Page 1 of 1

SRT kit - deadly accident in Switzerland

PostPosted: Jun 27, 2013 11:23 am
by paul
There's a new threas on UKCaving.com about an accident involving SRT in a cave in SWitzerland.

It appears that the cause of the accident was due to clipping into the the hand jammer using two snaplink karabiners.

See http://ukcaving.com/board/index.php?topic=15351.msg200407

Re: SRT kit - deadly accident in Switzerland

PostPosted: Jun 27, 2013 12:12 pm
by Scott McCrea
I don't read French or German, but looking at the pics and reading the UKCaving post and running the scenario thru my head, here's my best guess as to what happened. At the traverse knot/anchor, the handled ascender with the two, chained biners was moved past the knot. During this move, the biners got crossed-up. When she stood on the footloop to unload the short cowstail, the long cowstail disconnected, leaving her totally unattached.

As I have said often, simplify your gear. Get rid of "links" in your gear chain. The more link, the more things that need to be attended to and can go wrong.

Re: SRT kit - deadly accident in Switzerland

PostPosted: Jun 27, 2013 12:56 pm
by paul
Here's a translation of the probable cause of the accident:

"To pass the knots on a traverse line, the victim was putting all their weight on their footloops, thereby immobilizing the upper carabiner (one that connects the footloops to the hand jammer) in a vertical position. The second carabiner, on the cowstail is then unloaded and can move freely while remaining connected to the first carabiner.It is assumed that during the transfer of weight on the long runs, the second carabiner is positioned so as to open the gate of the first carabiner and is thus itself disconnected.

The victim disconnected the cowstail, and let out a cry of exclamation. They had neither the strength to hold on nor could reconnect the cowstail and dropped back. The loops on the footlops were tied too wide for they remained attached. Using a snaplink carabiner and no additional connector between the jammer and cowstail, has probably contributed to the fatal fall.

The picture (Figure 3, "Apparatus of the victim") shows the assumed original equipment configuration at the time of the accident. The offending carabiner gate offered a resistance to a relatively small opening, and this could exacerbate an already unfavorable combination."

In other words, because the caver had connected their footloops to their jammer using a non-locking carabiner and then (as is common in France and other European countries to use a long cowstail as a safety link from harness to hand jammer rather than a dedicated permanently attached safety link) clipped their cowstail carabiner straight into the first carabiner, when they were standing in their footloops with the jammer either attached directly to the traverse line or connected using the first carabiner, this carabiner became detached from the traverse line and that was their only form of attachement, So when the caver couldn't hold on, they fell.

The PDF document goes on to say:

"Conclusions and recommendations.

In the light of this fatal experience, we issue new recommendations:
You need to connect the cowstail directly to the jammer. For this reason most of the models are equipped with two holes for connecting separate footloops and safety link, according to the image in Figure 4, "jammer with two holes and maillon rapides."

Alternatively, as with many cavers here in the UK, have a separate dedicated safety link from harness central maillon to a locking carabiner or maillon rapide on the jammer together with footloops. Then you can use the long cowstail to clip in to the traverse line, just in case, when making moves on a traverse line where you must hang on it due to a lack of a ledge to stand on and want to stand in your footloops to aid progress. Of course having a dedicated safety link from harness maillon to hand jammer rather than just clipping in a long cowstail, prevents accidentally dropping the footloops and jammer.

Re: SRT kit - deadly accident in Switzerland

PostPosted: Jun 27, 2013 3:54 pm
by Extremeophile
It seems any of a number of modifications in gear and technique could have prevented this tragedy.

1. Connect the long cowstail biner directly to the upper ascender rather than the footloop biner. It's common knowledge in climbing to never clip non-locking biners together due to the risk of auto unclipping.
2. I use auto lockers on my cowstails. In this scenario it wouldn't have mattered, but if the long cowstail were clipped directly to upper ascender this would provide some extra security from coming unclipped. There is a greater risk of this problem on traverses.
3. Always maintain two points of contact for everything other than descending. This has been hotly debated in other threads, but here's yet another example that relying on a single point of contact while traversing, ascending, crossing rebelays, crossing knots, changeovers, etc. does not provide enough of a safety margin.

Re: SRT kit - deadly accident in Switzerland

PostPosted: Jun 27, 2013 4:18 pm
by Scott McCrea
Extremeophile wrote:3. Always maintain two points of contact for everything other than descending. This has been hotly debated in other threads, but here's yet another example that relying on a single point of contact while traversing, ascending, crossing rebelays, crossing knots, changeovers, etc. does not provide enough of a safety margin.

That was my first thought, too. But, I am imagining that she had two POA. It's just that the long tail came undone, unknowingly, when she stood to remove the short. I don't think this was technique problem. Well, other than using the biner-to-biner technique.

Re: SRT kit - deadly accident in Switzerland

PostPosted: Jun 27, 2013 5:13 pm
by Extremeophile
Scott McCrea wrote:
Extremeophile wrote:3. Always maintain two points of contact for everything other than descending. This has been hotly debated in other threads, but here's yet another example that relying on a single point of contact while traversing, ascending, crossing rebelays, crossing knots, changeovers, etc. does not provide enough of a safety margin.

That was my first thought, too. But, I am imagining that she had two POA. It's just that the long tail came undone, unknowingly, when she stood to remove the short. I don't think this was technique problem. Well, other than using the biner-to-biner technique.

It does seem that there was some coincidence in the timing of the long cowstail coming unclipped and removing the short cowstail in the process of crossing a knot. My point is that hanging from the short cowstail while clipping the ascender past the knot and standing in the footloop while clipping the short cowstail past the knot are both brief instances of not maintaining two points of attachment.

Re: SRT kit - deadly accident in Switzerland

PostPosted: Jun 27, 2013 5:52 pm
by Scott McCrea
So, are you saying we should carry three POA, so we can maintain two when passing a knot?

Re: SRT kit - deadly accident in Switzerland

PostPosted: Jun 28, 2013 9:22 am
by erin85
Having been caving in France a lot recently, I can tell you that almost no one uses locking biners to connect their ascender to their long cowstail. I have never seen it done, ever. It's because you use both cowstails very frequently, for a lot of moves that are not unknown but fairly exotic to the American crowd, and if they lock, it takes too long to manipulate them. Usually the only locking biner in the whole system is the one connecting the descender to the harness. When doing vires (horizontal traverse), you basically have one point of contact the majority of the time, it's just the way it's done, and no one seems to have a problem with it here. It's extremely easy to unload both cowstails, and unclip them both if you're not paying atttention. I think her error came from having the cowstail clipped into the footloop biner -- those should be independantly attached to the ascender, in my opinion. Dedicated saftey lines would be good -- been meaning to add one to my footloop at least, as you move that around a lot too in this style -- but quicklinks and auto-locking biners don't fit into the technical style over here.

Re: SRT kit - deadly accident in Switzerland

PostPosted: Jun 28, 2013 11:49 am
by Extremeophile
erin85 wrote:Having been caving in France a lot recently, I can tell you that almost no one uses locking biners to connect their ascender to their long cowstail. I have never seen it done, ever. It's because you use both cowstails very frequently, for a lot of moves that are not unknown but fairly exotic to the American crowd, and if they lock, it takes too long to manipulate them. Usually the only locking biner in the whole system is the one connecting the descender to the harness. When doing vires (horizontal traverse), you basically have one point of contact the majority of the time, it's just the way it's done, and no one seems to have a problem with it here. It's extremely easy to unload both cowstails, and unclip them both if you're not paying atttention. I think her error came from having the cowstail clipped into the footloop biner -- those should be independantly attached to the ascender, in my opinion. Dedicated saftey lines would be good -- been meaning to add one to my footloop at least, as you move that around a lot too in this style -- but quicklinks and auto-locking biners don't fit into the technical style over here.

I'm intrigued by these exotic moves. I'd like to learn more.

I don't agree that having a single point of contact on traverses is "just the way it's done". Obviously the victim in this incident had a problem with this style of caving, and even though the primary error may have been clipping the long cowstail to the footloop biner, maintaining a second point of attachment would have prevented that error from becoming fatal.

I also don't agree that the combinaiton of locking biners with Euro-style rigging takes too long. There are biners available that are small, have key-lock gates, and simple auto-locking mechanisms. These add a level of security, and your own argument about the ease of cowstails coming unclipped supports their use.

Perhaps a single point of attachment and the use of non-locking biners is commonly practiced in Europe, but there's a growing pile of dead bodies to support changing these conventions.

Re: SRT kit - deadly accident in Switzerland

PostPosted: Jun 29, 2013 9:11 am
by chh
I use a very similar system, though the carabiners on my footloop and on both ends of my cowstails are lockers. That being said, I rarely lock them, and usually not when doing more "exotic" movements. Erin85 is right, I often forgo this step for speed. I do, however, lock them when I think I should. There isn't a set of precise circumstances for this, just when I feel it is necessary. When traversing I also try to keep my short cowstail loaded before removing my long cowstail and placing it around a knot or bolt. I also try to avoid using my ascender on traverses if the angle at all allows it. FWIW, I would clip both my cowstail and footloop into the body of the ascender if my ascender had two carabiner sized holes. I use a Basic, so it doesn't. If I need faster work positioning with my upper ascender and cowstail and am not using my croll, and my shorty is involved somewhere else, I will also sometimes clip my long cowstail around the rope above the body of the ascender like a changeover. This keeps me into the rope directly with a single link, even if I need to remove the footloop with it's carabiner or even the ascender completely.
Anyway, there are a bunch of ways to make most any system safe with some thought and a lot of familiarity.
Most of the arguments for or against most SRT setups seem to revolve around regional differences. I believe this, more than anything, contributes to accidents more than a particular permutation of a frog setup. A caver learns one way to do things that is "correct" or "safe". Complacency sets in, and they believe they are safe when they aren't. Brains not dogma.
And now, I will release my inner nerd with a quote from Alastor "Mad Eye" Moody....
"CONSTANT VIGILANCE!"