I remember reading the results of some EDK roll tests that were posted on this site....probably within the last year. This was not long after the EDK had received much press in at least 2, if not more, accidents. In one accident (perhaps the Zion event that folks are referring too), the joined ropes had the knot come apart. Preliminary reports were that the knot was EDK, but perhaps not. Anyhow.

The upshot of the tests (multiple trials) were that the EDK was totally unpredictable in the amount of rolling that occurred. In one trial, it rolled twice before it seized hard. In another trial it rolled like 5 times, coming very close to the end of their tails (which were reasonably long, but probably 8" or less).

The conclusion of the testers upon completion, was that the EDK was unpredictable enough to be discarded as a regularly used knot.

I'll look around for that post where the test results were displayed. In any case, my memory serves accurately (for my person), so I don't need to re-read it to for myself. If anyone else has it, you might post it here for others to review.

Magazines are supposed to be authorities and definers

Hippie,

Perhaps now would be a good time to gather some mushrooms and hang out by the river with your family.

You can report back once you have something useful to say.

I did. You should experiment with being useful. I think it might be a much needed life-change for you. Better than the human door-stop that you are.

...EDK w/ wet ropes....knot rolls indefinitely and comes undone...

Stick to the facts about the topic at hand. That is, if you can provide any.

...are you just here to irritate me?

I'll try to fix the link?

I'm not exactly disagreeing with you.....but I've yet to see any review of the knot declare it to be failproof.

That is the very report that convinced me to switch to the EDK (overhand) for my default rappel knot. When the knot is tied properly it doesn't even roll under loads that it would be subjected to on a rappel. If it is tied badly it rolls, but tightens as it rolls, so that with standard 18-inch tails it has the same strength as when the knot is tied well. The only test in the report where the knot performed questionably is when it was tied in wet ropes of significantly different diameter, but even then it failed at > 1400 lb, a greater load then it would be subjected to on a rappel.

The one failure of the knot in the field noted in the report was in a knot that Gumby #1 untied and retied for no apparent reason, leaving only 6- to 8-in tails, and which looked "loose" to Gumby #2. In spite of the knot failing visual inspection Gumby #2 rapped on it anyway, and it failed. Hell, even the rangers who were called in used EDKs on the rescue.

Jay

The upshot of the tests (multiple trials) were that the EDK was totally unpredictable in the amount of rolling that occurred. In one trial, it rolled twice before it seized hard. In another trial it rolled like 5 times, coming very close to the end of their tails (which were reasonably long, but probably 8" or less).

The conclusion of the testers upon completion, was that the EDK was unpredictable enough to be discarded as a regularly used knot.

There's an enormous variability in the force it takes to start flipping the knot. I have no confidence that I've fully explored the parameters that affect the flipping, or that one test for each set of parameters creates much statistical confidence in the answer. As someone said, "it's not the average that's the concern, it's the standard deviation." What I do know is that I didn't find any combinations of factors for a well-dressed, well-preloaded, flat-overhand or flat-figure-eight knot with sufficient tails that would cause failure at less than body weight. I also know that millions of rappels have taken place on these knots without failures. I suspect that in each of the four accidents I mentioned above, the knots were too loose and the tails were too short, but it's also possible that they simply found a combination of rope sizes and conditions that doesn't hold as well as the ropes that I tested.

I think to use them for a rescue lowering or belay is to invite an accident. There are a few rescue teams who use them for lowering, and they have had no accidents yet that I know of. One team that I spoke to about my testing has now changed their protocols and discontinued their use of the knot.

I have personally used the flat-overhand on some rappels where I thought pulling the ropes across an edge might cause problems. Otherwise, I use a figure-eight follow-through knot with grapevine safeties. Most of the people I know use the flat-overhand, including Chris Harmston, who co-wrote the high-strength cord paper with me. I don't believe the flat-overhand will ever fail under body weight if it is tied well.

Sounds fine.....I'm not for or against the knot (disclaimer)

Here's the test I remembered.

http://www.xmission.com/...yer/testing/EDK.html

Quick summary (I didn't carefully pore through it)

EDK w/ dry ropes....knot held, ropes broke at a wide spread of pulls, during multiple tests. Anywhere from 1400-2100 lbs.

EDK w/ wet ropes....knot rolls indefinitely and comes undone.

(just for perspective)

Double Fishermans....ropes break at over 2500 lbs, even when tied improperly or ropes are wet.

I'm sure there is more out there. I'll let y'all decide for yourselves, but one thing is certain: The results are far from conclusive, and the EDK does not appear to be totally failproof. What that means in the real world is hard to say.

In my opinion, this constitutes a misreading of Moyers tests. I do hope that anyone who wants to argue with this perspective will read the comments below carefully before responding. We get into these pickles because people don't take the time required to understand complex issues.

Note that in order to achieve any of the knot loads reported by Moyers, the corresponding load to the rappel rope must be double Moyers knot load, and more or less constantly double Moyers knot load in order to correspond to the slow-pull testing method.

... involves a properly tied overhand EDK ...

... tests of the EDK for two ropes of unequal diameter, specifically an 11mm dynamic rope and an 8mm static rope. One test was with dry ropes and one with wet ropes. As KnudeNoggin has explained several times now in different threads, there are two ways to tie the overhand EDK in different diameter ropes, one of which is likely to be less resistant to rolling than the same-diameter knot. Unfortunately, we don't know which configuration Moyer[] tested, but his dry rope result was rolling at 1230 lbf, again a level that cannot be achieved by rappelling, with the 8mm line breaking at 1770 lbf. His wet rope result was 950 lbf, which is extremely unlikely, if not unachievable by rappelling.

From this I would say that the very worst of Moyer[']s tests, ...

No tests with so few trials can be considered to confirm the safety of a knot. As far as I know, such tests have never been performed on any knot.

The conclusion of the testers upon completion, was that the EDK was unpredictable enough to be discarded as a regularly used knot.

What I do know is that I didn't find any combinations of factors for a well-dressed, well-preloaded, [offset]-overhand or [offset]-figure-eight knot with sufficient tails that would cause failure at less than body weight. I also know that millions of rappels have taken place on these knots without failures. I suspect that in each of the four accidents I mentioned above, the knots were too loose and the tails were too short, but it's also possible that they simply found a combination of rope sizes and conditions that doesn't hold as well as the ropes that I tested.

In the rescue situation, two strands are not bearing the load and there can be two, perhaps three people plus a stretcher hanging on the rope, which will probably be a static rope and so capable of producing much higher forces if slack accumulates and there is a drop.

Double Fishermans....ropes break at over 2500 lbs, even when tied improperly or ropes are wet

In my opinion, this constitutes a misreading of Moyers tests. I do hope that anyone who wants to argue with this perspective will read the comments below carefully before responding. We get into these pickles because people don't take the time required to understand complex issues.

Moyers did 17 slow-pull tests on two versions of the EDK, the figure-8 version and the overhand version. The figure-8 version proved unreliable, leaving 6 tests on the EDK. Of these 6 tests, 4 were well-dressed and pretensioned, and 2 were on ropes of the same diameter. Of the two on ropes of the same diameter, only one was on dynamic ropes. So all the so-called variability and uncertainty are artifacts of the broad scope of the testing rather than any feature of the knot---as it is typically used---that could be detected by these tests.

Note that in order to achieve any of the knot loads reported by Moyers, the corresponding load to the rappel rope must be double Moyers knot load, and more or less constantly double Moyers knot load in order to correspond to the slow-pull testing method.

Only one of Moyers 17 test involves a properly tied overhand EDK on dynamic ropes of the same diameter. In that test the knot rolled at 1400 pounds, a level that is unachievable under any rappelling circumstances, and the rope broke at 2070 lbf.

In the second of the two tests of a properly tied EDK in two ropes of the same diameter, two static ropes were used and the EDK did not roll and broke at 2170 lbf.

Moyers also performed two tests of the EDK for two ropes of unequal diameter, specifically an 11mm dynamic rope and an 8mm static rope. One test was with dry ropes and one with wet ropes. As KnudeNoggin has explained several times now in different threads, there are two ways to tie the overhand EDK in different diameter ropes, one of which is likely to be less resistant to rolling than the same-diameter knot. Unfortunately, we don't know which configuration Moyers tested, but his dry rope result was rolling at 1230 lbf, again a level that cannot be achieved by rappelling, with the 8mm line breaking at 1770 lbf. His wet rope result was 950 lbf, which is extremely unlikely, if not unachievable by rappelling.

For those who are skeptical about this last claim, let me offer a quantified argument. Remember that the load to the rappel ropes has to be double the knot load, so the rappeller would have to produce 1900 lbf in order to load the knot with 950 lbf. Since the peak tension in a fall arrested by double ropes is about 40% higher than the tension that would occur if a single rope arrested the same fall, we are speaking here of a fall that would result in a peak load of 1360 lbf or about 6 kN if arrested by a single rope.

Let's say that our rope has a UIAA rating of 10 kN (which is higher than most ropes, but allows for the stiffening that is part of the aging process). Then fall factor required to load the knot with 950 lbf is about 0.6. To fall on rappel, the rappeller has to somehow accumulate slack and then drop. If the rappeller has rappelled a distance D feet from the anchor, stops on a ledge, and somehow contrives to pull up and through the rappel device S feet of slack, then a fall factor of 0.6 corresponds to the slack equation S=1.5D; you have to pull up and through the rappel device 50% more slack than your distance from the anchor in order to provide an instantaneous peak load to the anchor of 950 lbf, a load which caused the knot to roll in unequal diameter wet ropes under slow-pull conditions.

From this I would say that the very worst of Moyers tests, unequal diameter wet ropes joined by an overhand EDK that might possibly be tied the wrong way, although properly dressed and tensioned, produced rolling that cannot be achieved under any normal rappelling situations.

A qualification is in order: you could manage this by being 1 foot from the anchor and falling with 6 inches of slack. We don't know whether an instantaneous peak load will have the same effect on rolling as a more-or-less constantly applied slow-pull load; I would think not, but if so than here is a reason to have sufficiently long tails. It took 1300 lbf to roll the knot again, and this cannot be achieved by a rappeller.

No tests with so few trials can be considered to confirm the safety of a knot. As far as I know, such tests have never been performed on any knot. But we can say that there is nothing in Moyers tests to indicate any safety concerns about the a properly dressed and tensioned EDK for ordinary rappelling, especially in two ropes of the same diameter but even in the case of wet unequal diameter ropes.


I think Tree is remembering---well, misremembering---the parts that help to confirm the conclusion he has already reached. The part of Moyers report that I can find is his reponse to the claim that the data shows the EDK is safe:


Note that three of the four accidents involved the definitely discredited figure-8 version, and the one EDK accident (the only one that I know of that has ever been reported) involved some strange circumstances in which the knot was undone and retied for some reason just before it failed. Moreover, it is important to note that Moyers' tests do not contain any evidence of high variability in rolling loads for the properly dressed and tensioned overhand EDK. The variability he describes may or may not be an artifact of the range of testing situations he employed, but there aren't enough trials on the cases relevant to proper practice to tell anything about variability there. In other words, variability may be a problem, as it always is with so few trials, but one cannot take the variation in the extremely different testing circumstances as evidence for unseen variability in the relevant tests.

Trying to find something in Moyers that confirms Tree's recollection, we also have:

In the rescue situation, two strands are not bearing the load and there can be two, perhaps three people plus a stretcher hanging on the rope, which will probably be a static rope and so capable of producing much higher forces if slack accumulates and there is a drop. I can't see that this situation is one that bears on climbing practice. If we adopted the rescue perspective, we'd probably have to conclude that none of our belay devices are safe, for example.

Others have tested the EDK. No one that I have heard of has found any failure issues of concern for rappelling as it is done by climbing parties, as long as the knot is properly dressed and tensioned.

The one lesson we've learned from Moyers tests is that the rolling threshold for the EDK is far more sensitive to poor tying than the breaking threshold of the double fisherman's knot. But as long as climbers are aware of that sensitivity and are diligent about tying, I can't see any reason to view the EDK as unsafe, but I'll let Moyers have the last word here:

Whoah!

Rgold, that's a lot of words.

Let me simplify your post for you:

"Tree_hippie, you're a moron."

There you go. And you're welcome.

Looks like you've totally missed the reason that rgold's posts are among the few worth reading on rc.com. Anyone can call someone a moron (especially under the anonymity of the internet); a well-reasoned argument that doesn't resort to name-calling is refreshing and rare. I sort of doubt that rgold (or anyone else) wants you to "simplify" his posts.

JL

Thanks again for clarifying.

Moron.