Mammut Smart Alpine Belay Device failure mode

rob.calm wrote:Accidents from rope bunching can even occur with an ATC. A couple of months ago, a friend of mine had an incident because of the rope coiling up on the brake side of an ATC. He was lowering his partner when the rope on his side of the ATC formed a lump, which he didn’t notice as he was watching the climber he was lowering. When the lump hit his brake hand, it knocked it loose from the rope and the climber began to fall....

That accident has nothing to do with the device however. It is definitely a potential problem---tight kinks in the free end of the rope can knock the rope out of the brake hand when the rope is sliding as in lowering or rappelling.

I've seen a kink form and then jam in an ATC-XP, stopping a rappeller in mid-rappel and requiring a somewhat complicated rescue effort from me (the rappeller was hanging free and I was up above) to get to them and help free the jam. This incident really surprised me, but after asking around for a while, I did hear from some other people about the same thing happening to them.

The ATC-XP jamming only happened when the device was already loaded, so the effect was to prevent further motion for a person who was already hanging. The thing that makes the Smart (the name isn't sounding so good right now) failure mode particularly worrisome is the potential to jam during belaying motions when no load is applied. Stranding a leader in a bad position without the ability to pay out slack is an absolute nightmare, compounded by the fact that the vigorous jerking needed to free the device, if just a little bit too successful, might actually pull the leader off.

Can someone (more knowledgeable than I) comment on the choice of carabiner being used in combination with the Smart?

I use the Edelrid HMS Strike locker with my Smart and I've not experienced this issue yet.

I did read in the Mammut instructions that care must be taken to use an appropriately shaped locking biner with the Smart, but I haven't done any testing with multiple styles myself.

bearbreeder wrote:when not feeding the rope the non brake hand should also be shaking out any twists in the rope

I will say that I do this as second nature while belaying and it does help.

Sam,

I use the Wild Country Synergy HMS because it is a large round stock carabiner and I find it to be a quite smooth operation with the rope when used with the Smart.

I do not believe the type of carabiner matters in the issue being discussed. If the carabiner is in a certain position the rope has plenty of room to move through the center plate slot and into the other rope slot passage.

I've never encountered this myself either until I started messing around with the issue Jeremy explained. Everything has to be just right for this to happen, but as shown, it can happen.

Sam Byrne wrote:Can someone (more knowledgeable than I) comment on the choice of carabiner being used in combination with the Smart? I use the Edelrid HMS Strike locker with my Smart and I've not experienced this issue yet. I did read in the Mammut instructions that care must be taken to use an appropriately shaped locking biner with the Smart, but I haven't done any testing with multiple styles myself.

You want a fairly large HMS biner, no smaller than an attache and a very rounded barstock will help with smooth feeding. the old rocklock, old attache, the petzl william and sterling falcon all work well.

Interesting discussion. Is there a reason that Mammut (or some intrepid user) can't get rid of the middle plate between the slots? It doesn't seem to serve any important purpose. The slot prevents ropes from rubbing together when bringing up two climbers using guide mode, but is that a big concern?

EDIT: to clarify, just the bottom half -- not the whole slot/guide mode eyelet.

john l. wrote:Interesting discussion. Is there a reason that Mammut (or some intrepid user) can't get rid of the middle plate between the slots? It doesn't seem to serve any important purpose. The slot prevents ropes from rubbing together when bringing up two climbers using guide mode, but is that a big concern? EDIT: to clarify, just the bottom half -- not the whole slot/guide mode eyelet.

The MegaJul works on the same mechanics as the Smart, and has no bottom section, so sure. The problem would be that by taking out the bottom section you dramatically weaken the strength of the whole device when it is subjected to the bending forces from catching a fall. In short, the device could fold in half. But maybe it would be fine -- you'd have to do a stress analysis to find out.

BigFeet wrote:Sam, I use the Wild Country Synergy HMS because it is a large round stock carabiner and I find it to be a quite smooth operation with the rope when used with the Smart. I do not believe the type of carabiner matters in the issue being discussed. If the carabiner is in a certain position the rope has plenty of room to move through the center plate slot and into the other rope slot passage. I've never encountered this myself either until I started messing around with the issue Jeremy explained. Everything has to be just right for this to happen, but as shown, it can happen.

I'm curious about this now and will do some experimenting at home, but I wonder about the positioning of hands when this happens. When this failure occurs you're saying that it is due to feeding rope through. Can you clarify where each hand is at when this failure happens?

MDoody wrote: The MegaJul works on the same mechanics as the Smart, and has no bottom section, so sure. The problem would be that by taking out the bottom section you dramatically weaken the strength of the whole device when it is subjected to the bending forces from catching a fall. In short, the device could fold in half. But maybe it would be fine -- you'd have to do a stress analysis to find out.

Yeah, that's what I'm thinking, as well. Just to be clear, I was curious about removing the bottom half of the middle plate, not all three plates, so the effect on the strength of the device shouldn't be as dramatic. I think removing the bottom half of the whole device would cause the guide mode to stop working without a total redesign.

Sam Byrne wrote: When this failure occurs you're saying that it is due to feeding rope through. Can you clarify where each hand is at when this failure happens?

It appears to be an issue when the carabiner is at the top of the widest section of the carabiner slot and there is a very small amount of rope bight behind the carabiner. The carabiner is not under any tension at this time, but when the carabiner has tension reapplied from contact with the Smart, up and away slightly, the carabiner will drive the rope down and over the center plate edge into the adjacent rope passage.

I discovered this by trying to recreate Jeremy's situation. The rope was fed through the right side rope channel slot as normal so that my right hand was the brake. After pulling rope through the device, as you would on a normal lead belay, I brought my left hand to the thumb lever on the Smart to act as the brake while my right hand was free to shake out any rope twist. With the carabiner and rope in just the right position, once tension was applied to the carabiner from me repositioning my left hand to the Smart the carabiner forced the rope through. It did not take much effort, or force.

I believe it was a normal feeling/movement that when I slid my right hand down the rope to shake out the twist, my left hand on the Smart acting as the brake, and unconsciously to me at the time, pulled up and away with my left hand causing the carabiner to force the rope through.

I only caused the problem twice in multiple attempts, and again, the right conditions and orientation must be present. It took a bit of fiddling with different positions, etc. to duplicate the problem. For the majority of the time there was no problem. A few others, only a slight movement of the rope over to, but not crossing the center plate and cleared on its own.

john l. wrote: Yeah, that's what I'm thinking, as well. Just to be clear, I was curious about removing the bottom half of the middle plate, not all three plates, so the effect on the strength of the device shouldn't be as dramatic. I think removing the bottom half of the whole device would cause the guide mode to stop working without a total redesign.

I could be wrong but that piece of metal creates a housing for the rope and may influence breaking power for ropes of different diameters. I could be wrong, though. This piece of metal also keeps ropes from twisting around each other when rapping.

I think a better solution would be to make all or a portion of middle sheet of metal solid instead of cut out so that the rope physically couldn't go through. Theoretically one could just use some duck tape if they can fit their fingers in there.

eli poss wrote: I could be wrong but that piece of metal creates a housing for the rope and may influence breaking power for ropes of different diameters. I could be wrong, though. This piece of metal also keeps ropes from twisting around each other when rapping. I think a better solution would be to make all or a portion of middle sheet of metal solid instead of cut out so that the rope physically couldn't go through. Theoretically one could just use some duck tape if they can fit their fingers in there.

How would the carabiner be able to run through the device? You need the center plate slotted for the free movement of the carabiner. This allows movement of the carabiner to the brake position and will give room for the rope to feed through at the carabiner's rest position.

I believe the center plate is mainly for structural purposes, but will also create a separation for the ropes as a bonus. The braking component comes from the way the rope bends through the system and is pinched against a grooved bar/bearing at the top throat section.

Wow. I can't believe I didn't think about that.. DUH! That wouldn't work and I'm a dumbass.

Also I need to update the OP. This happened over a week ago so I was wrong on the details. This happen while I was being lowered after a lead. So the device WAS under tension at the time of failure. My belayer and I discussed the failure and agreed that it could happen on lead as well. Bear and a few others have confirmed that I think and provided pics.

rob.calm wrote:Accidents from rope bunching can even occur with an ATC. A couple of months ago, a friend of mine had an incident because of the rope coiling up on the brake side of an ATC. He was lowering his partner when the rope on his side of the ATC formed a lump, which he didn’t notice as he was watching the climber he was lowering. When the lump hit his brake hand, it knocked it loose from the rope and the climber began to fall.

Why did the belayer only have one hand below the device? Proper lowering is both hands below the device, with one hand always gripping the rope and moving with it, while the other hand slides. I would say this is clearly belayer error.

David Gibbs wrote: Why did the belayer only have one hand below the device? Proper lowering is both hands below the device, with one hand always gripping the rope and moving with it, while the other hand slides. I would say this is clearly belayer error.

No way! It was equipment failure. The rope formed a gnarl as it was curled. Ropes should be manufactured so they don't do that.

Rob.calm

So special thanks to BB for his input on this thread, I hope your new thread on the multiple failure modes of modern bealy devices is thorough. As for me I am returning to more classic type devises. Read tube. Simple wins!

Jeremy in Inyokern wrote:So special thanks to BB for his input on this thread, I hope your new thread on the multiple failure modes of modern bealy devices is thorough. As for me I am returning to more classic type devises. Read tube. Simple wins!

i dont want to dissuade folks from using assisted locking devices ...

when used PROPERLY and with full knowledge of the various "failure" modes they should be SAFER ... for one simple reason rockfall, especially with a grigri

however many if not most folks dont use em properly and dont understand the various failure modes ... with an ATC its simple, theres only so many well understood ways to drop you

i personally dont consider this "failure" mode a big deal if the smart use knows and PRACTICES how to clear it ...

however if the user doesnt know or doesnt practice the clearing method i posted last page its potentially very dangerous ... one should be able to instinctively clear the jam in a split second with the method i posted

my view is that the problem is caused by the central hole being bigger than the outer ones ... you can see this in the pics i posted where the inner ridge is not as deep as the outers

this allows the rope to slip in much easier and get stuck there

interestingly enough while the single channel smart may not be susceptible to this failure mode in regular use as it doesnt have the uneven holes .... but if it does get stuck yr screwed as the clearing method i listed wont work with the regular smart

if for some reason you cant clear it you need to go back to the "standard" belay device removal sequence ....

EVERYONE beyond the most supervise basic beginners who belays should know how to do this or some variation there off ... and PRACTICE it under load

even with ATCs and grigris if you get hair, clothing, etc stuck you may well need this method or some variation to clear it ... its a basic skill

"failure" modes that are not readily apparent is one of the reasons i dont consider these assisted locking devices suitable for beginners ... or to make up for basic skills

IMO they are for more advanced users who want some extra "safety" from rockfall or to ease the hangdogging of their 5.14 sport weenie partner ....

as one can see the steps to clear the alpine smart should it jam up is not the most "beginner friendly"

time to breed a few moah beahs

;)

That linoleum floor though...

Ray Pinpillage wrote:That linoleum floor though...

And that cheapo $80 (maple-syrup-dollahs) rope, too.