Loading Notifications...

/ SKILLS: Abseil Knots Explained

This topic has been archived, and won't accept reply postings.
UKC Articles 03 Oct 2016
Double Overhand knot, 3 kbFollowing recent online discussion stemming from a Rock & Ice article about appropriate abseil knots and some confusion over certain types of knot and their nicknames, we asked Jack Geldard to clarify the key points...

Read more
In reply to UKC Articles:

Very useful and clear information to a commonly asked question!
ericinbristol 03 Oct 2016
In reply to UKC Articles:

Best article on this I have every seen. Well done Jack.
Alex McCalman 03 Oct 2016
In reply to UKC Articles:

What's the consensus on using an alpine butterfly bend for abseiling? I've used it once or twice in fairly controlled situations, and it seems to be a perfect knot for it, but I've never met anyone else that's used it.
jsmcfarland 03 Oct 2016
In reply to UKC Articles:

here's your consensus: if it ain't broke, don't fix it. You should see the MP forums, every 3rd post is somebody asking if x,y,z knot is ok for abseilling, or 'inventing' knots for people to test............!
SenzuBean 04 Oct 2016
In reply to Alex McCalman:

> What's the consensus on using an alpine butterfly bend for abseiling? I've used it once or twice in fairly controlled situations, and it seems to be a perfect knot for it, but I've never met anyone else that's used it.

It doesn't have a 'flat' profile - two of the turns stick out. Whether that means they will catch onto things - who knows. It's also somewhat harder to tie than an overhand - which might make it less ideal for scary retreats. Might be good for icy ropes though
GrahamD 04 Oct 2016
In reply to SenzuBean:

I can tie an overhand in the dark wearing mittens. There aren't many other knots I could say that about !
Chris Craggs 04 Oct 2016
In reply to UKC Articles:
Being picky the 'tails 'on your double fisherman's are a bit OTT,


Chris



Post edited at 12:57
5
David Coley 04 Oct 2016
In reply to Alex McCalman:

> What's the consensus on using an alpine butterfly bend for abseiling? I've used it once or twice in fairly controlled situations, and it seems to be a perfect knot for it, but I've never met anyone else that's used it.

Well,

notableknotindex.webs.com/butt...

says this:

"Both the Butterfly Bend and the jam-prone Ashley Bend* can be mistakenly tied to form something that looks like the intended knot, but has poor security properties and can roll apart under certain loads in certain conditions. What makes this more dangerous is that the negative properties of this evil impostor are not usually apparent when setting the bend.

This evil or malignant imposter bend is discussed by Ashley in his Book of Knots as entry #1409 and as its more stable shifted form under entry #1408. The only difference between the two is that the more stable form has its free ends rotated or twisted relative to one another in the opposite direction from the insecure version.

Although the danger of tying this imposter is greater in the attempted Ashley Bend, one can imagine the "b" shape rotated counterclockwise, revealing that this same path of error for the Butterfly Bend is at least possible. The general knot form of the half hitch loop is another possible evil impostor for the Butterfly Bend and the Butterfly Loop, especially with some alternative tying methods for the loop.

If you feel that you may make such a mistake for these bends, the Zeppelin Bend is an excellent alternative, which has no serious impostor risk when tied with the "b" and "q" method, and may be immune to evil impostors regardless of method. The issue does not seem to affect the two common means of tying the Butterfly Loop"
David Coley 04 Oct 2016
In reply to jsmcfarland:

> here's your consensus: if it ain't broke, don't fix it. You should see the MP forums, every 3rd post is somebody asking if x,y,z knot is ok for abseilling, or 'inventing' knots for people to test............!

But that would mean The Flat Fish Sandwich would never get used
Jim 1003 04 Oct 2016
In reply to GrahamD:
> I can tie an overhand in the dark wearing mittens. There aren't many other knots I could say that about !

It doesn't really matter if you can do that or not, what is important is if it is going to come apart or KNOT
Post edited at 22:29
climber_Ken 05 Oct 2016
In reply to UKC Articles:

This good aadvice about abseil knots has been on the Needle Sports website for years. See: http://www.needlesports.com/content/abseil-knots.aspx
knudeNoggin 05 Oct 2016
In reply to climber_Ken:
What would be helpful --to reinforce "this good advice"-- would be tests showing the EDK-backed EDK (offset water knot...) tied in a few sloppy cases, and with specific flaws of (1) crossed strands, (2) some space between main and backing knot, (3) thick-thin rope misorientation, along with wet/frozen. I THINK that the knot will suffer these imperfections adequately, and that's a huge credit.

As for supposed flaw of the (venerable) offset water knot "Even when tied 100% correctly it is not as strong as some other knots.", that should be said of the recommended one just as well, but is in fact entirely irrelevant (and not a con : the point is to abseil, not to win a break-test contest). Why do people keep bringing such irrelevance up?

"double overhand" btw is a name for the component knot structure that the grapevine (dbl.fish.) uses, which is a strangle knot when tied around something (to bind --such as loose ends of a rope); it shouldn't be overloaded with yet another denotation here. (Similarly, "overhand knot" has older and fundamental senses best not muddied. The point in these abseil knots is that they are offset from the axis of tension, and enable the knotted rope to slide over irregular surfaces.)

Note that the article's presented offset fig.8 knot is not well set : for best results in this case, the tails should be pulled hard (to close the *choke* around the loaded ends), AND THEN the loaded ends should be pulled TOGETHER while holding the knot body to resist, so that the nipped tails will take a sharp bend. This will give a fighting chance to resist being pried apart. As it's set in the article's image, initial loading of the main lines will pull the knot open as they only then pull into a bend the two tails.

*kN*
DubyaJamesDubya 07 Oct 2016
In reply to UKC Articles:

I would use the double OH knot.
But I'm surprised the reef knot/double fisherman's combo isn't mentioned above as it is preferable to a straight double fisherman's.
1
LeeWood 08 Oct 2016
In reply to UKC Articles:

but the french are *still¨* advocating fig 8 for rappel

http://www.ffme.fr/fiches-ffme-techniques/page/confection-du-noeud-en-huit-jonction-de-corde.html

David Coley 08 Oct 2016
In reply to LeeWood:

> but the french are *still¨* advocating fig 8 for rappel


Having known someone who most probably died from a fig8 that rolled, this worries me massively. Especially as the tails are shown so short (although it does say 20-30cm, not everyone reads the text on around images).
r0x0r.wolfo 08 Oct 2016
In reply to David Coley:

To be fair 20cm for a figure 8 is not a lot either. A little more than half a ruler, a figure of 8 could roll off that tail fairly easily.
rgold 09 Oct 2016
In reply to David Coley:

Yeah and there's more from France; they clearly have not got the memo. For example,http://www.montagne-secu.com/rappel/ ; every shot shows a flat figure-eight, and http://cordevasion.com/noeud-de-jonction-descente-en-rappel/ , http://blogcrpa.fr/?tag=confection-du-noeud-en-huit-jonction-de-corde (scroll down).

As for using the EDK with different rope diameters, the DAV doesn't seem to think its a problem: see http://www.mountainproject.com/images/11/82/108401182_medium_e47dde.jpg . (Repost from MP)

An MP poster who goes by 20 kN tested the version of the EDK I use with an 11mm static rope and a 6mm cordelette. The cordelette broke at the clove hitch attaching it to the hydraulic cylinder at 5.28 kN and the knot didn't roll at all. But in this regard one has to be cautious in making conclusions, since a static rope might behave differently than a dynamic rope. However, his numbers are in line with the DAV numbers.

The point about my method is that there is pretty obviously a "right" and a "wrong" way to tie two different diameter ropes in an EDK in order to make rolling as hard as possible. Using just a single overhand knot in one strand to choke off rolling is significantly less bulky than tying two overhand knots back-to-back, a system that is often recommended but I think more likely to cause friction for the pull.

Here are some photos of the method. Frankly, I use it all the time, whether or not the ropes are of different diameters.
http://www.supertopo.com/photos/18/71/308672_1634_XL.jpg , http://www.supertopo.com/photos/18/71/308671_1199_XL.jpg , http://www.supertopo.com/photos/18/71/308673_8648_XL.jpg .
rgold 09 Oct 2016
In reply to UKC Articles:

I should add that the article mentions that all three knots considered are "weaker" when tied with different diameter ropes. Well sure, since one of the ropes in question is of a smaller diameter! In the BD tests published in Rock and Ice referred to in the article, the 10.2mm/8.1mm combination tied with the EDK tested at 13.8 kN, far beyond anything remotely attainable during rappelling or even taking a leader fall on the knotted ropes, whose energy-absorbing characteristics are required by the UIAA to be under 12 kN for maximal fall tests.

But Moyers never got results that good, by and large the EDK in his tests broke around 9kN, although if tied loosely with crossing strands, rolling started at the 90 kg figure mentioned in the article. Moyers considered rolling to be a failure, and arguably with good reason, but it is not the same as the knot breaking. The knot in question rolled three times and broke at 9 kN.

By the way, for most people, the 30cm tail recommendation is about the length of their forearm.
LeeWood 09 Oct 2016
In reply to rgold:

> they clearly have not got the memo

But - they have the same means to test knots and the same means for feedback and surveillance at all levels. Club structure and training in France is much tighter and more disciplined than UK.

I have shared a fig-8 knot with french teams - and of course itworks under normal dry conditions. NB. The uk failure mode is for stiff iced-up ropes.
rgold 09 Oct 2016
In reply to LeeWood:

The Zion http://www.supertopo.com/climbers-forum/4346/Death-in-Zion , Goat Wall http://www.rockandice.com/climbing-accidents/rappel-knot-fails-climber-falls-to-death-on-the-goat-wa... , tragedies did not occur in conditions of icing or wetness. The Raggeds Wilderness accident http://www.rockandice.com/climbing-accidents/euro-death-knot-mysteriously-fails was on an ice climb, so probably involved gear that was at least wet. But what failed in that accident was a 6mm sling tied with a flat eight. Although not widely reported, I think that both the EDK and flat eight should not be used with dyneema, which is much more slippery---even the double fisherman's knot is considered sketchy with dyneema and a triple fisherman's is recommended. Whether the sling in question was dyneema I do not know.

There was an accident in the Tetons in which an ordinary EDK was supposed to have failed in dry conditions, the only accident I've heard of that involved the EDK in dry conditions, rather than the flat eight, but what actually happened has been clouded by the fact that the knot was retied for some reason after the first person rappelled. That said, I believe Andy Kirkpatrick has an account of literally holding a rolling EDK together on wet ropes.

The problem with the flat eight is not only that it seems to roll more easily, but also that when it does roll it eats up a lot more tail than an EDK. It doesn't seem to have even a single advantage over the EDK so why the French still list it as an equivalent solution is utterly beyond me.
rgold 09 Oct 2016
In reply to rgold:

Here is the actual link to the German tests of EDK with different diameter (and materials) ropes that concluded there is no problem with the EDK in such situations. http://www.bergundsteigen.at/file.php/archiv/2009/2/print/48-53%20%28Hilfsleinen%29.pdf
David Coley 09 Oct 2016
In reply to rgold:

> Here is the actual link to the German tests of EDK with different diameter (and materials) ropes that concluded there is no problem with the EDK in such situations. http://www.bergundsteigen.at/file.php/archiv/2009/2/print/48-53%20%28Hilfsleinen%29.pdf

Any chance of a translation of the main table please.
Thanks
(Or Jim if he is around)
LeeWood 09 Oct 2016
In reply to rgold:

So why aren't the failures cropping up among french climbers - perhaps they're never so adventurous :/

Do any of your links to french sites explain their bias for fig-8 and mention bias against the overhand ?
rgold 09 Oct 2016
In reply to David Coley:

Column Headings:

1. Reepschnur diameter

2. Breaking Strength without Knot

3. Breaking Strength—Reepschnur tied with with EDK

4. Breaking Strength—Reepschur/10.2 mm climbing rope tied with EDK

5. Rolling begins in Reepschnur/Climbing rope EDK

Table entries (upper left data cell taken as (1,1) so not counting header cells):

(1, 5), (2,5) , (5,5), (6,5) No rolling happened.

(4,4) No breakage, knot rolled completely off
(6,3) Knots rolled off starting around 4 kN
(6,4) Mantle tore, then ?
rgold 09 Oct 2016
In reply to LeeWood:
I never used the term "bias for figure-8." The links I posted seem to treat both as equivalent, which seems like a mistake from the evidence we know about from multiple testers.

As to why "failures aren't cropping up among French climbers," I'm not even remotely close enough to the French climbing scene to have any idea whether or not such failures are happening. The statistical probability of a flat-8 rolling off the ends of a rope may be relatively small, but the tests tell us that it is even less likely for an EDK to roll catastrophically. Since the flat-eight has no advantages over the EDK and is more likely to fail, why would anyone put it on a par with a more secure option?
Post edited at 23:24
rgold 10 Oct 2016
In reply to LeeWood:
Ok, tried a little harder. Here are two videos from the Ecole Nationale de Ski et d'Alpinism (ENSA).

The first is a rappelling tutorial that clearly considers the flat 8 to be the standard; https://www.youtube.com/watch?v=N-jE-Da4P1U .

The second is on knots for slings, climbing rope/reepschnur combinations, and climbing rope/climbing rope junctions. https://www.youtube.com/watch?v=N-jE-Da4P1U . I'm not able to extract much meaning from spoken French without listening over and over, which I don't have the patience to do right now, so am not sure about what the video says. The numbers are, I think, in daN. The results do seem different from what's been found in the other tests I've seen, so I have to retract at least some of the certainty with which I expressed myself above.

The flat 8 seems to do even a touch better than the edk for joining two 7.3 mm ropes. A peculiarity of the tests is the relatively low loads, below 1 kN, at which rolling seems to occur in all these knots; this does not seem to be an observation of the other tests. Perhaps the velocity of the loading has something to do with the results? (In which case we'd need some standardization to even compare them...)

The knots joining reepschnur to climbing rope in the one test that doesn't jam the knot up against a quick link are tied in what I suggested earlier is the "wrong" way, thereby promoting rather than retarding the first roll.
Post edited at 00:58
LeeWood 10 Oct 2016
In reply to rgold:

V interesting ! they clearly have compared overhand & 8 - and show the 8 to be superior - fibre destruction before roll-off. What they don't touch on however is performance in icy conditions.

ps. 'bias' was my teminology
David Coley 10 Oct 2016
In reply to LeeWood:

I have one data point.
Tied a flat fig 8 and a edk in 8.3mm rope. Loaded with body weight from a tree in my garden. The 8 rolled, the edk didn't
rgold 11 Oct 2016
In reply to LeeWood:

I think "superiority" is a tricky thing. The flat-eight capsizes at lower loads and consumes twice (maybe more) as much tail when it does roll. That might be more dangerous than the ultimate breaking strength. Moreover, the knots were tested with tails that appear to be far shorter than the typically-recommended 30 cm. The EDK rolled off these shorter tails while the flat-eight broke. Longer tails wouldn't change the flat-eight result, but what would happen with the EDK?

Another interesting issue is rope diameter. Moyers tested 11mm ropes, the French tests were with 7.3 mm ropes, a diameter that is practically cordelette-sized. It is pretty clear that neither of these knots is good for sling material (which hasn't stopped them from being recommended for the purpose). Perhaps they are also not so good for climbing ropes that are nearly sling-sized?

I'm still wondering about the effect and relevance of loading rates. Perhaps Jim could enlighten us.

Xharlie 11 Oct 2016
In reply to UKC Articles:

I didn't see it mentioned, here, but there's another knot that I've heard called (incorrectly) a "double overhand" in the past, or a "European Double Death-Knot" - a somewhat unflattering name. You can see pictures of it on this South African climbing forum: http://www.climbing.co.za/forum/viewtopic.php?f=17&t=8520&p=47146&hilit=double+death+kno...

Basically, it is like the overhand (aka the European Death-Knot) but with an extra turn . The plain overhand has been the cause of deaths in the past but only because of short tails, bad dressing or incorrect use. The extra turn, forming the EDDK, prevents the knot from capsizing - apparently - while retaining the desirable flat edge of the ordinary overhand.

Anyway, it's what I use.

I don't like the double-fishermans - it gets stuck. The alpine butterfly tends to get stuck as well. I'd be quite happy with the overhand, but the extra turn is so easy, it doesn't make sense not to add it. (Certainly, one extra turn is easier than a whole second overhand.)
GrahamD 11 Oct 2016
In reply to Xharlie:

> The plain overhand has been the cause of deaths in the past but only because of short tails, bad dressing or incorrect use.

Do you have any citations for this ? I've never found anything that directly points to an overhand knot failure as being the cause of fatalities - yet the 'fact' seems to crop up as regularly as these threads.
James Gilbert 11 Oct 2016
In reply to LeeWood:

> but the french are *still¨* advocating fig 8 for rappel


On the same website they do also have the EDK as a possible knot for abseiling (with 40cm tails, again shown far too short in the photos).

http://www.ffme.fr/fiches-ffme-techniques/page/confection-du-noeud-simple-ou-de-plein-poing.html
Dave Garnett 11 Oct 2016
In reply to Xharlie:

> I don't like the double-fishermans - it gets stuck.

What do you mean - it's difficult to untie after loading or it gets caught as you pull the rope down?

I'm sticking to what I've always used - double fisherman's with stopper knots in the tails. I'm sure there are knots that are quicker to tie and easier to undo, but I tend to find that neither of those considerations seems so important as I ease over the edge of a f*cking great drop...

It's especially important if it's long, multipitch abseil, with the rope being pulled through and rethreaded through chains multiple times - you want something durable, stable and easily visually checked.

Anyway, if you pull the rope through the belay a bit so that the knot is just over the edge (which can be exciting but not dangerous if done properly) you can minimise the risk of a jammed rope. There's also a knack of slackening off a tight knot after use, but anyway I'd rather spend two minutes untying safely on the ground than two minutes hanging free on the way down wondering whether the knot will hold.

Stu Tyrrell 11 Oct 2016
In reply to Dave Garnett: Use the same with a reef knot in between, no problem untying.



LeeWood 11 Oct 2016
In reply to James Gilbert:

Ah yes - I'm still not familiar with french knot names. I note they recommend tails of 40cm here - while the obviously safer 'plein-poing' only needs 20-30 :o
jimtitt 11 Oct 2016
In reply to rgold:

Can´ t help you there, I don´ t do much knot testing. The only stuff I tested was ring loading on various tie-in knots (which is a similar loading and couldn´ t get the 8 to fail either.
I look at all the various tests and am just as confused a everyone else! The BMC are doing some testing but the result seem consistent with everyone elses- there is no definite answer!
I usually use a double-fishermans, most of the abseiling I do (and probably most people these days) the rock is so steep having a knot that rolls over edges seems unescessary. Also with the longer ropes nowadays I normally use just one one rope anyway.
Xharlie 11 Oct 2016
In reply to GrahamD:

I admit that I do not - I had always just assumed it as truth. I do know the EDK can roll or capsize, though, so assume that it could come apart if the tails weren't long enough. and the extra turn prevents that.

Anyway, your point is noted. Next time, I'll refrain from blithely stating that it has actually caused deaths and just say it can roll.

In reply to Dave Garnett: I mean that the DF gets stuck in cracks and on rock features, not that it is difficult to untie - although some do claim that, I have never struggled. Surely tying extra stopper knots is overkill and only makes it get jammed more often? Just a DF with even moderate tails should be absolutely bomber. (You can also isolate at DF with an Alpine Butterfly.)
Dave Garnett 11 Oct 2016
In reply to Xharlie:

> Surely tying extra stopper knots is overkill and only makes it get jammed more often? Just a DF with even moderate tails should be absolutely bomber. (You can also isolate at DF with an Alpine Butterfly.)

The only time I've ever seriously jammed ropes is by underestimating the friction round a tree (and nearly being benighted at Buoux), rather than jamming the knot but, yes, the stopper knots probably are overkill.

Then again, when it comes to just about the only 'one mistake and you're dead' in aspect of climbing other than soloing, I quite like overkill.
knudeNoggin 20 Oct 2016
In reply to rgold:
> The Zion ... , Goat Wall ... , tragedies did not occur in conditions of icing or wetness.

More importantly, we simply do not know many factors --even what knot was used, for sure. Tom Jones, a one-time climber now canyoneering much, carried a recommendation for the offset (it is not "flat"!) fig.8 even after the Zion tragedy; when I remarked about this to him, he was well apprised, claiming to have personally seen the particular ropes : "they were like cables!" (sooo stiff). Now, however, his site recommends the EDK-backed EDk.

I find it disappointing to read so many quick judgements based on so limited and SO LITTLE EXPLAINED snippets of *testing* : e.g., in the DAV quick chart, we have no idea of orientation (but must presume it is in the thinner-makes-the-choke way that I & you have championed for years (even w/images)). (Btw, ALL of those break values (73% - 79% - (the winner!) 92%) sure open my eyes (in surprise)!)

And one must realize --maybe it'll take some big images and test results to hammer the point-- that there are MANY FACTORS in the behavior of these knots : esp. the dressing (I've pointed to these shown fig.8s being poorly set), but other things not commonly realized (evidenced by which way the tails point). And loading in practice will see more of a shock-bounce on any flyping/rolling of such a knot than the steady-rate test-device's drop in tension and slow pick-up.

As to why there isn't a dearth of French climbers from loss due to consumption of Noued en Huit (offset, I think an aspect left untitled), it is a fair question, if the usage follows the apparent popularity of recommendations. (Interesting to read their recommending more length in tails for the OWK than Of8 : 30cm vs. 20cm!) Still, so many of the tests do show problems only well past reasonable abseil forces. I've read of some people supposedly getting the offset-8 (EDK-8) to roll by mere manual pulling, and reflect that I've not done it (the few times I've tried) with pulley-assisted loading.

*kN*
knudeNoggin 20 Oct 2016
In reply to Xharlie:
> but there's another knot that I've heard called (incorrectly) a "double overhand"

Actually, that's more "correct" than using the name for an EDK-backed EDK --the structure IS a dbl.oh.
A better name would be "offset strangle bend". (And that thread goes on to show Jost Gudelius's (et al.) offset grapevine bend and his orginally mis-named "triple fish.".)
What the first two knots miss is the efficiency >>of material (not tying!)>choking line I'd be quite happy with the overhand, but the extra turn is so easy, it doesn't make sense not to add it. (Certainly, one extra turn is easier than a whole second overhand.)

I'm leery of the potential for *slop* in the forming of the strangle. Whereas the making of a 2nd EDK is a simple repeat of the knotting just done, and even if both knots (the one to be offset loaded and its partner) are poorly tied, the "back-up" one gets LOADED QUA STOPPER --i.e., not pulled apart but pulled smack into the main knot. And, again, I think that testing will show this to be quite an accommodatingly effective knot, suffering all sorts of poor tying.
And, so, recommended for those times of duress --fatigue, darkness, haste-makes-waste, low oxygen, thick gloves, ... !

*kN*
Post edited at 01:53
knudeNoggin 20 Oct 2016
In reply to David Coley:
> Any chance of a translation of the main table please.

(on-line translation & guessing)

> The following table indicates the strength of connections of different rope diameters with Overhand teardrop (rap knot) [offset water knot!]. The data refer to % of the of the THINNER material.

(And the Seilschlupf = slippage column claims keiner (NONE !! (??)) for the thinnest two, and then £from 6kN£ for the 7mm; the 8mm I think is said to have exceeded testing device recording or maybe there was some slippaged?!)

I'm surprised (skeptical) of both the "no slipping" and high break strengths (even vs. rated vs actual --still seems (way) high!).
YMMV

*kN*
Post edited at 01:52
galpinos 29 Oct 2016
In reply to UKC Articles:

This popped up on the AMGA instagram feed from the Petzl Institute in SLC:

https://instagram.com/p/BMIQKtiAszy/
rgold 29 Oct 2016
In reply to galpinos:

Just for easy comparison, here is the French test I posted earlier: https://www.youtube.com/watch?v=N-jE-Da4P1U .

The difference in knot behavior seems significant to me, and makes it really hard to interpret these tests. It looks to me as if the French test employs a higher loading rate, and I think my earlier comments about the effect of loading rates on the test results are even more appropriate now: one has to ask whether the results are somehow intrinsic to the knot or an artifact of the loading rate, and if the loading rate has a significant effect, whether such tests are at all relevant to the climbing application of the knot.

Unless slack develops and there is something like a leader fall, rappelling on a dry well-tensioned and dressed EDK will not cause a failure. Wet, the worst performance in the Moyers tests was with unequal diameters, which saw the first roll at 950 lbf, a level possible if not highly likely from rappelling. We don't know what would happen if the extra overhand (in one strand as I've suggested, or in both strands as others have recommended) is added, but the tests by the user 20kN on MP suggest an improvement for the one-strand method.

A careful reading of the Moyers tests suggests that the figure eight is ok too, but is particularly sensitive to capsizing at low loads---the worst was at 290 lbf---if the knot isn't well-dressed and well-tensioned. Given this sensitivity to tying and the knot's ability to eat a lot of tail when it capsizes, there still seems to me to be no good reason to prefer it over the EDK, but it should be fine with long tails and proper tying, which is probably why there don't seem to be a lot of figure-eight accidents in France.
ex0 29 Oct 2016
In reply to UKC Articles:

Since when is a 'double overhand knot' 2 overhands tied next to each other rather than a single overhand with 2 bends?
galpinos 29 Oct 2016
In reply to ex0:

I thought the same. A double overhand and two overhands next to each other are very different knots.
ex0 29 Oct 2016
In reply to galpinos:

Ironic given the article name huh!
nb 30 Oct 2016
In reply to rgold:
Seems to me that the only conclusive evidence that can be taken from all these tests is:

Well-tied knot = good knot
Badly-tied knot = bad knot

The arguments here are all about which badly-tied knot is worse and in which of the multiple situations. The obvious solution is to always tie your knot well. If climbers can't do that when their life depends 100% on the quality of the knot, they're probably in the wrong sport (or slightly suicidal).

I live in France and have used either a simple overhand or a figure 8 for the past 30 years and so have the vast majority of my friends (I find the fig 8 is much easier to undo after multiple steep abseils). I see no reason to stop using them, but I do think these debates are useful in reminding people just how important it is to tie the knot properly.

Edited to add: with icy ropes it does seem sensible to double up the knot with another overhand. This can be done when using the fig 8 as well.
Post edited at 11:29
3
In reply to UKC Articles:

One point about the Double Overhand Knot that makes it safer than the Overhand in my opinion is that you have to think what you are doing to butt the two parts tightly together.

This is an extra safety factor that could be crucial if you are tired/scared/injured/in bad weather as you often are when these knots need tying.
knudeNoggin 01 Nov 2016
In reply, resp., to nb, rgold, stephen reid:

> The arguments here are all about which badly-tied knot is worse and in which of the multiple situations. The obvious solution is to always tie your knot well. If climbers can't do that when their life depends 100% on the quality of the knot, they're probably in the wrong sport (or slightly suicidal).

But this is unfair : fatigue & other forms of duress can challenge one to do things so well. Beyond this, the dressing and setting of these knots is nowhere well explained --the simplistic "pull all ends" misses key points (there is merit do setting the knot by pulling various ends in particular sequence!).

> ... : one has to ask whether the results are somehow intrinsic to the knot or an artifact of the loading rate, and if the loading rate has a significant effect, whether such tests are at all relevant to the climbing application of the knot.

Let me repeat (even within this thread!)::
I find it disappointing to read so many quick judgements based on so limited and SO LITTLE EXPLAINED snippets of *testing* : e.g., in the DAV quick chart, we have no idea of orientation (but must presume it is in the thinner-makes-the-choke way that I & you have championed for years (even w/images)). (Btw, ALL of those break values (73% - 79% - (the winner!) 92%) sure open my eyes (in surprise)!)

And one must realize --maybe it'll take some big images and test results to hammer the point-- that there are MANY FACTORS in the behavior of these knots : esp. the dressing (I've pointed to these shown fig.8s being poorly set), but other things not commonly realized (evidenced by which way the tails point). And loading in practice will see more of a shock-bounce on any flyping/rolling of such a knot than the steady-rate test-device's drop in tension and slow pick-up.

> One point about the Double Overhand Knot [<<- read "EDK-backed EDK"] that makes it safer than the Overhand [read "EDK"/"offset water knot"] in my opinion is that you have to think what you are doing to butt the two parts tightly together. This is an extra safety factor that could be crucial if you are tired/scared/injured/in bad weather as you often are when these knots need tying.

Rather, one doesn't even need to set the back-up knot snug to the main one, IMO --though that IS advisable. And esp. to your key point about tying conditions, the ability of the said knot combo to tolerate all sorts of imperfections (IMO --testing yet to show ...) is its very strong commendation to be at least "in one's toolbox" as an option, for at least such stressful times (or in cases of dubious rope combinations, unusual loads).

*kN*
nb 02 Nov 2016
In reply to knudeNoggin:

> But this is unfair : fatigue & other forms of duress can challenge one to do things so well.

It's got nothing to do with fairness. If you feel you're not capable of tying an abseil knot properly when fatigued and under duress then you shouldn't climb big routes - simple as. Everyone can f*ck up of course, but that's possible whatever knot you use.

To those unsure of whether a fig8 knot is safe to abseil off, you can test it pretty easily. Since there's no shock-loading involved you don't need a test-tower. Just tie two ropes together, clip them into the first bolt at your local climbing wall, tie a loop in them, drag a big crashpad underneath and start hanging your friends off it. If the knot is tied TIGHTLY with 20-30cm tails then I guarantee it will hold at least 5 of your porkiest mates - that's industrial safety standards met! Now keep piling on your friends and see what happens. At what point does it start rolling? You'll soon realise why there hasn't been hundreds of French climbers killed abseiling over the past few years!

Admittedly, the test is bit more complicated to run with icy ropes, but I bet most climbing walls have some kind of fridgy-freezer thing for cooling beers and making ice cubes. Knots rely on friction (that's why slippy Dyneema requires a triple fisherman's) so icy knots will defo hold less weight but I bet you'll still be able to hang a few people off it if it's been well-tied.

I'd be more worried about the state of a lot of abseil anchors in the UK - rusty pegs, saplings, stakes...

It's a funny old world!



3
rgold 02 Nov 2016
In reply to nb:

> If the knot is tied TIGHTLY with 20-30cm tails then I guarantee it will hold at least 5 of your porkiest mates - that's industrial safety standards met! Now keep piling on your friends and see what happens. At what point does it start rolling? You'll soon realise why there hasn't been hundreds of French climbers killed abseiling over the past few years!

I'm sure I'll be corrected if I'm wrong, but didn't David Coley do essentially this test and manage to roll a Figure-8 with nothing more than his own porky self?

The inconsistencies in the published testing data suggest that various subtle factors may be in play. The type and diameter of the ropes might matter, and the type of loading too, with cyclic loading an apparently unexplored contingency.

It is fair to surmise that the probability of either EDK or Figure 8 rolling is pretty low; that really isn't the point. The point is, is one of these knots more likely to roll than the other? Because we do have a choice.
nb 03 Nov 2016
In reply to rgold:

> The inconsistencies in the published testing data suggest that various subtle factors may be in play. The type and diameter of the ropes might matter, and the type of loading too, with cyclic loading an apparently unexplored contingency.

To be honest, I don't think I've ever seen such inconsistencies in any other tests anywhere in my entire life! The ENSA video (linked above) claims that the fig8 is stronger than the overhand and less liable to roll, whereas this article links to a test where the fig8 fails at 30kg! ENSA actually recommends the fig8 over the overhand in situations where you may get higher loading (such as when there is an avalanche risk) whereas this article says it is dangerous in any situation! It will be interesting to see what the BMC comes up with.

So when you have such inconsistant data coming from the 'experts' then you have a case for climbers doing their own testing. The beauty of this particular test is that it doesn't require shock-loading and so can be set up by any climber at a climbing wall with a bunch of mates. I'd do it myself but am stuck in London with no climbing gear for a few weeks.

But I will be very interested to hear from anyone who manages to roll a tightly tied fig8 off the end of a 20cm tail with 5 porkies hanging off it. Please cross the strands (my knots are never tidy!) but do tie it tight (pull hard on all 4 strands) and don't use dyneema or unreasonably skinny cord.
Chris Craggs 03 Nov 2016
In reply to nb:

> To be honest, I don't think I've ever seen such inconsistencies in any other tests anywhere in my entire life! The ENSA video (linked above) claims that the fig8 is stronger than the overhand and less liable to roll, whereas this article links to a test where the fig8 fails at 30kg!

I must admit I was amazed at this, two of us tried to pull a Fo8 apart - double body weight - and it didn't budge.

I use the overhand anyway: http://www.ukclimbing.com/images/dbpage.php?id=155389


Chris
Rick Graham 03 Nov 2016
In reply to nb:

>
> But I will be very interested to hear from anyone who manages to roll a tightly tied fig8 off the end of a 20cm tail with 5 porkies hanging off it. Please cross the strands (my knots are never tidy!) but do tie it tight (pull hard on all 4 strands) and don't use dyneema or unreasonably skinny cord.

But the problem is we do not often hang five climbers off an abseil anchor.

In a multi pitch situation, one climber goes first and sometimes varies the loading on the knot or maintains a constant load.
I have heard of accidents ( amazingly non fatal ) where the waiting climber has watched the abseil knot unravel in front of their face.
It would be nice to know why and which knot is most suitable.
galpinos 03 Nov 2016
In reply to Stephen Reid - Needle Sports:

Do you mean a double overhand, which is an overhand with and extra turn, or two single overhands next to each other?

The incorrect usage in the article is quite confusing.
In reply to UKC Articles:

Seeing that our research has been name-checked here, I figured I'd fill folks in with what we've been up to, what else we've got planned and my general feelings on the subject having done a fair bit of looking into this area.

(As an aside, I did the tests published on Needlesports website back when I ran the test lab at Lyon Equipment, in conjunction with Stephen Reid).

Getting an agreed nomenclature for the different knots would be a good starting point - it's made much more difficult when you realise that you're discussing a different knot to everyone else.

The next point is that, as has been pointed out already, whether people use the flat overhand bend or flat figure 8 bend or any other joining knot, failures of the joining knot are incredibly rare. To me, this suggests (and tests appear to bear this out) that these knots are usually safe but that perhaps there are some "edge case" scenarios in which they may become unstable and therefore dangerous.

With this is mind, testing we've done has moved away from the steady-state loading you get in a test machine. These machines are good for finding out how strong a knot is (as long as it remains stable) but aren't so useful when it comes to finding out whether a knot can become unstable and capsize or roll in normal abseiling use. What is really required are lots and lots of tests where people abseil and the knot is carefully observed to see how it behaves.

So far, we've concentrated our testing on the flat overhand bend because this is the knot of choice. We also felt that as a simpler knot, it might have fewer possible variables affecting how it is tied and performs. We tested new, slick ropes and old, worn ropes, wet and dry ropes, heavy abseilers and light ones. We mismatched rope diameters (8.6mm paired with 9.1mm), and after failing to get any knots to capsize started to tie the knots deliberately badly - loose and undressed.

The next step is to look at tests when the rope is cold and/or frozen, because temperature affects the properties of nylon considerably.

In some ways the results so far are reassuring, in that no failures have been witnessed. On the other hand, it would in some ways be good to have a lead, something to investigate to give a cause for failure. The big problem for forensic examination of joining knot failures is that the end result of a failure doesn't leave any evidence about how the knot was tied, or even which knot was tied, or even if any knot was tied at all.

So unfortunately, we can't give a definite answer at this stage, other than that the flat overhand bend seems pretty reliable and there is no evidence so far to suggest its use should be prohibited.
nb 03 Nov 2016
In reply to Rick Graham:

> It would be nice to know why and which knot is most suitable.

It would definitely be nice to know! I'm suggesting we work it out - no specialised equipment required here!

Instead of having 2 guys with an engineering qualification between them thinking up a few tests to do with whatever gear one of the manufacturers has thrown their way, we can have dozens of minds thinking up hundreds of ways of pulling a well-tied fig8 or overhand knot apart.

There are mechanical engineers, textile engineers, rope access workers, mathematicians, mountain guides and even the odd climber on these forums. There is plenty of potential for working this out, and since everyone can take video these days it's easy to communicate the results.

The internet has given us crowd-funding, why not crowd-testing!

And if anyone posts video of a well-tied fig8 coming apart with 5x body weight I'll post one of myself eating a hat!




David Coley 03 Nov 2016
In reply to Chris Craggs:

> I must admit I was amazed at this, two of us tried to pull a Fo8 apart - double body weight - and it didn't budge.


> Chris

I have rolled a fig8 with a little jumping up and down. My tests - from tree in garden - suggest that the issue is with dressing. Do a fig8 or an overhand well and neither is a problem. Do them loose and the fig8 rolls, whereas in my tests the overhand doesn't.
Chris Craggs 03 Nov 2016
In reply to David Coley:

> I have rolled a fig8 with a little jumping up and down. My tests - from tree in garden - suggest that the issue is with dressing. Do a fig8 or an overhand well and neither is a problem. Do them loose and the fig8 rolls, whereas in my tests the overhand doesn't.

I tried it again today with a loose F o 8, it half rolled and then locked. Rolling under a 30kg load still seams so unlikely unless the knot was so loose it was really no knot at all,

Chris
nb 03 Nov 2016
In reply to Dan Middleton, BMC:
Hi Dan

I was writing my comment when you posted, so my post is in no way a response to yours.

However given that you wrote...
> What is really required are lots and lots of tests where people abseil and the knot is carefully observed to see how it behaves.

... my idea just got more relevant!

Now I know that the BMC cannot encourage climbers to test abseil knots - even at 20cm above a crashpad - because the merest bruise would be cause for legal action, but there is nothing to prevent someone taking a personal initiative, seeing if they can pull a knot apart and posting a link to a video if they manage.
Post edited at 17:37
In reply to knudeNoggin:

>Rather, one doesn't even need to set the back-up knot snug to the main one, IMO --though that IS advisable.

If you don't the second knot is more likely to catch on something when you pull the ropes down.
In reply to galpinos:

I mean two overhand knots, one tied after the other, and butted up close two it. If you google double overhand knot what mainly seems to come up are photos of what climbers would call stopper knots.
GrahamD 03 Nov 2016
In reply to nb:

I did the test with a single overhand at my local climbing wall with the apparently absurd combination of a 10mm lead rope and 6mm prussic cord. It all held together fine which reassured me to carry on using the knot, even for ropes of slightly different diameters.
In reply to nb:

Obviously, if anyone was foolish enough to do this and did get a knot to fail, it would be most useful if the maximum amount of info was recorded, such as ropes used, temperature, images or video etc.

In reply to Chris:

I did the test where the flat figure of 8 failed at 30kg, and it was spooky. The knot just rolled over as the rope became taut in the test machine, and continued to roll and roll until the machine hit its end stop. It was just a normally tied knot, mind you this goes back over ten years, but I don't recall it being either badly tied or snugged down really tightly, just tied like a normal knot really.
In reply to nb:

> It's got nothing to do with fairness. If you feel you're not capable of tying an abseil knot properly when fatigued and under duress then you shouldn't climb big routes - simple as. Everyone can f*ck up of course, but that's possible whatever knot you use.

> To those unsure of whether a fig8 knot is safe to abseil off, you can test it pretty easily. Since there's no shock-loading involved you don't need a test-tower. Just tie two ropes together, clip them into the first bolt at your local climbing wall, tie a loop in them, drag a big crashpad underneath and start hanging your friends off it. If the knot is tied TIGHTLY with 20-30cm tails then I guarantee it will hold at least 5 of your porkiest mates - that's industrial safety standards met! Now keep piling on your friends and see what happens. At what point does it start rolling? You'll soon realise why there hasn't been hundreds of French climbers killed abseiling over the past few years!

> Admittedly, the test is bit more complicated to run with icy ropes, but I bet most climbing walls have some kind of fridgy-freezer thing for cooling beers and making ice cubes. Knots rely on friction (that's why slippy Dyneema requires a triple fisherman's) so icy knots will defo hold less weight but I bet you'll still be able to hang a few people off it if it's been well-tied.

> I'd be more worried about the state of a lot of abseil anchors in the UK - rusty pegs, saplings, stakes...

> It's a funny old world!

There have been quite a few incidences of Figure 8 unrolling. One happened to a friend of mine who fell a very long way and was lucky to survive. Both climbers were very experienced with a long history of climbing hard routes.

1
In reply to Dan Middleton, BMC:
> I did the test where the flat figure of 8 failed at 30kg, and it was spooky. The knot just rolled over as the rope became taut in the test machine, and continued to roll and roll until the machine hit its end stop. It was just a normally tied knot, mind you this goes back over ten years, but I don't recall it being either badly tied or snugged down really tightly, just tied like a normal knot really.

If these were the tests that you did with me Dan, then the all the knots were tied properly, i.e. well laid and pulled tight by hand. We did the tests on dry, wet and frozen ropes of 9mm diameter.

Here's what I wrote afterwards:

"...it was interesting to put the various knots to the test at Lyon Equipment's headquarters at Dent. Lyon's test rig maestro, Dan Middleton actually ran several tests, using different diameter ropes and wet ropes as well, but as the others did not produce significantly different results from the climber's point of view, we have simply reproduced the most telling results below (all the ropes used were brand new 9mm).


Knot Type, Result, Slippage on each rope (relaxed) to nearest £cm£
Frozen
Fig 8, Flipped at 300 Newtons, 11
Overhand, Didn't flip up to 1000 Newtons, 7
Double Overhand, Didn't flip up to 1000 Newtons, 5
Dry
Fig 8, Didn't flip up to 1000 Newtons, 5
Overhand, Didn't flip up to 1000 Newtons, 4
Double Overhand, Didn't flip up to 1000 Newtons, 3
(Sorry - not sure how to get this to lay out as a chart on UKC - but you can probably figure it out)

The first thing to note is that when the ropes were soaked and then frozen there was considerably more slippage than when they were dry. Second, and most importantly, in frozen ropes the Figure-of-Eight Knot flipped over itself at 300 Newtons or 30kg. As the weight of a male abseiler plus kit is likely to be around 80kg this is serious cause for concern. The test of course stopped after the first flip, but in a real situation the weight would be reapplied and the knot would be likely to flip again and again until it rolled off the end of the rope. This didn't happen to the same knot when the ropes were dry, but it looked as if it was likely to, and we were dealing with a perfectly tied knot ..."
Post edited at 18:59
Rick Graham 03 Nov 2016
In reply to GrahamD:
> I did the test with a single overhand at my local climbing wall with the apparently absurd combination of a 10mm lead rope and 6mm prussic cord. It all held together fine which reassured me to carry on using the knot, even for ropes of slightly different diameters.

Unfortunately a 10 + 6 mm combo is quite a common system amongst some climbers.

rgold has helpfully posted a lot of research, experimentation and advice about this combo.

Apparently the knot works a lot better depending on which side the 6mm is tied.

Hard to explain and understand, and crucially get correct every time on the hill.

Excellent advice and suggestions posted above by Dan, rgold and Steve Reid.
Post edited at 19:04
nb 03 Nov 2016
In reply to Stephen Reid - Needle Sports:

Hi Stephen

Can I just ask a couple of questions to better understand your results.

1) What was the reason for stopping the test after the first flip?

2) Why do your results stop at 100kg?

Thanks

Neil


knudeNoggin 03 Nov 2016
In reply to Stephen Reid - Needle Sports:

> If you don't (snug back-up EDK to first) ... more likely to catch ...

Perhaps so; my point though was about how many imperfect tyings this knot combination could endure w/o failure!

*kN*
knudeNoggin 03 Nov 2016
In reply to nb:
>> But this is unfair : fatigue & other forms of duress can challenge one to do things so well.

> It's got nothing to do with fairness. If you feel you're not capable of tying an abseil knot properly when ...

IMO, this is a bit arrogant; in any case, we'd like to have systems that cover broader vs. fewer situations, where practical. This point aside, there is no clear statement of what "tie properly" is, in detail. E.g., in the article under discussion, I've remarked about how the offset fig.8 is tied --i.e., that it has NOT been properly SET (the U-turn of the loaded parts should lie in the visual plane, not coming at the viewer as done, which betrays a looseness to be take up upon initial loading that will pry open the choke more than should be, giving a head start to flipping).

There are in rough terms three orientations of the loaded lines as they enter the offset water knot (EDK) & like knots. In the image of this article, the orientation is the one between extremes, the loaded parts entering and equally turning; at the two extremes, one of the loaded parts will make a loop in the visual plane to the other's arc. In the show orientation here, the tails run roughly aligned with the rap ropes; in the extremes, they will be roughly perpendicular to one or the other sides.

And beyond this variation, come the many variations of particular ropes used.
RGold has previously noted the issue of position of thick/thin ropes --that the thin should be which makes the *choke* of the loaded ends' entry. This point we've raised now over several years; but it remains one more detail to get right, whereas the simple EDK back-up knot as I've stated IMO will tolerate all sorts of imperfections of tying and yet hold.

*kN*
In reply to nb:

> Can I just ask a couple of questions to better understand your results.

> 1) What was the reason for stopping the test after the first flip?

> 2) Why do your results stop at 100kg?

Dan Middleton would be better placed to answer these questions than me but, from what I recall, it was due to the limitations of the rig we were using.

It was a shame that they stopped at 100kg as a large man with an alpine sack on bouncing on the rope could well exert more weight/force than that.

Never-the-less we were both really surprised when the figure-of-eight on icy ropes slipped so easily.

For what it's worth, we weren't conducting a rigorous scientific experiment. The situation was that the accident that I alluded to above had recently happened and I was surprised to hear that two such experienced climbers were using a figure of eight to join two ropes together but it turned out they were unaware of rumours then circulating as to how dangerous it was. For what it's worth, I had previously used a figure of eight many times as an abseil knot as its advantages over a double fisherman's to anyone doing multiple abs in the alps were obvious - easier to tie, less likely to get caught, easier to undo. It was definitely the knot a lot of people were using around 10 - 15 years ago until a few accidents or close misses were attributed to it, at which point I began using the overhand and then devised what I called the double overhand as a safer option. So there but for the grace of God etc...

When I heard of the accident, I realised that there were potentially a lot of climbers still using a figure of eight - hence the tests at Lyon Equipment (which they kindly conducted free) and the article on our website. The article was originally supposed to go in Summit magazine but they refused to publish it when they realised I had put it on the web - not sure why - to me it was really important to get the message across.
nb 03 Nov 2016

Open question: What concrete evidence is actually out there about a well-tied fig8 failing catastrophically?

Are there any tests which have separated the two ropes?

Has anybody actually witnessed a well-tied fig8 knot come apart during an abseil? (Genuine first person accounts only please).

Edit to add: my definition of a well-tied fig8 knot is all strands pulled tight and 20-30cm tails.
Post edited at 23:40
GrahamD 04 Nov 2016
In reply to Rick Graham:

> Unfortunately a 10 + 6 mm combo is quite a common system amongst some climbers.

Well I have to admit I was doing my test for strictly personal satisfaction and for abseiling this isn't a combo I'd ever use. Obviously it was just the wall rope I had and a prussic I'd forgotten to take off my harness. I was just curious as to what would happen with seriously mismatched ropes to reassure myself for the usual case (for me) of having slightly mismatched ropes.

Martin Hore 04 Nov 2016
In reply to UKC Articles:

I hope this hasn't been mentioned before as I've not read the whole thread.

I've understood for a fair time (as long as I've been using the Euro Death Knot or Flat Overhand for joining abseil ropes) that the Flat Fig8 rolls much more easily and is dangerous to use in this situation. I presume this is because of the way the two ropes under tension tend to break open the knot. What I can't understand is why belaying with the belay plate attached to your rope loop, as opposed to your belay loop, is considered good practice when you have used a Fig8 to tie in (I do it). The way the forces act on the knot when you hold a fall would appear to break open the knot in an identical way.

Anyone able to explain?

Martin
rgold 04 Nov 2016
In reply to Martin Hore:

If you belay using the rope loop, your tie-in knot (bowline or figure-8) should have a solid backup knot, in which case I don't think ring-loading has any untoward consequences.
nb 05 Nov 2016
In reply to nb:

> What concrete evidence is actually out there about a well-tied fig8 failing catastrophically?

By the dearth of answers to my questions, I can only conclude that either (1) everyone has got bored and f*cked off or (2) the video below is the most reliable source of information we have concerning the overhand vs fig8 debate. It shows a fig8 capsizing once at 167kg, again at 535kg and then blocking until the rope breaks at 750kg kilos. The overhand rolls off the end of the ropes with a high value of 530kg. (Remember when abseiling your weight will be equally shared between the knotted side of the ropes and the unknotted side, so you can basically double these values.)

Start watching at 3min30. https://www.youtube.com/watch?v=N-jE-Da4P1U

As for accident reports, I have yet to find an attributable eye-witness account of a well-tied fig 8 knot rolling off the end of a rope.

In the reports I found which were critical of using a fig8 to join ropes, it is only ever assumed that the knot was tied correctly. The conclusion drawn is that the knot must be weak when some rare and poorly-understood circumstances arise.

Now personally, if someone has done 1000 abseils on a knot and then on the 1001st it fails, I don’t instinctively think, ‘the knot must be flawed’, I think, ‘they must have f*cked-up badly’. Maybe this thought-reflex comes from a deep awareness that I too may f*ck-up badly one day.

Interestingly when a French climbing instructor fell to his death while rappelling at a crag in 2001 (the only French example I could find), the authorities concluded that the knot must have been tied badly, rather than that there was an intrinsic problem with the knot. Maybe the French are more willing to accept the fallibility of the human condition - they did define existentialist angst after all!

So, when these accidents started getting reported around 2000, the French response – coming through federations or government institutions - focused on educating people on how to tie the knot properly. The UK/USA response - mainly generated by the media or by hearsay - simply warned that the knot was dangerous. Interestingly this excellent BMC article on abseiling does not issue any warning about using the fig8 to join ropes.

https://www.thebmc.co.uk/abseiling-not-the-quickest-way-to-reach-the-ground

All this begs the question, ‘could this whole death-knot theory be a huge, collective figment of the imagination?’ Martin Hore’s comment certainly raises issues about the reasoning behind it.

While the ENSA test should by no means be regarded as ‘the word’ on abseil knots, it is the most credible info I have seen so far and seems to corroborate the results here https://user.xmission.com/~tmoyer/testing/EDK.html but with a very different conclusion.

It will be interesting to see the results of the BMC tests, particularly with regards to icy ropes. There are 2 issues that I see with icy ropes:

- There is less friction and so rope slippage will probably occur at lower loads.
- The knot will be more difficult to tighten.

I use my teeth to tighten knots on icy ropes. The jaw is amazingly strong and teeth grip so much better than fingers (I’ve even used them on my ice axes when my arms have given out!). There may even be a case for tying a back-up knot sometimes, although I’ve never felt the need to do so myself.



PS Here’s an even easier way for people to do their own DIY testing. Tie the end of a rope into an anchor (could even be a strong branch), tie a 20 cm fig8/overhand loop just underneath and then another loop just underneath again. Hang your mates from the lower loop (you only need 2½ mates now that all the weight is on a single length of rope). The middle loop will act like an offset fig8 joining 2 ropes but you stay attached to the same length of rope – easier and safer!

If you want to properly load the knot, tie one end to a tree-trunk and the other to the tow bar of your car. Don’t forget to take video!

To test icy ropes, take a short length of rope, soak it in water, shove it into the freezer compartment of your fridge, wait a few hours and off you go.




knudeNoggin 14 Nov 2016
In reply to nb:

> So, when these accidents started getting reported around 2000, the French response – coming through federations or government institutions - focused on educating people on how to tie the knot properly. The UK/USA response - mainly generated by the media or by hearsay - simply warned that the knot was dangerous.

OTOH, when one sees the occasional --thankfully infrequent-- tragedy citing the offset fig.8 as a possible culprit, and NOT seeing such reports for the offset water knot, there is a quite natural move to use the latter, and some suspicion about hidden issues with the former..

I will agree w/you on one point : DIY testing should be done by people --who will use their own particular ropes and not something else--, and should be easily effective given the loads (body weight, usually tempered by loading twin strands) of interest! It would for these DIY testers give some confidence, or as has been suggested in any cases of some concern, help raise the particular case to broader attention (and then further testing)!

*kN*

nb 08 Dec 2016
jon 08 Dec 2016
In reply to nb:

Hi Neil, what's RAD line?
WildCamper 08 Dec 2016
In reply to jon:
Rescue and Descent

its a crevasse rescue system for skiers
Post edited at 10:35
jon 08 Dec 2016
In reply to WildCamper:

I meant what kind of rope was involved.
nb 08 Dec 2016
In reply to jon:

Hi Jon, it's a static cord for ski-touring so quite specific - 6mm and very compressible. The guy from Petzl says that they tested the knots on dry ropes, wet ropes and icy ropes and in all cases the fig8 was their knot of choice. With icy ropes it rolled once then held til breakage.
nb 08 Dec 2016
In reply to WildCamper:

> Rescue and Descent
> its a crevasse rescue system for skiers

Not just crevasse rescue, really meant more for abseiling or installing a handrail. Can also be used for crevasse rescue but it's not designed as a rope for crossing seriously crevassed Chamonix-style terrain - these are the words of the Petzl marketing manager, not mine. Pretty vague in my opinion.
jon 08 Dec 2016
In reply to nb:

But did they get the same results with a couple of dynamic climbing ropes knotted together? Or is that what you are referring to in your second sentence?
nb 08 Dec 2016
In reply to jon

These results are only valid for the RADline rope. I'll do a bit of diy testing over the weekend and let you know what happens!
jon 09 Dec 2016
In reply to nb:

> In reply to jon

> I'll do a bit of diy testing over the weekend and let you know what happens!

Be careful!
Airtime! 12 Dec 2016
In reply to UKC Articles:

Really well written and informative. Good article.

This topic has been archived, and won't accept reply postings.