Author Topic: Knot Testing  (Read 26779 times)

PaulKruse

  • Full Member
  • ***
  • Posts: 60
Knot Testing
« on: January 13, 2007, 08:57:41 PM »
I’ve had a project on my action item list for some time at work to test the strength of various common knots, using at least a couple of different types of rope.  These need to be knots useful for the hanging of suspended scaffolding, or the rigging of vertical or horizontal lifelines. 

At the very least, I need to test the following:

Bowline (left and right handed versions)
Clove hitch (Backed up with single and double half hitches)
Figure-8 loop
Eye splice

I’ll have to use soft lay ¾ inch three strand twisted nylon.  I will probably also use a marine kernmantle.  I’m considering also using static and dynamic kernmantle rope from a rock-climbing source.

Since my boss is paying for the test, I need to limit the scope to what meets his objectives.  He is interested in making sure that our scaffolding and fall protection riggers are using the correct and the safest knots with the correct rope type.  I’ve got access to all sorts of test data compiled by other testing agencies, and have come to the conclusion that none of it really applies to our business.

We own a fully equipped commercial rigging shop and have all the equipment we need to do this correctly.  No jury-rigged set-ups in the kitchen.  I’ve tested many knots and ropes in the past, but have never made a systematic comparison of various ropes and knots.

What other suggestions would you make?

Paul Kruse

Fairlead

  • Sr. Member
  • *****
  • Posts: 328
  • IGKT Member since 1984 - IGKT Librarian
Re: Knot Testing
« Reply #1 on: January 13, 2007, 11:37:34 PM »
Hi Paul,
I would be interested to know how you use the Clove Hitch - and why.  Before I make further comment on backing it up etc.
I would also be interested in the test results of a "Double Overhand Noose" at the end of a rope using both a soft eye which is inserted in a karabiner and a hard eye inserted before being pulled up.  This knot will hold a hard eye very well, especially under load, which will protect the rope from excess wear from attachements etc.
Another knot about which some controversy exists is the strength of a simple Overhand Knot - said by some to be the weakest stopper knot - but 'rumoured' to have survived on one end when a bowline broke on the other end.
Wish I could be with you - alas I think I may be the wrong side of the 'pond'

Gordon

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 2722
Re: Knot Testing
« Reply #2 on: January 15, 2007, 06:12:12 AM »
Welcome, back, Paul!  Happy 007.

I’ve had a project on my action item list for some time at work to test the strength of various common knots,
using at least a couple of different types of rope.  These need to be knots useful for the hanging of suspended
scaffolding, or the rigging of vertical or horizontal lifelines. 
...
I’ll have to use soft lay ¾ inch three strand twisted nylon.
I will probably also use a marine kernmantle.
I’m considering also using static and dynamic kernmantle rope from a rock-climbing source.
How about some polypropylene or coextruded ("copolymer") PP & polyethylene (e.g., Ultra Blue,
PolySteel)?  What is typically used for this hanging scaffolding (seems polyester or some
combo rope of that with PP or coextruded fibres in the core would be used)?
Perhaps for a good compromise of shock absorption and durability, less stretch,
some of the recently conceived "gym ropes" used by rock climbers might work?
They area designed with thicker sheaths to endure repeated use in climbing
gyms, and sacrifice the full capability to hold a UIAA-defined severe fall with
the low impact forces of typical climbing rope.  They might be more beneficial
in "fall arrest" than some of what I've seen advertised for that--which tend to
be substantial (5/8" dia.) ropes of fairly inelastic material!  (Though one can
use a special, shock-absorbing lanyard to attach to such a line.)

Quote
At the very least, I need to test the following:

Bowline (left and right handed versions)
Clove hitch (Backed up with single and double half hitches)
Figure-8 loop
Eye splice
You can get as much variation in dressing/setting ONE type of Bowline
as you might in variants--and that might be a good thing to test, too.
E.g., one can dress the knot such that the end is pulled away from where
it will be drawn by loading, and one can set the collar tighter/looser.
Ditto for the Fig.8 (and Gordon's suggested Overhand) loopknot:  for
starters, which of its two ends will you load?  --might see if that makes
a difference (cf. Lyon Equipment's testing for the UK's HSE).
Though if few test cases are done, well, less can be concluded (Lyon's
few tests exemplify this, unfortunately).

There is a Bowline variation below--rightmost, "Janus"--which you might
try (and it would be nice to try an extension of this one where the end is
brought a 2nd time around the SPart and through the "rabbit hole"--it's
a matter of both further securing but of one more diameter of rope that
the loaded SPart wil bend around.



and a precise/specific orientation for the Fig.8 loopknot can be found here:
www.iland.net/~jbritton/KnotPhotoContributions.html
Updated Link > www.pssurvival.com/PS/Knots/Knot_Knowledge_Photo_Illustrations_2004.pdf
(This knot is tied in so many various dressings, it's amazing that folks think
some test results for "Fig.8" mean anything w/o knowing the details!)

Btw, what are the circumstances for use of the loopknots?  I.p., are there
some applications where the knot will be (or can be) tied completely
before being put in place (e.g., tying the Fig.8 in the bight rather than
"re-weaving" it)?  If so, that gives me some ideas for replacing the Fig.8!
(And I think I sent some of them to you--derivatives of #1425, sort of
symmetric Fig.9s.)

Also, try backing up the Clove with an Overhand stopper.

Quote
Since my boss is paying for the test, I need to limit the scope to what meets his objectives.
He is interested in making sure that our scaffolding and fall protection riggers are using the
correct and the safest knots with the correct rope type.  I’ve got access to all sorts of test data
compiled by other testing agencies, and have come to the conclusion that none of it really
applies to our business.

...  have all the equipment we need to do this correctly.

What other suggestions would you make?
I suspect that although you have "all sorts of test data," there isn't sufficient specification
about the structure of the knots for you to assuredly replicate the testing!  --that's typical.

I thus strongly suggest that you photograph the test specimens at the start and mid-way
to rupture, and ruptured, if sensible
.  It's the only way others can see into exactly what
your testing might mean.  Where feasible, it would also be wise to e.g. tie TWO loopknots
in a line, so to ensure that after the test you have one broken and one in the form that
it got from nearly breaking (which can be compared to the "mid-way" image to see what
changes in structure occurred).  Such information might lead to insights into how to dress
and set knots in particular ropes.

Also, what sort of loading do you expect in practice?  Slow-pull destruction testing is
probably not such a good model of expected use.

Cheers,
--dl*
====
« Last Edit: December 09, 2012, 08:39:01 PM by SS369 »

PaulKruse

  • Full Member
  • ***
  • Posts: 60
Re: Knot Testing
« Reply #3 on: January 15, 2007, 10:39:13 PM »
Gordon:
I inherited a crew of construction ironworkers and their unique customs.  On this side of the pond (USA), clove hitches backed up by a single or double half hitch are used extensively to secure a lifeline to a piece of steel structure, usually an I-beam.  When I first came to work here years ago, the knot worried me.  So I did some informal tests, and found that its strength was about the same as a standard bowline tied around the same beam.  Both break at about 75-80% of the rope’s published minimum breaking strength.  It seems to be a matter of personal preference to use a bowline or a clove hitch, but I will rarely see an ironworker use anything else for his upper connection.  Just the other day I saw an iron worker for another company use a figure 8 loop, but that is the first time in twenty years that I saw them use something different.

The second half hitch backing up the first half hitch does not seem to add any value to the knot.  Either jams into hatchet knots after being loaded, but that does not really matter.  The only way to load a VLL is to fall on it, and we would throw the rope out after any fall.

The rope being used is a rather expensive soft lay three strand twisted nylon, ¾ in diameter.  I don’t like it because it is about the least abrasion resistant rope available.  I’d like to change to a 5/8-inch static kernmantle, but that would be a serious paradigm change for these ironworkers, who tend to be a very traditional lot.  I won’t make any changes unless I have good data to indicate a change would be wise.

I’m not sure what a Double Overhand Loop is.  Got a picture?  I might know it by another name.

A simple overhand knot might be a good one to try.  I know from experience that it will hold very well, at least in the rope that we use.  But I don’t have any hard objective data.

One thing to keep in mind is that a knot that works well in one rope may not work so well in a different type of rope.  The reason we use this soft lay rope is that it will take and hold any knot well.  But that can be deceptive.  Knots you expect to be strong in a hard rope have may have much less strength in a soft lay rope.  The Alpine Butterfly has broken at only about 45% of the published minimum breaking strength in some of our earlier tests, for example.  The local SWAT team uses a very hard kernmantle, which they had tested with the same knot.  It worked very well, but the standard bowline proved to be very inferior – exactly the opposite that I found with our soft rope.  You really need to match the knot to the rope you are using.

When talking about percentages, you need to be very careful.  I use the minimum published breaking strength for the rope, which is much lower than its actual breaking strength.  That gives efficiency percentages that are artificially high.

Paul Kruse

PaulKruse

  • Full Member
  • ***
  • Posts: 60
Re: Knot Testing
« Reply #4 on: January 15, 2007, 10:59:30 PM »
Dan:

Thanks for the welcome back.  Good to be back.  I’ve been assigned to other projects for a while.

As always, you give me lots to think about.  I’ll have to do more thinking before answering you more completely, but here are few first thoughts:

We have traditionally used the most elastic rope available for both VLL’s and for hanging the scaffolding.  I’ll continue to do the same for the scaffolding.  We put a tremendous pre-load onto the support lines.  That keeps the platforms almost rock solid stable.  Remember that these are nothing more than boards hung out in space, so you need to rig them for stability.  Using elastic rope makes this very easy.

But I’d like to change to a very stiff rope for VLL’s.  The reason is that I don’t want people hitting the bottom when the rope stretches.  If I’m working with 500 feet over my head, I could get as much as 50 feet of stretch before arresting the fall and bouncing back up again – bungee style.

I put mechanical energy absorbers into the system, so I don’t have to rely on rope stretch to minimize arresting loads like a rock climber does.  While the legal limit is 1800 pounds, I design nearly all my systems to half that.  A fall on most of our systems would keep the actual loads down to about 750 pounds.  People who have fallen on systems like this say it is about like sitting down in an easy chair.  We use energy absorbers in all our lanyards, and sometimes use a second one attached at the head of the lifeline.

I never liked the 180-degree loop in the bitter end of the standard bowline.  I was planning to test a tucked Dutch Bowline.  I’ll spend some time thinking about your other suggestions.

These will be slow pulls, done on a hydraulic test rig.  I’ve done dynamic testing, but not to failure.  The ANSI standard calls for a 220-pound steel test weight, dropped six feet on a very rigid support.  All the knots stand up very well to that.  But loads without an energy absorber exceed 1800 pounds.

Anyway, I want to collect all suggestions and think this thing through well before starting any testing.  I don’t like to waste time in the shop trying to figure out what to do next.  You have given me some good ideas that I had not considered, and a few others that I need to consider some more before incorporating them into the test plan.

Paul Kruse

PaulKruse

  • Full Member
  • ***
  • Posts: 60
Re: Knot Testing
« Reply #5 on: January 15, 2007, 11:10:06 PM »
Another thing to think about:  If you don't apply the load into a nylon rope very slowly, the actual failure will often be caused by excessive heat in the nylon, rather than by an overload.  That is why a wet rope is stronger, the opposite of what the rope manufacturers publish in their manuals.  Yes, dry rope is stronger than wet; but wet knots and spices are stronger than dry because the water makes the knots and splices more efficient at reasonable load application rates.  I will be pulling dry rope quickly enough to get some heating, but not quickly enough for the heat to cause significant reduction in strength of the nylon.  The broken rope will feel warm to the touch, about 120 degrees F.

Paul

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 2722
Re: Knot Testing
« Reply #6 on: January 16, 2007, 08:02:37 AM »
clove hitches backed up by a single or double half hitch are used extensively to secure a lifeline to a piece of steel structure, usually an I-beam.  When I first came to work here years ago, the knot worried me.  So I did some informal tests, and found that its strength was about the same as a standard bowline tied around the same beam.  Both break at about 75-80% of the rope’s published minimum breaking strength.  It seems to be a matter of personal preference to use a bowline or a clove hitch, but I will rarely see an ironworker use anything else for his upper connection.  Just the other day I saw an iron worker for another company use a figure 8 loop, but that is the first time in twenty years that I saw them use something different.

The second half hitch backing up the first half hitch does not seem to add any value to the knot.  Either jams into hatchet knots after being loaded, but that does not really matter.  The only way to load a VLL is to fall on it, and we would throw the rope out after any fall.
Wow, I'm surprised that a Clove h. around an I-beam would not take a hit on the sharp edge,
as I'd expect it to distort to deliver the load largely  on the SPart to the edge, the rest of the
structure--even the HH.s--having given ground in the push-comes-to-shove heavy loading!?
(Whereas were the end tied off into a Bwl the point would be to distributed load to the four
legs of the Clove.)  I'm also surprised that the HHs.--esp. the single--would jam.  Is the
Clove set tight-snug around the beam?  (I'd think that one wouldn't try this with a loopknot,
though, except by making a round turn of the beam--a sort of friction grip--, were there
some concern about the attachment shifting position.)

As for using the Fig.8, it has, like most loopknots, a significant disadvantage of requiring
one to pre-form a part of the knot before sizing the eye.
You might try a Bwl in which the ...  \||/  Ah, heck, if you get 85% out of you Bowlined rope,
there's not enough lost strength to bother worrying over! (-;

Quote
I’d like to change to a 5/8-inch static kernmantle, but that would be ...
Much stiffer, as you note, and thus a challenge to knot.  Also, maybe it won't be as
well received by the I-beam edges (!)--better test that.  Re knot strength, cf.
www.caves.org/section/vertical/nh/50/knotrope.html.
(nb:  the bar graphs mix up (swap) the date for the 12.5 & 10.5 ropes.)

Quote
I’m not sure what a Double Overhand Loop is.  Got a picture?  I might know it by another name.
A simple overhand knot might be a good one to try. 
I suspect it's one & the same:  "Overhand Loopknot" I'd say.  --'bout the most obvious/simple thing
to quickly tie, and much used in commercial marine/fishing (sometimes as a mid-line loopknot).
NB:  there are TWO ends to this (as for any other trace/"re-woven" knot--check to see if it
matters which takes the load.  In the one case, the loaded twin will bear into and so be cushioned
by its twin; in the other case, it will pull away from its twin into other parts of the knot.  YMMV ?!

Quote
it worked very well, but the standard bowline proved to be very inferior
It would be hard to keep the Bwl tied in such rope.  The EBDB I pictured above can work
in such stuff, where one wants to avoid 1dia. bends.

Quote
When talking about percentages, you need to be very careful.  I use the minimum published breaking strength
for the rope, which is much lower than its actual breaking strength.  That gives efficiency percentages that are artificially high.
Ah, hmmm, and so that "85%" above was maybe 65%?  WHY do this, if you know the real strength?
What I was going to suggest above but then cut it off with the remark about "85%" is to make
a Bwl where the end is brought around a 2nd time to form a 2nd collar--gives you two eyes
on the I-beam (so less abuse), and 4 vs. 2 dia. of rope through the central, nipping loop of the
SPart, which should increase strength (for what it's worth).

And retiring the life line after ANY fall?  --something climbers certainly don't do!
And if you're providing such soft impact forces, well, ... send the old rope up here ... !!

 ::)

PaulKruse

  • Full Member
  • ***
  • Posts: 60
Re: Knot Testing
« Reply #7 on: January 16, 2007, 05:22:56 PM »
Dan:

I don't think you understand how we combine the Clove Hitch with a Double Half Hitch.  Give me an email address and I'll send you a picture of you like.  I think I even have a picture with it tied around an I-beam with 5000 pounds suspended from it.  Very little rope damage and it held very well.  I don't know how to post a picture here, but you would be welcome to post it if you like.

I'll get to the rest of the topics you bring up later when I get more time.

Paul

Willeke

  • Sr. Member
  • *****
  • Posts: 429
  • knopen . ismijnhobby . nl
    • Willeke's knotted Ideas
Re: Knot Testing
« Reply #8 on: January 16, 2007, 06:10:16 PM »
Paul,
When you type your answer in the window just underneat it is Additional Options.

Click on that and you get some options and a bar in which you can enter your picture.
I have only used this way with pictures in the memory of my computer.
It has some limits:
Allowed file types: txt, doc, pdf, jpg, gif, mpg, png
Maximum attachment size allowed: 128 KB, per post: 4

If you have a picture on internet you can use the insert image (first button above the smileys) and add the url inbetween the [img][/img) tags.

I hope this helps.
I can not add much to this discussion, being a fancy tyer, but I am reading it and try to understand.

Willleke
"Never underestimate what a simple person can do with clever tools,
nor what a clever person can do with simple tools." - Ian Fieggen

Writer of A booklet on lanyards, available from IGKT supplies.

PaulKruse

  • Full Member
  • ***
  • Posts: 60
Re: Knot Testing
« Reply #9 on: January 16, 2007, 09:13:54 PM »
Willleke:

Thanks for the hint.  I'll give it a try.  This is a picture of how we would attache a 3/4 inch nylong rope to an I-beam using a Clove Hitch backed up by a double half hitch.  The tape on the bitter end is a seal so we know who tied the knot.  You can see that the 5000 pound test weight did damage the rope, but only a little.

Also, it matters nothing in this discussion that your are a fancy knot tyer.  Speak up and ask questions when you have them.

Paul Kruse

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 2722
Re: Knot Testing
« Reply #10 on: January 17, 2007, 07:15:26 AM »
Thanks, Paul.  Though it's pretty hard to discern the ins & outs of the
Clove hitch; it appears to be leaning towards, but not all the way
to, the far side of the beam (I think some part of the wrap can be
seen yet going maybe 3cm-2" farther away before bending around
the beam)--which implies that much of the load is put into the Clove's
cross part and by that to the near side.  Hard to see the Half-hitches.

As for the Bowline, that's one of those "wrong"/"Dutch" ones of lore
--end on the outside (one can discern the untensioned wiggle in it).
(I'm surprised there weren't Bowline picketers protesting ...   ;D  )

--dl*
====

PaulKruse

  • Full Member
  • ***
  • Posts: 60
Re: Knot Testing
« Reply #11 on: January 17, 2007, 01:08:38 PM »
Dan:
I think you understand it correctly.  If you look at a Clove Hitch, you see a point where the rope crosses over between the two half hitches that make up the Clove Hitch.  That point is burried under the Double Half Hitch that backs it up.

About half our riggers tie the left hand BL.  Since they test out at the same strength as a right hand BL, I don't make a big deal out of it.  In fact, you can make the Dutch BL more secure by tucking the bitter end under the collar, so that it comes ot parallel to the working end.  That is the way I tie it whenever it is going be supporting my weight and balance, like when I rappel.  Our ironworkers call the left hand or Dutch BL, the "Bastard Bowline."  They call my version the Tucked Bastard.  The TB can be dressed into all sorts of interesting knots, including a useful slip knot that can be locked up at will.  It is also one of those "flip of the wrist" knots that can be tied in about two seconds like the Fast BL or the Tug Boat BL.  That means it can also be tied quickly on a bight.  The standard BL does not tuck so well, and it can't do all these other things, either.

Also:  We have about 2400 people certified to work at unprotected heights, and about 30 certified to rig special fall protection for them.  In addition, we normally have maybe a thousand subcontractors working at heights.  In the last 20 years, I can remember only about a dozen times when someone fell on their fall protection PPE, and they were all subcontractors who provide their own rope and rigging.  Any fall will initiate a very detailed safety investigation, which will impound all equipment involved in the incident.  So given how rare a fall is, throwing the rope away is not a great expense.

Paul

Fairlead

  • Sr. Member
  • *****
  • Posts: 328
  • IGKT Member since 1984 - IGKT Librarian
Re: Knot Testing
« Reply #12 on: January 17, 2007, 01:12:25 PM »
You have both miss-read what I wrote - I referred to a Double Overhand NOOSE not a Loop.
The difference being quite important in that the Noose slides, the Loop does not.
It has also been called a Poacher's Noose.
Tied by turning the working end back alongside the standing part to form a bight (which becomes the noose) then tie a Double Overhand knot using the so called 'grapevine method' in the working end around the standing part.  The tighter you pull up the Double Overhand the more friction on the standing part.  Makes a handy clamp when gluing too!
A firm favourite in my knotting repertior.

Gordon

PaulKruse

  • Full Member
  • ***
  • Posts: 60
Re: Knot Testing
« Reply #13 on: January 17, 2007, 03:36:34 PM »
Gordon:

Thanks for the clarification.  Sounds like what I know as a Packaging Knot.  Used to be in the old Boy Scout manual when I was a kid, 40 years ago.  Great for binding packages.  Works great for a clamp, too.  Except that version used a Figure-8 rather than an Overhand knot to make the noose.  That difference makes it very easy to lock it down with a single half hitch over the bitter end after you pull it tight.

Where would you use this knot in the rigging of either suspended scaffolding or fall protection PPE?

Paul Kruse

Fairlead

  • Sr. Member
  • *****
  • Posts: 328
  • IGKT Member since 1984 - IGKT Librarian
Re: Knot Testing
« Reply #14 on: January 17, 2007, 04:14:22 PM »
lNot really Paul - Here is a scan from my book