UKC

Thought this might be of interest, given the popularity of autobelays. 

https://www.theguardian.com/australia-news/2021/oct/13/man-dies-after-falli...

Initial suggestions seem to be of failure of the device itself rather than the more usual user error.  Hopefully some more definitive information will appear sooner rather than later.  Are SafeWork a popular autobelay make?

I'm guessing there isn't any information on the relative safety of autobelays compared to human belayers.

The 'SafeWork' mentioned in the article is a government department in NSW investigating the incident.

The climbing centre's instagram page reported 3 days ago that it was auto belay failure, and they've removed all auto belays and will never use them again in that centre.  No mention of the brand of device though.

Scary stuff... 

Ah - I'd read it as the manufacturer, despite several readings of different accounts. Good that it's an independent body.  Use auto belays a lot, and whilst I'm not going to sweat one incident, it'd be good to know the manufacturer.

I wonder if anyone has any insight into whether autobelays are safety checked and if so how often? At the end of the article is states:

"The man’s death comes just two days after the gym was allowed to reopen after the easing of Covid-19 restrictions in NSW. "

It doesn't seem totally unlikely that in the faff of re-opening perhaps a check was missed or perhaps a service of the equipment was delayed?

I believe they are serviced and checked annually (in the UK at least), I saw a social media post from a wall the other day saying as much (basically warning that there are no auto belays while they have their annual service)

Yeah there's qyuite a few companies in the UK who carry out installation and servicing, looks like it's pretty mandatory to maintain certifications https://trublueclimbing.com/service-my-device/annual-service 

I'd guess auto belay failure is a less common cause of accident than (human) belayer error, so don't let this put you off.

Well, at my local wall all of the auto belays were taken away and serviced recently. I will continue to use them, the problem with global reporting is it reports accidents from across the world where different safety rules may apply, also no sense of scale of issue, how many and how often are they used without incident? Also it would limit me to just bouldering some visits!

I would hope they are checked more frequently than annually!

I would imagine serviced (at least) annually, but I would also expect they would be checked daily from the ground and something like weekly for a a close up visual inspection at height.

Would be interesting to hear if this is something like the standard regime at most walls from people in the know.

Any insight on the dislikes? There's plenty of potential reasons for delays to get inspectors/service engineers out: social distancing/virus spread limitation by minimising visitors on premises, backlogs because of a sudden surge in demand after restrictions ending, staff having 50 other things to do when re-opening, human error with dates shifting etc.

I mean even MoTs got an extension and likelihood of injury/death on the road as a result of not inspecting vehicles is considerably higher than failing autobelay.

I imagine the first thing any investigation will turn up is whether any checks and/or services were missed or pushed back.

No dislike from me, but i don't think it is helpful at this time that your were speculating checks could have been missed or servicing delayed.

Speculation, unless you have some relevant inside knowledge that most of us don't, can be unhelpful, and can quite easily go from obvious speculation as in your post, to someone repeating it as a suposed fact.

Also don't worry about dislikes.

I don't, just curious really as I didn't think I'd said anything particularly disagreeable and thought I might be missing something glaringly obvious. I can live with people being oversensitive about speculation.

Or appropriately sensitive.  You don't think your speculation was disagreeable, others will see it differently and arguably rightly so.

Whether that's the reason for you picking up a few dislikes I couldn't say.  Now that you've mentioned it you'll probably pick up a few more just because you mentioned it.

Might just have been your way of phrasing it, I would have thought that the proximity in time to the reopening would make everything relevant to do with the reopening an obvious area for investigation.

Obviously that may have nothing to do with the cause of the accident, but then you only said it was a possibility.

Can only speak for my preferred wall in Berlin: 4 autobelays, each one sent back to the makers in Sweden every 3 months for inspection. Seemed like they were being over-precautious when I heard that, but now I'm glad of it. 

I don't know about auto belays but our fall restraint lanyards and harnesses at work need a mandatory inspection every 6 months. Lifting gear ranges from 6 months to a year depending on what it is. Everything has to have a pre use check each time it's used as well.

I would like to think auto belays get an annual service, a 6 month inspection and a daily check by the wall but who knows.

To add, sometimes mechanical things fail. I have serviced and overhauled hundreds of air powered hoists that lift 500kg over the years. Each one was also weight tested to 125% after the overhaul.

We had 1 fail. It failed safe so it didn't drop the load and when it was stripped it was found the drive shaft had snapped. It was sent off for lab analysis and found to be a flaw in the metal and there was no way we or the manufacturers could have known that without destructive testing. The manufacturer does destructive testing on x amount from every batch but sometimes shit happens. 

Maybe this - i.e failing safe - is something that autobelay manufacturers should consider incorporating in their designs?  Then again, maybe they already do.  As you say: sometimes sh1t just happens.

yup, checked and maintained by and independent party seems to be the common practice in Europe (and recommended by the manufacturer).  But the center themselves might be required to do more checks. In fact, my local centers seem to check that every autobelay seems operational before the open… every day. Sure it’s not the full check done yearly. But climb a bit and drop to see that they are working.

This reminds me of an autobelay accident in Helsinki two years ago or so, where the autobelay had an invisible manufacturing fault, causing the device to fail after several years of use. The climber luckily fell from a pretty short height and was not seriously injured.

The main takeaway from this incident was the autobelay system in question was perhaps a bit too fancy, with a air-over-hydraulics unit that consists of body, hydraulic rams, pulleys, pressurized tank and a steel cable - meaning inspection of the whole setup was basically impossible and the fault in one part that was not supposed to ever need service was never noticed.

The gym removed all autobelays of this type after the accident and replaced them with TruBlu devices, which I guess are simpler in construction and therefore easier to inspect & service.

How it was at the UK wall where I worked was that they would be given a visual inspection every month, every 6 months they would be taken down and they would receive a detailed inspection and every year they would be sent to the manufacturer to be inspected and serviced. There could be other checks I'm forgetting about, this was a few years ago. 

In the Sydney incident the strap itself snapped, rather than the mechanical device.

He fell near the top, somehow the strap broke, and he may have suffered the head injuries hitting the wall on the way down, in addition to the obvious impact, that I don't know.

Apparently there was a similar catastrophic failure of the strap on the same brand device in Czech Republic(?) not long ago.

Do you have sources for this? I can't find much info beyond the initial article.

Sad to hear, and makes you think as I was using the auto-belays at my local wall last night.

I was just letting go when I got to the top, but I used to down-climb the easiest route, the theory being that I was rarely testing the system.

Think I will start doing that again ...

//www.abaris.co.uk/Trublue-Auto-Belay?gclid=EAIaIQobChMIh5yntfbq8wIV5-jtCh18EAFkEAQYASABEgJJm_D_BwE

Gives some information (inc user/maintenance manual).

They cost £2500! Work on magnets (Lenz theory).

I think this type are at Awesome Stockport, don’t know about Sydney.

The maintenance guide says there should be a weekly inspection, which includes cleaning dust etc off the unit. Should be fairly easy to see as a punter if your local wall with TruBlu units are following the schedule!

As far as I know Trublue autobelays are the only kind we have in the UK. 

There did used to be Redpoint as well, but there was an accident at the Westway back in 2007 (a non-fatal one fortunately) when one of those failed.  The manufacturer never was able to determine exactly what had gone wrong with it, and in the end they recalled them all and pulled out of the market altogether.  (Autobelays never having been a particularly important part of their business as far as they were concerned.)

Once you've successfully clipped in correctly, I bet you're still statistically safer on an autobelay than with a human belayer.  Though perhaps YMMV, depending on your belayer.

That sounds surprising. So the webbing broke? You would think that would be something that would be easy to inspect, with the webbing getting worn down over time and replaced well before it would get too abraided. Unless of course it somehow got cut over something sharp. Hopefully a thorough investigation will be performed and the results presented in due time.

You can get fairly jarring falls on an autobelay if you speed climb an easy route or jump for something and as a result get some slack because it can't take in fast enough.  There's less 'bounce' in the system than on a rope or one of the special autobelays for speed climbing.  I wonder if the cumulative effect on the webbing is significant if somebody sets an easy route which is good for speed climbing or a route with a move which can done as a dyno.

This is I think one of the major weaknesses of the system not  talked about. There doesn't appear to be any 'give' in the system beyond the braking capacity of the eddy current system, which can be really harsh at high speeds. Hence I would imagine with slack out (webbing caught over hold on the final section or it just got blocked entering the auto belay) and you could get falls which are the equivalent of factor 2+. I would expect some of the severity of the fall should be seen on the harness or autobelay system itself if we're talking 18kn which is the breaking limit of the TruBlu XL webbing suggesting the belay loop would break first if the webbing were in good condition.

Since autobelays tend to be attached at the top of the wall I think you'd struggle to have a factor 2 fall.

And whilst you can "out climb" an autobelay you do have to go very fast. You'd really have to be trying. And as I understood things a fall with slack in the system wasn't a problem for the webbing, but more an issue of it damaging the autobelay mechanism. I'm also led to believe that on a Tru-blu system this just causes you to get stuck on the wall rather than causes you to fall. 

Of course.

But it was a closely-connected friend so I won't say more. I know that's annoying and normally I wouldn't make such a post ("I have a secret!!") but in this thread I saw speculation that I knew was inaccurate so I felt I should just put out there what happened, as I think it's important to know.

Everyone should feel free to totally disregard it if they wish.

I didn't say it would be a factor two, i said 'equivalent to' because of the static nature of the rope. 

e.g. to explain, imagine I had a rope which was half static for the first 4m and then dynamic for the last 4m. 

I clip a bolt at 4m above the ground and climb on another four meters. 

If I fall, there is 8m of "rope" out and I fall 4m so it's a factor 0.5 fall.

But the first 4m of static means that effectively there is no stretch in that section so it's like I'm tied into the bolt at 4m. Hence the equivalent force is that of a fall factor 1. 

I'm thinking that there is very little elastic nature in the webbing and if I fall near the top, only a very small amount of that webbing is 'out' to take the fall. 

As I mentioned, it's not a case of outrunning the autobelay but perhaps getting it caught on a hold or the webbing wrapping strangely and getting blocked into the entrance of the autobelay. 

Not to mention on the climber. Nobody survives 18kn.

Those training for speed climbing competitions? Would've guessed this is mainly done on auto-belays.

Theres a specific auto belay for that, not the standard Tru Blue.

Here is a link to the manual for Trublue, the most common autobelay in use in the UK in my experience. Page 27 outlines the manufacturer recommended inspection regime (daily functional inspection from the ground, weekly inspection of the whole unit (either at height or bringing the unit to the ground) and annual servicing from a manufacturer approved service agent.

The manual also gives advice on webbing inspection and what to look for to identify damage.

While there isn't enough information to say what caused this specific accident, in the larger picture, human error accounts for the largest majority of accidents with autobelays so the message to all users should be to check, double check and check again

Is this mainly people failing to clip in at the start? I'm trying to envisage how you can have an accident once you've done so.

Which is true, but if you do a bit of speed climbing for fun you also tend to quite like climbing easy normal routes fast as a warm up and it's not that hard to outpace a Tru Blue enough to get a small amount of slack.   They used to have a sign up in Ratho saying no speed climbing on the normal autobelays. 

If you jump for stuff and miss when going dynamically on a Tru Blue you get a bit of a jar.  This is what I posed about before.  My concern was if there is a route which encourages dynamic climbing and so there is a fair number of falls like that there may be a cumulative effect on the webbing.

I witnessed this at the local wall. A good climber was warming up and realised at about 12m that he'd forgotten to clip in; he was instantly paralysed and started shaking. The hall's regular route setter was amazing; he first shouted at everybody to pile up any mats they could find, then pulled a toprope and grabbed a belayer before tying in and leaving a 2m tail onto which he tied a screw gate. He basically ran up (and clipped) the adjacent route, then clipped first the climber's belay loop with the screw gate and then  a quickdraw just above the climber into the middle of the tail. He belayed himself with a couple of draws, untied, and the climber was lowered. Getting himself down wasn't a problem. The whole thing took about 30 sec; maybe he'd thought the whole thing through in leisure moments, maybe he'd had training, but the order (the mats!) and speed of doing things were impressive.

To answer your second sentence - me too, discounting mechanical/material failure, unless the route being attempted is way to one side of the autobelay. 

Most walls I've seen autobelays in have a sort of plastic "mat" thing to which the auto itself is clipped, about 1.5m high off the floor, the effect of which is to "block" the first few footholds so you'd notice you weren't clipped in.  Seems a good idea.

That sounds like an amazing response. To do that so quickly and effectively must mean they had thought that through or trained for that scenario.

Thinking about it now, the likelyhood of it being used is pretty small, but I would hope all walls would have a plan and training in place to cover such eventualities.

I had the impression that was a direct response to known instances of people not clipping in and accidents occurring. A good simple solution which you'd think would cut out most potential future incidents.

But I wondered if there were any other ways of accidents occurring. I've a vague memory of something to do with someone coming onto the autobelay really gradually and it failing to engage properly as a result, but might well be remembering wrong or it could possibly have been an issue with an early design?

Yes mainly. The issue is exacerbated compared to climbing with a partner because there is no opportunity for a buddy check

Most (maybe even all?) walls now have some type of barrier over the bottom metre or so that the autobelay is clipped on to when not in use. This is there to reduce the likelihood of a climber forgetting to clip in as they would have to pull the barrier out the way to get on the route

There are some other hazards that autobelays present that aren't nearly as big an issue on roped climbs. The biggest one is snagging during descent where something catches on a hold (e.g. a gear loop or the elastics at the back of a harness) and the climber gets hung up as a result. This is why all guidance for autobelays bans helmets from being used on them

Arithmetic fail?

It's not a factor anything fall, as you've just hit the deck

On several occasions I've seen that the spring loaded crab on the autobelay hasn't rotated back into the lock position so you would effectively be climbing on a snapgate.  I think this happens a lot because chalk and gunk gets in the mechanism on heavily used autobelays and I always double check by pushing on the gate.

I think you might be thinking of a Grigri - perhaps being used for top-rope soloing?  But they have also occasionally been known to fail to lock under those circumstances when being used to belay conventionally (in which case the climber might deck if the belayer isn't holding the dead rope).

An autobelay is essentially a fixed-rate (ish) automatic descender, you can accelerate up to that speed of descent as gradually as you like but it won't let you accelerate beyond it.

Yes, you might be right.

In case of True blue they under go an annual service with only one Centre in the UK signed off to do this.

Each unit comes with a manual that describes a 6 month check carried out by the owner as well as daily and weekly checks.

The manual also says each user should carry out a function check before use.

So check the carabiner locks and the tape retracts.

Also pictures of when to stop using tapes due to different types of ware and tear.

The true blue units have 2 separate systems the retraction works on a sort of coil spring that pulls the tape in. The lowering is then done by a magnetic clutch system.

There is a small chance that the coil spring could snap, it is wound metal. But if this happened the unit would still lower the climber to the ground safely.

The idea behind the magnetic system is it's harder to fail as with some of the components other companies use. 

The double locking crabs with two gates certainly seem to be an improvement.