Author Topic: Knot breaking strengths and theories - again  (Read 54884 times)

DerekSmith

  • IGKT Member
  • Sr. Member
  • *****
  • Posts: 1518
  • Knot Botherer
    • ALbion Alliance
Re: Knot breaking strengths and theories - again
« Reply #30 on: July 03, 2007, 08:34:58 PM »
Then came a gratuitous diversion into major whining,
which serves no purpose but to depress interest.  (This was briefly relocated to its own thread, where its lament
could be indulged, but apparently someone wants this whining to intrude where others are interested in another
topic and have to read it!  Heckuva way to (ab)use a forum! 

--dl*
====

I agree with much you have said Dan and apologise for for my whining intrusion.  I have deleted my "gratuitous diversion" so that others do not have to read it and hopefully you will continue to collate your list of test reports.

Derek
« Last Edit: July 03, 2007, 08:38:48 PM by DerekSmith »

agent_smith

  • Sr. Member
  • *****
  • Posts: 993
Re: Knot breaking strengths and theories - again
« Reply #31 on: July 04, 2007, 03:26:52 PM »
I have a great deal of respect for the IGKT and would very much like to be able to access knot information - rather like a knowledge bank.

My angle has - and probably always will be - knots used in human life support (ie mission critical applications).

I respect those who use knots for decorative purposes, but that is not my personal area of interest. If you tie a decorative knot incorrectly, there wouldn't be catastrophic failure with attendant death or disablement.

In climbing, abseiling, caving, mountaineering and rescue applications, get the knot wrong and you could be staring down the barrel of a real catastrophe.

Here are the knots that I would like to put to the guild as a starting point for analysis:


   KNOT                    ABoK No.       CATEGORY           ALTERNATIVE NAMES

1. Figure 8 loop                # 1047                End line knot           Figure 8 on the bight, Figure 8 follow-through
2. Re-threaded figure 8        #1047             End line knot           Rewoven figure 8
3. Double figure 8 loop         #1085              End line knot           ?Bunny ears? (a truly ridiculous name)
4. Clove hitch                    #1178                 Mid line knot
5. Double fishermans knot  #498 & #1415     Joining knot (bend)   'Double English knot' & a 'Grapevine knot'
6. Prusik Hitch                    #1763                 Sliding friction knot
7. French prusik           #1758, #1764    Sliding friction knot     'Machard Tresse'
8. Tape knot                   #1412               Joining knot (bend)   Ring bend, Water knot
9. Alpine butterfly knot       #331 & #1053   Mid line knot            Linemans loop, Butterfly
10. Munter hitch          #1195 & #1818   Load control hitch     ?Italian hitch? or ?HMS? or ?Mezzo Barcaiolo?
11. Rosendahl bend        N/A              Joining knot (bend)   Zeppelin bend


Note: Knot number #11 is controversial. There are conflicting ideas about which 'bend' is optimal in mission critical applications (ie if the bend fails, death will result in accordance with Newtons laws of physics).

Not all user groups would have need of all 11 knots. Classes of users who undertake multi-pitch lead climbing on cliffs would arguably have need of all 11. Indoor 'plastic pullers' would use only one or two knots at best.

Mine rescue teams would use most but not all of the 11 knots (the tape knot is not used by many Australian rescue teams on account of the wide availability of heavy duty sewn web round slings).

I'm going to re-assert the 4 S's which are:
1. Suitability to task
2. Stability (very important)
3. Security (very important)
4. Strength (not as important serials #2 and #3)

Dan Lehman might have a word or two to comment about this... but serial #1 is important to me. In any situation, there will be a knot that is best suited for the job. For example, there are instances where a clove hitch or a butterfly knot might work, but one might give optimum performance. In another example, the optimum knot (ie bend) for uniting two ropes may be in contention - double fishermans, Rosendahl, offset over-hand bend (OOB) also known notoriously as the 'Euro death knot'. In mission critical applications, serial #2 & serial #3 is of paramount importance. But, someone might need to properly define these concepts.

There is much needed data on 'slide & grip' hitches (ie sliding friction hitches) such as the Prusik and French prusik.

In Australia, controversy is raging at present on the optimum diameter of accessory cord to form a hitch around the parent rope. Some argue the diameter ratio is 50% - that is, a cord used to tie a prusik hitch will reach optimum performance when it is 50% of the parent rope diameter.

There appears to be general agreement that three (3) wraps provides optimal grab in the prusik hitch class.

French prusiks require a greater number of wraps - the exact number is dependent on the length of the loop and the cord/rope diameter ratio. Generally, if the wraps are too few or too loose, there is a risk of unexpected slippage with potential slippage.

Here is another link for you Dan: http://www.forestry.gov.uk/forestry/kirn-5lebje
Report date: 2002
Title: Ropes and Friction Hitches used in Tree Climbing operations
Author: Paolo Bavaresco, Treevolution


Its a bit slow to load, so it might be best to right click and choose "save target as". Some interesting results in there...

Here is another report by Ron Farmer in Australia: http://www.chockstone.org/TechTips/ProtectingAnAbseil.htm

Again, some interesting conclusions drawn.

These are prime examples where the tester was ignorant of the existence of the IGKT.

I personally am an advocate for using 6mm diameter accessory cord across the range of friction hitch species. I am also a made keen rock jock climber and prefer to climb with double ropes (ie I use two 'half' ropes - each 9mm diameter).

I have found that 6mm cord grabs and reforms well under cyclic loading on parent rope diameters in the 9-11mm range.

I would be keen to to receive feedback and wisdom from the IGKT on this subject...


agent smith

« Last Edit: July 05, 2007, 12:40:26 AM by agent_smith »

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 3767
Re: Knot breaking strengths and theories - again
« Reply #32 on: July 20, 2007, 07:22:26 PM »
My angle has - and probably always will be - knots used in human life support (ie mission critical applications).
I respect those who use knots for decorative purposes, but that is not my personal area of interest.
If you tie a decorative knot incorrectly, there wouldn't be catastrophic failure with attendant death or disablement.
Don't forget other practical uses of knots, such as commercial fishing--which unlike angling uses ropes of equal
and greater size than life-support applications; although in most cases knot failures in ComFish shouldn't result
in personal injury, you can imagine that those whose very engagement in the whole activity DOES have some
considerable (asserted to be highest ...) risk, and whose livelihoods depend on results, do not have great
tolerance for cordage failure, BUT, OTOH, have some experience and that also can be edifying.  (I think that
were some new rigger to make a trawl of lobster pots and then those trawls be pulled up losing all the pots
that it might become a personal-injury matter!)

Quote
Here are the knots that I would like to put to the guild as a starting point for analysis:
   KNOT                    ABoK No.       CATEGORY           ALTERNATIVE NAMES

1. Figure 8 loop                # 1047                End line knot           Figure 8 on the bight, Figure 8 follow-through
2. Re-threaded figure 8        #1047             End line knot           Rewoven figure 8
3. Double figure 8 loop         #1085              End line knot           ?Bunny ears? (a truly ridiculous name)
4. Clove hitch                    #1178                 Mid line knot
5. Double fishermans knot  #498 & #1415     Joining knot (bend)   'Double English knot' & a 'Grapevine knot'
6. Prusik Hitch                    #1763                 Sliding friction knot
7. French prusik           #1758, #1764    Sliding friction knot     'Machard Tresse'
8. Tape knot                   #1412               Joining knot (bend)   Ring bend, Water knot
9. Alpine butterfly knot       #331 & #1053   Mid line knot            Linemans loop, Butterfly
10. Munter hitch          #1195 & #1818   Load control hitch     ?Italian hitch? or ?HMS? or ?Mezzo Barcaiolo?
11. Rosendahl bend        N/A              Joining knot (bend)   Zeppelin bend

I regard #1 & 2 as identical knots, different tying methods.  (An interesting field study would be
to see if in fact the so-tied knots actually do generally result in identical knots, or, rather, if each
method has a bias toward one vs. another version of the Fig.8.  E.g., the most natural/easy way
to begin "re-threading" is a way that could lead to the Weak Perfect form; and sometimes it
seems that torsion build-up in tying in the bight leads to an imperfect completion.)  Interesting
that the Fig.9 is missing, though LoaL asserts that it's regarded as dominant in the UK.

Hmmm, I wonder at not having some variation of Two Half-hitches in the set:  what if one needs
to tie off w/o access to the rope end?  2HH w/stopper finish (Slip-knot, for bight tying), or the
Anchor hitch w/similar stopping, makes a handy way to effect this.  One can also tie a variation
of Bowline.  --or mere Hhitch w/bight through bight ... & Slip-knot finish.

Quote
Note: Knot number #11 is controversial. There are conflicting ideas about which 'bend' is optimal
One can re-tuck ends, which might both bump strength and further remove loosening => untying risk.

Quote
(the tape knot is not used by many Australian rescue teams on account of the wide availability of heavy duty sewn web round slings).
And they join any such slings by metal?  I'm still waiting for someone to figure out that a multiple "girth hitching"
(a name I do NOT like for this, but it's in the parlance currently) of slings makes seeming sense, at least until/if
testing shows some reason to avoid it.

Quote
I'm going to re-assert the 4 S's which are:
1. Suitability to task
2. Stability (very important)
3. Security (very important)
4. Strength (not as important serials #2 and #3)
I guess #1 is pretty much the obvious:  if one is going to tie two rope ends together (say,
to make an endless loop/sling), a stopper knot or hitch is not going to be used.  It otherwise
is the entire consideration!  As to the importance of the others, one might challenge the
idea that a knot for tying in needs to be stable, if one keeps the eye small--i.e., that
the chance of any "abnormal" loading is effectively removed.  In the same way, I could
see that security in the sense of untensioned security might be deemed irrelevant
where the task at hand didn't have any untensioned period--and were it some heavy
loading, untiability might be a key characteristic (slack-security, stability, material efficiency
all being put aside).
But this direction entails knowing knotting more than merely being able to index from
a list of predetermined tasks into a set of knots, really.

Quote
For example, there are instances where a clove hitch or a butterfly knot might work,
but one might give optimum performance.
... or not; or the trade-offs might be too hard to evaluate to any reasonable degree of clarity.
(or even depend upon material!)

Quote
Here is another link for you Dan: http://www.forestry.gov.uk/forestry/kirn-5lebje
Report date: 2002
Title: Ropes and Friction Hitches used in Tree Climbing operations
Author: Paolo Bavaresco, Treevolution
Yes, I need to submit some analysis & questions re this to Paolo.  I find some of the results
confusing/inconsistent (but might be misunderstanding the test configuration).

Now, about critiquing these various test reports, that remains on the to-do list.
A general critique is that they don't show exact knots/knot-geometry at play; this means
that one can only guess at the significance of the test results.  On break tests, a general
criticism (or wish item) is that they don't say where the break occurs.  E.g., for a Sheet Bend,
is it usually the bight or the loop part that breaks?  --that much would be easily told
(which rope ...), as opposed to exactly where in the bight/loop it broke.

--dl*
====