UKH

I occasionally use this knot as an alternative to Bunny's ears when rigging an abseil. The top two arms go to the anchors. Not sure if it has a name. I find it easier to adjust and it uses less rope than bunny's ears when the anchors are far apart. 

You tie it by forming a figure of eight on a bight and then partially undoing the knot. The knot looks very neat with everything pulling in the right direction, but is it compromised by the fact you have partially undone a tried and tested figure of eight, will it slip or roll if one arm fails, for example?

I should add I have used it many times without incident, so I guess I have convinced myself it is safe enough, but interested to hear thoughts.

Looks like it could quite easily invert. Why not just use an alpine butterfly?

(p.s. I'm all for devising and using new knots, I climbed for years on "Scott's locked bowline" bowline after figuring it out myself, then stopping for years as I thought it was silly to trust a knot that's not had extensive testing...) then a few years later I saw "Scott" had also worked out the same knot, named it and the Internet had widely tested it with videos etc.

So now I'm back on it! 

Looks like it could be very unsafe if the top rh strand happened to get cut/seriously damaged by rockfall. Have you tried tying it with just a very short stump top r, then clipping top l and bouncing around on it a bit at a safe height?

I'd not want to hang on that if the right hand anchor strand gets cut. Bit of bouncing and you're gone.

Or even more simply, just an overhand.

An important consideration is do other people know it.

At best... you'll occasionally get irritating people "correcting" your alpine butterfly and "helpfully" telling you how to tie it properly.

At worst... other people will be unfamiliar with your knot and not able to inspect it. Having other user/users of the rope inspect stuff as they go, is always useful extra pairs of eyes for safety.

Set it up at home with a bit of accessory cord and hang some weights on it. Cut the single, right hand strand and see what happens.

I'd stick with an alpine butterfly, but I think you're right - in practice the only real advantage of it over a simple overhand is that it's prettier.  (People say the AB is easier to untie than an overhand if it's been heavily loaded, but I'm not at all convinced that is actually the case.)

Regarding the OP, that's a knot that I would call a "directional figure-8" (though I can't remember where I got that name from), and it's my knot of choice for making the loop on the standing part that makes the main 'pulley' of a trucker's hitch.  Which is not a climbing thing, obvs.

It could well be fine for all I know, but this isn't something I'd use for climbing/abseiling/whatever.  And FWIW, I don't that "I've done this a hundred times now without incident" would be nearly sufficient testing to show that it is safe.  Safe is when it won't fail once in a hundred thousand uses or more, and I think it's a common heuristic trap to confuse "this is perfectly safe" with "I've been doing this a while and have got away with it so far".

Pretty sure I used this shoddily once to back up one stake to another to abseil. They very quickly 'self-equalised' as soon as my weight was over the edge and a bit of poo came out!

Isn't that an in-line figure of 8?

https://www.101knots.com/directional-figure-8.html

I used one last night actually, to rig a belay at the top of a climb.  It had two bits of gear, one piece a long way away so equalising it with a sling wasn't really possible.

Happy to be told that I shouldn't have but I did learn it from an instructional book.

What's on the end of the rope going out the bottom of the shot? If that's what you're abseiling on, then why aren't you simply going from top right to where your knot is, up to top left and back to the knot position, and simply tying a big overhand on the bight to give you a masterpoint with clip-in loop. Put an HMS on it and rig a releasable abseil using an italian hitch? Far better to prevent a big problem resulting from a jammed device than try to deal with it!

In terms of 'is the knot OK' - in a safe place, test it to beyond the maximum load* you could exert on it against a known 'good' method three times - once with both anchors in place, then again with left anchor out and again with right anchor out. It looks like it will be fine as the top loop locks all three lines coming out the top. It may well be really hard to undo compared to other knots using more strands though.

*equivalent of 2 people bouncing on it. I think a single aid climber can generate ~3kn (300kg) by bounce testing aid gear , so you'd be looking at 600kg.

I've just tried that and it does jam up without slipping through. Even if you poke the stump as you call it through to cause a further undoing step it still holds. However hardly a conclusive test and if the knot does pull through it's disaster. You could mitigate against this by tying an overhand in the LH loop which prevents that loop from detaching if the knot unravels due to the RH arm being cut near the knot.

However I think the general consensus so far is that people don't approve and there are safer methods which is useful feedback.

I suppose the fact that this knot isn't widely used tells you all you need to know. It does look neat though doesn't it?

Yes abseil rope at bottom. Understood what you said until you mentioned a releaseable abseil?

I assumed they meant attaching the abseil rope using a tied off Italian hitch so that it can be released if necessary.

Yes that's the chappy. Your diagram uses a different method to tie it than mine but the result is the same. I think we are being told to be careful as it isn't a particularly failure tolerant knot.

It looks pretty equivalent, though tied in a different way as nikoid described it and I'm struggling to figure out whether it ends up exactly the same or just very similar. In any case, the warnings on that website about how directional loading is important to the integrity of the knot would seem like a potential concern for its use in this context (if something else failed).

For this usage? Or you just learned the knot from a book and then adopted it for this purpose yourself? 

Even if it does turn out to be safe, which wouldn't astonish me, I'm not sure I see any particular advantage to it over just tying an overhand that's even simpler to tie and even simpler to adjust.

Which strikes me as quite weird advice in the context of setting up an abseil for personal use. Don't think I've ever seen it done outside of group abseils in a professional context (where there's normally a second rope that's the real safety line anyway). You'd need loads of spare rope to make it truly useful (which was exactly what Nikoid didn't have in the first place), stuck abseil devices should be pretty uncommon in the context of competent climbers and the option to release is only any use for the first of the two climbers down in any case (while there's still someone at the top to release it). But unless you've got enough slack behind the tie-off to lower the stuck climber to the ground, what help is it anyway?

I guess it's still a hazard worth considering but maybe the answer is to be good about carrying prusiks and knowing how to use them to get weight off the belay device rather than faffing with releasable abseils.

I thought it was a bit odd too but it's the only explanation I could think of.

Certainly wasn't disputing your assessment, that's how I read it too and I can't see how else it could be interpreted. Guess my post was really a reply to Toerag rather than you.

Rightly or wrongly, I've survived 20 years of climbing using about 5 knots (fig 8, clove hitch, overhand, bowline, prussic).     

I'm definitely a fan of keep it simple!

How are you making your prussic loops?  

You got me there Dave!

Ok, 6 knots.

Maybe 7 if I count the barrel knot (this is not going as I expected). 

I been climbing 56 years

Bowline, fishermans, stopper, overhand, clove, larksfoot, italian, various prussiks for me

If either the top right strand OR the loop got cut, the result would be the same, considering it's just one strand passing through the knot body.

As mentioned, there are much better and safer ways of creating anchors.

I totally agree with keeping it simple, I just couldn't resist pointing out the missing knot! The only other one I regularly use is the alpine butterfly.

Might help to be able to lower the stuck party a bit to a ledge, or something, to sort the problem out, but, otherwise, I agree with you - not worth the faff for personal abseiling and I only use it when instructing groups.

That's only 3 knots

That appealed to my pedantic nature

I've used this knot for years. I use it when setting up a fixed abseil (eg to get down to sea cliffs) where one of the anchors is a long way back from the other. Bunny ears would use too much rope and would be fiddly to adjust to the right length.  I can't now recall where I learned it, but it was suggested for this sort of situation in climbing. I'm surprised more people aren't familiar with it as it seems to be quite widely used.

I prefer it to the Alpine Butterfly because I find it easier to tie a long loop.  In order to get the correct angles and equalise it properly the knot sometime has to be quite a distance from the anchor and the loop may have to be quite long, sometime perhaps a couple of metres.   It is also easier to adjust than the AB.

This is usually set up at the top of the crag where, barring a meteorite strike, there is no risk of a strand being cut by stonefall. When used for abseiling there is no risk of loading the wrong end as it is naturally set up to use the correct strand for the ab, and the other ends are secured to anchors.  Even if one of the anchors were to fail this would not load the vulnerable strand.

I suppose it might be possible to tie the knot with the loop pointing down so the load would then come onto the wrong strand. However this is very obvious as the loop is then very clearly pointing in the wrong direction, away from the anchor.  It is very easy to see whether or not it is correctly tied.

It is not a substitute for the Alpine Butterfly, which in most situations is a more versatile knot, but I find it is very useful for this particular purpose.

Hi Luke,

Yes, for precisely this usage.  I didn't just see it in a random knot book and decide to use it.

I know you are right.  However, it looks nice.  It took me a while to learn it.  Not many other people know it.  And as my wife will tell you - I am a bit neurodiverse.

In this setup the ends of the rope coming out of a simple overhand are pulling in opposite directions. I suspect this weakens it considerably (although in practice the rope is not going to break under normal loads).  With the directional figure 8 the pull is in line, running through the knot rather than pulling it apart.  

I can somewhat see what you mean but it strikes me as fundamentally similar to the loading of the overhand when it's used to join abseil ropes, so I'm not sure I buy it as a disadvantage.

It's probably not an issue in practice, but the directional figure 8 is neater.

I'm curious - what book was it, and who's the author?

You just need enough slack to get them to somewhere where they can take their weight off their jammed device - a ledge, or somewhere they can put a bit of gear in, or a bolt they can clip. So, on something free-hanging all the way down, yes, you'd need loads of rope, but for anything else 6 feet is almost certainly enough.  And yes, it's only really useful for the first person down, but also not impossible that it couldn't be useful for the second.   The simple fact is that it is a sensible thing to do for virtually no negative aspect, just like arranging belays so you're not part of the system.

Yes, always carry a prusik or two (for all sorts of reasons) . However, putting a prusik on and using it when your head is held 2 inches away from your device because a bundle of your hair has gone through it and you're hanging onto the dead rope to stop things getting worse isn't going to be easy or possibly even possible! I've seen a girl with chin-length braids manage to get her hair stuck in her device once, it doesn't have to be long hair to cause trouble.

Fair enough. You've convinced me that it has more potential value than I thought.

Though I think this neglects the complexity cost. Most abseil setups are fundamentally pretty simple, so harder to get wrong and easier to see if you have. The addition of the tie-off definitely adds more places to make a catastrophic error and also increases the chance of missing one. Which is particularly critical with abseiling where you know for certain that you'll be relying entirely on the system. 

So I respect your reasoning, but I don't think I'll be applying it myself or suggesting it to others. Certainly not routinely.

It was this one https://www.falconguides.com/9780762790043/rock-climbing-the-amga-single-pi...

The manual for the Association of Mountain Guides of America.

You may be wondering why someone on a UK website is citing an American book.  I can't remember how I got it actually.  I expect it was a gift or something.  

Ta.  Not one I'd heard of before, nor the authors. 

Though I did find a John Long book very useful when I was first learning to place gear, and from your link there it looks like one of the authors might have co-written an updated version of that with him.  The version I had was quite evangelical about using a cordlette to build a belay, took me a while to figure out that it was almost always better for me to just use the ropes.