Grant Prattley test of 'anti-Bowline' (aka Eskimo Bowline)

[agent_smith]

KNOT TEST REPORT: (Summary from a blog on his website)

Tester details: Grant Prattley
Title of report: Sheet bend Vs Bowline
Link: http://overtheedgerescue.com/technical-rope/sheet-bend-vs-bowline-knot/
Company: Over the edge rescue
Application context: Recreational climbing/abseiling/caving/canyoning
Publication date: 05 October 2017
Testing lab type: Presumed 'Pseudo lab' (not a certified lab and not accredited by a third party agency)

The author has made critical errors in describing a 'sheet bend'. The author attempts to explain and differentiate between the 3 sub-categories of knots:
[ ] Knot
[ ] Bend
[ ] Hitch
His headlining remark is; "The way to tell at a glance if you have a sheet bend is the bight is on the inside of the loop formed compared with the outside for a bowline".

The author fails to recognize that his alleged 'sheet bend' isn't a 'bend' (ie an end-to-end joining knot) and is in fact an anti-Bowline (an eye knot; aka 'loop knot'). The anti-Bowline is also known as an 'Eskimo Bowline'.
Loading profile is key - and load is directed via the SPart (standing part) and not as a round sling with an end-to-end joining knot creating the link to form the sling.
It is possible to surmise that Grant Prattley noticed that the core (e nub) of the 'anti-Bowline' appeared identical to #1431 Sheet bend - and tried to build his contention in this way. But, in the photos he tendered on his website, loading was via the SPart as you would expect in an abseiling setup.

It is difficult to ascertain precise details about the test and what actually was tested. It is unclear.

Furthermore, this is another example of the default mindset of 'pull-it-to-failure' (ie probe the MBS yield point of the knots). This pull-to-failure mindset permeates virtually the entire rope rescue community. The author (Grant Prattley) doesn't appear to have any other way of conceptualizing knot performance except via pull-it-till-it-breaks; and the knot that wins the contest is by default, declared to be 'superior'.
Given the purported application of the knot was abseiling (canyoning), the nominal load would be the weight of 1 person. How much load is the knot required to resist? Is Grant applying a SF (safety factor) to the knot, which therefore raises the expectation of resistance to load induced failure?

Grant Prattley fails to recognize the concept of 'ring-loading' - and that in fact, an 'anti-Bowline' is resistant to this type of loading profile. Furthermore, he doesn't understand basic knot anatomy such as; eye leg, eye, collar, SPart, etc. For example, in a common #1010 Bowline, the collar forms around the SPart. In contrast, the collar forms around the ongoing eye leg on an anti-Bowline. He also appears to be confused between an end-to-end joining knot (ie bend) and an eye knot (aka 'loop knot').

It appears that the testing was carried out 'off-premises' - possibly at a 'pseudo lab'? If so, equipment and testing methodology would not be required to meet strict expectations of scientific rigor.

It is difficult to draw any conclusions from Grant Prattley's arguments and test setup. Nothing is clear and the entire premise underpinning his arguments appear to be flawed.

His testing reads as follows:

Testing

We completed two tests (testing Prattley and Mandeno, July 2005) of the single sheet bend pulled as a knot,

What does this mean exactly?
There is no diagram or schematic to clarify the test method and configuration.

His conclusion reads as follows:

Conclusions and thoughts

    Pulled as a knot the sheet bend performs poorly as it inverts and slips.

Again, it is difficult to extract a precise understanding of what he means by this statement.
In my view, it would be difficult for another tester to try to repeat Grant Prattley's test to either confirm or refute his results.

[Harold Kahl]

That is horrifying, especially for somebody who uses knots for climbing, and is an instructor to boot.

[agent_smith]

I don't want to appear to be too harsh in my criticism of Grant Prattley's test regime.
However, he is publishing to the world - and holding himself out as being an 'expert' roping technician.

Because of that, he should be held to higher standard than a hobbyist/enthusiast tester.

In fairness to Grant, it would seem that he noticed that the core of the 'anti-Bowline' (aka 'Eskimo Bowline') appeared identical to the core of #1431 Sheet bend.
However, the 'anti-Bowline' is an eye knot (aka 'loop' knot).
And the Sheet bend is an end-to-end joining knot.
The loading profile is different in each case.

When loaded as an eye knot, all tension force is injected via the SPart.
Newton tells us that both eye legs (as a pair) will provide a counter-balancing force that is equal and opposite to the force at the SPart.
My point is that when the anti-Bowline is loaded as an eye knot, force enters the knot core from 3 directions (via 3 rope parts).
In #1431 Sheet bend, force enters the knot core via 2 directions (via 2 rope parts).

Also, it is implied that the context is canyoning (ie abseiling) - and so the nominal load on the knot would be the body weight of just one person. And so, testing the knot to its MBS yield point doesn't prove much. Of greater importance is knot stability and security.

The loading profile of the knots in Grant's test is unclear - he did not supply clear and unambiguous photos/diagrams showing how the knots were loaded. It would be difficult for another tester to try to repeat Grant's test to either confirm or refute his results.

[Harold Kahl]

One thing that bothered me about that article is that it seemed to imply that an unsecured 1010 bowline was okay for life support. Even I know that's a bad idea, and I've never used a rope for climbing. Maybe that's not what he was saying, but somebody could certainly have gotten that idea.

[agent_smith]

per Harold;

One thing that bothered me about that article is that it seemed to imply that an unsecured 1010 bowline was okay for life support.

In fairness to Grant Prattley, I would hesitate to construct this intention or meaning to his test report.
In my view, the 2 photos at the top of the web page were given simply to allow a side-by-side comparison of the two structures.

It would be a stretch to assume that he is suggesting that #1010 Bowline is suitable for human fall-arrest (ie life support). But, I think what you are suggesting is that the mere presence of the #1010 photo might mislead some novices into thinking that this structure is better than the companion image to the right?

[Dan_Lehman]

I don't want to appear to be too harsh in my criticism of Grant Prattley's test regime.
...
 It would be difficult for another tester to try to repeat Grant's test to either confirm or refute his results.

Rather, I think that you can :: that he loads each knot
qua eye knot (as pictured, around a rock) --and merely
mistakes what the Eskimo bowline should be called.

I'm surprised that that knot capsized!?  --at 3kn & 4kn,
respectively of his aged 8mm & 10mm low-elongation ropes.
You've presented images of Bluewater rope strained through
four loads to 10kn, to show its deformation thus, so you have
crossed the capsizing threshold he had, by double.  (We can
wonder how well he set the knot, if you get results of stability
well in excess of 4kn.)

Separately, if capsizing is to be expected --at significant forces--,
we might wonder if there was even a *benefit* to it, for use
catching a dynamic load --that ol' "tarbuck hitch" alleged magic,
or might there be frictional damage!?


--dl*
====