Soapbox

As I hate to abseil at the best of times I tend to be fairly keen on safety.
I read this article, and am concerned about the use of a larks footed sling used to attach the belay plate. Dynema (the material used for most slings these days) has a low melting point, and due to the bouncy nature of nervous abseilers, there is a real risk of melting the knot due to the friction created as the knot comes tight. There has always been warnings about connecting nylon to dynema using a larks foot in a dynamic envoriment for this reason.
I've not been on one of these "best practise" courses, but surely this method can't be the recommended one.... can it?

I normally connect my belay plate straight to my belay loop directly with a short prussik underneath attached to a leg loop (so short that it can't get caught in the belay plate.)

Daniel,

until recently I would have agreed with you, I always attached my belay 'krab directly to the harness loop. Recently, though, I attended a climbers rope course at PYB and the method described by Jules was the one preferred by our instructor. This is also appearing in current literature on the subject and I think the aim is to get the belay plate away from your body as this improves your control of the descent. The leg loop prussik thing is still considered best practice I think.

FH

Some years ago, on an Alpine crevasse rescue course, we were advised that when absieling, it is wise to use a short prussic attached by a screwgate crab to the rear of your harness, with the prussic knot attached below the friction device. While moving down the prussic knot is held to allow the rope to run. In the event of a rock/ice fall knocking you unconcious, the hold on the prussic knot is released and the knot crabs the rope and stops the decent.
At the time we were using either a simple figure eight decender or a crab with an Italian Hitch. With a complete set of rock climbing gear this may be unnecessary, but when hut-to-hut touring in the alps, you need to keep the weight as low as possible, and this system seemed to be very effective.

There was an interesting debate about exactly this on UKC recently. www.ukclimbing.com/forums/t.php?t=177687

I don't usually extend the sling if it's a straightforward abseil because I don't think it makes a lit of difference. However, when abseiling with a figure of 8 it would be easier to see if the krab was cross-loaded and that's especially relevant because of the very low forces that it takes for a fig 8 to twist over the gate and break the sheath by leverage. That's also a good reason for using a larksfoot instead of a krab at the lower end.

I think it's a mistake to present an article on 'advanced' technique before presenting one on the basics.

The points on not bouncing, knotting the end of the rope and adding extra friction are perfectly valid (though I'd suggest the first two points are 'abseiling basics').

i) it doesn't make any attempt to explain how the belay device is used to control the rate of descent, which is the fundamental thing you're doing in an abseil. (Okay, this _is_ basic stuff.)

ii) I'm not desperately happy at the use of an overhand knot; they can roll, leaving the belay device out of reach, which might be a problem if it jams.

iii) no mention is made of an autobloc to hold the dead rope in the event that the abseiler lets go (either injured or a novice error).


I don't actually have too much of a problem with the cow's tail; I'm fairly happy with that. Let's look at some numbers:

Assume the climber weighs 100kg. Assume, in bouncing, he manages to achieve an acceleration of 2g. This gives a force of 2kN.

If the knot pulls tight, let's say it slips 1cm. Thus, the energy acquired by the knot is F*d = 2000 * 0.01 = 20J. Since the knot will only tighten once, this is a one-off input.

A 300mm 15mm dyneema sling weighs 20g. I'd guess the length of the knot is 30mm, so weighs 2g. The specific heat capacity of polyethylene (dyneema) is surprisingly high, at 2kJ/kg/K, or 2J/g/K, so the expected temperature rise is thus

energy/(mass*SHC) = 20/(2*2) = 5K

i.e. a 5 degree celcius rise in temperature.

Feel free to pick holes in this 'back-of-an-envelope' calculation.

I've just noticed that what I said above about using a larksfoot rather than a krab at the bottom end doesn't make sense because the fig 8 can't lever over it in that position anyway. Of course a krab at the bottom end of an extender could be cross-loaded without being noticed, but without the leverage the forces won't be enough on their own to cause failure.
My other point about extending to prevent leverage still stands though.

Captain;

I agree with a lot of your points, but not the one about the overhand knot; at least not in the way described. I'm assuming you mean the way it's used to make an isolation loop to the larkfoot?

You'd have to really pull the legs of the knot apart hard transversely to achieve a roll, and even so, the overhand is far less prone to this than the fig 8. In the situation described, slippage would have non-serious consequenses even if it was several inches.

> You'd have to really pull the legs of the knot apart hard transversely to achieve a roll

Doesn't slipping a krab into the loop do just that? Since the two sides of the sling are around the krab, they must be at a tangent to the krab, and thus are loaded at almost 180 degrees.

I must admit that I almost removed the comment, and did say I wasn't desperately happy. I'm also not desperately unhappy... I just don't like overhands, and I'm ashamed to say that it's probably irrational...

I'm happy to bow to your superior knowledge, and take the point back. Just shows how much care you need to take when preparing articles or posts on the subject...

I can see what you're getting at and it's a perfectly valid point, but I think in practise the force will be just to one side of the knot and the other side will remain in a fairly true position. I've sometimes wondered if it would be possible to roll a fig 8 tied in to a harness if you had slings clipped to the loop as anchors, as it seems to me that in some circumstances that could create a transverse load on the knot if the leader fell. But I've decided that it's so improbable as to be not worth worrying about.

Please don't bow to me! And please carry on thinking outside the box. It's good to discuss these things.

That UK Climbing post was actually started by yours truly, Alison! ;-) And it provides some useful answers.

To precis, best-practive seems to suggest that because the lark's foot should be pulled tight around your harnesses strong points before starting the abseil, and henceforth subjected to a fairly uniform load, the danger of 'sawing' is minimal. And adding further links to the system in the shape of an extra krab would only increase the number of potential weak links.

As for extending the belay plate, I prefer this option as it leaves plenty of room for the 'autoblock' prusik*, without it getting tangled up in the plate.

Current wisdom also advises that prusiks are attached to the belay loop rather than leg loop, as the latter option can lead to inversion when wearing a heavy pack.

(* IGNORE the fools who say prusiks are a waste of time - they take 30 seconds to set up and the people who don't use them are often the same ones who end up dead at the bottom of the cliff. So there's an element of personal choice in how safely you climb? Try explaining to your dead mate's mother that your friend was happy to gamble)

Oh yeah, one other thing I meant to say... Knotting the end of the rope is sensible, but I would recommend knotting each strand separately. Tying the two together increases the possibility of the rope catching on something on the way down, in my experience.

All great tips in general, though. Just remember that not all people on this website have even climbed before - more basic info might be a sensible idea.

The only problem with the overhand knot is untying it after its been loaded, tying knots in Dynema is ok it will weaken it much more than nylon though but not enough to worry us.
I think it’s a bit of an old wifes tale that it can snap the sling when loaded due to heat generated by friction, anyway most Dynema climbing slings are 50% nylon.
As far as the setup is concerned for me I would not abseil with out a prusik below the abseil device, its certainly much easier to operate the device if you extend it with a short sling but in reality I usually just clip straight into my harness loop. I always have a Purcell Prusik on my harness instead of a cows tail
as I find this more flexible and in the case of dynema sling its more dynamic if you fall on it.
My only other comment would be in the first pic they have the abseil rope looped through a rigging rope obviously if you were to retrieve the abseil rope you would damage the rigging rope.

I think Richard is right about it being an old wives tale that it can snap the sling through friction; at least in the case in the article. I thing the concerns come from the fact that under some circumstances teh larksfoot can be very weak and this message gets over-simplified.
Years ago it used to be known as the "seamstresses knot" because to cut the thread they used to tie a larksfoot on a bar on the the back of a chair or similar and give a sharp tug. The thread would break at exactly the crossover point on the knot. Someone once tried to explain why this is in terms of opposing forces, but I forget the details now. That's why it's not recommended in certain situations; especially where a sudden force is likely to come on the knot. In the relatively low and constant forces described in the article I don't see it as an issue.

There will always be room for compromise between maximum safety at one end of the scale; and sufficient safety at the other. And also for calculated risk when people embrace that through choice or necessity.

Larks-foot a 120cm sling to the belay loop.

I must say that the use of a larks-foot to tie onto the belay loop does not fill me with confidence. I was taught and still teach new climbers to tie the rope onto the harness by threading the rope/ropes through the belay loop and belt. The same would apply if abseiling, the krab would be clipped over both.

The prussik is a 'Best Practice' issue as far as I am concerned

All in all I suppose it's a belt and braces thing

Very interesting to read your feedback on the article folks. Apologies it's taken me a while to respond; I posted the article and have been away since. I've certainly learnt from this article so thanks for your comments. I want to make the point that there will always be pros and cons of any method and I'm presenting ideas here. Being dogmatic regarding technique is rarely a good idea but I'll always be careful to describe techniques that are safe. If the articles generate some quality discussion and makes us think about what we do and why then that's a good thing. I have responded to all the points below and updates will be made early next week:

>> Daniel Sells' and Robbie Williams' point re. larks-foot dynema sling onto belay loop

Never seen a noticable increase in temperature at the larks-foot knot at the end of an abseil and do not believe that a larks-foot knot tightening during an abseil will generate enough heat to damage a dyneema sling.

Friction of nylon components on other nylon components is an important safety issue in climbing. The best example is if you were to set up a bottom rope using a sling as the anchor and run the rope through the sling at the top rather than through a krab. As you lowered the climber down to the ground there is a risk that the sling would melt through. We are not talking about that amount of friction here; as Captain P shows and I can demonstrate with similar maths if you want. Compared to other knots, a larks-foot is relatively weak but if your slings are in good condition (and they should be!) they are strong enough to use in this configuration with 'normal' i.e. 80-100kg loads. Plas y Brenin have taught this method for some time and I've used it for years.

(cont. below)

>> Captain Paranoia's points re. presenting an article on advanced technique before the basics:

Point taken Captain but I think that learning the fundamentals from an experienced friend or an instructor is important and won't be presenting an article on the basics of abseiling. Take anchor selection for example; one of the biggest risks in abseiling. Although there are books that describe how to test and use an anchor I still see inexperienced climbers/mountaineers selecting some awful anchors on training courses etc. To be honest, I wouldn't want to try to present these fundamental skills on the internet and I guess that's why I made clear at the start of the article: "The fundamentals of anchor selection, rigging the rope, abseil technique and backing up abseils are best learnt practically from an experienced friend or an instructor. However, assuming that you're happy with the basics, it might be worth having a think about some of this…".

i.) For the same reason I have not described how to control the brake rope etc.

ii.) I've never seen the overhand roll more than a cm up the sling in this situation. Shortening slings with overhand knots is accepted practice. When direct belaying we often shorten a sling with an overhand knot and clip the belay device (HMS krab/Italian hitch or auto-locking belay device e.g. Reverso) into the loop formed. I know you sort of re-tracted the point but it's important that I responded.

iii.) I would encourage using some method of backing-up your abseil in case you lose control of the brake rope. The prussik does get a passing mention when I'm describing setting up the abseil at the start. In fact, there is one on the rope in the first pic but my hand's covering it (you can see the krab attached to my leg loop). The problem is that I didn't want to have to get into a discussion about the relative merits of prussik loops/shunts/length and diameter of prussik loop/different types/discussion of whether to attach to leg loop or belay loop etc. so I chose not to discuss it. As I say at the beginning of the article "backing up abseils are best learnt practically..."

I will make it more clear at the start of the article what I am NOT discussing and also add that I do recommend using a prussik or similar device to back-up your abseil.

(cont. below)

In tree work we always back up an abseil with a prussik it's recommended practise and safe as houses.

>> Gorilla Pie re. knotting end of ropes

Yes - definitely agree with knotting the ends of the rope individually when abseiling on two ropes. Allows any twists to spin out of the ropes at the bottom. Will update the article as such.

>> Richard Gale's point re. untying the overhand knot after abseiling.

Yep - this can be a problem when using small diameter slings e.g. 8mm dyneema, when a sling is wet or you're wearing gloves. I use a 12mm diameter sling for this which is fairly easy to untie.

>> Richard Gale's point re. pulling the abseil rope through the static rigging rope anchor.

Since there is no load on the rope when it is retrieved through the rigging rope there's limited risk of either ropes being damaged as the abseil rope is pulled through. It's important to check the condition of a sling or length of rigging rope before you abseil off it but pulling an unloaded rope through after abseiling is not in itself an issue.

Captain P and Gorilla Pie - thanks for your general comments on the article. Will bear them in mind in future. Hope you all had a good weekend whatever you were up to.

Cheers,
Jules.

Re Robbie's point;

>I was taught and still teach new climbers to tie the rope onto the harness by threading the rope/ropes through the belay loop and belt. The same would apply if abseiling, the krab would be clipped over both.

The central loop of a harness is specifically designed by manufactureres to be safe when used for abseiling and unless you have a very unusual make of harness, the normal way to abseil is from the central loop alone; whether that is attached by a krab or a sling.
It is not usually necessary to clip the krab through the belt or through the leg-loop attachment when abseiling, and on some models of harness this could result in three-way loading on the krab or pull it into the transverse position.