You might be on to something.

Jay

Assuming that these configurations are equally likely, the probability of both slings becoming configured in the failure configuration is just 1 in 4.

I just pulled one of those off of a friends sling yesterday. She had it installed correctly, but wanted to use the biner in the end for something else. She pulled it out leaving the String on the runner. I could see how you might clip into that incorrectly upon a later date.

I agree that it is unsatisfying to suppose that each mis-configured sling an independent event with a 1-in-1000 chance of happening. But then again, a systematic error alone doesn't completely explain the accident either:

I suspect that most climbers, usually, test their anchor system with more than enough weight to break a Petzl STRING (Petzl appears to capitalize the product name) in this configuration, before committing to it.

Even allowing for the possibility of some systematic error on her part...

I think that one also needs to also suppose that Karen adopted her flawed routine recently enough that she had not yet discovered this possibility.

Also, although it is a bit of an aside, if one wishes to apply more precise statistics, it is worth noting that repeatedly doing something that is unsafe in a way that allows a single anchor sling to fail like this with probability p, gives a probability of p/(2-p) that both STRINGs break, when at least one of them finally does break. (That's assuming that every time one weights the anchor, both slings are weighted enough that any mis-configured STRINGs do break.) So, even repeatedly clipping anchor bolts in a way that gives a 1/2 chance of clipping each bolt in an unsafe way (such as what Jay describes) will, 2 times out of 3, result in only one of the STRINGs breaking.

Never girth hitch or tie knots with those skinny dynema/spectra slings. They can act like a wire saw and cut through themselfs or nylon.

I may be missing something.

No, no one tests their connection to a sport anchor, ever. At most, you inspect the bolts, but you assume implicitly that the draws or slings you're connecting to the anchors are 100% reliable. The only way to test the slings would be to rely 100% on the belayer, but it's the other way around: you use your complete confidence in the slings (and the bolts) to test the belayer!

Knots in dynema slings typically reduce their peak strength by almost half. While it most likely wont kill you it is best to be avoided.


I'm sorry Jay but you do seem to make baiting comments with regularity. I most certainly do test my connections to sport anchors for body weight before going off belay.

There have been numerous death reported on this site when people fail to test their new safety attachment with body weight before removing their other safety. This happen alot in rappel accidents.

So do you also test your harness, your rope, and every carabiner on your rack before you rely on them, or do you actually assume that these elements of the protection system will work as advertised, and as they have every other time you've relied on them?

Jay

No, no one tests their connection to a sport anchor, ever.

Never untie from the rope, even when threading the shuts at the top.

No.

But I'm not sure why that is relevent.

Because her connector to the anchor fails. No one checks their connector to the anchor.

You can't check every element of the system. You have to assume that some element is the bombproof gold standard. The only way to check that the connectors are ok is to rely 100% on the belayer; but, in my opinion, the belayer is much more likely to fail than two independent connections to the anchor. So, by testing the anchor by reference to the belayer, you would actually be increasing the chances of an accident occurring.

Knots in dynema slings typically reduce their peak strength by almost half. While it most likely wont kill you it is best to be avoided.

I wholeheartedly disagree.

No, no one tests their connection to a sport anchor, ever. At most, you inspect the bolts, but you assume implicitly that the draws or slings you're connecting to the anchors are 100% reliable. The only way to test the slings would be to rely 100% on the belayer, but it's the other way around: you use your complete confidence in the slings (and the bolts) to test the belayer!


I think that sherpa79 has identified, as confirmed by ensonik, a highly plausible way that the slings could have become configured in this failure mode. The mechanism is systematic insofar as it involves clipping the "Stringed" biner into a gear loop, and random insofar as happening to catch a single strand of the sling while clipping.


I think that one needs to consider that Karen might have adopted a flawed routine recently, but the fact that—if sherpa79's mechanism (or one similar) is correct—there can be a large random element to the mechanism, could mean that Karen had been following the same practice for a long time. In sherpa79's mechanism, for instance, one could clip the Stringed biner to their gear loop many times without ever catching the correct single strand needed to cause the failure (and it is just one of the strands; catch the other, and you're fine. And moreover, whichever strand you catch, the result looks similar, so actually having caught the wrong strand before, and having experienced no problem, could lead one to shrug off the dangerous configuration.)


Maybe it's the beer I just drank, but I don't follow this at all. If the probability of making an error with either sling is p, and the errors are independent, then the probability of making an error with both slings is p^2.

Jay

No, no one tests their connection to a sport anchor, ever.

Jay, anyone who does not weight their attachment to the anchor before yelling Off and again weight their rappell before unclipping from the anchor is liveing on borrowed time.

Possible that you sport climbers get into a few bad habits due to the routine of yo yo ing?

strings and like devices place a serious flaw in the system when attached to any sling that isn't stitched shut like a dog bone.

I wholeheartedly disagree. Knots will decrease the strength of any sling. A dyneenma sling is rated to around 22 KN, a force I do NOT want to ever feel. In fact, half of that will probably cause your body damage. There is no reason to avoid using material other than 7mm nylon cord for applications such as anchors, which require bends and knots in the rope. People do this all the time, and it is safe. Feel free to back up your reasoning with something, I offer this

http://www.alpinist.com/doc/ALP18/newswire-dyneema-broken-sling-report

Although that really shouldn't be needed.