that the forces inside the knot s nub are not evenly distributed, and this might lead to the formation of some weaker areas/links.
   In general, it is believed that the most vulnerable areas are situated along points of smaller curvature, and there is a handful of studies that relate curvature at those points with strength. See (1), (2), and references therein.
   I believe that the double eight knot was chosen because of its resistance to slippage, not because of its greater strength. In the climbing world, there was never a systematic, exhaustive strength test of all the different variations of this knot - or of any dressing variation of the few knots used there...Not one of the many rope manufacturer of climbing ropes, did ever a systematic strength study of those different variations of the double eight bend, and all that is taught to the students in climbing schools, is to tie only the standard variation. Who can prove that some other ,"ugly" variation, is not stronger and safer than the most frequently used one ? So much for the claimed "attention to details" in this supposedly technology-oriented field ! It is the progress in the material sciences that saves the ass of the climbers, not their scientific literacy.

1) http://arxiv.org/ftp/physics/papers/0103/0103016.pdf
2) http://www.lmm.jussieu.fr/~audoly/publi/ACNKnots-07.pdf
Title: Re: Dressing a figure-eight loop?
Post by: DDK on September 14, 2011, 08:28:52 PM
(1) The technique I use is to first lay the figure eight flat and have the working end leave the figure eight going away (down) from me.

(2) Then I rethread the working end back through the figure eight having it lay always on TOP of the rope it is following back producing a top and bottom pair.

(3) Once the rethreading is complete, partially tighten by first pulling on the working end (which should be the one on TOP of the pair) AND the leg of the loop which is connected to the BOTTOM of the pair.  (This is the tightening of the inner wraps of the knot, i.e. the inner Thief Knot)

(4) Then pull on the standing part (which was on the BOTTOM of the pair) AND the leg of the loop which was connected to the TOP of the pair. (This is the tightening of the outer wraps of the knot)

This process automatically correctly dresses the loop and produces it in what I believe is termed the "strong" form (standing part connected to outer wrap of the knot).

(personally, I prefer the "weak" form which is produced by transposing the words TOP and BOTTOM above and reversing the two tightening steps)

DDK
Title: Re: Dressing a figure-eight loop?
Post by: Dan_Lehman on September 16, 2011, 05:53:46 AM
I've been told how to tie a figure-eight loop many times
but I've never been told how to properly dress it, if there is such a way.
Just following through and pulling everything tight seems to lead to several rather ugly variants :)

Yes, isn't that peculiar --that there is much advice to "dress
and set the knot properly," but no specific guidance on what
this might be!?  In a great many presentations of this knot
(or is it just 2-3, repeatedly echoed?), it is not even indicated
which of the ends should be the one loaded, which the tail.

In practice, all sorts of things seem to be tied & tried,
and so far, I'm unaware of any serious shortcomings to any
of them.  Still one might wonder at some vulnerability yet
to be tickled, and further ... .

Of the set of 4 versions of the knot that Xarax points to in
another thread, I call the topmost one "the perfect form",
and the loading of the left white end vs the right side's
orange, "the strong form", with the "weak form" being
the other loading (orange vs. white, left to right).  --this
for the end-2-end knot; for the eye knot ("loop knot"),
one side will have both ends loaded --legs of the eye.

Xarax's knot-C is what Dave Merchant recommends as being
up to 10% strronger, and easier to untie.  I find it an
unseemly and not-so-easily oriented version.  Btw, my
"strong form" arose from a similar assertion of greater
strength (of like magnitude), by someone who could
be believed to be attentive to the knot (and who at least
had presented an unambiguous form, and pointed to the
difference of which end was loaded!).

I recall one fellow who did some informal knot-breaking
in a contest of end-2-end knots, where each test specimen
was configured with two such knots (in a knot-A vs. knot-B
competition), and the ends were terminated with some form
of figure 8 eyeknot --I don't recall him having the details.
In all cases, one of the end-2-end knots broke, never these
eye knots!  And this was with not only the fairly strong knots
such as fisherman's, butterfly bend, & the Zeppelin but
also double fisherman's, blood knot, and even one that
OUGHT to have been equal, twin fig.8 bend !!?

--dl*
====
Title: Re: Dressing a figure-eight loop?
Post by: knot4u on September 18, 2011, 03:57:03 AM
Related issue, do you see any teachers in the climbing world teaching any alternatives to the Figure 8 Follow Through?

For example, is the Zeppelin Loop ever taught?  It's less bulky, easier to dress, quicker to tie, easier to untie, perhaps more secure, and probably stronger.  One disadvantage is the Zeppelin Loop is more difficult to learn for most people, but the benefits outweigh this disadvantage if the student can learn it.
Title: Re: Dressing a figure-eight loop?
Post by: SS369 on September 18, 2011, 02:56:33 PM
Hello knot4u,

yes, to answer your question, some do teach alternatives to the "standard" re-threaded 8 loop as a tie in loop.
There are threads here that go on about the many variants of the bowline for this purpose.

On the topic: I think that so long as the dressing is neat, paths "railroad tracking" each other in the final tightened knot that the security and strength< for that's worth, is essentially the same.

Quote
In regard to the Figure 8, On Rope (page 45 , col 2) states : According to Montgomery, it is possible to tie this knot incorrectly, resulting in about a 10 % strength loss. Knot destructive tests and Ashley confirm that this is not the case. Whether the standing line travels the inside or the outside path, it really makes no difference. [unquote]

SS
Title: Re: Dressing a figure-eight loop?
Post by: DDK on September 18, 2011, 03:24:09 PM
@dl & @SS

I would like to be more sure of what you mean by 10% increase (decrease) in strength.  The usual interpretation would be that if the original knot had a strength of 200 pounds, than an increase in the knot strength by 10% would mean that the new knot had a strength of 220 pounds.  Equivalently, if an original knot had a strength of 40% of the ultimate breaking strength of the material in which it was tied, then the new knot would have a strength of 44% of the ultimate breaking strength of that same material.  Just making sure we are talking apples to apples - thanks.

DDK
Title: Re: Dressing a figure-eight loop?
Post by: SS369 on September 18, 2011, 04:07:39 PM
I think that the subject, the 10% Reduction in strength, is a guess at best not proven sufficiently. But, I do think that what is meant (I could be very wrong) is that if the correctly dressed re-threaded figure 8 loop had a so called strength of 40% of the un-knotted line that the incorrectly dressed affair would then be at 30% of the breaking rate.

SS
Title: Re: Dressing a figure-eight loop?
Post by: xarax on September 18, 2011, 05:45:00 PM
paths "railroad tracking" each other

   Do you mean that this description is an unambiguous one ( is sufficient to help us dress the knot in one, the best, way), or that any dressing where the paths can be described as such, is adequately strong ?
Title: Re: Dressing a figure-eight loop?
Post by: SS369 on September 18, 2011, 08:59:00 PM
Paths "railroad tracking" is a one-timer. For this topic/knot only. Use at your own risk. lol

It may or may not have relevance to other knots and their proper dressing.

It is a term used by "decorative" knotters, braiders and such to communicate that the path of another line follow alongside, in essence parallel the first line's path.
Although I think a picture notated would be infinitely better, yes, I do think this is an adequate descriptor for the dressing of the re-threaded figure 8 loop.
That is if anyone understands it.
;-)

SS
Title: Re: Dressing a figure-eight loop?
Post by: xarax on September 18, 2011, 11:21:32 PM
   Railroad tracking, or roller coaster tracking ? That is the question...
   It is amusing that a car that runs on the (parallel) tracks of the "common" figure 8 A, d'Artagnan bend (1) makes 2 inversions ( it turns upside down two times), so it looks more like a roller coaster car, while the car that runs on the (also parallel) tracks of the figure 8 B, Athos bend, tilts but does not turns upside down, so it looks more like a railroad car. ( A car on the tracks of the C, Porthos bend, makes 2 inversions -like at A-, while at the tracks of the D, Aramis bend, makes only 1.)

1) http://igkt.net/sm/index.php?topic=2198.0
Title: Re: Dressing a figure-eight loop?
Post by: SS369 on September 18, 2011, 11:26:58 PM
Roller-coasting, in the final dressed form, may be more apt, but it makes me dizzy! ;-)

SS
Title: Re: Dressing a figure-eight loop?
Post by: DDK on September 19, 2011, 04:00:31 AM
There is something to be said for a bend which can be easily identified as having been tied correctly.  This would seem to favor the "perfect" forms of the Figure 8 Bends and Loops.  Along with our love affair with symmetry, this likely explains the popularity of teaching the "perfect" forms.

As far as "strong" vs. "weak" for the Figure 8 Bend, I find the "weak" form more natural when tying from the Thief Knot.  In addition, I do not care for the distortion in the "strong" form where the outer collars try to "overrun" the inner collars.  Admittedly, if the "strong" form is actually stronger, this "distortion" is likely part of the reason.

DDK

Title: Re: Dressing a figure-eight loop?
Post by: xarax on September 19, 2011, 09:47:30 AM
Along with our love affair with symmetry, this likely explains the popularity of teaching the "perfect" forms.

  There 17 symmetric figure 8 bends, to my latest counting... :) (1) .I doubt that one can prove, with any simple means, that any one of them is more symmetric from the others. The figure 8 bends presented at (2) are symmetric, too ! All those bends can be tied as easily as any one of them, because they are formed by the same number of tucks..I  am afraid symmetry by itself can not help us decide which are the best forms. We need detailed  tests, that are missing. The popularity of teaching the so-called "perfect" forms can only be explained by ignorance from the part of the teachers - and the well known attitude of all ignorant teachers to hide their ignorance under some rug...Are we also going to pretend that we know, when we simply do not, and keep parroting the same old wrong things ? Knotting is not going to be promoted that way, that is for sure.

1) http://igkt.net/sm/index.php?topic=2198.0
2) http://igkt.net/sm/index.php?topic=3148.0
Title: Re: Dressing a figure-eight loop?
Post by: DDK on September 19, 2011, 04:24:06 PM
Along with our love affair with symmetry, this likely explains the popularity of teaching the "perfect" forms.

  There 17 symmetric figure 8 bends, to my latest counting... :) (1) .I doubt that one can prove, with any simple means, that any one of them is more symmetric from the others. The figure 8 bends presented at (2) are symmetric, too ! All those bends can be tied as easily as any one of them, because they are formed by the same number of tucks..I  am afraid symmetry by itself can not help us decide which are the best forms. We need detailed  tests, that are missing. The popularity of teaching the so-called "perfect" forms can only be explained by ignorance from the part of the teachers - and the well known attitude of all ignorant teachers to hide their ignorance under some rug...Are we also going to pretend that we know, when we simply do not, and keep parroting the same old wrong things ? Knotting is not going to be promoted that way, that is for sure.

1) http://igkt.net/sm/index.php?topic=2198.0
2) http://igkt.net/sm/index.php?topic=3148.0


I disagree with your assessment of the ease of tying and identifying a specific non-"perfect" form of ABOK 1411 (Figure 8 Bend).   It is the additional twists which make them less easy to tie and identify and give them the appearance of being less compact in my opinion.  In the absence of other more compelling factors to differentiate the bends, the properties I mention are more than sufficient for preferring and teaching the "perfect" form.

As far as what teachers teach, it is more-or-less expected that they will teach the "tried and true" (meaning tested and it worked, repeatedly), especially if life or limb is at risk.  There is no other reasonable alternative and has nothing to do with their ignorance or pretending to know everything.  While you see ignorance, I see the current state-of-the-art.

DDK
Title: Re: Dressing a figure-eight loop?
Post by: xarax on September 19, 2011, 04:58:40 PM
I disagree with your assessment of the ease of tying and identifying a specific non-"perfect" form of ABOK 1411 (Figure 8 Bend).   It is the additional twists which make them less easy to tie and identify...

You disagree, obviously, but you are wrong...As I have said in
http://igkt.net/sm/index.php?topic=3618.msg20727#msg20727

the "other", "not-perfect" forms of the Figure * bends have less twists, not more, and are at least as easily to tie as any of them, because they are tied by the same number of tucks. The so-called "perfect form"should need more objective reasons to justify its name...So, because those properties you claim do not exist, and there is "absence of other more compelling factors to differentiate the bends", the arbitrary use of the name "perfect form" for any single one of those 17 bends, is a misnomer.

As far as what teachers teach, it is more-or-less expected that they will teach the "tried and true" (meaning tested and it worked, repeatedly), especially if life or limb is at risk.  There is no other reasonable alternative and has nothing to do with their ignorance or pretending to know everything.  While you see ignorance, I see the current state-of-the-art.

   Those 17 bends are not tried, so we can not tell if they are truly more safe or not. I know that it is "expected" that the teachers teach only what they know, but they should teach also that this is only what they know, and not the "best" or "perfect, as they always do ! The "alternative"e is to test those bends, and prove which is safer, and how much safer it is. So, teaching only the so-called "best" or perfect" form of the many figure 8 bends, is only a proof of ignorance and pretension of omniscience. While you see current state-of-art (sic), I see only ignorance and easy parroting. Fortunately, progress of materials save the poor pupils of those "teachers", but this has a side effect : Those people continue to walk around telling everybody that, in the field of their "speciality", anything that should and could be known is already known, so further research is useless.  
Title: Re: Dressing a figure-eight loop?
Post by: DDK on September 19, 2011, 06:33:34 PM
. . . the "other", "not-perfect" forms of the Figure * bends have less twists, not more, . . .

I disagree and base my observation on the use of a single strap to model the "railroad tracking" rope pair and bend.  This process quickly reveals the additional twists in the non-"perfect" ABOK 1411 bends.  edit: tighten the strap to most easily see the effect of the additional twists
  
Those 17 bends are not tried, so we can not tell if they are truly more safe or not. I know that it is "expected" that the teachers teach only what they know, but they should teach also that this is only what they know, and not the "best" or "perfect, as they always do ! The "alternative"e is to test those bends, and prove which is safer, and how much safer it is. So, teaching only the so-called "best" or perfect" form of the many figure 8 bends, is only a proof of ignorance and pretension of omniscience. While you see current state-of-art (sic), I see only ignorance and easy parroting. Fortunately, progress of materials save the poor pupils of those "teachers", but this has a side effect : Those people continue to walk around telling everybody that, in the field of their "speciality", anything that should and could be known is already known, so further research is useless.  

It would seem somewhat tedious to append "by the way, this is our current state-of-the-art and understanding and there might be other things better, but, because we haven't found them yet, I can't tell you about them" to the end of every statement.  It is my experience that this caveat is implicitly understood in most fields of expertise.

DDK
Title: Re: Dressing a figure-eight loop?
Post by: xarax on September 19, 2011, 06:44:42 PM
I disagree and base my observation on the use of a single strap to model the "railroad tracking" rope pair and bend.  This process quickly reveals the additional twists in the non-"perfect" ABOK 1411 bends.

a car that runs on the (parallel) tracks of the "common" figure 8 A, d'Artagnan bend (1) makes 2 inversions ( it turns upside down two times), so it looks more like a roller coaster car, while the car that runs on the (also parallel) tracks of the figure 8 B, Athos bend, tilts but does not turns upside down, so it looks more like a railroad car. ( A car on the tracks of the C, Porthos bend, makes 2 inversions -like at A-, while at the tracks of the D, Aramis bend, makes only 1.)


  I would be glad to be able to test the state-of-art roller coasters ! Because I see no state-of-any-art-whatsoever in those 19 century , steam-age railroad wrong assumptions about the figure 8 bend !
Title: Re: Dressing a figure-eight loop?
Post by: DDK on September 19, 2011, 06:51:31 PM
I just had a thought which might explain different points of views on the number of twists.  If one were to wrap a strap smoothly around a pole, one might say there are no twists, i.e. "perfect" wrapping and another might say there are the same number of twists as wraps.  I have taken the former view where it appears xarax has taken the latter.  FWIW

DDK

edit #1:  the "perfect" forms of ABOK 1411 are smoothly wrapped while the non-"perfect" bends are not.  edit #2: It occurs to me that the amount of rope-to-rope contact will likely be greater in the "perfect" form of ABOK 1411.
Title: Re: Dressing a figure-eight loop?
Post by: xarax on September 19, 2011, 07:20:27 PM
the "perfect" forms of ABOK 1411 are smoothly wrapped while the non-"perfect" bends are not.

   Nice try. Close, but no cigar... :) The A bend is similar to the C, in the number of wrapping, be them counted your way or mine...And there are MANY more fig 8 bends to test your theory...
   Please, do not try to find other, secondary things to prove the uniqueness of your Standards...The amount of rope-to-rope contact can be measured very easily, and then we can see if it is greater at some bend than at all the others...I bet the so-called "perfect" form will be left far behind in such a race - but this should not be considered as a proof or as disproof about its safety. There are so many other things that should be taken into account... I think that only detailed laboratory tests will solve our problem here.
Title: Re: Dressing a figure-eight loop?
Post by: DDK on September 19, 2011, 07:53:56 PM
the "perfect" forms of ABOK 1411 are smoothly wrapped while the non-"perfect" bends are not.

   Nice try. Close, but no cigar... :) The A bend is similar to the C, in the number of wrapping, be them counted your way or mine...

When tied in a single strap, the C bend shows obvious distortion on tightening and must be constrained from contorting to a more smooth wrapping.  As I mentioned in reply #20, how one numbers the twists will depend on your point of view.  With the example I gave, that is smoothly wrapping a pole with a strap, one could easily put in opposing twists at different points in the wrapping and maintain the same number of twists as the smooth wrapping, but, certainly, it would be much uglier and not as smooth.  This is your C bend.

DDK

Title: Re: Dressing a figure-eight loop?
Post by: xarax on September 19, 2011, 08:25:44 PM
   When tied in a single strap, the C bend shows obvious distortion on tightening and must be constrained from contorting to a more smooth wrapping. ...how one numbers the twists will depend on your point of view.

  Now you move again to other considerations...First,, there was the "additional twists", that, at the C bend, happened to be as many as the ones at the A bend. Then, there was the "amount of rope-to-rope contact", that turned out to be less in the A bend than in other bends. Now, it is the "distortion that must be constrained"...
  I know you try hard to prove your point, but you keep changing points of view, and this might not be the best strategy ! My gut feeling is that there must be a unique, simple factor that should be able to sufficiently single out A, d Artagnan, from the three musketeers, and possibly from the other bends as well, but we are not going to figure it out by such desperate attempts. Try to think in more abstract terms, may be thinking how the rope that forms each link would "see", "feel", and "understand" the knot, i.e. think in terms of a local, intrinsic, rather than a global, outside view of the tangled configuration.   
Title: Re: Dressing a figure-eight loop?
Post by: SS369 on September 19, 2011, 08:50:48 PM
May I suggest the use of notated pictures for clarity of statements about the dressing of a figure 8 loop?

SS
Title: Re: Dressing a figure-eight loop?
Post by: xarax on September 19, 2011, 09:04:31 PM
   We are talking about different dressings of the ABoK 1411 bend here, and if we can really prove that the A bend is the "perfect" one indeed,  the "Gold Standard". See (1), and pictures attached there, and here. Unforunately, we can attach only 4 pictures in ech post, and the different variations of the fig.8 bend are many more !  :)

1) http://igkt.net/sm/index.php?topic=2198.0
Title: Re: Dressing a figure-eight loop?
Post by: SS369 on September 19, 2011, 09:42:30 PM
Are we going to assume that what works as the proper dressing in a bend will also be best practice for a loop? Figure 8 for both instances.
The loading is a wee bit different in the two. And different in the loop application as determined by a few factors, e.g., diameter of structure that the loop encompasses, unstable moving load (such as  climber tied in or swing), etc.
So if you had to pick just one proper dressing, of the Figure 8 loop, that would most satisfy all scenarios of usage, which would it be and why?.

SS
Title: Re: Dressing a figure-eight loop?
Post by: xarax on September 19, 2011, 10:22:29 PM
Are we going to assume that what works as the proper dressing in a bend will also be best practice for a loop?

   This never happens. When the loadings are different, the knots are different. Having said that, the high symmetry of the fig.8 configuration should allow us examine both of them in parallel, at least partially. The more subtle things you mention, "diameter of structure that the loop encompasses, unstable moving load", are beyond my abilities to examine, I am afraid.
   I would be happy if, at the end of the day, it is proved the best dressing for the bend would be the best dressing for the loop, too - but, of course, I can not be sure about it...It would be nice if we have one mental picture for both knots, would nt it ?
   So, I think we have first to examine the bend, and only afterwards, with all the knoledge we would have aquired during this endeavour, to examine the more special and difficult case, the loop.