Knot Limited Sliding X - Failure Mode?

I've bee doing some reading on the great debate about sliding x anchor's, trying to draw my own conclusions about what I feel is safe.

Generally speaking a Sliding X with Limiting Knots seems to be well regarded as equalized, redundant, and with limited extension that, except potentially in some rare, severe cases, has an insignificant impact.

My question though, is what happens if the sliding X sling does fail. Some testing I found from Black Diamond shows the X with limiter knots fails at the knot at over 20 kN.

The force it holds is great! (with a new sling, I'm assuming.) However, it's the "Fails at the knot" that concerns me.

When the sling fails at the knot, is it still redundant? You basically have no tail on your overhand knot. Assuming a properly dressed knot, does the knot generally continue to hold, having been set by the fall, or will it unwind itself, causing total anchor failure?

Does anyone know of any study showing how this anchor setup fails? I haven't been able to find it.

If it fails at the knot I'm pretty confident the anchor will fail. Basically shock loading onto a flat overhand on some thin webbing with no tails, seems unlikely to hold.

That's generally my thought, but I'd like to understand if someone has actually tested the theory.

Brian L. wrote: Generally speaking a Sliding X with Limiting Knots seems to be be regarded as worthless.

There, corrected that for you.

Some actual testing that might answer your question

dmmclimbing.com/knowledge/s…

Almost any anchor will fail at the knot at that's the weakest point. If it fails at 20Kn then it doesn't really matter because you wouldn't survive the fall if it caught you anyway.

Dan Bachen wrote:Some actual testing that might answer your question dmmclimbing.com/knowledge/s…

Thanks, I have seen that, unfortunately they aren't looking at a sliding x with limiter knots, but a regular siding x, and aren't testing to failure.

As far as extenion goes, and impact force, they are lookig at a static system without a dynamic element in it (the rope). One of those rare, severe cases I refered to. FWIW testing done by John Long showed the dynamic element basically negated any impact from a short extension.

ajpowers wrote:Almost any anchor will fail at the knot at that's the weakest point. If it fails at 20Kn then it doesn't really matter because you wouldn't survive the fall if it caught you anyway.

That's true. This is more of an academic exercise, specifically targeted towards proving, or disproving the "is redundant" assumption most people have about this set up.

Although a compromised sling, perhaps due to aging/uv degradation that wasn't recognized and retired could fail at a lower margin.

A company called Rigging for Rescue did testing on sliding x's using slings. What they found is that when a sliding x is shock loaded from a sudden fall, the sliding x tends to lock and not provide much equalization - most of the force is on one of the anchors. However, if the sliding x was constructed of accessory cord, the equalization worked much better.

Nathanael wrote:If it fails at the knot I'm pretty confident the anchor will fail. Basically shock loading onto a flat overhand on some thin webbing with no tails, seems unlikely to hold.

When I was an EMT on a Search & Rescue team, our medical director said that "things start to break at 15 kN" - with "things" being bones, solid organs, etc. so what Nathaneal says is true.

Rigging for Rescue also noted the same thing. The standards in the rescue industry are designed to keep worst case scenarios of gear failures way below this level.

You don't happen to have a link to that article do you?

I have heard about the lock up issue, but have also heard, in addition to type of sling material/cord that biner choice can mitigate this as well ( using a round stock pear biner that slides easier). But I'd like to see what the study says itself concerning that, if anything.

Clipped into a single strand with a 12mm round stock karabiner the load split between the two anchors in a 90° Vee is:-

9mm rope 1.74:1
8mm cord 1.66:1
7mm cord 1.53:1
6mm cord 1.48:1
4mm cord 1.40:1
12mm Dyneema hybrid tape 1.48:1
10mm Dyneema 1.40:1
8mm Dyneema 1.39:1
16mm Nylon tape 1.67:1

If you use two karabiners, one in each strand without the X twist then the results are about the same, slightly higher but of no real significance.

Using a single karabiner as a sliding X then with 8mm cord the load imbalance rises from 1.66:1 to 1.89:1.
Cord has the regrettable tendency to twist under load during to it´s construction, in some of our testing especially with longer legs the sliding X locked up completely due to this.

We have performed over 800 pull and drop tests on all the popular belay building systems and see no reason ever to use a sliding X or any other type of "equalising" system. It is not taught in the UK or Germany.

Hi Jim,

Who is "we" your post refers to? Is that study published anywhere?

If I am reading this right you're basically claiming even an "ideal" equalized anchor doesn't equalize the load. How does this compare to an anchor built with a master point? Especially if the master point isn't ideally "equalized" (or the load vector is off the ideal pull vector).

Also, do you have data that compares the sliding x with the webbing single strand drops?

And, fwiw, what were the test conditions?

I believe the thought today is that "equalization" is almost, if not impossible, to achieve. There will always be more force/load going to one arm over another. If this is the case, distributing the force/load over multiple points so that a redundancy is achieved and no extension can happen are a better alternative.

I see a sliding X with limiter knots as basically an over-complicated setup that gains no advantage. A simple over-hand power-point setup is a better choice.

David Gibbs wrote:I see a sliding X with limiter knots as basically an over-complicated setup that gains no advantage. A simple over-hand power-point setup is a better choice.

Yep

David Gibbs wrote:I see a sliding X with limiter knots as basically an over-complicated setup that gains no advantage. A simple over-hand power-point setup is a better choice.

Don't take this the wronng way, but I'd prefer this thread not to de-evolve into a he said she said about what your preference is for a particular system. That's been hashed and rehashed too many times already.

Please keep discussion on topic to the OP, or provide data to back up your claims (like Jim provoded).

BigFeet wrote: I believe the thought today is that "equalization" is almost, if not impossible to achieve. There will always be more force/load going to one arm over another. If this is the case, distributing the force/load over multiple points so that a redundancy is achieved and no extension can happen are a better alternative.

What I'm really getting at with my question to Jim is even if the slidingg X doesn't perfectly equalize, how does it compare to systems that basically admit to not equalizing. If a master point anchor has a ratio of 3:1 on the bolts, or even more, maybe there is still a benefit. Especially in the case when the master point isn't loaded ideally. I dont have any data to make that judgement, but it sounds like Jim might.

Brian L. wrote: maybe there is still a benefit. Especially in the case when the master point isn't loaded ideally. I dont have any data to make that judgement, but it sounds like Jim might.

The benefit is no extension, and redundancy.

If you have good pro or bolts, does it matter that you have equalization? Load/force distribution is what both of the anchors we are discussing have, but only one has no extension.

Jim may be able to help out with the "data", but I don't see the need to look at it to make my judgement call on this one.

Yeah, but testing shows the small extension, with a dynamic element in the system (rope) has a negligible impact to force on the anchor. In that case isn't it better to distribute load, which has less chance of causing anchor failure in the first place?

Every test that shows high impact force on the anchor from extension is using a static system.

Isn't this the point where one would chime in with...."YER GONNA DIE!!"

Brian L. wrote:Yeah, but testing shows the small extension, with a dynamic element in the system (rope) has a negligible impact to force on the anchor...

Yeah, but can you not also build your anchor with a dynamic element and have a master point? Does it take up anymore time?

I do not know the specific numbers, but I would much rather have a small pendulum swing on one arm of an anchor than a downward, jolting, and instant change of direction in force/load. You be the judge, Chief.

I'm not really sure what you mean by saying the rope is also involved with a master point. The rope, in the case of a knot limited slding x basically eliminates the extension factor (ref: John Long in his book Climbing Anchors).

The rope doesn't make a non-equalized anchor equalized. So you're still putting all the force on one piece. And fwiw there is still swing/extension it's just not much.

Now, show me that a sliding x doesn't equalize any better than a master point, and maybe you have a point. But you need to back up opinion with fact.

Since it's been shown the extension isn't a factor, I'll go with the option less likely to fail in the first place. You can decide for yourself, chief.

Brian L. wrote:I'm not really sure what you mean by saying the rope is also involved with a master point.

Where in my post do you see the word "rope"?

Brian L wrote:The rope, in the case of a knot limited slding x basically eliminates the extension factor (ref: John Long in his book Climbing Anchors).

No, it does not. The limiting knots limit the extension, not eliminate it.

John Long's book... yeah, I've read it, along with many others that fill my book shelf.

Brian L wrote: The rope doesn't make a non-equalized anchor equalized. So you're still putting all the force on one piece. And fwiw there is still swing/extension it's just not much.

Correct and incorrect. As I stated, in an earlier post, you cannot achieve " equalization " - too many variables to contend with. The load/force is not all on just one piece but distributed across multiple points. You are wasting time trying to obtain equalization. At this point in time, and until someone comes up with a better way, the best you can do is distribute the force/load over multiple points to give yourself redundancy.

Brian L wrote: Now, show me that a sliding x doesn't equalize any better than a master point, and maybe you have a point.

The point is that you don't need to worry about equalization. Think about redundancy and solid pro.

Brian L wrote:But you need to back up opinion with fact.

You mean like you have been doing so far? You seem pretty sure of yourself already, so is there really a need to?

Brian L wrote: Since it's been shown the extension isn't a factor, I'll go with the option less likely to fail in the first place. You can decide for yourself, chief.

Then why even ask a question on a forum in the first place? I mean, this is where people's experience and knowledge are expressed. Wisdom, obtained through experience, can be quite useful. If you don't want other people's opinions then I would suggest you not ask questions and just lurk around.

What if your sliding x anchor gets chopped inside the limiter knots? Is it now less likely to fail over an anchor with a master point?