Author Topic: Test Repetition/=Verification (flogging assertions!)  (Read 3981 times)

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 4072
Test Repetition/=Verification (flogging assertions!)
« on: January 30, 2018, 04:12:22 AM »
Roo wrote:
Quote
On the last page of your Bowline Analysis paper, you recommend four loops for life support.  So I picked the simplest one*, which is the Lee Zep Bowline.

I tied and firmly set the Lee Zep Bowline to one end of a carabiner, and a Gnat Hitch on the other end.   Both were tied in Bluewater II rope with a short 3" tail.  Keeping the carabiner in the middle, and my hands an equal distance from each knot, I flogged the rope against the ground until the Lee Zep Bowline came undone after 22 impacts.  The Gnat Hitch remained firmly tied.

I repeated the test with a traditional Figure 8 Loop against a Gnat Hitch.  The Figure 8 Loop came undone after 52 impacts.  The Gnat Hitch remained firmly tied.
Most of us probably read this and thought the Roo
was smoking the Bluewater II, and seeing gnats everywhere.

However, he doesn't assert mysterious, divining methods,
but using almost household materials for some of us.  BWII
is an old product, with many imitators.

Sooo, can we repeat the test and confirm his results?   ::)
(Differences in flogging are to be expected --some will
be kinder & gentler, others more alt.Right.)

 ;)

agent_smith

  • Sr. Member
  • *****
  • Posts: 1390
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #1 on: January 30, 2018, 06:45:12 AM »
This is just simply a case of roo misunderstanding the purpose / application of the Lee  zep Bowline within a dual clip-in system for indoor (ie artificial surfaces) climbing.
In reference to the attached photo - one can see that the Lee zep Bowline (LZB) is contained within a closed system...that is, the tail exits from the 'LZB' and then continues on to terminate as a noose to the second carabiner.

'Flogging' is not a valid test of the 'LZB' (in this context) in my view.

roo seems to be determined to bring more attention to the gnat hitch (noose) - but to what end I wonder?

The gnat hitch (noose) is unsuitable as a tie-in knot to a harness for rock climbing applications because it acts like a noose. Tie-in knots for rock climbing must be eye knots (which have a fixed eye). Rock climbing harnesses have no metal hard attachment points (no 'D' rings) - its all constructed from textile materials.

In relation to the Lee Zep Bowline, I recommend it as part of a dual clip-in system for indoor (artificial) climbing gyms - configured as per the attached photo. It is a replacement for the #1053 Butterfly eye knot (which has a tendency to jam after a month or so of hard use in a commercial climbing gym).

EDIT NOTE: Flogging a knot by striking it against the ground repeatedly to induce failure does not reflect the loading profile of the LZB in the configuration I recommended. However, 'flogging' a knot might provide some insight into how a knot behaves when in a collision with sudden deceleration and impact against a hard surface. This might be of interest say to assess the resilience of a hitch used to moor a dinghy or small boat to a wharf which is exposed to gusting wind and breaking waves. The oscillation of the wave energy and wind might create somewhat similar conditions to 'flogging'. So i am saying that 'flogging' a knot has a place and might reveal information. Perhaps a gnat (noose) might be a good choice to tether a dinghy under such conditions?
« Last Edit: February 05, 2018, 07:51:49 AM by agent_smith »

roo

  • Sr. Member
  • *****
  • Posts: 1874
    • The Notable Knot Index
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #2 on: January 30, 2018, 04:02:18 PM »
Here's a summary of the cursory tests from that thread:

Quote
I tied and firmly set the Lee Zep Bowline to one end of a carabiner, and a Gnat Hitch on the other end.   Both were tied in Bluewater II rope with a short 3" tail.  Keeping the carabiner in the middle, and my hands an equal distance from each knot, I flogged the rope against the ground until the Lee Zep Bowline came undone after 22 impacts.  The Gnat Hitch remained firmly tied.

I repeated the test with a traditional Figure 8 Loop against a Gnat Hitch.  The Figure 8 Loop came undone after 52 impacts.  The Gnat Hitch remained firmly tied.

[...]

The same test repeated with your EBSB Bowline against a Gnat Hitch (all conditions the same):  The EBSB came undone after 66 impacts/shakes.  The Gnat Hitch remained firmly tied.

[...]

Anyway, since you brought up the #409 noose, I again repeated the test with Bluewater II, and the firmly-set #409 (with a slightly longer tail than 3") came undone after 44 shakes/impacts.  The Gnat Hitch (with a 3" tail) on the other side of the carabiner remained firmly tied.

I'm all for producing more data points & numbers.  For anyone who is interested, I tried to keep the flogging amplitude to about 1 foot off the ground before each impact.  My hand grip was spaced slightly beyond shoulder width.

update:  Gnat Hitch versus Gnat Hitch: 112 impacts/shakes one came undone.
« Last Edit: February 04, 2018, 06:24:03 AM by roo »
If you wish to add a troll to your ignore list, click "Profile" then "Buddies/Ignore List".


Harold Kahl

  • Full Member
  • ***
  • Posts: 73
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #3 on: January 30, 2018, 04:44:34 PM »

The gnat hitch (noose) is unsuitable as a tie-in knot to a harness for climbing applications because it acts like a noose. Tie-in knots must be eye knots (which have a fixed eye).

According to this video, the tie-in knot to a carabiner is required to be a cinching anchor knot, and this is a requirement of ANSI Z133.1-2006, to prevent accidental opening or side loading of the carabiner.
https://www.youtube.com/watch?v=-fZH4DrnHnM

ETA: Never mind. I see you have explained this on the other thread.
« Last Edit: January 30, 2018, 04:49:05 PM by Harold Kahl »

roo

  • Sr. Member
  • *****
  • Posts: 1874
    • The Notable Knot Index
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #4 on: February 04, 2018, 06:19:43 AM »
Here's a summary of the cursory tests from that thread:

Quote
I tied and firmly set the Lee Zep Bowline to one end of a carabiner, and a Gnat Hitch on the other end.   Both were tied in Bluewater II rope with a short 3" tail.  Keeping the carabiner in the middle, and my hands an equal distance from each knot, I flogged the rope against the ground until the Lee Zep Bowline came undone after 22 impacts.  The Gnat Hitch remained firmly tied.

I repeated the test with a traditional Figure 8 Loop against a Gnat Hitch.  The Figure 8 Loop came undone after 52 impacts.  The Gnat Hitch remained firmly tied.

[...]

The same test repeated with your EBSB Bowline against a Gnat Hitch (all conditions the same):  The EBSB came undone after 66 impacts/shakes.  The Gnat Hitch remained firmly tied.

[...]

Anyway, since you brought up the #409 noose, I again repeated the test with Bluewater II, and the firmly-set #409 (with a slightly longer tail than 3") came undone after 44 shakes/impacts.  The Gnat Hitch (with a 3" tail) on the other side of the carabiner remained firmly tied.

I'm all for producing more data points & numbers.  For anyone who is interested, I tried to keep the flogging amplitude to about 1 foot off the ground before each impact.  My hand grip was spaced slightly beyond shoulder width.
It seemed like it would be fitting to see what would happen if this test pitted a Gnat Hitch against another Gnat Hitch.  The result:  one came undone after 112 impacts/shakes.  This little hitch sure punches above its weight, though I'm not exactly sure why.
« Last Edit: February 04, 2018, 06:21:25 AM by roo »
If you wish to add a troll to your ignore list, click "Profile" then "Buddies/Ignore List".


agent_smith

  • Sr. Member
  • *****
  • Posts: 1390
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #5 on: February 04, 2018, 07:22:29 AM »
Quote
According to this video, the tie-in knot to a carabiner is required to be a cinching anchor knot, and this is a requirement of ANSI Z133.1-2006, to prevent accidental opening or side loading of the carabiner.

+ your edit where you seem to acknowledge another explanatory post.

Hi Harold, I would like to know what is on your mind.
That is, when you decided to post a reply in this thread, what was the underlying purpose of your post?
EDIT NOTE: (all good - explanation received 06 Feb 2018 - thank you),

I looked at the video you posted - it has nothing to do with tying a knot into a rock climbers harness in preparation for lead climbing.
Your video is specifically related to tree climbing techniques used by 'arborists'.

Tying-in for lead climbing on rock surfaces and rope ascension systems for tree climbing are 2 completely different applications.

Now, you did have an end note which seems to retract/contradict your video post - but, if this is the case, it strikes me as odd that you didn't delete your post altogether...you have left it as is.
The question in my mind is 'Why?'

I looked at another of your posts where you persist with trying to claim that an experimental eye knot (derived from #1024) is insecure and even posted a video attempting to somehow challenge and contradict my initial comment that the eye knot appears to be secure (when properly dressed and cinched tight in human-rated climbing cordage). Alan Lee then posted a video to show how you might have inadvertently arrived at your false positive conclusions (please refer to Alan Lee's video post).

Again - I am left wondering what is the issue?

So Harold, in lead climbing applications, it is critically important that no connector (eg carabiner)  is used to create the rope attachment to the harness. If a lead climber did try to use a carabiner to create a rope attachment to his/her harness, the outcome could be fatal. If you refer to climbing harness manufacturers websites and instructions, you will find that all of them (without exception) only advise using a fixed eye knot (never a noose). The most widely recommended tie-in knot is #1047 F8 eye knot. Having said that, it is also acceptable to use one of the secure Bowlines (but NOT the common #1010 Bowline which is insecure).

I challenge you to find a rock climbing harness manufacturer that recommends a tie-in knot that acts like a noose. In the same way, I challenge you to find a manufacturer that recommend using a carabiner as a means to attach a dynamic climbing rope to a harness (in preparation for lead climbing). Now, if you fail to find such a manufacturer, could you please be courteous and post back in this thread with your findings. This would help close this matter and put it to rest. I repeat again that it must be a rock climbing harness manufacturer - and you would have to specifically look for 'tie-in knot' in preparation for lead climbing. Note that this is distinct from tree climbing applications.

NOTE: In some exceptional and rare circumstances, some highly experienced speed climbers (elite climbing athletes) who are trying to set new world record speed ascents of notable routes such as the 'nose of El Capitan') - have experimented with dual locking carabiners as a tie-in interface. The intent is to speed up the belay change-overs when one climber swaps leads with the other. This tactic can and does save time in setting new speed records but, it comes at a price. And that is, they increase the risk of catastrophic failure - which they hedge their bets on mitigating that risk by using dual connectors. Keep in mind that these guys are top-level athletes who know and understand the risks of what they are doing and are willing to face the consequences. These guys would never suggest to the world at large that using connectors to tie-in is safe for general use (because it isn't).

Here is a link to a climber taking a fall (just to illustrate what an extreme fall in climbing can be like - I've never personally had falls this big - biggest I have experienced is 20m):
Link: https://www.reddit.com/r/climbing/comments/6rjg5v/this_will_go_down_as_one_of_the_greatest_rock/
Note that a noose or a carabiner attachment would be unwise in falls of this magnitude.

Harold - do you understand what I have written?
« Last Edit: February 06, 2018, 04:47:38 AM by agent_smith »

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 4072
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #6 on: February 04, 2018, 10:52:20 PM »
Quote
According to this video, the tie-in knot to a carabiner is required to be a cinching anchor knot, and this is a requirement of ANSI Z133.1-2006, to prevent accidental opening or side loading of the carabiner.
... [gratuitous verbosity deleted]

Harold - do you understand what I have written?

I doubt many have read all of that.
The simple point to make/reiterate is that the cinched
requirement was ROPE-to-METAL,
and not rope-to-harness/fabric.
(How many words was that?)

 ???

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 4072
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #7 on: February 04, 2018, 10:55:03 PM »
'Flogging' a knot (or any rope structure) is not a valid test method in my view.
What IS, for slack-security, then?

Irrespective of merit, the *team-building* exercise
of several folks trying the testing --whatever its worth--
and comparing results should be a Good Thing.  And
several of us have ropes of the target sort, too (not
to exclude other cordage, though).

--> might need some more specific guidance from Roo
on his exact technique (not to preclude similar ones,
but to better understand, esp. if results differ).


--dl*
====

agent_smith

  • Sr. Member
  • *****
  • Posts: 1390
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #8 on: February 04, 2018, 11:02:17 PM »
Quote
The simple point to make/reiterate is that the cinched
requirement was ROPE-to-METAL,

Actually Dan, sorry - but no; that was not the original issue (starting to obfuscate things).
Interesting attempt to provide an escape path for those who suggest that a noose can be used on a metal carabiner (of course it can!). I had already pointed that out long ago.
Did you see my reference to IRATA code of practice - and the diagrams showing #409 and #1120? 'Cows tail' lanyards are typically terminated with a knot that acts like a noose (rope to carabiner contact). Refer to attached photo of termination knot (#409) which functions as a 'noose'. The termination knot exists at the end of the 'cows tail' lanyard (not at the harnesses attachment interface).

The assertion was that a tie-in knot that acts like a noose can be tied to a rock climbers harness - textile-to-textile.
And it is here that a the line is drawn in the sand. Rock climbing/ mountaineering harnesses dont have metal connective hard points. You can't go textile-to-textile with a noose structure to arrest free-falls.

Its okay - I understand why you want to try to provide an escape route for roo.
I can see that there might be some miss-understanding of the differences between industrial full-body harnesses and recreational climbing harnesses.
Most industrial full body harnesses have metal hard 'D ring' attachment points.
Recreational rock climbing harnesses dont have metal D ring hard attachment points (its all constructed from textile material to save weight).

Be that as it may, in a free-fall arrest situation, OHS / WHS legislation applies at a workplace...and you are normally required to use an energy absorbing lanyard (which uses 'tear-web' stitched material that absorbs energy and is destroyed in the process).

PS Some more words for you.
Flogging is not a test method that is exactly repeatable by others. How do you define the tempo and strike angle, velocity and impact force of the flogging so it can be reproduced? In all the published test data I have seen, flogging doesn't appear in the data sets. I wonder why?
« Last Edit: February 05, 2018, 04:14:45 AM by agent_smith »

agent_smith

  • Sr. Member
  • *****
  • Posts: 1390
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #9 on: February 05, 2018, 12:58:27 AM »
Quote
I flogged the rope against the ground until the Lee Zep Bowline came undone after 22 impacts.  The Gnat Hitch remained firmly tied.

In my mind, this 'test result' doesn't prove that the Lee Zep Bowline is somehow a lesser knot structure. If you are going to test something - you need to devise a test method that produces valid results against explicit performance criteria.

Okay - so we beat the crap out of a knot and carabiner by smashing it against the ground (dirt, concrete, grass??). Lets assume concrete.
We just keep smashing it until we induce failure.
roo claims that his gnat (noose) wins the contest.
Is this a real-world test of an eye knot used in a climbing harness?

Here is analogy - lets pull #1410 to its MBS yield point (ie break it). What does this prove?
Lets flog #1410 on a concrete surface until something fails. What does this prove?
The point here is that you have to set up valid test conditions in order to reliably collect data from which you can draw valid conclusions.

The gnat (noose) does have a place - and it would be great when used in a proper context.
That context is not rock climbing (eg tying a dynamic rope into a textile climbing harness).

The gnat noose is a remarkable structure - all kudos to roo for its discovery.
But it needs to be used in the proper context/application (which isn't as a tie-in knot to a harness for rock climbing).


agent_smith

  • Sr. Member
  • *****
  • Posts: 1390
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #10 on: February 05, 2018, 07:51:13 AM »
Quote
What IS, for slack-security, then?

Another way to ask the question is:  Is 'flogging' a valid method to investigate a particular knot's slack-security (or vulnerability to cyclic loading events)?

The answer could be both 'yes' and 'no'.
It depends on what you are trying to determine - that is, what are you trying to test?

For a certified test lab, I can foresee difficulty in designing and documenting a method of beating a knot specimen vigorously against a hard surface.
For a backyard test, beating a knot against a hard surface might provide some insight into how the knot behaves in a rapidly decelerating + whiplash reference frame.
So this might (for example) be a valid test for investigating a secure way to tether a boat/dinghy.
« Last Edit: February 05, 2018, 07:53:40 AM by agent_smith »

Dan_Lehman

  • Sr. Member
  • *****
  • Posts: 4072
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #11 on: February 06, 2018, 01:06:58 AM »
Quote
What IS, for slack-security, then?

Another way to ask the question is: ...

The question was asked point on.  If one is going to address
concerns about security --relative security of various structures--,
how so?

We've not heard details of what has been casually called "flogging",
but my current surmise is that it's a not-so-hard flipping about of
a knotted structure, to see what the hoped-to-be-fairly varied
jostling/pressing/shaking can induce.  Rockclimbing can foresee
(for some climbs, and not for others!) instances of "between-a-rock=
and-a-soft-place (= body) rubbing, with some risk there of snagging
on a nubbin/flake.  (DAV or UIAA Pitt Schubert (& Heinz) have that
case of the water-knot being pulled open upon a rock point impaling
the exterior strand to hold it while the knot position moves, pulling
out the tail (seems to need more movment than should come, IMO).)

Some movement/shaking might get a bit of *push* if the eye's
legs don't quickly equalize (i.e., rather than rotation, one leg
takes minor compressive/pushing force).

Recall how surprised I was to --by chance (<-why was it not
foreseen & deliberate?!)-- weakly cyclically stress an OWK
and see bit-by-bit slippage-out of the exterior/choking strand !!
(something your favored added tuck precludes)

Quote
Flogging is not a test method that is exactly repeatable by others.
How do you define the tempo and strike angle, velocity and impact force
of the flogging so it can be reproduced? In all the published test data
I have seen, flogging doesn't appear in the data sets. I wonder why?
And what does?  (nothing)
It's not so formal a thing to expect some close match of results;
and one might benefit from the very lack of exactly repeatable
movements --i.e., get a broader coverage of the possibilities.
(I read e.g. the not in the Dave Richards testing of 3 kernmantle
ropes that he made a point of being the sole knot tyer, for some
presumed like tying
   (note that this actually presumes that his tying is all so exact,
     itself --he might not know how he is differing
        (just as Evans Starzinger I think had no idea that
         in his testing of fig.8 eyeknots he'd loaded variously
         both tails (weak & strong form, as I call 'em) !)
    ).
TO WHICH I thought "so, your results only apply to YOU,
and not the rest of us, then?!
)

Rather, I think that a batch of folks trying to "shake out"
security aspects, and having variance in methods, could
do well.  Some structures might survive and others fail
and then maybe, yes, some mixed results.  We expect
to have not quite exact materials, after all.  And we want
broader than just-THIS-rope-&-movement indications.
(And I expect to regard results, even if nicely tabulated
*per flog*, in more of a good / okay / poor sort of rating.)

The bowline has long pleased marine folks (though
now some books hint --well, heck, even Ashley carried
such a hint and offered that tucking...-- of a need for
additional measures.  But were the sailor clan to deal
with climbing rope (& flogging), they'd see a problem
new to them, but right before their eyes.


--dl*
====

agent_smith

  • Sr. Member
  • *****
  • Posts: 1390
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #12 on: February 06, 2018, 01:32:24 AM »
To address your original post:

Quote
Most of us probably read this and thought the Roo
was smoking the Bluewater II, and seeing gnats everywhere.
I'm surprised roo didn't report you to the moderators for this comment.
He apparently reported me for expressing a concern that he might remove himself from Earth's gene pool and be listed on 'Darwin Awards' if he uses a gnat (noose) as a tie-in to a rock climbing harness.
I genuinely do not wish to see any harm come to roo (or anyone).

Quote
However, he doesn't assert mysterious, divining methods,
but using almost household materials for some of us.  BWII
is an old product, with many imitators.

Thats right - its a backyard test.
There is nothing intrinsically wrong with 'backyard' testing.
But, a certified test lab would have difficulty devising a methodology using 'flogging' that is repeatable in other labs.

Quote
Sooo, can we repeat the test and confirm his results?   ::)
(Differences in flogging are to be expected --some will
be kinder & gentler, others more alt.Right.)
The answer to this is complex.
In a 'backyard' test environment, results will likely vary.
In a certified test lab - I'm not sure how 'flogging' would be defined and then measured (so it can be repeated in other test labs).
You would have to specify:
[ ] frequency
[ ] impact force generated on impacting surface (if the knot specimen is to make contact with a 'surface') - link: http://www.amti.biz/fps-guide.aspx (maybe flog the knot against a 'force plate')
[ ] speed imparted to knot specimen (maybe in the form of velocity)
[ ] strike angle (eg glancing blow, perpendicular blow to surface)
...list is not exhaustive.
When publishing papers that report test results - it has to employ the 'scientific method'. In which case you are open to criticism from the scientific community and your paper will be subject to peer review.
If you're just reporting backyard tests, you need to be clear about that - and particularly that your 'results' may not have been peer reviewed. However, backyard testers are not held to the same burden of scientific rigor.

Quote
The question was asked point on.  If one is going to address
concerns about security --relative security of various structures--,
how so?

Its known as 'reversing the onus of proof'.
Legal practitioners do this all the time.
One party or the other has the 'burden' to provide the evidence to prove or disprove a claim.
Thats the analogy I was giving you.
Like saying; "Prove that 2+2=4"
And then reverse the onus of proof by saying; "Prove that 2+2 isn't equal to 4"

So to extend this analogy:
Prove that direct nylon-to-nylon contact in a high energy free-fall will cause damage to the textile material.
Or; Prove that direct nylon-to-nylon contact will not cause damage to textile material in a free fall event.

There is sufficient evidence about the damaging effects of direct nylon-to-nylon contact during a free-fall event - where one material is 'sawing' across the other.
Tie-in knots for rock climbing harnesses are fixed eye knots (not nooses). We do not see widespread damage of climbing harnesses and ropes when eye knots are used as a tie-in.
Is this sufficient proof that a fixed 'eye knot' does not cause significant damage during a free-fall event?
Empirically, we can answer in the affirmative.


« Last Edit: February 06, 2018, 02:33:29 AM by agent_smith »

Harold Kahl

  • Full Member
  • ***
  • Posts: 73
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #13 on: February 06, 2018, 02:40:15 AM »

Hi Harold, I would like to know what is on your mind.
That is, when you decided to post a reply in this thread, what was the underlying purpose of your post?
I posit that you were attempting to contradict me in some way - to possibly gain satisfaction?

Jeez, don't be so paranoid. I posted it because I was looking for clarification. Having seen the video, I was confused, because it seemed to be a contradiction. I didn't go back and delete the post, because the subject of tying to a carabiner had not been discussed in this thread, I don't think. So I thought it might be useful for another novice like me who is not intimately familiar with tying to a harness.

agent_smith

  • Sr. Member
  • *****
  • Posts: 1390
Re: Test Repetition/=Verification (flogging assertions!)
« Reply #14 on: February 06, 2018, 02:50:18 AM »
All good Harold (not 'paranoid' at all)!

I was just curious why you left your videos posted and didn't delete them or modify them in some way to reflect the proper context of discussion.
Both seem misplaced - and sorry to say...they do appear to be trying to making a counter-point.

But, the IGKT forum is for making points though... :)

It still seems just a little bit odd that you haven't commented on Alan Lee's video (different thread) where he shows how you may have stumbled on the alleged failure mode of the 'unknown eye knot'.
With respect to the tree climbing video - the application is completely different - its not a rock climbers harness and it is not in the context of lead climbing.

Anyhow, all good and I look forward to more posts from you!