Author Topic: Gripping Hitches: A Proposed Rating System  (Read 17793 times)

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 3768
Re: Gripping Hitches: A Proposed Rating System
« Reply #45 on: February 27, 2012, 07:01:26 AM »
1) one limits material considerations and so loses insight to learn how such factors affect gripping;

I am afraid we have to limit the materials, so we have comparable numbers in all the other things.
When we will have some numbers, that show that, in one common material, at some load,
the A gripping hitch with X coils slips, while the B gripping hitch that uses the same number
of coils does not, then we can proceed and see if this relation is reversed when those hitches
are tied with/on another material.

This sounds as though you think it's necessary for many
people to generate your "20 cases" rather than you yourself
(or any one/few of others)?!  Whereas I find repetition within
anyone's reach, but variety of materials is less easily got,
but more so by collective action.

Quote
I would rather aim for more common loading,

   I have asked you which loading ( as a percentage of the "absolute", knot- or rope- breaking one)
you do consider as the "more common", but you have not answered.

I have answered that percentage is a fools gold of value,
and that a "common load" is just that --of absolute force,
say, 100#.  Perhaps a set of common loads can be agreed,
such as some measured drop of barbell weights for shock;
some body weight stressing; and some pulley-assisted
higher loading.


Quote
by varying the materials

   I think that this is what is "unrealistic" indeed - as the possible materials
( taking into account composition, surface friction, structure, patterns of fiber
braiding, etc ) are innumerable ! And I also believe that this "reasonable" requirement,
the noble form of which is the dictum "material matters",  has often been used as
an excuse to us knot tyers, so we never start testing anything ! ( and keep blah-blahing...).

And if the results cannot endure the variety of materials,
of what use are they?  (Awkward wording : but I mean that
if one thinks testing must be so narrow, because these differences
would have effect, how valuable will narrow results be, then --as
the world of usage will differ?!)

Again to the first point made above, why should several
of us use, e.g., "hardware-store solid-braid 1/4-inch nylon"
--i.e., the same/common material-- and run tests ?  Is it
that we each can only do unique knots, and must join our
disparate results by a common material?  Otherwise, I see
little benefit; or we are double-checking each other, whom
we don't trust to get it right.

It will help to have some at-least-similar circumstances (along
with different ones) for some bit of *sanity* checking, verification,
replication.  But I think we stand to learn more by use of a variety
of conditions than of trying to mimic one narrow one.

As for any of this being an "excuse", well, that's silly.  (We can
delay for plenty of other excuses --don't need any more for that.)


--dl*
====

xarax

  • Sr. Member
  • *****
  • Posts: 2781
Re: Gripping Hitches: A Proposed Rating System
« Reply #46 on: February 27, 2012, 04:00:36 PM »
   you think it's necessary for many people to generate your "20 cases" rather than you yourself (or any one/few of others)?!   
...variety of materials[/u] is less easily got,

   We can easily repeat a test 20 times - it is not 20 times more difficult than a single test !  :) Yes, I think it is necessary for anybody who can test, to test, and test each knot 20 times.
   Me, personally, knows next to nothing about materials ! I could not explore this vast "variety of materials" even if I had the time to do it...because I do not have the required knowledge and experience...A collective endeavor could possibly attract/persuade a rope manufacturer to contribute to such an effort, for marketing purposes - if for nothing else.

percentage is a fools gold of value
 a set of common loads can be agreed, such as
some measured drop of barbell weights for shock;
some body weight stressing;
and some pulley-assisted higher loading.

   I see that you have a certain difficulty with numbers... :) Which is this "common" load? What relation will it have with the material or the diameter of the rope ?  How much those "barbel weight" would weight ? Dropped from which height ? How much mechanical advantage is needed, on what force ? How many pulleys ? Which higher loading ? 
   ("Body weight" stressing would only count for dressing and pre-tensioning the knot, I suppose. Also, fortunately or not, there is not ONE body weight - and my body weight is not enough even for a tight enough pre-tensioning, I am afraid !  :) )

And if the results cannot endure the variety of materials, of what use are they?

  Differences of different knots due to their own structure - that remains the same- will lead to better "informed analysis"- useful when the same knots are tied with/on different materials, I guess.

why should several of us use, e.g., "hardware-store solid-braid 1/4-inch nylon"--i.e., the same/common material-- and run tests ?

   The same question was asked about languages - not materials - some time ago, in a place named Babel, if I remember right... :)
« Last Edit: February 27, 2012, 04:15:20 PM by xarax »
This is not a knot.

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 3768
Re: Gripping Hitches: A Proposed Rating System
« Reply #47 on: February 27, 2012, 08:05:07 PM »
   you think it's necessary for many people to generate your "20 cases" rather than you yourself (or any one/few of others)?!   
[in contrast to => ...variety of materials is less easily got,

We can easily repeat a test 20 times --it is not 20 times more difficult than a single test !  :)
Yes, I think it is necessary for anybody who can test, to test, and test each knot 20 times.

That's not my point : by insisting that everyone do this testing
with Rope-X and Ojbect-Y, what is the point of all this repetition,
or was it by 5 testing 4 times each we reach the desired 20?!
Why is it not sufficient that any one person test (however...)many
times with a particular circumstance, and have added into that
result those of others testing the same structure under different
circumstances?  (And we are likely to naturally come to having
a variety rather than so similar circumstances --though we might
strive to find some commonality, ALSO, among us or pairs of us,
to use for repeated-testing sort of verification.)

I can see this : with some common circumstance, each person
can test unique knots and --presuming such testing to be well done--
thereby we get immediate comparative values.  And some new
knot can be presented via this common testing value at the
start.  Still, we lose if there is no variety of circumstance to help
outline a structure's behavior.

.:.  It's not either/or, we can work towards both.

Quote
I could not explore this vast "variety of materials" even if I had the time to do it...
because I do not have the required knowledge and experience...
A collective endeavor could possibly attract/persuade a rope manufacturer
to contribute to such an effort, for marketing purposes - if for nothing else.

But you have some collection of kernmantle cordage,
to which I think Knot4U has more mundane solid-braid
cordage, and I have misc. laid & braided flotsam-jetsam
marine cordage (and several of us have various kernmantle,
and solid-braid, and laid).
Again, this is to my point : YOU (or any of us) cannot explore
the vast domain, but collectively we can cover some contrasting
circumstances, and get some results that can fuel theories of
knot mechanics to be tested in such circumstances with our
variety of knots.  --which won't come if choosing a single
circumstance.


Quote
percentage is a fools gold of value
 a set of common loads can be agreed, such as
some measured drop of barbell weights for shock;
some body weight stressing;
and some pulley-assisted higher loading.

   I see that you have a certain difficulty with numbers... :)
Which is this "common" load?
What relation will it have with the material or the diameter of the rope ?
How much those "barbel weight" would weight ?
Dropped from which height ?
How much mechanical advantage is needed, on what force ?
How many pulleys ?
Which higher loading ? 
   ("Body weight" stressing would only count for dressing and pre-tensioning the knot, I suppose. Also, fortunately or not, there is not ONE body weight - and my body weight is not enough even for a tight enough pre-tensioning, I am afraid !  :) )

Problem with numbers?
All of your questions can be answered,
possibly with some set of desired values
(say, static wgt. of <whover's lightest?>
/ 125# / 200#)
Dropped heights can be agreed, again with
perhaps some couple/few : 5-10-20cm ?
"Higher loading" might be left more "implementation-dependent"
to accommodate various circumstances --allowing that
we'll be considering/comparing somewhat different
forces per case.  (Where some curious difference of
behaviors arises, effort can be made to draw the testing
closer together in circumstances, to explore this.)

"What relation ... on material or diameter of rope?"
Exactly the sort of useful information one might get,
only by having such variety in testing.  E.g., we might
find that we can hold, *statically*, 125# in smaller
but not larger diameters; and that *dynamic* loading
shows greater/lesser difference between diameters?!

CONDITIONING (dressing/setting) of the knots pre-test
will be something to attend to; how to do this with some
transferable commonality is a challenge.  Clever applying
of dead weights is a seeming way to achieve uniformity,
but maybe not so easily/practically done?!  (Certainly,
"pull hard as you can" guidance gets different results with
me pre-/post-accident!)

Quote
And if the results cannot endure the variety of materials, of what use are they?

  Differences of different knots due to their own structure --that remains the same--
will lead to better "informed analysis" --useful when the same knots are tied with/on different materials, I guess.

[nb: I'm choosing ' --<parenthetical>-- ' for dashes, so to bind the
typographical mark to the associated text; trad. way of surrounding
the dash (and "em dash" is used here, and should be '--' w/o special
typographical, long-dash figure) can be harder to discern esp. where
the assoc.'d text spans multiple lines!   :)  ]


Yes, we want to understand structural effects.  I think that
we'll get to this as well or better with a variety of conditions.
I'm not against looking for one or some common ones,
but do want to have a variety in play.


Quote
why should several of us use, e.g., "hardware-store solid-braid 1/4-inch nylon"--i.e., the same/common material-- and run tests ?

   The same question was asked about languages - not materials - some time ago, in a place named Babel, if I remember right... :)

Linguists are now urgently concerned about the LOSS
of languages, and the loss with them of their unique
presentations of how humanity perceives the world
and expresses itself!


--dl*
====

xarax

  • Sr. Member
  • *****
  • Posts: 2781
Re: Gripping Hitches: A Proposed Rating System
« Reply #48 on: February 27, 2012, 09:35:47 PM »
Why is it not sufficient that any one person test (however...) many times with a particular circumstance, and have added into that result those of others testing the same structure under different circumstances?

   It is sufficient- provided the "however many" times, would be at least 20 times !  :) That is exactly, and all what I have said. Otherwise, I am afraid that me, for one, I am not going to believe anything, because I do not believe in once-happening miracles... :) ( As "circumstances", I read here other materials, longer specimen, other types of loadings (for example, dynamic loading),different construction (laid, solid or not , single or double braided, kernmantle, etc,)

with some common circumstance, each person can test unique knots and --presuming such testing to be well done-- thereby we get immediate comparative values.
...any of us cannot explore the vast domain, but collectively we can cover some contrasting circumstances, and get some results that can fuel theories of knot mechanics to be tested in such circumstances with our variety of knots.  --which won't come if choosing a single circumstance.

"Bingo" !  :)

you have some collection of kernmantle cordage,

   Only about 2 meters of each, I am afraid...( which length I was begging to acquire from the stores...) With my home-made universal test machine. I need 2-2.5 meters for one test, so for 20 tests I need 40-50 meters - and if I test also wet ropes, as I am afraid I will be obliged to do if I use nylon ropes, I need 100 meters...FOR EACH KNOT.

All of your questions can be answered, possibly with some set of desired values

   That is what we need...Now, as I believe a special area of practical knotting is at the outdoors/rescue activities, where a greater margin of safety is requited, I always think in terms of ropes and not cords - so I am talking about greater diameters, weights and heights than you are...That is why I have proposed the "round" 1/2' , easily available in stores.

CONDITIONING (dressing/setting) of the knots pre-test will be something to attend to; how to do this with some transferable commonality is a challenge.  Clever applying of dead weights is a seeming way to achieve uniformity, but maybe not so easily/practically done?!

Measure the force with which you pull the ends by the use of an instrument...There are plenty of cheap, reliable electronic instruments around.

Linguists are now urgently concerned about the LOSS of languages, and the loss with them of their unique presentations of how humanity perceives the world
and expresses itself!

First, do not confuse the more objective "perceives the world", with the more subjective "expresses itself". We want to perceive, and understand, this little corner of the world that is named "practical knots" here...Now:
1. Linguistics have proven that all the languages of the "humanity" have a common structure, that has also shaped the human brain. If you do not learn at least one language when you are a child, you are not going to be able to use any language when you will grow up... because your brain will not be able to re-trace the path of your ancestors, of how human brain has evolved in the course of human pre-history.
2. There is one and only language that has proven to be the most powerful tool the human race has discovered, and that language is mathematics - i.e. numbers, among other things.
   Let us rely on some f.. numbers, and leave the extensive linguistic gymnastics... They can harm the brain more than any accident can harm the muscles...
This is not a knot.

richardpeterson

  • Jr. Member
  • **
  • Posts: 20
Re: Gripping Hitches: A Proposed Rating System
« Reply #49 on: March 02, 2012, 07:54:29 PM »
I saw some discussion here about different people testing in different circumstances. I don't know if the following suggestion has been put forth, but I'll make it anyway.
 
Perhaps if different people are going to perform testing with different materials and under different circumstances, they could all perform a baseline test, so that their results could be more meaningfully compared.
 
For instance - suppose each person follows this procedure along with their other tests:
 
Using the pole and line you will be testing with, tie a rolling hitch (the ABOK 1734 version) with the two turns at the bottom, using the lower end coming from the two turns as the standing part. Tighten or pretension as you will for your other tests. Mark a spot below the bottom of the rolling hitch a distance of one pole diameter, so you can see when the hitch has slipped by one pole diameter. Slowly load the standing end until the hitch slips to your spot. Record the maximum force the line sustained. Repeat this test 10 times and post the results along with any other test results you perform on other hitches.
 
The point is not the details of my test. It doesn't have to be a rolling hitch. It doesn't have to be one pole diameter slippage. It doesn't have to be that actual test at all.
 
I'm just proposing a point of reference that will be reproducible to some degree.
 
I'm not suggesting that the other tests should measure the force required to make the hitches slip. I'm saying that some kind of standard baseline test could help account for differences in materials and testing procedures.
 
That way, we get something like this (again - fiction here):
 
Quote from: knot_tester_X
Hitch X required 6 wraps to stay fast under maximum loading. My mean rolling hitch slip force was 20lbs

Quote from: knot_tester_Y
Hitch Y required 5 wraps to stay fast under maximum loading. My mean rolling hitch slip force was 100lbs

That doesn't exactly make for an apples-to-apples comparison, but I do think it would go a long way in helping make sense of heterogenous tests. Certainly the closer two people's baseline test values are, the more apt I am to directly compare their other results.

xarax

  • Sr. Member
  • *****
  • Posts: 2781
Re: Gripping Hitches: A Proposed Rating System
« Reply #50 on: March 02, 2012, 08:18:01 PM »
Perhaps if different people are going to perform testing with different materials and under different circumstances, they could all perform a baseline test, so that their results could be more meaningfully compared.

   That is exactly what Dan Lehman had proposed. My humble opinion is that we should start testing the different structures - rather than the different materials -, because I believe that differences in structures reveal more things about knots than differences in materials. So we have more to learn by studying first the many knot structures we have, and then try to proceed to the vast ( and expensive !  :) ) area of different materials.

Tighten or pretension as you will for your other tests.

   We should be careful here, because I think that pre-tensioning differences might not be "erased" afterwards, during normal loading... ( at least at some more complex knots, and in certain degree ). Pre-tensioning is a part and parcel of the dressing procedure - and different dressings are expected to lead to different results.

Repeat this test 10 times

  Make it 20, so a 5% ( a commonly believed important percentage number...) corresponds to one exception.

 I hope that you will stay in the Forum, and start testing - we NEED people, desperately ! Welcome !
« Last Edit: March 02, 2012, 08:19:02 PM by xarax »
This is not a knot.

richardpeterson

  • Jr. Member
  • **
  • Posts: 20
Re: Gripping Hitches: A Proposed Rating System
« Reply #51 on: March 02, 2012, 09:06:45 PM »
I hope that you will stay in the Forum, and start testing - we NEED people, desperately ! Welcome !

I plan to stay around, but I don't have the resources or expertise to do much more than casual testing, providing anecdotal results.
 
But I can contribute ideas!

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 3768
Re: Gripping Hitches: A Proposed Rating System
« Reply #52 on: March 04, 2012, 04:53:11 AM »
Perhaps if different people are going to perform testing with different materials and under different circumstances, they could all perform a baseline test, so that their results could be more meaningfully compared.

   That is exactly what Dan Lehman had proposed.

No, it goes further, in this point-of-reference testing for
comparisons.  (I see more value in the different-conditions
testing --vs. identical ones & varied structures--; but I'd not
put forward this common-structure baseline idea --and it's
a good one.)

I'm also not gung-ho about the 20-tests count : I really
hope that variety isn't so much that one would feel the
need for this (other than by some lust of "numbers"),
and might question the full conditions & rigor to go to
this extreme vs. doing other things.  (I.e., one might
question the method of measurement --accuracy...--
per case; and 20 times a dubious thing doesn't make
it much better, if at all.)

?!  a compromise of sorts re repetition :
that testing first make greater strides in diversity of everything
AND THEN --for some special cases of interest-- there might
be seen some few cases to get more rigorous with.
(This is assuming that the first course is not found to be
a load of slop, but in fact some decent insight into what
is working, what is changing how things work (changes
in structure, changes in conditions).  The 2nd course
becomes a focus on some particularly interesting cases;
maybe it's only done by a couple testers (even one!) who
have key ingredients.)

 - - - - - - - - - -

Something to do : agree on materials --cordage (should
be plural, for all of us (2 is plural  :) ) & objects.


--dl*
====

xarax

  • Sr. Member
  • *****
  • Posts: 2781
Re: Gripping Hitches: A Proposed Rating System
« Reply #53 on: March 04, 2012, 06:16:33 AM »
the first course
The 2nd course becomes a focus on some particularly interesting cases;

   Two, or even more courses / rounds is a good idea. If we have some tug-of-war type of tests - testing pairs of bends, for example - I had thought of even more "economical" systems, where only few of the contenders are going to be tested more times - like the systems used at chess tournaments.
This is not a knot.

DerekSmith

  • IGKT Member
  • Sr. Member
  • *****
  • Posts: 1518
  • Knot Botherer
    • ALbion Alliance
Re: Gripping Hitches: A Proposed Rating System
« Reply #54 on: March 04, 2012, 10:15:34 PM »

Elongation

Elongation is the amount by which the coils move from being as circular as possible under maximum loading (i.e., as much loading as the user can give it). In other words, elongation is the tendency for a gripping hitch to resist spreading out. Here's the formula...

Elongation = Average shift distance of coils = (Sum of shift distance of coils along pipe) / (Number of coils)

Excellent proposal k4u, but I do have one issue with the 'Elongation' parameter.

I think you are viewing your assessment of a knot based on the fact that the majority of these knots treat all the wraps as one, and if they distort one coil then they generally distort all the coils to more or less the same extent.  Certainly it is valid to judge these knots negatively if they distort the loops, because it is a sign that force is being transmitted (via the distorted loop) to the very back of the hitch, and therefor it is risking sliding it along the pole.

However, the KC hitch was designed to create grip, yet leave the all important anchor turns to grip while free from any dragging forces.  In the KC hitch, the first loop or two are intended to open in order to create the all important cord tension into the pole using the scissor action coupled with the huge mechanical advantage gained in the first few degrees of opening of a turn.

How would you score the KC hitch with your proposed scheme?

Also, this example of the KC Sling hitch, has three turns out, then three turns back - is this a six turn count, or because it is made with a doubled cord, is this a 12 turn hitch?  And does this score the knot poorly in your proposed system?

Derek


knot4u

  • Sr. Member
  • *****
  • Posts: 1076
Re: Gripping Hitches: A Proposed Rating System
« Reply #55 on: March 05, 2012, 12:01:49 AM »
On further thought, I don't think the proposal in the original post is definitive enough. I'm looking for a test in which the meaningfulness of the data cannot be denied.

So, I revise my thinking to go in an entire different direction. What I propose is a pole that allows a computer to measure the orthogonal force at any given point on the pole. That way, the measuring system not only measures coils and elongation, but it also measures the magnitude and position of forces orthogonal to the pole surface.

Once we have more definite data on orthogonal forces, we can directly calculate the amount of friction a hitch is imparting to a pole and where exactly the friction is taking place. The goal here is to be more scientific and to remove, as much as possible, how people "feel" a particular hitch may be behaving. Yeah, we're venturing off into electrical engineering, but it's certainly doable.
« Last Edit: March 05, 2012, 10:19:35 PM by knot4u »

richardpeterson

  • Jr. Member
  • **
  • Posts: 20
Re: Gripping Hitches: A Proposed Rating System
« Reply #56 on: March 05, 2012, 02:57:02 AM »

I have a few questions here, so I can better understand this idea.


What sorts of decisions do you hope people will make based on the ratings? To me, that seems like it would be the main driver of the kind of rating that is devised.


Also - why is the number of turns the basis for efficiency? Is it because more turns are presumed to be more work for the knot tyer, and thus less desirable?


Could efficiency alternately be measured in terms of the amount of line consumed by the knot?  I'm sure hitches that wrap at an angle consume quite a bit more line per wrap. The hitch that holds fast with the least amount of line could be considered the most efficient.  Of course, line length would have to be measured in multiples of a single 360 degree wrap, in order to account for the diameter of the line and pole (as opposed to measuring absolute length).  So a half-hitch would consume one unit of line.

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 3768
Re: Gripping Hitches: A Proposed Rating System
« Reply #57 on: March 05, 2012, 09:37:02 PM »
I have a few questions here, so I can better understand this idea.
What sorts of decisions do you hope people will make based on the ratings?
To me, that seems like it would be the main driver of the kind of rating that is devised.

At this inchoate point of things, I think that we can't
say much about the future; we want to make some
reasonable start, and with the experience of whatever
that is, we might see to some improved system --perhaps
motivated by some potential *outside* use of the ratings.
I think that the initial desire is just a way to *grade*/order
structures in terms of their efficacy.

Quote
Also - why is the number of turns the basis for efficiency?
s it because more turns are presumed to be more work for the knot tyer, ...

I'll presume to speak also for Xarax in saying that it's
believed that in general the build-up of many wraps will
ultimately succeed in gripping, but such structures are
not only tedious to tie (typically) but also wasteful of
material.  So, naturally, we'd like to see what structures
make best use of material, delivering equalized or maybe
some cleverly biased loading into parts such that a good
grip comes more cheaply of material (esp.), and maybe
of tying effort as well.

Quote
Could efficiency alternat[ive]ly be measured in terms of the amount of line consumed by the knot?
...

Within some bit of rough measurement, this will largely
be the same as counting wraps.  It is a nice datum one
can figure for knots, though, and somewhere I have some
notes about ratios of material-consumed / rope-diameter
for eye knots.  (It can be misleading as to how much material
is needed for tying the knot --i.e., one might need to work
with more for Knot-M than for Knot-P even though M when
finished consumes less.)


--dl*
====