UKC

I've been aware of this potential problem for some time, and looked out for it carefully, but it's never actually happened - although I can see how it could. I generally use quite skinny ropes, but the devices I use (or have used) have minimum diameter recommendations well below these.

In the past I've used BD ATC Guide, and WC Pro Guide Lite, both of which are suitable for ropes down to 7.7mm according to the manufacturers. Currently, my main device is a Giga Jul which is ok down to 7.1mm, apparently. My skinniest doubles are Beal Ice Lines 8.1mm, skinniest single is Edelrid Swift Protect 8.9mm. I will shortly be on a trip climbing sport multi pitch, using the Swift as the main rope, pairing with one Ice Line for abseils.

You would think that if manufacturers rate their devices for a certain range of diameters, this would include all normal usages, including guide mode, and there should be no possibility of the ropes squeezing past one another in the slots; however, I've heard reports of this problem occurring with ropes considerably thicker than the official minimums (which are very skinny indeed).

Any experiences/thoughts? Thanks.

In reply to bpmclimb:

A few years ago on here, Jim Titt said the normal failure mode was the ropes inverting under load, typically at about 4kn I think. So guide mode only for seconds not leader falls.

Experienced it personally for the first time a few months ago, double 8mm in bd atc xp guide, clipped to belay loop as bottom clamp for prussiking up.

Old rope , gone soft , so possibly easier to squeeze  past under body weight.

Edit, 4kn is probably max load for any rope /plate combo. On a test rig ropes get frightenly skinny.

In reply to bpmclimb:

The old Trango B52 was bad for it, was not that uncommon at all. I think the failures were at 2Kn if I remember correctly. The trick is to not use guide mode on a single rope only on doubles.

In reply to ColdWill:

possibly this is why the old Reverso had a metal atachement rather than wire iot prevent the carabiner flipping.

In reply to bpmclimb:

Can you elaborate? I can't work out how this might happen

In reply to nufkin:

Well, guide mode locks by the weighted climber's rope pushing down on the dead strand. With enough force, and with a couple of mm of space when using a skinny rope, the loaded rope can squeeze down past the dead rope, so they end up upside-down in the slot, in which case the locking doesn't work.

In reply to bpmclimb:

Looking on mountain project, it mentions Jim's testing around ten years ago. From memory most ropes invert and slip at 3-4 kn, the best was 10.5mm semi static at 6kn. 

Petzl reverso manual has a diagram warning of slip mode potential.

In reply to Rick Graham:

One should note that "most" relates primarily to the rope diameter, I only tested initially because I had one of the early trendy devices (probably the MegaJul) and was trying it with a Beal Iceline. Hanging on it I tried to arrange a bodyweight release and flailing around to get my foot in the sling it dumped me onto the ground. Reliable locking looks different. The "clip the carabiner through" solution only prevents one failure mode, not the one I had.

In reply to Rick Graham:

One should note that "most" relates primarily to the rope diameter, I only tested initially because I had one of the early trendy devices (probably the MegaJul) and was trying it with a Beal Iceline. Hanging on it I tried to arrange a bodyweight release and flailing around to get my foot in the sling it dumped me onto the ground. Reliable locking looks different. The "clip the carabiner through" solution only prevents one failure mode, not the one I had.

In reply to bpmclimb:

i've also read of at least one incident when the ropes squeezed parallel enough to jam (but not fully switch and release) and the belayer had a very hard timer unjamming the device even after the second got their weight off the ropes.

The BD Alpine Guide gadget has thinner slots and might be the best bet.  But maybe there is also more friction hauling the ropes through...

In reply to bpmclimb: Does anyone have an image of what an inverted setup looks like? Or better still of it happening?

In reply to bpmclimb:

I have a Kong Gigi (flat plate, big holes) that I use on occasions where I know I will be using guide mode a lot, usually multipitch, and the downward force of pulling a fat rope through a reverso / pivot would eventually trigger my tendonitis. 

I have had it on a single occasion where it twisted around, but only with the kong gigi (which is very similar to lots of other flat belay plates out there), I've never had it happen on a reverso/pivot style device. 

From what I recall, the issue can only occur when you are belaying a single strand of rope, not if you are belaying from two ropes, as it required the carabiner to twist 180 degrees too which cannot happen if there are two ropes clipped in.  There is a fix you can do to stop the carabiner rotating - clipping both live and dead strands with the rear carabiner, best seen in the photo on this site.

https://www.vettamountainguides.com/journal/2015/4/3/the-kong-gigi-a-guides...

In reply to bpmclimb:

I have 8mm double ropes and I don't trust them in any normal ATC guide, even when not in guide mode. I ended up buying the ATC Alpine guide specifically for these ropes just to be safe.

I reckon you'd be okay with it in guide mode however. I have had the rope invert in a normal ATC before but if anything it just makes it lock. Un-inverting is a massive pain however.

In reply to wbo2:

In these devices the belay device hangs from a carabiner using a ring attached to one end.  Then a bight of rope is stuck through each slot and, behind, a 2nd carabiner is used to keep them in the device.  The device and ropes are orientated so the climber's rope lies on top of the belay's rope.  If the climber falls their rope presses on the belayer's rope providing enough resistance to lock the device.

The word "invert" is badly chosen here because it doesn't bring to mind what happens.  What actually happens is the (thin) climber's rope squeezes past the (thin) belayer's rope in the slots so the ropes are crossed or partially crossed. AFAIK this simply jams the device making it harder to release in use rather than stopping the braking effect.  It has to be remembered that in all these devices releasing the rope under tension (from a fallen climber) is hard and a PITA.

As a regular user of both thin ropes (7.5mm) and guide mode plates (old reverso 4 with the lightened side plates, BD ATC guide and DMM reverso) I've never had this happen.  I attribute this to never letting slack build up on the climber's side and isolating the guide plate with doubled sling from the anchor (so that it can orientate itself as it wishes).

I like these devices a lot and find the biggest hassles to be (1) running out of screwgates on long climbs - you need 2 more each and it's easy to forget this and (2) the backing carabiner wandering into an irritating orientation which is easy to avoid unless you aren't paying attention.

In reply to gravy: Thanks, get it now I like guide mode quite a lot, and can see this would be a royal p.i.t.a. tho' I've never had it happen to me though I can see how it could.

The shape of the pivot as an example would make this harder as I recall the slot is V-shaped? I don't have mine to hand.

In reply to gravy:

I think the word 'invert' does capture what is happening in the scenario the OP is asking about. Have a quick read of the link I posted, there's some good pictures accompanying the text.   

With thin enough rope/cord any single strand guide mode setup on a BD ATC or similar can invert where the brake hand ends up on top and the climber side ends up on the bottom - taking it out of the autoblock configuration.  

This does require the crab to twist though, so it is impossible (AFAIK) when belaying two ropes, and there is a work around detailed on that link. Even with a classic pivot or reverso, the belay keeper wire has enough play on it to allow rotation of the rear carabiner. 

You are right, there is a half way PITA where they jam up which isn't ideal, however the true inverting problem is indeed a big problem.