UKC

https://americanalpineclub.org/news/2024/9/11/the-prescriptionseptember

Reading through the full report is worth it, but I've been climbing over 30 years and always pretty switched on - at first reading all the magazines and then 25 years on UKC and other web forums, and I don't think I've ever heard of an accident where the failure mechanism was even suspected of being like this. Has anyone else?

It has been discussed before, but the report shows the amazing and sensitive reports that are done by the American Alpine Club into accidents, including fatal ones, that all climbers can learn from or at least understand better the risks we are all taking. I wish something similar was possible in the UK. I follow the Facebook accounts of the local to me Peak District mountain rescue teams, and from time to time see reports of climbing accidents that they have attended. Sometimes the report will hint at what happened, other times not. I understand this is not something the the MRTs should or have the capacity to do, but it's a shame we don't have follow up reports like the AAC ones to help understand what goes wrong and how we can protect ourselves and others. And I know about the BMC self reporting function but it's not really comparable. 

Take care all. 

In reply to TobyA:

The picture of the Petzl Spirit I found shows a raised "knuckly" on the rope end - this would in freak circumstances create a significant high pressure pinch point on the rope.  I recoilled at the idea of that shape (in ignorance of course as I'm not a designer.  It looks as if it is intended to keep the rope close to the long axis of the krab but like any irregularity it looks to me as if it could act unexpectedly.  I am just old fashioned and ignorant and speculating but I'd definitely avoid a krab with a "live/loaded" end pointy discontinuity like that.  It's also possible that I have seen a bad image that does not do the krab justice.

It's just as possible that the other side of the krab pressed the rope against an irregularity of the rock surface as well.

This is assuming the initial inferences leading to the conclusion in the OP reference are correct

In reply to TobyA:

The AAC annual accident books are interesting and sobering reading. I completely agree that something similar would be useful over here.

In reply to TobyA:

I think a lot of people in the UK have a very different perspective on death. I suspect it has something to do with traditional religious associations or something.

But the tendency is for everyone to say RIP (whatever that is supposed to mean! - you don't rest when you're dead), and move on without mentioning the incident again.

There is a strong message that if you mention details of the incident you are being disrespectful.

In reply to TobyA:

Having read the report I still can't visualise what they say happened. Perhaps someone could recreate the circumstances on a test rig because if there is a danger it would be good to understand what it is.

In reply to BruceM:

Publishing details is something that should really only happen with the consent of the friends and family, but we should be trying to shift our culture so this becomes more normal. It's really valuable to discuss incidents like this.

For the avoidance of doubt, if I die in a climbing accident I'd like everyone to be discussing it afterwards. At least someone else could learn something from my mistakes.

In reply to TobyA:

I think the title is rather misleading because the Krab didn't cut the rope the excessive force did. (if I understand what happened correctly)

In reply to DubyaJamesDubya:

I thought the proposed explanation was that the rope forces "clamped" the carabiner down hard against the slab, pinching the rope tight.

With the rope no longer free flowing at the crab, it became a fixed anchor point, and potentially a sharp one due to the shape of the carabiner profile.

This sharp fixed-anchor cut the rope.

Scary.

In reply to TobyA:

I highly doubt the Karabiner had anything to do with the rope breaking, just acted as a pinch point. my finger of suspicion would lie with the condition of the rope and that it was subject to a heavy fall over a sharp edge 

In reply to DubyaJamesDubya:

Reading the analysis it sounds like the rope was pinched between the carabiner and the rock and cut at that point. Even a factor two fall onto the rope shouldn't have generated enough force to break it (11ish kN as measured by hownot2). They also say that all of the bundles of core strands were chopped at the same length, as opposed to different lengths as you'd expect if it pulled over an edge.

I wonder if the sheath towards the end of the rope had excessive wear?

In reply to pancakeandchips:

It read to me like the karabiner may have been pinching the rope at a point precisely at the lip of an overhang or overlap, which I suspect may have been reponsible not just for the pinch effect to have been more pronounced (than would be expected against a flat wall) but potentially also for the introduction of a sharp rock edge right at the point of load against the krab.

I'm often aware of the danger of side-loading a karabiner (when the quickdraw lays down a slabby section of rock but the bottom krab projects into space half over an overlap) but I'd never considered the potential for the rope to get trapped right at the point of the overlap.

In reply to TobyA:

I have probably failed to understand something but is there a chance that they are saying that the rope cut the rope? I have heard this mentioned vaguely in warnings about climbing using ropes as "twins" (two ropes through every QD)

In reply to john arran:

Ditto. Seems like the takeaway is to extend over the edge* and for those equipping bolted routes to be mindful of this failure mode.

*Though one can quickly imagine a scenario where a long sling arcs along a sharp edge with equally dire results…

In reply to JLS:

Except that with a longer sling the rope would no longer get pinched, the fall would no longer be effectively factor 2, and the peak load would be very much lower.

In reply to TobyA:

The way I read this is that the proximity of the karabiner to the edge was decisive. If it had been laying on the slab, away from the edge, then more of the force on it would have been parallel to the surface of the slab, with less downward force to trap the belay strand between the krab and the slab - possibly resulting in a different outcome.

Am I understanding this right ?

A tragic incident. Twenty four is awfully young.

In reply to wercat:

The report describes the rope as being pinched underneath the carabiner so those circumstances seem a possibility. However, Rob Chisnall’s note talks about the implications of T or H cross-section carabiners in general rather than the knuckly on the Spirit. Here’s a photo of a Spirit from Petzl’s site:

In reply to john arran:

Indeed, but I’d still have thought there was still a risk cutting the sling in a particular adverse scenario. I find it hard to say whether this would be less of a risk than this “pinch effect” reoccurring.

In reply to TobyA:

What a horrible, horrible story.

In reply to TobyA:

In days gone by, before mechanical devices, alpine clutches were often used to secure lines etc... which generally do exactly what many people are saying was the cause. I've dropped test on some with low stretch rope in the mid 90s. Braided teyrelene, which was half the strength of modern ropes, but never had a failure.

It's tough to speculate, but I will, I suspect it's rock, edge and rope position. Anyone who has rigged caving will naturally always look for clean lines, thinking about rope and karabiner position once loaded. 

In reply to Blue Straggler:

No, single rope. To my reading it clearly said that the krab was the cutting edge and they have tried to understand how that happened when normally it doesn't.

In reply to BrumClimber:

"No rock edge was involved in cutting the rope, and no rope sheath material was observed on the rock."

In reply to TobyA:

Something doesn't quite add up in the description. What they're describing is not a cutting action but snapping due to the forces involved.

In reply to TobyA:

 I'm a little wary of how definitive that statement from the report is, and some of the others, given the level of analysis that's been done and the expertise of the person doing it. As far as I can make out, the main report writer is just a normal climbing instructor who happens to work in that area and went along to take a look. No disrespect to the guy, I'm sure he's a great instructor and a very experienced climber who would have immensely valuable advice on climbing in general. But there's no way he or any other normal instructor has significant experience of accidents like this because accidents like this are vanishingly rare. So with the limited evidence he's got after the fact from speaking to a single traumatised witness and seeing the few bits of gear left in the rock, how can it possibly be justified to say that a rock edge definitely wasn't involved? I'm not saying I know better, that would be absurd. What he's come up with might well be a plausible hypothesis, I just don't think it can be considered any more certain than that.

In reply to Luke90:

When you hear hoof beats coming up behind you, don't assume it's a zebra.

In reply to pancakeandchips:

What if those learned journalists from the daily mail decide to launch a discussion about your accident? 

In reply to BruceM:

I find it interesting that if someone dies climbing in the US they are willing to publicly dig into the causes of it, to help prevent future occurrences, but if someone is shot it's just "forts and pears"

In reply to DaveHK:

Well, Luke's point – if I understand it correctly – might be that the report wasn't exactly up to the DAV's or ÖEAV's Safety Committee industry standards of investigation.

The cut rope was sent to some random youtuber instead of a lab, after all.

In reply to Frank R.:

Yes, I was agreeing with him. My reference to zebras was directed at the guy who came up with the cutting explanation.

In reply to Frank R.:

This was tragic - just 24 years old. I suspect it will be forever impossible to fully account for how a rope could snap like that. There are three points I will make though:

1) uK double rope technique, particularly on trad is intended, partly at least, to minimise/eliminate the risk of something like this happening.

2) Falling off, particularly on easier trad, should still not be considered as a sensible 1st choice for retreat. There are always going to be a lot of unpredictable forces and hazards involved. On easier trad surely attempting to downclimb is a better option, even if it does finally result in a fall?

3)I don't suppose the decline of the use of nuts - a low-cost option to facilitate retreat - helps with this mindset either.

In reply to Rob Exile Ward:

I must say, that was my first thought but perhaps he felt he couldn't downclimb. 

In reply to TobyA:

Thanks Toby, I see now that my post was not clear. I know this was a single rope incident, I just meant I vaguely remember some talk about how rope moving across rope under tension can cut right through. 

In reply to Blue Straggler:

Wouldn‘t the ends of the cut strands show melt signs in this case?

In reply to Frank R.:

He‘s not exactly random; he has close to 300k subscribers, works closely with Mammut, has access to their testing equipment, and has apparently never had issues with any of the manufacturers whose gear he tests.

In reply to McHeath:

You’re thinking of a different YouTube guy.

In reply to TobyA:

A quick image to show what I understand happened. 

The climber fell a fair distance and the rope flowed through the carabiner below. When the climber passed the carabiner, the rope dragged over the carabiner and seems to have trapped the rope below the carabiner against the rock (and potentially the rope on the rope like guide mode, but I fine this unlikely). 

Because the fall could have been 20ft and the 'pinch' could have caught 5ft below the carabiner, it's potentially a factor 4 fall (and possibly more). 

The sheath would be pinched first and the core strands may have pulled through some before the rope was finally 'fixed'. As the rope cut, the strands would have 'sucked' back into the sheath. The clean cut would basically be from a narrow edge acting as the point at which the rope snapped (hence shotgun noise)

In reply to McHeath:

A few 'ifs', but if this was a uk death, doubtful cause then the coroner could seek an expert witness. It's a near certainty they'd then return the incident site, with new equipment and try to replicate the incident based on statements from those present ie other climbers and the rescuers. You'd then test a few probable set ups (as you'd never know for certain), using a dummy etc.. and or use indoor test facilities and drop weights over edges with the karabiner in various positions. Lyon Equipment's training set up at Shap could be used in the uk for example. There are likely similar USA places. Ideally when it's a known death, friends and MR shouldn't mess with any equipment too much, trying to preserve any clues as much as possible, mark how much rope was paid out at the time, plastic bag the karabiners in question, photograph the rock area immediately. Obviously this is in an ideal world, but it shouldn't really be treated any differently to say a car crash on a motorway, gather the evidence safely and quickly, whilst keeping those doing it safe. 

Imho ropes are surprisingly hard to break, to cut through a rope with a rope, needs quite a bit of work/load/time see-sawing etc.   Whilst there is stress/strain/compression forcing a rope under load around a small radius again ropes aren't just cut or severed easily. The easiest way to cut a rope in the hills is if it's secured just after a sharp edge(a few cm) so there won't be any stretch and the same point is loaded and absorbs all the impact.(think how easy it is to cut a rope with a knife under load compared to just holding it) If you secure a load a couple metres back or up from an edge, as you load it the rope stretches, so the same point isn't continuously on the edge as it slips over it.

In reply to DubyaJamesDubya:

The combination of the rope, rock and krab pinched the rope like a buckle, isolating a very short piece of rope, increasing the arresting force, pinching and acutely bending a very highly loaded piece of rope which failed as a result.

I've seen gear positioned/loaded like that and been wary of it but never figured it'd actually snap. Grim.

Jk

In reply to timparkin:

I think your probably quite close to my thinking, any noise could be just from a whip like crack after it broke and contracted. 

In reply to 

Could this sort of pinching potentially being more likely with an extended alpine draw? It's often quite hard to get the orientation of these correct as they might have twists that you don't notice or don't bother to correct.

In reply to McHeath:

Having 300K subscribers makes you more qualified than an official forensic expert, accredited UIAA labs or a DAV safety committee engineer since when???

Those actual experts handled incidents and accidents for decades, worked with the police (in Germany, where the plods take accidents resulting in deaths pretty seriously), textile and materials science university departments, the manufacturers. Heck, most of them basically wrote the safety norms and tests we all take for granted now.

They know how to analyse micrographs of cut, abraded, torn or melted ropes, because they had all done it before. They have done elemental mapping imagery before to find traces of chemical contamination from acids. All of that while still learning new things, as the field isn't really fixed.

And even the actual experts show how hard can it be to test something properly – remember the early 2002 UIAA/EN norm for edge‑cut‑resistant ropes? And that it was shelved just three years later, when the test protocol turned out to not be reliably replicable amongst different labs?

Because the tiny, micron differences between the test edges meant the same rope would pass the edge test at one lab, and fail at another? Microscopic differences they thought wouldn't matter, as the test edges were pretty expensively machined to quite the precision?

Well, this youtuber is building his own abrasion test equipment from random abrasive things bought on the internet, and I am totally seeing him doing "perfectly scientific" edge‑cut rope drop tests probably next year, ignoring all the technical difficulties of such. Guess what, that's potentially 300K subscribers following bad science and bad engineering.

This youtube guy had already been proven wrong on some of his more controversial videos, IIRC. Can't be arsed to delve through all his obnoxious vids to find it again, sorry, but it was something about fall factors or whatever. His conclusion did contradict much more extensive UIAA and French tests, if I remember it well.

No thanks. The simple pull test he did there? Perhaps, that might be still up to his level. If he's following the protocols, calibrating his tools, et cetera. Yet a proper lab could likely do much better, and he basically thrashed some of the evidence for views. That's what I find wrong. Even if he still got his pull test gear calibrated and protocols set‑up properly, which is quite a moot point.

Don't get me wrong – "citizen science" can be great, but there are limits to it. There was a youtuber who built his own XRF, and even calibrated it as much as possible and included the proper radiation shielding. A fun project, I guess. It did work, actually. Yet I probably wouldn't go to him if I wanted to check for chemical rope damage by elemental mapping or imaging, or check some import cinnamon for signs of lead chromate adulteration.

There was a youtuber who tested old photographic lenses which used thorium glass back then for radioactivity with a Geiger. Nice. Yet I wouldn't go to him if I wanted to assess to total Bq activity of such a lens, as that's the job for a nationally accredited lab with very expensive gamma spectrometry equipment and equations to properly estimate the total activity from the spectra.

Guess what, the accredited lab doing the hard work found quite different results than the amateurs on youtube using their cheap Geigers, with some of the youtubers' conclusions totally wrong and potentially dangerous.

Perhaps the worst example of all was the recent DIY "biochemist" group promising an easy way to synthesise expensive medications on the cheap on your kitchen table, using a few 3D‑printed parts and some sous vide pots for reactors. Like meth cookers, they completely ignored all the issues of separating enantiomers, purification, testing of the product (no LC‑MS or NMR there at all!), or even the source and purity of the reagents and precursors they wanted to use (they were not synthesising complex molecules from just the basic calcium carbide feedstock stuff and advancing from there, obviously). Totally crazy, yet it got a lot of views.

Sorry, but any youtuber who's mainly just promoting themselves and their merch – without being really meticulous about their scientific or testing protocols right from the start of their channel – nowadays simply gets the smirk treatment and my perhaps slightly boomer "a random youtuber" monicker from me...

In reply to ExiledScot:

It's the US of A, remember. IIRC the coroners there are simply elected without any forensic qualification required, unpaid and often just the local county's funeral home operators, without any forensic expertise needed whatsoever.

You are thinking UK, or maybe even Germany level of investigation. Yes, that would have been likely quite different in the expertise drawn into a wrongful death or accident investigation. Over there, unlikely. Unless the poor family had enough money for a civil investigation, which doesn't seem to be the case.

In reply to TobyA:

i had an unusually hard catch once when i was doing some fall practice climbing brant direct to get a fixed line on one of the routes to the left. basically i had what happened in the report; the live rope pulled down the clip, the crab was right on a relatively flat piece of rock so my weight on the rope effectively pinned the carabiner to the rock, stopping much free rope from running through. rachel was belaying and initially i couldn't wrap my head around how shed caught me so hard. the gear was not extended, just a long draw, extending it would have done the trick though. im slightly more aware of the issue now! its unlikely but not that hard to recreate. very unusual to have it sever a rope though.... 

In reply to JLS:

You´re right, I was mixing him up with the "Hard is easy" guy who goes about his stuff a lot more scientifically.

In reply to Frank R.:

See above: I was confusing "Hownot2" with "Hard is easy".

Points taken though.

In reply to TobyA:

If I'm understanding this correctly, the climbers end of the 9.4mm rope became almost static due to being clamped close to where it entered the carabiner, resulting in the force of the fall being concentrated fully on a very tight bend in the now almost static rope where it crossed the carabiner.

This scenario doesn't sound too uncommon to me. 

In reply to TobyA: Is it clear if the edge believed to have cut the rope is on the inside of the krab, or on the side, squeeze cutting the rope against the rock?

In reply to wbo2:

No, that couldn't possibly be clear from the minimal evidence available.

In reply to timparkin:

I'm struggling to get my head around your maths there, could you try and re-explain. I can't understand how you can exceed factor 2. You can't fall more than twice the distance to your last runner can you (in terms of the FF calculation - not counting the stretch here which is only considered after)? Even if you fall onto and past the belay, the physical length of the rope is not long enough to go any further.

In reply to Holdtickler:

Fall factor is distance fallen divided by length of rope that holds the fall. If the rope feeds back through the runner as you fall, then pinches tight at the runner, the length that holds the fall can become very short.

In reply to timparkin:

The belayer also states that they took in two to four feet of slack during the fall which could also increase the Fall Factor.
As this is an open analysis of the accident, I'm just wondering if letting out slack as part of a soft catch could have prevented the accident as the rope would be in continuous motion. 

In reply to FactorXXX:

Yes I thought taking in probably made things worse. You want the rate of rope dropping back through the krab to be slower than the falling climber, which I think it would be if the krab was lying against the rock as described (and the belayer didn't take in). 

Incidents like this do make me wonder about the wisdom of "getting used to" falling - which I never have btw!

In reply to Blue Straggler:

This is the risk not with using ropes as twins, but if you have half ropes and are inconsistent about using them either as halfs or twins over the course of a route.

The risk is that in the event of a fall, the ropes will move at different speeds through each crab increasing the risk of one of the ropes burning through the other.

The recommendation is that over the course of a route, you use the ropes either as twins (both ropes through each piece) or as halves (one rope through each piece), and don't change your mind.

In reply to FactorXXX:

There would be enough rope in the system stretching, taking any slack etc that in almost all circumstances there would be some motion in the rope EXCEPT this is the occasional circumstance the rope is trapped, and would have no motion even with a soft catch.

I don't think is very relevant to getting used to falling on hard routes.  

In reply to Darkinbad:

ok, with you now. cheers for clearing that up for me.

In reply to TobyA:

That is scary.

So the crab pinched the rope on the overhang/slab preventing it moving freely and focused a factor 2 fall directly onto the now fixed bit of rope running over the raised ridge on the side of the crab cutting it?

I’d be interested in seeing that tested and understanding how sharp an edge needs to be to cut against force.

So sad.

Toby

In reply to Toby_W:

I think the consensus is the fall factor was greater than 2. The much higher forces generated means a relatively blunt edge (on the krab) can sever the rope. 

In reply to wbo2:

Falling off a VS is very different from falling off a steep 7a.

In reply to TobyA:FWIW I don't think the concept of fall factor would be relevant/helpful in this case . As the rope would be 'locked' at the karabiner it's theorised as a straight fall of a certain mass, on a mildly elastic cord onto a fixed point effectively acting as a blunt blade.

In reply to Toby_W:

I'm wondering if there is any way the sharp edges of the solid gate, where it is cut away at the base to form the hinge, could be involved. Under normal circumstances the shape of the krab keeps the rope from loading these. As unlikely as it sounds, those edges are sharper than the working edge in question.

In reply to planetmarshall:

Thank you! 

In reply to TobyA: A tragic incident. It seems their conclusion is that the rope broke on the krab but there is no mention of how new or worn the krab in question was or did I miss that? Anyway doesn’t seem like good news for Petzl. If I owned any Petzl Spirits I wouldn’t be using them for climbing anymore.

In reply to TobyA:

I think "analysis" is a bit of a misnomer here. "Speculation on a tragedy" might be more accurate.

Very sad for all those who knew the climber and disquieting for climbers in general.

In reply to dr evil:

Tight bends weaken ropes, we know that, it's why certain knots perform worse than others. Pretty much all quickdraw krabs have comparable bend radii where the rope sits. My understanding of the report is the combination of the high fall factor, high force, generated by the unlucky pinching isolating a short length of rope and the acute bend overloaded the rope in the bend. It's not a brand/model specific issue and I think 'cut' is misleading, 'locally weakened through bending' might be a more appropriate idea.

Jk

In reply to jkarran:

I'd be looking at how beat the end of the rope was if the guy was regularly falling on it as it broke six or seven feet out.

In reply to jimtitt:

It's a thought but sport ropes take a huge beating at the ends yet rope snap stories are rare enough to still catch our attention.

Jk

In reply to McHeath:

At risk of this thread about a tragic incident being sidetracked, I would like to offer some more positive coverage for the work Ryan does on his "how not 2" channel. He is the first to make it clear that he is not a scientist and not an engineer, but that the shear amount of testing he and his team has done, always using multiple samples should give a pretty good comparison in the general performance of different materials, gear and set ups. Where possible he tests using both dynamic and pull test setups and focuses any analysis on real world use cases. His background in slack lining may give a slightly different viewpoint on some matters, given the very high direction specific forces involved.

I this case it may have been prudent to not send him the whole sample of incident rope in case any other labs wanted to test it but, I can't remember if he covers this in his video. 

There was a long thread discussing this accident on MP.  If you are interested in reading more you can find that thread here: 

https://www.mountainproject.com/forum/topic/124721273/seneca-rocks-accident...

In addition to being tragic for everybody involved, especially the belayer and the family of the victim, it was also a strange and therefore frightening; perhaps that's why so many people are still coming to terms with what took place.  

For my part, I think the explanation proposed in the report makes sense; a very rare but possible combination of factors lead to the rope being pinched under and cut by the carabiner.  If there's a take away, it's perhaps to consider double ropes, or to think more closely about where and how  carabiners sit against the rock, bearing in mind that the orientation of all the components can likely will change under load.  Of course, maybe there's a different explanation.  

Stepping back from this particular accident, I think the question of how to discuss accidents in important.  

In my view, there is no "correct" way to discuss accidents; there are just different perspectives.

Some people feel very strongly that all analysis, discussion, or speculation should be left to the professionals and limited to the family of victims.  Others see great value in talking about things publicly, sharing opinions, analysing details, and so on.  

The tension between these viewpoints is often seen on the "Accidents and injuries" forum on MP.  That forum is exists to discuss and learn from accidents, but there are always voices pushing back against these goals. In the last few years, analysis usually takes place in the "Accidents and injuries forum," while celebrating the lives of victims takes place in the "Memorial" forums.  There is a place for both.  

I've noticed that on UKC there seems to be less willingness to discuss  accidents publicly; as soon as people begin to analyse accidents, the conversation is steered away from speculation, or the threads are closed.  I don't know if this reflects a particular approach to risk and safety in the UK, or a greater respect for privacy; perhaps the answer is combination of both. I'm not judging--this is just an observation, and perhaps I'm not correct. 

Personally (and once again this is only my opinion) I think discussing accidents publicly can be extremely beneficial for many reasons.  We can learn from accidents, discussions help people process fear and uncertainty, and they encourage a culture of transparency and openess, something which is often lacking in climbing. 

Here's another relevant thread from MP, which discusses mistakes and near-misses that could have lead to serious accidents.  I think discussion like this are really important for all climbers, young and old. 

https://www.mountainproject.com/forum/topic/126203278/scariest-momentsmista...

Anyway, the discussion about how to discuss accidents is important. 

Please discuss!

In reply to brunoschull:

It's this interpretation that doesn't make sense to me. I don't think you could describe anything being "cut" except by a sharp edge, and unless a krab has been notably worn with use, I don't think the radius at any part of any standard karabiner could ever be described as a sharp edge.

Of course, it's possible for a rope to "snap" under a particularly heavy load, and the smaller the radius of the krab, the lower the load at which this would be expected, but that's not the way it's described and it isn't what is meant by a rope being "cut".

If the rope was "cut" at all, I would think it would have to have been by a sharp bit of rock that the krab very unluckily happened to pinch the rope against. Wouldn't have needed to have been more than a tiny sharp pebble in exactly the wrong place.

The reason this is important is that the wording as quoted above suggests that the design of this, and potentially other, karabiners might be partly or wholly responsible for a risk of ropes being cut in the event of some falls. I don't see any justification for this as a possibiity and therefore I don't think the use of the phrase "cut by" is justified in relation to the karabiner. Indeed I think its use has potential to mislead climbers into believing their gear may not be safe when there seems no reason at all to draw that inference from the information presented.

In reply to john arran:

the edges of those cutout lightweight karabiners look pretty sharp to me with a radius of curvature of approx 1mm?

In reply to john arran:

There is some merit to the discussion – rope‑worn carabiners with grooves are definitely known to cause rope cuts in a fall, just like any other sharp edge (even if the carabiner itself almost never breaks in that grooved spot, unlike some people feared). Though that doesn't seem to be the case here.

And thinner diameter carabiners might do slightly more damage to ropes in frequent falls, as written by DAV and UIAA safety committee ex‑president Pit Schubert:

Ropes face another danger. As the weight of carabiners is reduced, their body diameter also decreases. The smallest diameters in lightweight carabiners today are around 7.5 mm. Frequent load stress from falls can be detrimental to the ropes.

After all, the standard for ropes is designed for a diameter of 10 mm. Only heavier carabiners have a body diameter of this size.

Despite the smaller diameter of lightweight carabiners, the rope cannot break, but with each fall‑related load, the rope is slightly damaged. This damage can become significant when under strain over a sharp edge.

Anyone who falls frequently should ensure that the carabiner body has a sufficient diameter.

Mind you, that's only my own fast and loose translation from German from a pretty early edition of his book Safety on the Ice and Rock. And he still stresses that even while repeated falls on a thinner diameter carabiner might do more damage to the rope, it still shouldn't cut through under any normal circumstances. More stiff? Yes. Sheath damage? Possibly. Core cut? Unlikely.

And he explicitly says that a rope cut should be still nearly impossible just on the carabiner, but if the rope was already damaged and pinched at that exact spot, potentially with a sharp mineral crystal in the mix, it could have been a very rare Swiss cheese type accident where all the things that could go wrong went wrong.

Unfortunately, we won't ever know, as no proper incident analysis had been done. So I quite agree with your view that we shouldn't just jump to conclusions.

In reply to TobyA:

I've had a lot of discussions about this.  I don't buy the analysis that the carabiner pinched the rope causing a factor-2 (or maybe factor 2+, more about that in moment) and the rope broke over the carabiner.  I think you can do factor-2 drops over a carabiner all day and the rope won't break.

The particular climb the party was on is just loaded with sharp edges. I posted a picture of the spot where the accident happened in the Mountain Project thread referenced by others above, and the opportunities for cutting are evident.  The analysis rejects breaking over a sharp edge because no rope fibers were observed on the rock and some rope fibers were observed on the carabiner.  I don't think such observations are even remotely decisive; there's a host of reasons why such rope fibers might have simply blown away or perhaps the observers didn't happen to look in the right place. 

There's also the possibility of chemical damage.  I know someone who had a rope break on an upper-belayed fall and the analysis determined that chemical damage was involved.  The owner had no knowledge of any occasion when that could have happened, but throwing your ropes down in parking lot might be all it takes.

The question about whether the fall could have been greater than factor 2 has been raised.  The answer is yes, considering that the belayer pulled in some slack. But you have to assume that the carabiner-pinching mechanism took all the rope between the protection and the belayer out of the picture.  To see how this can happen, suppose the climber falls from 5 feet above the pro, the belayer hauls in 2 feet of slack, and the carabiner pinches off the rope so that only the rope from carabiner to falling leader is available for energy absorbtion.  In this scenario, the climber falls 8 feet (not 10 feet) but because of the pulled-in slack there is just 3 feet of rope available.  In this case we have a fall factor of H/L=8/3=2.7.

In reply to Thugitty Jugitty:

That's the bit I picked up on - the shape of the cross-section.

Another variant of the Spirit (here: https://www.sierra.com/petzl-spirit-carabiner-straight-gate~p~6845n/) shows it with a fairly triangular cross section, which "could" have been enough of an edge to cut the rope.

In reply to brunoschull:

You misunderstand the concerns here. Speculation just after a fatal or highly serious accident has happened often has poor information quality and can really hurt relatives and friends who are struggling enough with the immediate shock; in such situations it's best to await better information.  The internet response immediately following such tragic but unclear events is a human one but it's effectively rubberknecking.

I don't know anyone here who doesn't favour proper accident analysis and few who object to discussion of that analysis (I'd ask how else do those with misunderstandings or incorrect ideas learn?). I also don’t know of anyone here objecting to climbers who post about their own accidents to inform others.

You can judge who is serious about accident analysis by those who have at least some of the many excellent written volumes on their bookshelves. For the many who claim to care online it's odd such sources are always small print runs.

rgold's perspective is always appreciated and highly valuable; he has a deep wealth of knowledge and experience.  It's certaintly true that we don't know exactly what happened, and it always makes sense to consider more common things first (like the rope cutting on a sharp edge of rock) before entertaining more unlikely things (that the rope was cut when it was pinched under the carabiner).

That said, I don't think we should dismiss the accident report created by people, also highly knowledgable and experienced, who went to the site and inspected the equipment carefully, talked to the belayer and other witnesses, and so on. 

The scenario they describe, while unlikely, seems plausible to me.

I think it's important to emphasize that the proposed mechanism does not simply involve, "The carabiner cutting the rope."  Instead, it involves a series of steps. 

• The fall pulls the top strand of the rope tight against the carabiner and begins to pinch the lower strand.
• The contact between the lower strand and the rock (perhaps some tiny feature or crystal) cuts the rope sheath longitudinally.  As Ryan shows on HowNOT2, cutting a rope longitudinally only a few centimeters results in the entire sheath separating, because of the way the strands are woven.  The core strands are now exposed, and the strength of the rope is significantly diminished.  Once again, as Ryan explains in HowNOT2, he measures about a 30-40 % reduction in the strength of the rope without the sheath. 
• As the forces grow higher, the top strand pinches down completely on the new exposed core strands and locks them in place.  This is why the belayer felt no force. 
• The fall factor is high, probably close to 2, maybe a little higher, maybe a little lower.  Regardless, it's high enough to cut the rope, especially as it was previously diminished in strength.  All the core strands in sever in one audible bang, like the sound of a gunshot. 
• The cross section of the carabioner may have played a role in this, including the reduced diameter, and the profile. However, the great number of factors that had to line up "perfectly" for this to occur make it extremely unlikley that such a scenario would occur again, and, therefore, that we should call carabiners like the Petzl Spirit "unsafe" or stop using them. 

It would definitely be interesting to see if Ryan, or anybody else with the equipment and capacity, could attempt to simulate this sequence of events with different rope angles, carabiners with different cross sections, and so forth. 

Just a shocking and sad accident all around. 

In reply to BruceM:

Climbing has a lot in common with flying - you've got something up in the air that can't easily be back on the ground again immediately in every case and that has to be made safe - and flying is much better at being open and objective about sharing outcomes from accidents and learning from them.  Climbing could certainly get much better at it.

In reply to Offwidth:

I agree. Uninformed speculation is never helpful, and not only may it distress the relatives and friends but it cannot shed any useful light on what happened or what might be done to prevent something similar happening again. That can only come from a proper investigation.  However the results should then be available, to discuss and learn from.

That is what the published article attempted to do. Although there are justifiable concerns over the way that was conducted, it did attempt to establish the facts and come up with an explanation. Subsequent discussion and speculation has been based on that, rather than on uninformed assumptions made by those who just read about it on the internet.

I doubt whether a UK report would have given the names and other details of the climbers, but that seems to be the style of American journalism.

In reply to Howard J:

The problem in producing a UK wide report has always been who would pay for it and who would coordinate it. Both are significant issues. The BMC volunteer efforts are an excellent start but are a distance from the US or European reports.

https://services.thebmc.co.uk/modules/incident-reporting/

In this particular sad event the analysis has happened, so discussion  is reasonable.

I still think the key lessons are not being learnt however well the information is produced; otherwise the Yosar analysis detail would have shifted significantly.

http://www.bluebison.net/yosar/alive.htm

In reply to Frank R.:

Again, this is related to the bit I picked up on from Rob Chisnall's comments in the article.

You can make the carabiner lighter by making the bar thinner, but that would reduce the radius for the rope to run over. Modern forging techniques allow you to move that metal into more complex shapes that maintain strength and increase the radius for the rope to run over in the bends, such as "H" (more of an "I", really), "T" and triangular cross-sections (as in some Spirit krabs, but also seen in other brands).

This "could" result in a side profile that would be "sharp" enough, with enough force applied, to cut a rope.

In this case, the pinch of the rope reduced the amount of rope available to absorb the fall (compounded, possibly, by the belayer taking in some of it) and therefore increased the sideways loading on the krab enough to do just that, when trapped against a solid surface.

It looks (without more detailed analysis) to be a very rare set of factors coming together and resulting in a tragic acident.

One thing I've not seen is whether there was any potential for the karabiner to have swung sideways in the event, resulting in more of a "slicing" action than a pure downwards "chopping" action. If sideways movement was possible, it might have been down to which way the gate was facing and therefore if the rope was trapped against the spine of the krab or able to roll out of the way. Another avenue of testing.

In reply to brunoschull:

I think informed discussion about accidents is important. The problem I have with a lot of these threads is that we don't have informed discussion, we just have idle speculation. If we don't have the facts of an accident and an expert analysis of the failure mechanism, what is the value of the the discussion? What weight should we give the lessons learnt?

In this specific case, my concern lies in some of the statements of "fact", e.g. "No rock edge was involved in cutting the rope, and no rope sheath material was observed on the rock". I cannot understand how this can be stated with such certainty. Even if the sheath was cut by the krab, the core strands may have been cut by the rock edge below when exposed by the cut sheath contracting. For a climb that from the MP thread images and rgolds comments above is littered with sharp edges, that is a bold statement.

My other issue is the lack of investigation. Again, for example, where was the analysis of the broken fibers? Did they show the classic mushroom end of a high speed tensile break? Maybe they had more of a clean fracture surface typical of a lateral pressure with a high stress concentration?

Textile failure in a dynamic fall is a complicated topic and a meaningful discussion requires enough analysis to make it worthwhile.

(Edited for typos and to add I am the BMC rep on CEN TC136 WG5 and the UIAA Safety Commission and the leader of the UIAA working group on a rope cutting/abrasion test so am pretty interested in this stuff.....)

In reply to TobyA:

To several, a factor 2 fall won't break a rope, it's not even a indication of impact force, just a rough equation on how much impact the climber will feel when you factor in rope stretch / absorption. A long factor 2 fall will require the rope to absorb more energy than shorter, then you get into time graphs where the energy is absorbed over so many milliseconds and what the peak force was. It's the peak force which breaks things. 

Taking the extreme example of a longish via ferrata but not on proper gear, using slings, and instant stop, limited absorption with high peak force.

So if the belayer took in or paid out a little, it's unlikely to effect the outcome. But think about the time you've protected an overlap making sure the krab isn't sitting right on the edge, you want it high to protect you so use a short extender, the rope will then be running under tension against the rock. Rounded grit likely fine, rougher or sharper potentially not. 

It wouldn't be hard to put a rope in an alpine clutch, pull with load meter and measure the failure force. Then back work with climbers weight, height fallen etc.. to either confirm if krab severing the rope theory is even viable or not. 

In reply to brunoschull:

Yes, a good explanation - I agree with you that's plausible.

Interestingly, I was recently messing around tensioning a static abseil rope to use as a Tyrolean traverse. Using a 4:1 pulley system with 4-5 adults pulling on it and gripping the rope with various devices.

When we used an old Wild Country Ropeman Mk1 (the style with the ridged cam:

https://absolute-snow-content.cdn.rlab.net/original/63bc6b76-3c34-4435-b70b...), the force being applied atually caused it to cut the sheath with an audible bang (not quite a gunshot). Given the tension, the sheath sprang back to expose 2-3 feet of core. We stopped pulling at that point.

Certainly pleased I've got the mk2 version as my emergency kit. Put me off wanting to use the mk3 or mk4, too!

In reply to LastBoyScout:

First ropeman had overly aggressive teeth, and the forces to tension lines across gorges start to get quite silly once you get up towards the 60 or 70 degree mark. 

If you take a ball park 100kg of pull force per person, times the people, the ratio, less a bit for friction you're putting a serious load on that device. 

Offwidth wrote:

You misunderstand the concerns here. Speculation just after a fatal or highly serious accident has happened often has poor information quality and can really hurt relatives and friends who are struggling enough with the immediate shock; in such situations it's best to await better information.  The internet response immediately following such tragic but unclear events is a human one but it's effectively rubberknecking.

I understand your concerns.  I think there are good reasons to limit discussion and withold speculation out of respect for the magnitude of tradgedy, and for the feelings of family, friends, and those immediately involved.  However, there are other perspectives.

There are many ways to process trauma, many ways to grieve, many ways to show respect.  See this thread for discussion of a recent fatal climbing accident in California, where people close to the victim reached out for information, as speculation and discussion continued.

https://www.mountainproject.com/forum/topic/126800181/death-at-lovers-leap

Howard J wrote:

Uninformed speculation is never helpful, and not only may it distress the relatives and friends but it cannot shed any useful light on what happened or what might be done to prevent something similar happening again. That can only come from a proper investigation. 

I respectfully disagree.  I think that public discussion, including uninformed speculation, can be exceedingly helpful for several reasons.

First, even if information is limited at the start, the very process of discussion can contain important lessons.  For example, as discussion unfolds, participants can identify what information is missing, construct different scenarios that could have resulted in whatever events took place, share details about and lessons from similar accidents.  There’s great instructive value in this process.

Moreover, having a place to discuss accidents openly can help encourage a culture of greater transparency surrounding accidents.  I think it’s not a coincidene that in the United States MP has a dedicated “Accidents and injuries” forum, and that the American Alpine Club produces the highly regarded Accidents in North American Climbing.  These things go hand-in-hand.  It's not a competitom—whose climbing culture is better—but I think that open discussion forums and publications like ANAC are a step in the right direction. 

The information produced by bodies like the the DAV or ENSA is very good, but it’s also very general and not disseminated widely.  

Can you immagine how beneficial it would be to have an open forum to discuss accidents and incidents in the French Alps, for example?

A friend of mine with great experience in profesional risk assessment in climbing moved to Grenoble, and as part of his adjustment to European life he contacted the French authorities and inquired about the possibilities of creating something like the ANAC in France.  You can imagine their response: 1) We have this information but why would we share it?  2) Who are you?  3) Why do you think we need to change anything about our mountain culture?   I think such inquiry would be met with an equivalent response in Germany, Italy, Austria, and so on.  The culture of sharing information and discussing accidents openly and without judgement is just lacking. 

In the interest of promoting a culture of greater transparency and humility surrounding accidents, I think we should publicly disucss accidents more and not less. 

In reply to ExiledScot:

We reckoned we were putting maybe 800-900kg on the rope.

We started off using 6mm prussic cord to tension the rope and we didn't snap that.

In reply to LastBoyScout:

Depending on the knot and using correct proportional diameter, a prussik will usually slip before it breaks. Classic is best for this, klemheist obviously the worst, french in the middle, because of the way they tighten.

I'd think you were putting more force than that in, but it is hard to say, as it's subjective on how hard you pull and efficiency loses depending the pulleys used. 

In reply to brunoschull:

Looking through the speculation and uncertainty on this thread, as well as the stats Offwidth posted above about experience not really correlating with accidents I really question the value in such analysis.  I mean what have we learnt from this thread regarding safe practice going forward? I would suggest pretty much nothing, or at least nothing that wasn't allready widly known. I don't doubt very rarely an accident happens that highlights somthing we can genuinely all learn from.  Allthough  TBH, I can't really think of an example. The sad truth is that the vast vast majority of accidents are due to well known factors, worn gear, rope running over sharp edges abbing off the end of the rope, incorrectly tied knotts and so on.

In reply to ebdon:

I completely agree that this particular example is too unclear, and perhaps too unlikely, to take any concrete lessons from. But how did those well-known factors become well-known in the first place other than by them being diagnosed when people made the errors.

In reply to Luke90:

Good question, I don't know, but it would seem in the UK the 99% of causes of climbing accidents are pretty well known despite the lack if systematic reporting, maybe we have learnt it all from the DAV or Americans? Somehow that seems unlikely. I guess my point is how much do we actually learn from a fatal accident where someone didn't finish their knott, or the rope ran over an edge or they didn't tie a knott in the end of the rope? The fact these are common suggests not a lot.

Has anyone here changed there behaviour/ practice after reading a report? Personally I have only started doing things differently after my own near misses and deep down I will admit I only look at accident reports so I can think 'well that won't happen to me' I suspect I'm not alone in this.

In reply to ebdon:

I mean what have we learnt from this thread regarding safe practice going forward?'

I suggested 3 possible lessons upthread. They may not be fashionable (at the moment), but they still seem valid to me.

In reply to Rob Exile Ward:

This is exactly my point - the lesson of don't fall off on trad, especially where there's lots of edges or it's slabby, or double ropes are less likely to be cut then singles are very common knowledge. It's obvious. We haven't learnt anything we didn't allready know. 

Now perhaps Luke90 is right and we only know these things through prior meticulous reporting. I don't know. I just don't think it's actually that useful what is in effect pointing out the bledin' obvious. 

In reply to ebdon:

As someone who actually reads ANAC, it has made me slightly obsessive about buddy checks, knotting the ends of ropes and double checking everything abseiling.

And as someone who has been climbing for years, being comfy in the environment can lead to complacency which can lead to errors. A lot of the fatalities are due to pretty basic errors.

That is evident in the stories told in ANAC. I would prefer not to become one of those stories.

In reply to brunoschull:

It would be interesting to calculate the pinching force - we know the upper (climber) rope cannot go below the surface of the slab.  I would hazard a guess that it's pretty low* if the rope is free to run through the krab, and theoretically it was and there was a reasonable amount of rope out. That low pinching force* is unlikely to be able to stop the rope running through the krab which is being pulled by a high force (the falling climber).  Therefore the rope must have had some force stopping it from running out i.e. a snag between the climber and the belayer.  Pure inertia of that length of rope would seem unlikely to be able to provide that force.*

Do we know what the sheath looked like?  When you say longitudinally, do you mean that the rope was pulled over a sharp crystal which effectively cut the sheath like undoing a zip, or gutting a fish? If this is the case we would expect to be able to reconstruct the sheath to work that out.

Any boater / climber knows the belayer will feel very little force if there's enough friction in the system.  Chances are that the combination of friction* from the V notch and relatively instantaneous coming tight of the rope* was enough to create the peak force able to break the rope.

There must have been an edge of some sort or previous weakness (i.e. a 'stress raiser') involved in the core severing, the odds of them all breaking at the same distance are so slim otherwise.

Agreed. pending further investigation this would seem to fall into the realm of the 'freak accident'.

Absolutely.

*I've not even attempted to do any maths, I'm hypothesising.  It would be really useful for technical bodies to do testing to replicate the scenario to see just what forces can be generated by running ropes over edges / bulges.  We're all worried about loading ropes over edges between the last piece and the climber when falling (and rightly so), but should we also be worried about rope drag caused by edges, cracks and bulges en-route creating high fall factors?  Many of us will have experienced stuck ropes or tremendous drag resulting in the need to untie from one of our half ropes in order to progress. Would we have been in the same situation had we then fallen off on our remaining rope?

In reply to ebdon:

It’s not necessary meticulous reporting that is required for the layperson, but I think having an idea of what is causing accidents and near misses to happen, or gear to fail, is pretty helpful.

It can identify, for example, if there is a steeper learning curve for a particular belay device or of the new method of doing something is actually less effective than the old method etc.

How that info then gets passed back to the layperson is then the key. The AAC started the ANAC book as they found that stories had a lot larger impact that statistics. 

In reply to Toerag:

Of course we should be worried about rope drag!

The more rope drag in the system, the higher the force on the climber and the top piece of gear, and the lower the force on the belayer. The drag effectively reduces the length of rope over which the force can act.

Minimising drag minimised the chance of your top piece ripping.

In reply to ebdon:

Well perhaps you haven't but maybe someone else less well informed has.

I have learned there is a suspected failure mode that involves the rope being pinched between the rock and a carabiner. It doesn't really matter if it actually happen in this case or not, I still think it worthwhile that I now have in the back of my mind the concept that somehow trapping a short length of rope could potentially be a very bad thing.

In reply to ebdon:

sounds like you're undermining your own argument here. 

Others might be less lucky. Not a learning method I'd encourage for widespread adoption.

In reply to daWalt:

No its not. Im not sugesting it is. It is however how most humans behave.

In reply to Frank R.:

Sounds to me like you have a fundamental misunderstanding of what his channel offers.

I'm sure if there was a 'proper lab' that would pull test thousands of samples of all sorts of things from all over for free and put it up on youtube they'd get plenty of followers, but as far as I know there isn't, so it's more a choice between him pulling them or nobody.

In reply to ebdon:

No, I don't think you're alone in this.  It's human nature to read accident reports and comb through them to find some kind of obvious mistake.  It's part of the fear processing and the rationalization to keep climbing.   

And yet, if there is one valuable lesson we can distill from all these accidents, it's the 100 % certainty that if you climb long enough you will make mistakes, including obvious mistakes.  Hopefully, you get lucky, and these mistakes don't kill you.

Maybe that's the real value of public accident discussions; the reminder to remain constantly vigilant, and the humility to accept that no matter how expereinced you are, you will make mistakes.

Don't get me wrong--obviously I think the details of accidents discussions are important.  But it's the awareness that such discussions foster which is most important.  

I agree.  There are people here commenting on what Ryan does with no understanding of how he approaches things.  Sure he seems like some weed-smoking, slack-lining "bro" but he's added a huge amount of knowledge to the climbing world, and arguably his findings have been disseminated to more people than any of the DAV reports (for example) in the last decade.  In this sense, he's harnessed modern social media to improve awareness and safety more than any other official body or individual. 

I'll never forget one of his older videos demonstrating how fast rope can move through an ATC when under tension.  Scary stuff, but really important to see clearly. 

In reply to brunoschull:

I don't disagree with what you are saying other than immediately after an accident with no clear information; which in my view is a bad idea unless the family and friends want that done. I've witnessed depressing impacts on relatives where unfair accusations and bizarre misinformation surfaces following serious incidents and deaths. My view was also informed by Mountain Rescue services,  who were particularly cynical about press reports. In a particular case where I was present people on UKC forums accused accident victims of incompetence based on incorrect information, while one of the victims was in hospital struggling to survive. Any possible benefits of early discussion where facts are unknown simply don't come close to justifying such risk of further damage.

In reply to Offwidth:

The difference in the US and German/European approach and that of the UK is generally speaking the police are involved at an early stage, in the USA it will be the Sherriffs office and for example in Germany (in Italy it is the Guardia Finanza) we have a specific mountain police force involved in rescue and accident analysis with specialist investigators, even in the non-mountain areas each police force has a trained investigator to look at accidents in climbing walls/gyms. So when the press get involved they early on get a statement from the police who then involve the testing labs and specialists/DAV as they will be giving evidence to the coronors court. In the UK it seems a bit more vague!

In reply to Summit Else:

No, the reality is it's all been tested a thousend times and understood, he appeals to the YouTube generation who aren't bothered to read and understand things which can get more complicated than Ryan wants to believe. Dumbing down is the phrase.

In reply to jimtitt:

FTFY

However we do sometimes have the odd excellent reflective news article, from the likes of Ed Douglas.

In reply to brunoschull:

People have said ATCs were slick since before Youtube was even invented. It didn't need a video. 

In reply to jimtitt:

You sometimes share results of testing you've done with us on here. Undoubtedly more could be said about those topics, but you quite rightly limit the volume and complexity of your analysis to suit the medium. Why is it horrendous dumbing down when he does it but not when you do?

And to be clear, I appreciate both, this isn't a criticism of what you post on here. I like interesting results and analysis wherever they're coming from. And it's interesting on video too. It doesn't replace proper scientific analysis, but it's certainly more accessible, and that's fine as long as people recognise the limitations. I've not seen tons of his output, but in what I have seen he doesn't seem to particularly overstate the significance of his results or make outlandish claims, and I'd certainly have difficulty finding any other source for some of the information he's sharing, which makes it a clear benefit to the climbing community in my view.

In reply to brunoschull:

Is this actually true? I watch quite a lot of his videos, I normally find them entertaining, but I'm not convinced that there's much knowledge adding actually going on (for climbing anyway, I can't comment on slacklining or canyoning).

I'm thinking more about the break tests here, rather than e.g. the how to big wall videos which are more subjective (and imo, not very good). I do like the measuring of cams he's got on his shop's website.

He's definitely harnessed modern social media very well and is getting information out to people better than the official bodies. 

In reply to Fellover:

I think information sharing has changed. Int days of olde people chatted in club huts, down the pub, sat in Pete's for hours etc.. now people sit in laybys on their own van lifing, in between wild swimming and climbing. There's much less face to face social knowledge sharing.

In reply to Luke90:

Popularising complex subjects is difficult, first you actually have to understand them.

In reply to TobyA:

The question has been raised about whether discussions like the one here teach us anything we didn't already know, and this has been extended to the value of accident reports in general.

I think that there is a powerful emotional need to understand the causes of a tragedy, and this extends far beyond climbing accidents.  There seems to be a deep human desire to grasp what has caused bad outcomes, quite apart from the practical value such understanding might confer.  Perhaps it is part of a grieving process, perhaps it is a (misguided) way of convincing ourselves a similar thing won't happen to us, or maybe there is more to it then we understand.

One of the things accident reports and their discussions do is to put the hard stamp of reality on otherwise hypothetical concerns. Sure, we all "know" that rope drag can raise peak fall loads and that double ropes provide additional safety on sharp ground, but those hypotheticals can take on far more urgent meanings when we find just how far from hypothetical they can be.  Another role, already mentioned here, is that we eventually learn of accidents happening to people "just like us," not some othered category that can be dismissed as inexperienced, incompetent, unprepared, inattentive, or complacent (not that there aren't plenty in those categories too).

As for the discussions, my experience is almost the opposite of the argument that they only confirm what we already know.  At least for me, the discussions often challenge unquestioned assumptions and provide perspectives that hadn't occurred to me.  Does any of this make me a safer climber? I don't know.  It makes me more aware of the fact---also obvious---that we operate in a dangerous environment and that things can go from entertaining to disastrous in an instant. It continually updates Whymper's classic warning,

Climb if you will, but remember that courage and strength are naught without prudence, and that a momentary negligence may destroy the happiness of a lifetime. Do nothing in haste; look well to each step, and from the beginning think what may be the end.

It took a great tragedy for Whymper to arrive at that view.  The accident reports and their discussions gives us an opportunity to achieve such clarity without having to be engulfed in tragedy ourselves.

In reply to Summit Else:

I gave you plenty of examples from other countries how it can be done properly, it's usually the police investigators and forensic experts (which would't do in the USA, fractured as it is), but they also employ experts from the likes of DAV or ÖEAV. Perhaps AAC could do it? It's in their interest, after all.

And both DAV & ÖEAV often take interest directly, if they think an incident needs investigating or something about it could potentially advance climbing safety. They have done so plenty of times in the past.

In reply to brunoschull:

Are you seriously arguing about the 300K subscribers again?!? Your argument, taken to the extreme, is like saying that the fact that climate change or COVID deniers have "disseminated their 'findings'" to more people on social media than any of the IPCC or NHS bodies is actually a good thing for climate and virus safety...

A single potentially wrong video of his is a safety nightmare, that's why (and I vaguely remember a few that might have been exactly that, claiming bold claims that haven't really worked out in the end).

Especially given his "bro" manner and subscriber counts. I might even argue he did worse for the safety of climbing than most other vloggers, given his "bro" manner and lackadaisical approach to testing protocols, especially at the very beginning of his youtuber star "carrier" (you know, when his test cells were pretty much just random junk off aliexpress or wherever?)

And no, even there you are still wrong. Where do you think the various UIAA/EN norms that got updated in the last decade are from? From the work of like DAV, ÖEAV and other experts. Otherwise your Klettersteig kit would still be an old knotted rope slotted through a Sticht plate (unsafe in so many ways), or corrosion‑prone bolts and anchors would be still used at the seaside, or whatever. While you might not see the results of their work on click‑bait youtube videos ("Are our VF kits ☠️–traps? Watch to find out more after the advert, or Yer Gonna Die!!!")

Has the youtuber actually done any groundbreaking tests that had uncovered any new truths or any faults with previously accepted techniques and actually advanced safety in a consistent and replicable manner?

Like testing the impact forces on the anchor, belayer and gear with various belay techniques on multi‑pitch? Things most American climbers on MP still don't understand at all? Like when a HMS belay off the anchor might be actually the safer thing to do? I think ENSA did a video and paper on that and ÖEAV or DAV published a lengthy report on the same as well.

Or at least stuff like snow anchor testing in various types of snow? Including improvised ones? But all done in a proper scientific and replicable manner, as much as the mountains allow? Ice screw placement testing in actual lab ice? Micrographs of cut ropes and how to reasonably tell if the rope cut on a hard edge, rock or snapped some other way? The way you can usually tell from them if a rope was deliberately cut by a knife when under tension, a hard edge on metal gear, sewn through by abrasion on rock in a swing, melted through with another rope or simply cut on a hard rock edge on impact without swinging abrasion at all?

The thing is, proper science isn't that much flashy at all and doesn't use clickbait video titles just to get money or sell merch (predatory journals' papers obviously excepted). One can get a bit lackadaisical with the titles when popularising the science (think Scott Manley for a rocketry example), but never with the actual numbers it's based on. Doing otherwise is still dumbing down the matter.

I have here right now a three volume book by the former DAV and UIAA safety commission president. A somewhat older edition, so slightly outdated on modern standards and their subsequent findings. Full of anecdotes from all their investigations and testing. Only in German and a few other non‑English languages, sorry. The writer would actually make a really good scientific vlogger now, had they been born into a different generation – they are really witty, able to describe horrific accidents with both humour and pity (while keeping it short) and most importantly, keep to the point and the technical facts. Just their work over the decades (and that of their colleagues) has done more for mountain safety than a few dozen of slick bro youtubers combined.

In reply to Frank R.:

"Only in German and a few other non‑English languages, sorry."

I "liked" your post because passion and a vitriol fueled screed always make for good reading.

No need to apologize for the languages: I live in a polyglot nation (Switzerland) and speak and read German, Spanish, French, and English. 

The DAV and ENSA materials are very good, but they haven't found a way to disseminate the information widely to the public, and they don't react quickly enough to current events.   

Ryan fills that gap, as do public discusion forums. 

Carry on.

In reply to jimtitt:

Which is why all the good science popularisers are in such a short supply. And why I really enjoy the writings of the ones that I consider really good – they make the very hard effort to understand enough of a very complex subject, while still dumbing it down enough for the average reader to understand at least the gist of it. Without leaving out all the important bits, of course!

A good general science background is a must for that, I think. Some statistics and maths background the second (or possibly even the first, as you can't understand science without that). General humanities just for the Latin and broadening of the mind. Biology or whatever the field one might want to report on. Enough of physics, because there is always some. Et cetera.

Plus of course good and concise mind and understandable writing. Bridging the disciplines can be really hard at times. Think more like a translator, if you had to translate through several intermediaries from some rare Pidgin, including a step through Google Translate

In reply to Frank R.:

Do you have any examples of videos or claims he's made recently that you think were dangerously wrong? Because I must have only seen a tiny subset of his output by occasionally clicking an interesting-looking link. Everything I've seen showed the tests he was actually doing and made pretty modest claims. But you and JimTitt keep insinuating that he's actively getting stuff wrong and given the limited proportion of his output I've seen, I'm quite prepared to accept I might not have seen that, or might have seen it without recognising the error given my own lack of expertise. Care to share?

In reply to brunoschull:

[...]

Well, I gathered as much from your nick – "Bruno" in your nick was the hint. But my "Only in German and a few other non‑English languages, sorry" was simply a statement that Schubert's books, or even most of the publicised DAV or ENSA reports just haven't been translated to English at all. While I can read German or French (somewhat) fine, plenty of other UKC posters might not. So my apology was aimed mostly at them, as they might not have been aware of their existence at all.

In reply to brunoschull:

Hah, apologies for that. I thought I had toned that down, and it's quite OT for the actual thread anyway.

They could certainly do better at that, even if they do try at times (some of their recent videos were pretty informative). The point was, they can't be as speculative as any youtuber, given their reputation. A video out from ENSA or DAV better be rock solid, as they usually are. Which can be a problem in this fast‑moving social media world. And the ones they do put out tend to be (rock solid, in my opinion), but they can't just wildly speculate...

That's where we might disagree – I'd rather have a rock solid evidence (even if that's not always possible, given the forensics can depend on such random occurrences as even tiny crystal undoing the water knot as in some past documented incidents), than just some speculation by a famous internet personality backed by just a single pull test or the left rope (still without at least micrographs of it).

Sure, they might have actually done a lot of good, but given their wide audience, getting a single thing wrong would be even worse.

In reply to Luke90:

Well, he did a pretty lame and 100% click‑bait video on shock loading slings around five or so years ago. Using pretty cheap load cells without any real time precision and a very low sampling rate (totally inadequate for the test), even admitting it in the video itself. Yet uploaded it as the total bees' knees and a hard truth, ignoring all the inconvenient questions.

Tells you something about the guy's credibility, doesn't it. Sure, a soft squishy mass will always behave differently than an equivalent rigid mass, but that's what figurines like DAV used in similar tests (one who scared the night cleaner out of their mind once, being suspended by its neck at the staircase) are for.

Basically sold that video as a "scientific" truth that any shock loading of slings doesn't ever exist at all, just to get the notoriety and the views from a controversial theme. In my view, he actually contested well‑proven knowledge just to get subscribers and views at the start of his youtube carrier. Any red signs there yet? Plenty of them for me, saying the guy is not reliable at all.

Tells you all about their scientific creds. At least it does for me. Even if they might have gotten better since then, their background is forever tainted by that, simple as that.

youtube.com/watch?v=nr3YBDnOI8Q&

In reply to ebdon:

Not putting rubber retainers on sling draws.....

In reply to john arran:

Cutting is done by the tightness of the radius of an edge and the force applied. 

Knives only seem sharp at our eyes ability to determine. The edge radius changes for a knife depending on how sharp it is. 

Cut basically means "The radius of the change in direction of the rope combined with the force applied caused the item to break at that point"

In reply to Toerag:

A guide mode belay works on the same principle and obviously works well - unless I misuderstand your point.

In reply to jimtitt:

Here's a concrete example.

1) The scientific lab doing 'proper' science

DMM do a test on falling onto slings and say that climbers can break sling tethers with a fall factor 1 .

2) The YouTube nerd 

Actually takes a fall and shows that the forces are massively less

The first caused so much repeated disinformation in the community. The second showed that in a real world fall, the forces are substantially less. 

As a research engineer myself having worked on country wide disaster reporting, the difference between private individuals and research labs is often just the name. Especially in niche areas.

Ryan may not be as rigourous as most but as real world observations go, they are far from useless. 

You can say that the sampling rate wasn't high enough but repeated falls showed little variation, rate of force development can be seen just by pausing a 30fps video feed, etc. 

And someone mentioned the DAV reports on this... where are they? Can I see them and check the methodolgy myself etc?

In reply to timparkin:

A guide plate differs in that the climbers rope is forced to pinch the belayers rope by the shape of the device (unless it is too thin and slips past the belay rope....)

In the proposed scenario (which I am still skeptical about), the climbers rope is meant to have pinned the belayers rope without anything guiding it, so I can only assume it is not sat directly above the belay rope (very unlikely for this to happen and for the ropes to stay above each other if a dynamic fall), but has pinned the belay rope via the krab. From the images I have seen and the description, this seems unlikely and would be dependent on the route of the rope from the belayer to the cam's krab providing enough drag and therefore friction.

However, as there has been little investigation and no testing, we shall never know.

In reply to Frank R.:

I'd rather just be able to see the single sample test with my own eyes. Ryan points out repeatedly that these are single sample tests and the test conditions are completely clear (unlike many 'real' tests). 

People's ideas that 'official' scientific results are 100% perfect is ludicrous. There are so many mistakes in research that it's quite laughable. The consensus builds up over time and many discoveries have been made or at least repeated on the citizen science level. 

Perhaps DAV etc should react to some of Ryan's popular videos to help people understand them instead of hiding behind paywalls and 'authority'. 

In reply to brunoschull:

I have always been aware of sharp edges, but have never before considered the potential for the rope-end carabiner of a runner lying on a low angled slab above a steeper section, to cause a pinch point that could possibly cut a rope. (Or contribute to cutting a rope, if the rock-type was indeed also a factor.)

So this is eye-opening.

This situation applies to trad and sport climbing. The last three sport climbing venues I visited had slabs above steeper sections. Although, in sport, you don't get to choose where your pro is, you are relying on the bolter knowing about, and placing bolts in ways, that reduce the possibility of this potential issue.

In this incident the poor climber, sadly, did everything pretty much right. Including extending the runner of that fateful piece. Ironically, if he had back-clipped that runner (and the gate hadn't released), the pinch effect wouldn't have happened (or been as bad).

So this incident raises questions about good practice: perhaps introducing knowledge of that pinch idea into standard basic climbing teaching? (As important as knowledge about back-clipping?)

And in fact, why should we ever use single ropes in any form of climbing? Fire crews never rely on non-redundant systems. And when top-rope soloing, we always stress the importance of redundancy -- multiple rope strands and devices.

So threads like this and the Mountain Project one are invaluable. As are, of course, good accident reports.

I really wish some company or authority with the ability could run through a bunch of tests to try to explore that runner on low-angled slab pinch idea. Are lightweight profiled carabiners potentially a bad idea in this situation?

Finally, I think everyone involved in wondering how this could have happened, feels very sad for the climber, his belayer, and family and friends. At the time, he, like any of us, would have thought he was doing everything by the book.

It's just that perhaps the book isn't quite finished yet. And if that were me. I would hope at the very least, that this incident would help add yet another potentially life-saving chapter.

In reply to timparkin:

Edit to add: I don't think the famous DMM tests are a good example of 'proper science'. I noticed a while ago they seem to have removed the video and results table/article from their website, maybe they'd rather people forget about them!

I do agree that the DMM tests were not really very helpful, but Ryan (HowNot2) hasn't actually done an equivalent test that I'm aware of (the one Frank linked above is quite different).

Regardless, we already knew that the DMM tests were not reflective of reality because there are several reported (in an informal sense, mainly on the Mountain Project forum) incidents of aid climbers taking >FF1 falls onto daisies and not breaking them and not being particularly hurt themselves. I do acknowledge that the information is not well publicised.

I do think the Ryan bashing from some people goes a bit far, he is mostly very open about the fact he's not 'scientific'. I like what he does in general.

I'm not convinced that is sufficient evidence to say the sampling rate was high enough (presume we're talking about the sling/impact force video Frank linked), but it's probably not worth getting into here.

I'd also be interested in DAV reports on dropping human-esque masses onto slings if they exist. I do find it very frustrating that DAV stuff is all in German (yes I am a useless typical English person that only speaks English). Probably find it especially frustrating as when I was remotely involved in academia all the papers I needed to read were in English regardless of where they were written.

I also find it very frustrating that Ryan hasn't replicated the DMM sling drop tests using a human, it's basically the only thing I want him to do, he's been more or less well enough set up enough to do it for years now, yet he has not (obviously he can do what he wants, but I'm allowed to be frustrated by his choices :p ).

In reply to timparkin:

Ryan showed that in his scenario the forces were much less, he didn't explore (either by testing or establishing enough data to allow him to apply theory) other " real world" scenarios and he has no interest in doing so or learning from others who have. His aim is fun videos to suit his audience's attention span and derive his income from these, the boring stuff isn't his thing.

All DAV tests are available in their archives, that's the boring bit.

In reply to BruceM:

No, that would be a huge overreaction at this stage. I don't think the evidence that pinching caused the cut in this case is particularly convincing, as discussed upthread. And even if it was what happened, it's going to be vanishingly rare, depending as it must on a really unlikely combination of circumstances.

Back clipping is an easy and common mistake to make, is vastly more likely to cause a problem, and is trivial to teach people to avoid. This potential new consideration is a rare scenario, is unlikely to cause a problem even when it is present in principle, and would be hard to teach people how to avoid. What actual steps would you take to avoid this? If anything, it perhaps adds to the case for double ropes on climbs like this one, but I think there are already stronger and better-proven reasons to consider that.

If better evidence becomes available in the future that this is a real concern, and the combination of circumstances required to cause it becomes well understood, it might become something worth teaching and sharing knowledge on for more advanced climbers who already have the basics down. There's no way it ever rises to the level of being a beginner-level consideration like back-clipping.

Right now, it's nothing more than a curiosity for us climbing nerds to be intrigued by and maybe a case for more testing to see whether the hypothesis can be reproduced.

In reply to jimtitt:

And why should it be?! His videos share some useful information and provide some entertainment as well. He doesn't pretend to be a scientific authority. I'll give you that some of his titles have been a bit clickbait-y and that's something I'd prefer not to see, but it's the nature of the beast. 

There's a place for both approaches in the world, and we'd be worse off with only one or the other. If we could somehow force YouTubers to all take a rigorous scientific approach to their testing and never simplify or entertain, that would probably reduce rather than increase the knowledge shared, because almost nobody would be watching!

In reply to BruceM:

Yes, but you can choose to place a longer quickdraw, or to extend one, which seems to be a good lesson from this accident.

In reply to Fellover:

Not sure about FF2 sling drop tests, but DAV did use a very realistic and very expensive automotive crash‑test mannequin for some drop tests in the past. Maybe to ascertain what's the effect of deformation of harnesses and soft(ish) bodies on the absorption of impact forces generated or something like that (IIRC, I don't have the books with me).

It's pretty clear that a 80 kg human body behaves slightly differently than a 80 kg steel weight, although for some strange reason, nobody ever volunteers for the real drop test at forces strong enough to break spines, so steel weights it is (and in the UIAA/EN‑892 norm, also because they offer repeatability).

As I remember it, they broke the spine of the figurine several times, which was something like $1,000 a pop for that particular part ($4,000 in today's money).

Fun fact – in‑between tests, they once "stored" the fully dressed figurine hanging by its neck in a stairwell, having no space elsewhere. The night guard or cleaner almost got a cardiac arrest on finding it...

In reply to Frank R.:

The most recent UIAA Safety Commission Plenary meeting was hosted by CAI in Padua. We were shown some pretty terrifying footage of the last time the Safety Commission visited Padua. They were throwing themselves (Commission delegates and CAI members) off a makeshift platform to test the efficacy of the Italian hitch when belaying. They were taking some impressive falls for the cause!

Thankfully CAI have since invested in a couple of drop towers that meant none of us has to step up and put our bodies on the line.

In reply to Frank R.:

The 80kg steel mass represents a 100kg climber (it's really 102kg if I remember rightly) and based on tests done on parachute harnesses with later climbing gear tests to lower impacts and then factored up to givae the current soft body equivelent mass. It's probably more accurate at higher-impact falls than short ones but that's erring on the safe side so good enough. With small forces the human body is very squishy but this effect soon disappears, the falls from height guys have done loads of work on this. For the sling tests the DAV use a very heavy truck tyre which gives a reasonable value apparently.

However!!! We do know that in a real "real life" situation somebody (squishy body included) managed to break a karabiner in a chain of three quickdraws when he slipped, the subsequent re-enactment giving a force of 27kN.

In reply to galpinos:

That film is cool, I was shown it by one of the guys involved. It's one of those historic things the modern climber doesn't understand was part of the basic research which was done many decades ago which are the foundations of the modern standards. The "high priest" of climbing safety, Pit Schubert who lead the DAV research for thirty years was actually partially disabled as a result of an early belaying accident, Werner Munter who popularised the Italian hitch in the USA had actually demonstrated the "Munter Hitch" (basically a half hitch around ones shoulder) previously to the UIAA, this smashed his collarbone to pieces and after he had recovered he became an immediate convert to what we call the Italian hitch.

An interesting historical aside; the original climbing rope standard was introduced in the UK and instead of a crude weight dropping actually used an electromagnetic linear accelerator which is what the BSI had available, the drop tower being later adopted on financial grounds. As part of a project I built a similar test rig to replicate 50m falls, it was awe inspiring when you triggered it (not as frightening as our first attempt which involved a 14ton tractor driving past at 50km/hr) but it did what we wanted.

In reply to jimtitt:

Pit was before my time but the stories persist and both Pit and Vittorio were remembered fondly in Padua. There are still some of the old guard left, Marco Boaiti from Kong, and I overlapped with Jean Frank Charlet. It is easy to forget the efforts of those that went before us!

I'd have liked to have seen the tractor effort!

In reply to Fellover:

BTW, upon further reading, DAV did their first controlled FF1 impact force test between a (live) human and an equivalent steel weight well back in the late 1960s. Not belayed, falling onto a dynamic rope fixed at anchor.

The test with the rope found out the impact force around 50% higher for the steel weight (that number shouldn't be considered as authoritative for all drops, obviously!), with a shorter and higher peak force. Unfortunately for the volunteer (Schubert himself), the FF1 test still found him with a pretty painful lumbar area spraining...

Tätigkeitsbericht des DAV-Sicherheitskreises 1969-70, DAV, München.

So yes, everybody was well aware of the difference between a "squishy" and a solid weight, five decades before some youtuber's videos.

That doesn't really mean that the daft youtuber was right in his totally incorrect and bombastic claims about "all falls on slings are perfectly safe". It just means that there is an additional safety factor in humans being squishy, but that squishiness in us can under the wrong circumstances also manifest as pain, long‑term damage over many falls, severe injury or in extreme cases, even the spine snapping and subsequent disability or (actually documented) death.

Not just from the sling itself snapping and you decking (it might potentially hold even a FF2), but the climber's body snapping. In an uncontrolled fall (e.g. rock to a helmet), when the body might invert torso and backside down. Or even the carabiner snapping. Several documented cases of non‑locking carabiners breaking in a near FF2 fall even when the sling held – break location consistent with an open gate failure, likely from unfortunate recoil of the gate induced by the higher and shorter load on a sling).

Photo exhibit (source DAV):

1. Impact force in time in FF1 fall on fixed dynamic rope, solid steel weight (a) vs. live human (b). Live human still had lumbar pain. One sample only.
2. Automotive crash‑test mannequin, drop test on a dynamic rope, "unconscious" (not controlling the fall) – "broken" spine, crushed lumbar vertebrae, torn stomach. "Dead".

In reply to galpinos:

Me too! Sounds epic, especially for any elf & safety officers

In reply to jimtitt:

I finally got to my Schubert books and added a few real world cases – of carabiners and such snapping on near FF2 falls that he cites there – to my post above. There was one even from the UK, with a doubled sling extended by a quickdraw. Climbed above anchor and fell – the quickdraw carabiner broke (likely open gate failure from gate whiplash), poor climber decked, ending up with heavy injuries.

In reply to jimtitt:

The difference in results with a squishy body isn't to do with different forces it's to do with rate of force development. 

e.g. imagine a short drop onto a 120cm dyneema sling. 

Perhaps the extension of the sling with an 80kg weight attached is 1cm (or less perhaps) and so the force is massive. 

A dangling squishy human probably extends by about 10-20cm considering the distribution of limbs etc. Hence possibly 10% of the force. 

In reply to Frank R.:

But the detail missing in that test is the distance fallen and the distance of braking. For a rope on a larger distance, lets say a few meters, the differences between humans and rope would make sense to be 50% or thereabouts

A fall for a distance of 0.12m onto a dyneema sling is a VERY, VERY different story. 

Did he say that? Can you give me a timestamp on the video please?

So you're saying the forces on Ryan weren't what he reported and were substantially high, possibly up to 10kn+? Or are you saying something else? Perhaps you should supply a link to the test case he actually talked about rather than some very different test case.

So we're talking about a couple of very different scenarios (which you haven't provided details of) and a re-enactment that obviously doesn't match the original scenario (unless you're missing out lots of details. 

Whereas Ryans is fully documented and can be examined visually. There's a potential error in the sampling rate, which was detailed at the time.  I think we can presume he didn't experience 27kn

So unless someone starts sharing more useful data so that normal people can access it, people will end up looking at Ryan's results and making their own conclusions. 

I'd still like to know where Ryan has catastrophically misrepresented himself causing potential deaths (which seems to be implied)

 

In reply to galpinos:

We were trying to measure the effect of heat on belay devices in long falls so built an anchor with an ATC in the foundations of a building and I stood there with my leather welders gloves on at the ready, as the McCormick roared by I dropped the rope and dived to the side! Then hastened to the workshop for a schnapps or two.

My "particle accelerator" was more fun, a 14m long steel girder with an overhead crane sled and a lot of bungee cord (we had a 100m drum) which we tensioned strand after strand with a power winch then let it go, the last 5m was a sand brake. It was impressive to say the least! It was actually an idea from the guy who did the definive work on what we call screamers for the US military aviation on cargo restraints in accidents as with his vertical drop rig he couldn't get enough speed so started adding bungee cord to get to what he wanted.

In reply to timparkin:

You are struggling here! It isn't what Ryan says, it's what he doesn't say because he doesn't know that is important. He survived his "real world" test, in another scenario he may not have.

In reply to jimtitt:

Not just what he doesn't say, also what he explicitly implies by his clickbait titles and speech.After

After all, Betteridge's law of headlines almost always applies to daft Yt videos using clickbait titles ending with a question mark...

https://en.wikipedia.org/wiki/Betteridge%27s_law_of_headlines

Maybe he could have climbed up to the anchor (or even above it) for a repeat of his "test" to finally debunk the "myth"? I am sure the guy would have the balls for that! For bonus points, let him close his eyes without a countdown so he doesn't know the exact moment of his drop. For more extra brownies, let him try with a non‑locking biner on the sling, or even a chain of quickdraws...

Blind, that's how we did some pretty instructive different belaying methods' drop tests at my club (with just a sandbag for weight, for pretty obvious reasons).

In reply to jimtitt:OK - can you give 

OK - educate me. What is really important that isn't plainly obvious? 

In reply to Frank R.:

Not sure what you're trying to say - Click bait just exists, if you can't cope with it, don't watch. However, the clickbait isn't lies in this case, it's possibly exaggeration for edutainment sake. 

If only DAV could get the attention and do everybody a favour by sharing what they've got. Even better, just video what they do and publish it complete so people can see for themselves. 


Explain the problems that a vague clickbait title causes, or the dangers that he creates?

In reply to Frank R.:

exactly - so they don't cause a problem because everybody understands them

This idea that marketing is bad is ridiculous - we should all be hair shirt scientists publishing in pay to play journals and just passing references back and forth. I'm an ex research engineer with a PhD in mathematical engineering modelling (motors, generators, transformers + circuits - force, electromagnetic fields etc in the time domain) and have peer reviewed journals in my time (thankless and the standard is abysmal). I'd much rather see something obvious and practical that I can clearly assess than take for granted some half assed, badly documented publish or perish clickbait.

In reply to jimtitt:

One of the problems with Ryan's test of the failure of one leg of a self-equalizing sling is that it doesn't seem to correspond to situations of fundamental concern.  The tests assume the climber is hanging on the anchor and one of the anchor points fails.  The amount of rope he had in most of the tests suggests his concern was for someone jumaring up a fixed rope attached to the anchor, but in the final test, he clips directly into the anchor sling, which is a bit more realistic. 

What isn't realistic is that the climber is hanging on the anchor minding their own business and suddenly one of the anchor points blows. The worst-case scenario in which shock-loading might be a concern is that the leader takes a factor-2 fall onto the belay and rips the belayer off the stance.  This involves much higher loads than Ryan is testing and so we have no idea from his results about the role of anchor extension in the cases where the issue is most consequential.

All this is fine, except for the fact that he titles the video "Is shock-loading a myth?" and then, by implication if not by direct statements, implies that it is.

[If I may be allowed a personal rant, I find the term "shock loading" to be undefined and so open to whatever anyone thinks it should mean.  The only definition that makes any sense to me is that a shock load is applied to an anchor after an object has fallen freely for some distance before loading the system.  This being the case, every leader fall shock-loads the system, there's absolutely nothing mythical about it. What Ryan is testing---inadequately according to my comments above---is whether extension in a self-equalizing anchor can produce loads high enough to injure the anchored climber or break the sling. He says no, if you are just hanging on the anchor when one of the anchor points fails.]

In reply to timparkin:

Apologies, but I recently learned that to discuss further with people resorting to ad hominem attacks ("are you on drugs?") and deliberately misquoting ("he's not risking his life enough", which I never said) one's posts in a direct discussion is pretty much futile, unless one brings the three billy goats along to the bridge. Ta and kindly enjoy your ranting into a mirror.

In reply to Frank R.:

OK - I retract but that series of escalating examples seemed ridiculous to me and completely besides the point. 

In reply to rgold:

Succinct, nicely written, to the point and raises the very exact points I had about the video all along. That is that its title implies it's safe for all the similar but slightly different generic situations, instead of just his exact sui generis situation (or the situations he doesn't mention, as Jim noted). Thanks.

In reply to rgold:

Hmm... I took the example to be fundamentally what it was, a mock up of the equivalent fall 'type' as the DMM examples. e.g. Someone reaching up above a belay and their foot popping for example. Is this what you mean by the most consequential example? (p.s. I think the first examples with more rope involved was easing into something quite scary)

Yes, pretty much. With the lack of definition of shock loading and the irrational fear around it, it seems to have reach mythical status and meet the definition of something that is a myth (i.e. not the 'real' that people seem to understand it as). 

Or if you just fall when above an anchor - the more realistic scenaria

In reply to Frank R.:

I suspect you can generalise the particular from the fall and materials involved into other useful situations. What other situations did you have in mind?

My generalised point about Ryan's example is that if we take the test case published and wideley dispersed

1) that a factor one onto a sling will break the sling

The to address it, we just need someone to try the same factor one onto a sling but use a body (or equivalent) to introduce another data point. 

2) just do it

I'm not sure how much extra examples or research are needed in this case. There's a misconception and an actual example to see what happens. 

All this talk of extra situations he doesn't cover seems all 'nice to have' but I can't see how it detracts from the fundamental of what he did? Test scenario is self explanatory, conclusions aren't over stated. What dangers have been introduced to the community? Essentiall, the DMM information caused a fear of using certain scenarios that could itself cause a danger when people ignore what are actually quite safe scenarios.

I'm honestly really trying to understand all these extra examples that are being talked about? I completely agree that Ryan couldn't really write a technical paper given what he's done and would struggle to get it peer reviewed but is this really important?

p.s. I'm not being contrary - I think some of these things are really important and I've learned some from Ryan's website that I haven't learned elsewhere (and I've tried to find things in DAV and subscribe to the Alpine Clubs accident reports etc). 

If the information shared publicly is so bad (I don't think it is but I could imagine something better) then DAV/BMC/Alpine club should be reviewing these things. 

Here's a simple example I would love to know about. In a fall situation using a sliding X, how much equalisation do you get in a slow pull vs a drop tower? 

Here's another - in a well equalised anchor (lets say three points in a W with overhand master), how much does leg length affect actual equalisation? Is a nylon sling a lot better with it's slight give? 

I've tried to find some of this information and failed (which means it's either exists and is badly indexed or just isn't findable at all). Thoughts?

In reply to Frank R.:

Thanks for giving some information on the DAV testing

In reply to timparkin:

All right, retraction accepted – I'll try again then.

I think I explicitly ruled out "publish or perish" papers when I mentioned predatory journals way up above. Nobody likes such poor "science", of course, and it's a growing problem in bio fields and others. I don't really think DAV or UIAA or ENSA is of that kind. Sure, climbing safety science is complicated and more difficult to model or replicate, as exact replication of something like a random single quartz crystal undoing a loaded water knot in a documented accident is almost impossible. Yet it still has some standards, with those and the knowledge improving over the time.

Sorry, but you are not exactly his intended audience. With your engineering background, you would probably understand the forces and implications of his test and its limitations in all the possible real world scenarios, unlike its many other viewers. As in you might think it obvious the test means just what it means, falling on one half of doubled a sling anchor in one anchor point failure out of two from below the anchor when one point of the anchor fails without any reason (as they "cut" it). Others simply won't, and would easily extrapolate the title to any fall on sling at anchor, even above it.

Now imagine some totally random, less‑technically‑educated viewer seeing its clickbait title, watching the video without really understanding the fall distances in it – or the forces involved – at all and getting the idea that all sling "shock‑loading" is a myth. It might be a surprise to you, but I know plenty of people whose take of that video would be just like that.

Even if an experiment is done correctly (which I still doubt a bit in this case), it's the communication of the results that matters. Hence why science communication is a field in itself, requiring a good mix of understanding of both the humanistic and technical fields, and preferably no need for click‑bait.

Contrast that again to the science communications videos from the likes of Scott Manley, who tends to  be on the interesting side of explaining rocketry to utter laics, but still being usually pretty correct and succinct without too much click‑bait, while still having nearly 2M subscribers. Or many similar other channels.

As for the cases I mentioned, Pit Schubert's books are "freely" available online, if you search enough. Or just buy the latest amended edition in print. As are some of the DAV safety commission reports (although only the later ones). Sadly, mostly in Deutsch. Though some of their more important findings have been published in English, as were ENSA's and others.

Schubert used to be both a great communicator and really good tester, by my impression. All his decades of safety engineering as the head of DAV safety commission and others point to that, even without invoking any "authority fallacy". If you are any interested in safety in climbing, it's still highly recommended. Even if some of the things he mentions in the very earliest editions have been amended since (the field constantly evolves, after all).

In reply to timparkin:

Fall into a sling from at or above the anchor, obviously. He didn't test at all for that (for some pretty obvious reasons), yet there are plenty of cases where that resulted in decking or injury. Which comes against the whole vibe and title of the video, even if he only tried the fall well below the anchor. Again, plenty of viewers would think that's perfectly safe, given his video. Even if he doesn't mention that exact scenario. It's not about what's being said, it's about what's not said.

In reply to timparkin:

I'll dig the answers out tomorrow when I have time on my office computer.

In reply to Frank R.:

Did you watch the same video as me?  His conclusion was (direct quote)
"Shock loading as a concept is not a myth, but is it dangerous?  Based on what I feel, literally in my body, when we did this experiment, I don't believe it is when you add other components into the system - like a dynamic rope."

In reply to Frank R.:

In reply to Summit Else:

I haven't watched the video, so may have missed some important context, and I recognise that 'shock loading' may itself not have a strict definition, but on the face of it I'd say with the addition of dynamic rope into the system it would be hard to continue describing it as shock loading. Indeed, that's precisely why tethers are now routinely made from rope rather than tape.

So is the above quote just saying 'shock loading isn't dangerous when you stop it being shock loading by using dynamic rope'? 

In reply to john arran:

Yes.

The context is the question:
Tying limiting knots in a sliding X sling anchor reduces potential extension if one piece fails, at the cost of reducing the breaking load of the sling by ~50%.  Is the benefit of reduced extension (and thus avoiding 'shock loading') sufficient to justify the reduction in ultimate strength?  [show your working]

In reply to jimtitt:

Or today!

There is no fundamental difference in equalisation between the four different test methods I used which ranged over the entire speed range from static to drop testing.

The sliding X is inherently poor anyway, using 8mm cord (other materials give different values) the load-split across the two anchors is 1.675:1 without the X and 1.89:1 with the X.

There are plenty of theoretical discussion on the effect leg-length has on fixed-point anchors BUT extensive testing actually shows we are so bad at tying them the stretch is utterly irrelevant. Marc Beverly's experiments with number of very experienced guides in perfect surroundings (indoors) tying 24 three-point anchors showed it is a lottery how equalised they are, the bad performers getting 6 results with less than 15% load on one anchor. McKently who did the same tests got his worst performer with a load distribution of 53%/10%/37%.

I also tested this with a number of random climbers, sometimes achieving zero load on one anchor. I did try pull-testing with 8mm cord and being stretchier makes no fundamental difference, the take-up in the knot being overwhelmingly greater.

In reply to timparkin:

Interesting thread. Having read all of this and the MP thread, I think only a couple of points have been missed.

One is the knot used to tie in with. High fall factor falls onto short lengths of rope have been extensively studied due to their use as 'cow's tail' lanyards in rope access and caving. With the right knots they can reliably reduce impact forces to manageable levels (~6kN in FF2). Fig-8s are very good whereas the worst knot in this respect is the bowline. Sewn loops do not do this at all - so the modern rope lanyards are little better than slings and need care (although I haven't drop tested them so don't know whether the adjustable element (eg Petzl Adjust) is reliably able to limit forces.

Second is the question of whether the rock or karabiner was responsible for cutting the rope, or whether it simply failed due to overloading. The test rigs required to induce a rope to fail in the middle of a loaded section are quite impressive - two large (~200m dia) smoothly threaded capstans with the rope wrapped multiple times and the threads adding extra friction. In any other situation the rope fails where it forms a tight radius, either in a knot or around a connector, and it fails from the outside inwards. The tight bend means the fibres running around the inside carry very little load compared to the outside, so those on the outside fail first and the the failure propagates inward. In this situation the pinch point against the rock provides a concentration of force for the failure to begin from -  so in the same way cams work on smooth rock you don't even need a sharp crystal (although it would help). You need a much tighter radius than a karabiner - an obviously sharp edge - to overcome this effect and instead cut outwards. So I think the karabiner can be exonerated here.

Edelrid have done some excellent research in recent years on rope failure over edges, the big takeaway is that resistance is not linear with force and decreases rapidly with higher loads. Similarly small decreases in rope diameter bring large decreases in cross-sectional area and hence significantly lowered cut resistance.

I haven't any data but I would expect it to be the same, unless your drop test starts with it not sitting neatly, in which case anything could happen.

Here's another - in a well equalised anchor (lets say three points in a W with overhand master), how much does leg length affect actual equalisation? Is a nylon sling a lot better with it's slight give? 

In my experience the leg with the least stretch takes most of the weight. So either the shortest leg or the one with more strands. Unless the angles are large or uneven. Yes, the more stretch the more equalisation, but assuming you mean nylon sling vs dyneema, not enough. Building the belay with long links from the climbing rope, with small angles, does work. Cordelette is better than slings but static not dynamic, and rarely long enough.

In a 'W' the centre is both the shortest and two strands, so will take the majority of the weight to the point where the outer two are probably taking very little. However they do of course provide shock-free redundancy in the event of failure of the central point, which is probably a better way of thinking about equalisation a lot of the time.

In reply to Adam Long:

Yeah that makes a lot of sense... I'd be really interested in some abstracted figures into the body as a loosely connected mass and how it reacts to various deceleration profiles. e.g. the layards above would be vary by the length of lanyway (one FF2 isn't equal to another FF2 as the deceleration distance would be different based on tether length). 

In some of the conferences I went to, bodies had been modelled with finite element and dyanamic force modelling routines. These were being used in car testing. It would be amazing to get a bit of time on them to say model fall distances from 20cm to 2m on static tethers. 

One of the really interesting results I saw on Ryan's modelling was that what feels like a soft factor 0.5 on a fair amount of rope was actually a lot harder than the same fall on a short amount of rope (still FF0.5). Presumably because the body has time to brace and provide it's own muscular stiffening frame. Hence the fall is on a stiff body - whereas a short fall hits the body without tension and hence has loostely connected floppy mass, less force but it feels worse because it hits sensitive parts of the body. 

...

Yeah, some of this I've seen in Ryan's tests and it seems like equalisation is a myth too (another of his clickbait titles). 

Thanks for the reply!

In reply to jimtitt:

That's great Jim and thank you very much - It's on a parr with some of what I've seen tested. 1.89:1 is pretty atrocious and it's also variable on intial conditions and load angle. Nice to know I don't need the nylon slings I initially thought might be useful for this. 

Is this documented anywhere? (the sliding X in particular). I was surprised that drop testing wasn't worse. 

In reply to timparkin:

It's a complicated subject! No it's not published, it was a project funded by a climbing equipment manufacturer and the data belongs to them, I can release some of it but the rest is theirs, the only person who got my wrote-up was rgold (who posts here) because a) he understands it and b) because he was doing the technical stuff for a new edition of a well-known American book on anchoring which got it mostly wrong. It took about 18months and hundreds of drop and pull tests to finally get to grips with the whole theme.

I used four basic test methods depending which was most convenient or applicable and naturally compared them initially because particularly drop tests are time consuming and expensive, the simplest is a slide test where you set up the anchor with a suitable weight (I use 20kg) and slide the masterpoint well to the side and let it go then measure the relative forces on the anchors. Another is set up the anchor the same but on a backboard which rotates then tilt this until the master karabiner reaches equilibrium, these are easy as you don't need computer time to intepret load-cell readouts, just a simple vector measurement. The third way is drop and slide so a simple combination but dynamic.

Full-on drop tests are another game altogether, lifting a 90kg block up and dropping is work (even with a powered winch) and destructive and gets doubly so when you start working on three-point anchors with one point failing, you are lucky to get three runs in on a Sunday afternoon before it's time to pull out the data from three strain guages and open a beer. And stuff breaks, both McKently and I got to scenarios where brand new slings broke on impact (the DAV as well if I remember rightly). But they reveal the fundamental flaw in the self-equalising concept, if you build a three-point anchor and one point fails then the other two equalise and share the load. Or they don't. If the failure was the karabiner this might be true, when it is any other part the karabiner and it's associated junk goes down to the master point karabiner and locks it solid, kaboom that was the next in line taking the full impact.

Which raises the philosophical argument which proved to be the death of the whole equalising argument. The climber (with the exception of battling for your life in a snowstorm near the top of the Eiger N Face) choose to belay or not. Since the end objective is to build an anchor which will hold the worst case but we know that they will be lucky to get within roughly 60% load sharing and from other research that climbers judgement of cam/nut placements is notoriously poor (otherwise we would only use one anyway) then basically we are just hedging our bets, adding in the complication of extension when one fails with an unknown increase in the force on the remaining piece(s). Which begs the question of why are you building a belay why relies on guesswork and unknown physical principles. So back to two two independent points, both hopefully capable of holding the full force and forget the macrame, you can't make a silk purse from a sows ear.

In reply to jimtitt:

Exactly. You would hope that any one of the pieces you put in as an anchor would hold in the case of a fall, but if any one of them didn't, you'd want to give the others the best chance of doing so by not subjecting them to additional load due to the first one popping.

In reply to jimtitt:

Yep - my conclusions exactly from available data. Thank you very much for sharing!

In reply to Adam Long:

That's interesting. Thanks.

But you are suggesting that the carabiner crimping down on the rope against [even quite smooth] rock could be enough for the rope to fail.

Which is what the accident investigation team proposed.

And that is still pretty scary. Irrespective of whether the carabiner surface itself or the rock did the "cutting".

After reading through the article and this thread, I have learnt little new apart from some interesting details of Jim's testing.

Sorry if this sounds harsh but its standard advice and knowledge that

Ensure your rope runs cleanly through quickdrawers both when climbing and in a fall so as it not be trapped against the rock. 

Fall factors more than 2 are possible if the rope traps behind a flake or spike, though the probable resulting rope cut may be caused either the high fall factor forces and/or simple slicing.

I am.disappointed that the mods have not thought to edit the thread title, as others have pointed out it is extremely unlikely that the krab was to blame, just its position was unfortunate. So near krab not by krab.

In reply to Rick Graham:

Not everyone learns the above starting out so it's worth repeating as you have done. I would add that on starting out on trad (and easy sport tbh), it is best to go by the old 'the leader must not fall' rule, then adapt that as you get more experience.

In reply to Rick Graham:

Apolgies Rick, I think there is considerably more here than your simplistic interpretation.

Do we think that the climber did not ensure his ropes were running cleanly?  Do we think that there was any way he could have known or expected that his rope would get pinched?  I think that the mechanism was unprecedent, and therefore not as easy to mediate against as you suggest. 

Some people have suggested that the rope was not pinched by the carabiner, but trapped in some kind of V groove lower on the route, or just by friction against the rock, but there is no evidence of that at all.  There is no evidence of any edge that would lead to "simple slicing."  And it would take a huge amount of force to cut a rope in the middle without some sort of mechanism.  So the proposed explanation that the rope cut when it was pinched under the carabiner makes the most sense.

I would say "under the krab" might be the most accurate, but really all the fuss about the word "cut" is pedantic word parsing.  I think thaat the language used in the official report makes sense--the rope was cut by the carabiner or cut when it was pinched under the carabiner. 

Did the cross section of the carabiner play a part here?  We don't know.  More testing would be great.

Do we think the same accident would have occured with an older style, larger diameter, 10 mm, smooth, rounded carabiner? 

My guess (and it's just a guess) is no.  But again more testing would be interesting.

Shout out to Ryan: "Do you want to pull test some ropes pinched under carabiners with varying profiles?"

In reply to Rick Graham:

While this might be the case, in the original post I just went from the title of the AAC article (and indeed my understanding of what they say in the article): "Fall On Rock | Carabiner Cut Rope

Monongahela National Forest, Seneca Rocks"

In reply to brunoschull:

My guess is probably yes. A thicker rope would be a more likely route to a better result.

I made the point that the karabiner should not be blamed because I think your point is exactly the wrong takeaway here. I've got lots of quickdraw krabs, some over twenty years old, and they all fit the forged curved 'T' profile to some extent. My only krabs which are fit a significantly 'smoother, rounder' profile are HMS belay krabs. You would struggle to go to a shop and build a rack of quickdraws with krabs that offer any certain advantages to the ones used here. Whereas it's very easy to buy a thicker rope, or better, double ropes.

For me, the big takeaway here is to be very wary of using skinny singles on complex trad terrain.

In reply to brunoschull:

My guess is probably yes. A thicker rope would be a more likely route to a better result.

I made the point that the karabiner should not be blamed because I think your point is exactly the wrong takeaway here. I've got lots of quickdraw krabs, some over twenty years old, and they all fit the forged curved 'T' profile to some extent. My only krabs which are fit a significantly 'smoother, rounder' profile are HMS belay krabs. You would struggle to go to a shop and build a rack of quickdraws with krabs that offer any certain advantages to the ones used here. Whereas it's very easy to buy a thicker rope, or better, double ropes.

For me, the big takeaway here is to be very wary of using skinny singles on complex trad terrain.

In reply to Adam Long:

I think you posted twice, Andy?  Maybe delete one?

Your suggestion (use a thicker rope) makes sense. 

Personally, I really don't know if the carabiner played a role here, but I'm going to go with "Yes," it did affect the outcome, and "No", this would not have happened with a round stock, larger diameter carabiner. 

Just to be clear, I'm not suggesting that everybody should run out and buy HMS carabiners for their quickdraws *that would be silly) nor that the Petzl Spirit is poorly designed in some way.  Most of my carabiners are DMM Alpha Trads, but I think I have some Spirits mixed in there, and I certaintly won't throw them away.  For many applications I'll also continue using my Attache locking carabiners (which also have a bit of a T or I profile).

So, my point is not that the carabiner "caused"this accident, but that the carabiner might have been one factor among many that contributed to the very rare "perfect storm" of events. 

In reply to brunoschull:

So what have we learnt? I've read most of the thread and as far as I can tell, we don't actually seem to know what happened in the incident beyond the fact that a rope broke.

The rest of your post is just questions, not things we've learnt. It would be good to know what did actually happen, but sometimes it's just not possible to know exactly what happened.

In reply to Fellover:

I have to say, if this thread is ment to be a shining example of the benefits of analysing fatal accidents it's an absolutely crap one. Nearly 200 posts in, a metric shittonne of pretty baseless speculation  a YouTube video and an AAC report, no one seems any the wiser as what caused this incident. 

I know it's not a popular view but I really do think a lot of this is frankly glorified rubbernecking.

In reply to ebdon:

As i said early on upthread, the issue here is that there has been insufficient investigation to conclusively state a failure mechanism, and the conclusion reached in the report is insufficiently justified. Had the report been more thorough, the need/desire/want for answers would not have lead to so much speculation.

In reply to ebdon:

Au contraire!  The very act of discussing this has made us all more aware of climbing and risk than we would be otherwise.  The lessons in this thread go way beyond this accident.

In reply to brunoschull:

Really? So what we have learnt is climbing is a dangerous activity? Quelle surprise.

In reply to brunoschull:

Mais Oui 

plus ça change plus c’est la même chose 

In reply to TobyA:

Re several posts back.

Climbing isn't supposed to be a dangerous activity. That's why we have quality gear, good practice, standards and thorough testing.

The rope in the accident was 9.5-9.7. Not really a skinny single these days. I've been using a 8.7 quite a bit.

I'm grateful for this thread and the parent MP thread.

Please keep it up. If some people don't get anything out of it, they should stop reading it

In reply to galpinos:

It’s a shame the rope was not sent to a lab for analysis. The degree of wear and whether the fibres were chemically compromised in any way would be relevant to drawing a meaningful conclusion.

In reply to BruceM:

'Climbing isn't supposed to be a dangerous activity.'

It's not 'supposed' to be anything. It is though intrinsically dangerous; and however much we might 'invest' in good gear, or technical knowledge, it will remain so.  If people want the safest possible version they're stuck with climbing walls.

In reply to mrjonathanr:

Also inspection of the “cut” fibres. The state of the end of the fibres would get us the failure mechanism which could then give us better insight into the cause. 

In reply to Rick Graham:

Rick,

While I agree with most of your post, I took something away from Toby's post and the attached articles.

I've climbed for many years and always been wary of clipping gear where the krab might lie on the rock so I extend it . However on a few occasions when I've been 10- 20 feet above the runner and looked down I've realised that the krab was sitting on a slab because either the sling wasn't long enough or the line I took changed the rope direction. So I've created a dangerous situation if I fall. Nothing new there!

What I hadn't fully realised in my pursuit for lighter gear was that how sharp the H cross section is on some of my krabs.  I had never really considered it. I'd always focused on load strength.  When I read this I replaced the krabs on half a dozen of my slings with older krabs with a much more rounded cross section. It doesn't mean that the rope won't be damaged in if I fall, but it seems like a simple precaution that could increase my chances.

cheers

Phil

In reply to biggianthead:

Is there any factual evidence that the H profile is actually sharp enough to cut a rope?
Has it been tested?
A simple test would be to put a karabiner and rope in a vice and slowly close the vice to see if the karabiner cuts the rope.  Not very scientific, but would give a good initial indicator.
The above test could be improved with the addition of force gauges, etc.

In reply to biggianthead:

I keep thinking that the risk could be reduced further if we had access to carabiners with double the thickness spines and gates, so the rope end horizontal section was raised off the rock by 6mm+.

You could have a couple of quickdraws just for those slab above drop situations. Because that's what I see at all my sport crags a lot.

Heavy, but could be a lifesaver. And you only need one or two. Like a screamer, or pulley crab, just for certain situations.

In reply to ebdon:

Quite the reverse. There have been many interesting posts relating to the testing regime and stds we take for granted. I, for one, will be thinking much more carefully about any situation where the rope is near an edge. The thread hasn't been focussed just on the original incident which has only made it better. You have a very different definition of rubbernecking to me.

In reply to BruceM:

That sounds like a job for a clip on bit of plastic. I'm sure someone with a 3d printer will knock you up a prototype. It'll make millions I tell you!

In reply to Holdtickler:

Somewhat rarely, I got zero feedback on proposing the possibility of the sharp edges of the gate hinge could have played a part. I can't imagine that it would be very likely but with the mystery of the accident, I think the other sharp edges should be at least considered. 

Is there a rare situation where a krab could be loaded that way?

In reply to FactorXXX:

dislike from me

Your experiment won't tell you much, because the strength of polymers is highly dependent on the rate at which strain is applied, which is why the gold standard for climbing gear is drop testing not tensile testing.

In reply to Toby

With regards to incident reporting, it hasn’t really gained traction here in the UK, I think for a variety of reasons - a missed opportunity to help keep people safe. If we didn’t have the capacity to lean from mistakes we wouldn’t be here as a species. 
 

But get this: The Association of Canadian Mountain Guides had to withdraw it’s professionally mandated incident reporting system on the instructions of their insurers - lability concerns of course. And it’s subsequently happened here in the UK with at least one professional association, which is just staggering beyond belief. 

 

In reply to Rich W Parker:

an example from another potentially hazardous sport governing body, the British Hang gliding and Paragliding Association, who actively encourage pilots to report any incident with potential for others to learn, including near misses/equipment issues etc. Serious/fatal incidents are dealt with separately by formal investigation by the BHPA as required by the Air Accident Investigation Branch.

https://www.bhpa.co.uk/safety/incidents/

https://www.bhpa.co.uk/safety/investigations/

In reply to Jon Greengrass:

The test I suggest is simply one that effectively crushes a section of a standard climbing rope with a particular karabiner.  Totally static and with no elongation of the rope along its axis involved.
If the proposed failure mode is correct, then it should be relatively easy to ascertain if a modern profile karabiner is more likely to cut a rope as opposed to a traditional one.
If deemed necessary, The test could be expanded by doing the same test with tension applied to the rope and using force gauges to measure the compression force applied to the rope.
The important factor in all of these these tests is comparing various profile karabiners using the known same test conditions.  It's worth remembering at this point that the propsed failure mode categorically stated that the profile of the karabiner played a significant role in the failure.
Not saying my proposed test is perfect, but do enough of them with known variances applied to all of the karabiner profile types and I reckon you'll quickly build up a picture as to the validity of the proposed failure mode.  

In reply to Tom the tall:

When it comes to incident analysis, I think models to aspire to are on-line video case studies published by the Air Safety Institute.  Many of these reports have things of interest---in a general way---for climbers and others who have to make decisions in risky terrain.  For an example, have a look at youtube.com/watch?v=8PBUVMCbmFQ&.

In reply to FactorXXX:

This is where the definition of "cut" and "edge" becomes problematic. Let's say an edge is something with a certain radius at it edge (anything can be an edge in this case although I would say you probably need <40 angle and a radius of 5mm or less for most people to accept it as one)

1) you have just pressure cutting, i.e. a flat item (rope) on a surface with the cutting edge 

2) you have the bending of an item over the edge (i.e. rope changing direction over the edge)

I suppose if you hard pull a knot, even the rope could be considered an edge as it forces a small bend radius on the other bit of rope in the knot and enough force causes it to break.  

So in the example, I think the idea is that you have two potential things happening

1) the rope getting trapped between the carabiner and the rock

2) the bend radius of the carabiner

The first is where the potential T or I section bar might have an effect. Personally I can't see the amount of force applied by the carabiner to the rock being that high - even if the rock was at slabby 30 degrees, the force applied would be half of the fall force at most, and that's presuming the carabiner was exactly at the edge of the rock. If the carabiner was above the edge, the edge would limit the forces applied substantially

The second is your standard "pull hard on a rope around a carabiner and at some point it will break".

The fact is it's a really freaky accident and the probably learning is to think about how the carabiner will sit if you take a fall - "'ware the edge"!

In reply to Offwidth:

In my case  it's an opportunity to discuss testing with some people who really know what they're talking about. The fact is the accident has been reported on so no 'better' information is really going to arrive. Given the potential questions raised from the report I think it's beneficial to see how it could be interpreted. 

There's also the old adage that "if you want to good answer on the internet, publicly state the wrong one and an expert will quickly arrive for you (thanks Jim, rgold frank)

The small print runs generate massive prices. Perhaps if the establishments wanted people to know more about accidents, they would make a cheaper online only version available? 

In any academic speciality, there's a hierarchy of summarisation that is needed as the reader is less and less familiar with the intricacies of the core material. There are few people who can appreciate the maths, physics, material science, etc. to make the most of the raw source material. So quite often it's useful to be able to get a 'rough' summarisation of these details. 

Plus, like with Jim's excellent comments above, the raw data is proprietary but the author is free to share summaries. You can't buy that online, you have to engage with the author when you have a chance. 

In reply to timparkin:

I was replying to what bruno said more generally about UKC accident posts and as such you are selectively quoting me completely out of context.

https://www.ukclimbing.com/forums/rock_talk/rope_cut_by_karabiner_-_analysi...

As for what you do say... the likes of Jim, rgold, galpinos, Adam and Frank don't always reply, so occasionally complete b*ll*cks goes unchallenged and often threads just don't leave a clear message. When I had one of my first major rants about UKC dislikes (for contra-indicating high quality posts), one of the things that triggered me was more people disliking than liking an excellent informed technical safety post of Jim's.

The John Dill link is online, short and written in plain language and it's vital messages are important way beyond Yosemite, but are still widely ignored.

http://www.bluebison.net/yosar/alive.htm

In reply to timparkin:

One should note that Pit Schuberts (he unfortunately passed away this year) books are an extended compendium of over 200 articles which were published in the DAV magazine and the safety-orientated publication issued by the combined German/Austrian/Swiss alpine association which are all available at no cost online or in the DAV archives. The books themselves aren't expensive at €29.99 per volume and were a private project. He also wrote a number of books and pamphlets which are part of the coursework for a number of qualifications. We discussed translating his books into English about ten years ago but didn't see enough potential interest to justify the expense, the American market is difficult.

Personally I don't involve UKC threads too much, most information I post on Mountain Project (earlier on RockClimbing.com) because they allow graphics and illustrations embedded in posts, I also don't particularly like repeating the same stuff time after time, I put it out there once, not every two years for lazy people who can't search properly.

This is one of the problems Ryan has, he says it's never been tested but reality is a) it has been done and b) he didn't bother to look. One of his better efforts was when he first built his drop tower and wanted to look at the difference in breaking force of karabiners between pull-tests and drop-tests (because pull tests aren't "real world). He was understandably miffed when presented with a vastly more accurate set of results, the reason why they are different and why the standards use pull-testing, the testing itself being freely available online for anyone capable of searching the internet for more than two minutes. Along with a whole row of other interesting research into climbing equipment. His other problem is with proprietry information, the manufacturers have immense amounts of testing and research data pn a level he couldn't cope with and they keep it, put in public it would anyway be misinterpreted, they would have to spend inordinate amounts of time explaining and defending it and also did the research to use it for their commercial advantage.

There is a wall between the manufacturers and the internet forums and this was built by the forums themselves, when (as was the case of probably the last industry person to post on UKC on a technical matter) the forum accuses directly a senior management figure of outright lying so they sell more products they stop communicating. Peer-to- peer discussions over a few beers at a trade show is where it all happens (and they can get more hard-core than you'd imagine, the industry has a lot to lose and have no qualms in telling another company their product is crap and please take it off the market). It is different if someone from the more traditional media channels get involved, they have years or decades of experience, might have at least a grasp of the concepts involved and know if they get it wrong they don't get invited back, a half-wit on the internet only wants the instant gratification of three likes on a forum.

In reply to jimtitt:

Thanks for taking the  time to write that long informative  post. 

What a lot of climbers possibly do not appreciate is how much free time goes into all the preparation for the standards that have made climbing gear as safe and fit for purpose as it is today. Also modern best practice recommendations in how to use it.

I think its possibly fair to state that  a lot of climbing hardware manufactures subsidise the cost of gear with higher mark ups or overall profit on industrial product.  In other words,  climbers are getting it on the cheap.

In reply to brunoschull and others: I had a look at some Petzl Spirits today - I don't own any but the bloke I was climbing with had them.  I do have some Alpha trads, phantoms and others. and I would say the spirits were amongst the most round for internal bend, and the I beam shape was less sharp for the 'side impact' scenario than the alternatives.  I certainly wouldn't thrown them out

In reply to Offwidth:

Can you at least explain the relevance to what you/we were talking about as if I'm going to read 8,500 words it would (which I have before but haven't memorised) it would be nice to know what I'm looking for.

p.s. A generalisation that "The internet response immediately following such tragic but unclear events is a human one but it's effectively rubberknecking." doesn't really need much context (and it didn't seem to help when I reread it) and it effectively reduced all such conversation to obscenity. Whether your intention was to shut people up because we're all rubbernecking or not, it felt like it.

In reply to FactorXXX:

The problem is, UIAA and ÖEAV plus DAV & ENSA already tried to do such a test, scientifically, back a decade or two ago. Any micrometer difference of the edge surface lead to wildly different results.

Only to find out that any micrometer or even nanometer differences of their very specifically designed and very accurately produced artificial edge test specimen didn't really work, as the edge accuracy simply wasn't same enough among all the different labs.

As in the very same rope tested in one lab passed the edge cut test, but flunked it in another, and vice versa. Not really what you want as a certification body.

Turns out testing edge cut resistance of ropes with enough repeatability is really difficult. Current UIAA Safety Commission members active here here might add to that history, but it wasn't really simple at all...

In reply to Frank R.:

I did mention it up thread but may the current leader of the UIAA Safety Commision working group for a new cutting/abrasion resistance test.

Suffice to say, it is not an easy test to devise to meet the repeatability and reproducibility required, but also ensure that test is a suitable approximation of the real life scenario.

In reply to timparkin:

Threads can take unexpected directions.

As I said I was replying to what I saw as an unfair claim about UKC accidents threads in general (albeit true sometimes specifically in some posts). I really think immediate discussion following an accident when based on a largely unknown scenario gains us little as a community but can harm loved ones of the victim (something I've painfully witnessed). I wish speculation wouldn't happen but I have no power to stop it beyond urging caution in such circumstances. The press at times are guilty of shockingly bad behaviour in such circumstances, that would be deleted by moderators here.

In contrast, where those involved give first hand evidence immediately following an accident, we are in a different situation and I think most here welcome that, and discussion on it (I certainly do). Equally, after time taken to analyse an accident, I also think most here welcome discussion (as do I). Hence, I think bruno was wrong to generalise but I can understand his concerns.

In regarding 'rubberknecking' comments as a potential 'obscenity' you seem to be failing to understand that is at least for most of us a hard-to-change result of our psychology. It's true that wintertree puts a fair case for individual exceptions for some, but in my experience he is unusual .... he certainly seems a better man than I.... I've certainly been guilty too many times on a motorway.... and because of this I have to remind myself to focus on paying attention to my driving situation. Focus is important as the distraction can and sometimes does lead to accidents on the other carriageway. 

Back to Dill's summary report of analysis: it shows us failing to recognise our psychology can get us into trouble. It is why those 80% of easily preventable yosemite deaths is such a hard situation to change. Across all accident analysis I'm aware of, it is what kills most experienced climbers. I repeatedly copy it here in the hope some will take actions that might help reduce such risk: buddy check more rigorously; always double check before an abseil and tie knots on the end of abseil ropes; remind ourselves to stay fully focussed on easy movement on highly exposed terrain; have the right equipment for unpredicted but not uncommon occurrences (especially emergency clothing where weather can change quickly).

Now back to the main theme of this thread we have a specific technical safety question in an unusual situation where sadly a fatality occured. I think the analysis linked in the OP is made in good faith but the conclusions are far from confirmed (for reasons given by those who know this area much better than I do). I welcome this thread as I think most here will. I'd encourage people to engage on this thread; even if, on other threads when information is much less clear, I would hope people would think carefully about what they say.

In reply to jimtitt:

I think this is one of the things that I (and I imagine Ryan) find really annoying. That's essentially discussion behind closed doors and I'm sure it happens in every industry (not blaming you or the people involved), but I want discussion in the open where I can read it/watch it, because I'm interested.

Ryan seems like the only person doing very open testing (admittedly not very 'well' from a scientific pov), presented in the English language. There's a market of people out there who are interested in that and companies/any English language group have failed to give those people anything.

Before it gets brought up, I am not interested enough/don't have enough free time to learn German and therefore access DAV stuff, sorry.

In reply to Fellover:

That is the function of the climbers associations, some do it better than others.

In reply to Offwidth:

I am with Offwidth here. This 0.1% (if that) incident has generated a huge amount of discussion, precisely because it so unusual. But the 80% is where the most stands to be gained (and most easily gained) in reducing injuries and fatalities. That isn't to say I won't look twice at how my draws lie against the rock, but things like buddy checks are likely to have much greater payoff.

There was a great thread on near-misses on MP recently (to which I was able to add a couple of examples) which really brought this home, and brought home how close many of us have probably come to adding to these statistics.

In reply to Fellover:

With a little patience and the knowledge of what the climbing terms should be, you can do very well with Google Translate.

In reply to Darkinbad:

I'd add using half ropes on complex terrain (for redundancy, reduced rope drag and improved escape options) and making sure any extension is well clear below edges (inc. allowing for any subsequent sideways movement) are to me pretty standard safety precautions I've known about from early in my climbing that may have prevented this accident (and any need to worry especially about the unusual scenario).

Some US trad approaches always looked very odd from my UK trad background: climbers carrying a second rope on their back up wandering climbs for rapping off; often carrying way too much gear for routes well within their grade limits; linking pitches inappropriately when we might split pitches for communication (where the leader or second might need it). With all that extra weight and inevitably worse rope drag it's not something I could ever change to without massively dropping the grades I wanted to climb.

In reply to Offwidth:

Totally agree with "US trad approaches."  I think the prevalence of granite and desert sandstone climbing has a lot to do with the prevalence of single ropes, that together with translating big wall techniques to other climbing contexts where the big wall approach makes little sense.  The silly fad of ever-longer pitches using ever-longer ropes and all the hassles and drawbacks of carrying a tag line for full-length rappels rather than climbing with twins or half ropes are as perplexing to me as a US climber as they are to Offwidth.