[i_a_coops]

You could use some Sugru and some cling film to make a perfectly shaped bung!

 

[X]

Do we get a mention in the FA list when you climb this project?

[i_a_coops]

Do we get a mention in the FA list when you climb this project?

Your faith is touching! I've only just convinced myself that I actually want to do it instead of going and trying something easier/safer/with nice shiny bolts in it. Tying in on the sharp end feels a way off yet 

But also yes, I think if I wrote it up confusingly enough I could make it look like everyone on here seconded it to really confuse climbing-history.org 

[remus]

But also yes, I think if I wrote it up confusingly enough I could make it look like everyone on here seconded it to really confuse climbing-history.org 

  give me a heads up when you're close so I can book in a few weeks of Dev time to make the necessary tweaks to ch.

[Moo]

Last time I used sugru and clingfilm to make perfectly shaped bung I ended up in A&E.

[i_a_coops]

Ok re the revolver, I think it's back-of-envelope physics time.

I think the extra distance fallen doesn't definitely/obviously translate into extra force on the gear. The extra distance gone down by the climber is going to be pretty much the same as the extra distance that the belayer gets lifted (?), so if the climber and belayer are roughly the same weight there's actually NOT more gravitational potential energy to be dissipated(?).

But, the extra distance fallen I think does mean the fall lasts longer (does it? with absolutely zero friction in any of the carabiners and similar weight climber + belayer I think the climber would get to a constant speed and not stop until the belayer hit the first piece of gear, whereas with infinite friction the climber would stop quickly. I can't see a reason there'd be a turning point in the graph of friction against fall time?)

Also, with no friction between the rope and the top biner, the entire rope will stretch equally, meaning the rope stretches more (?) and therefore more of the energy goes into that rather than impacting the rock. With infinite friction, the only bit of rope that stretches is the bit of rope between the climber and the last piece of gear, which in essence means a higher fall factor which is definitely worse.

The only scenario I can think of where lower friction increases the force on the top piece is where the belayer is still stood on the ground and the climber is sitting on the rope, where the friction force is supporting some of the climber's weight and taking the same amount of weight off the belayer. However if the belayer is lifted off the ground in the fall then after the fall I think the top piece is still taking the total weight of the climber and the belayer once everything has come to rest regardless of the coefficient of friction in the top biner

  

[i_a_coops]

(Obviously this is physics rather than engineering, I wouldn't be completely shocked if the variability of the efficiency of the pulley at different loads and/or weird boinging somehow meant that the peak force was actually higher with a revolver. I just think the first order approximation suggests otherwise    )

[Fultonius]

There's some posts on supertopo about this, and someone made a spreadsheet model that demonstrates it - the general gist seemed to be for low force falls it reduced peak force, but for high force falls (assuming the pulley actually rotates, which is debatable) it increases the peak force due to the 2:1 pulley effect.

It's counter-intuitive, but because the pulley increases the load on the belayer, the total load goes up (load on anchor is climber + belayer).

I'd love to see those guys at HowNot2 do some testing on this. (being and engineer I want it proven in the dirty unpredictable real world...)

Reminds me that I had an idea for a sky hook that I've never got round to testing...

[X]

There's some posts on supertopo about this, and someone made a spreadsheet model that demonstrates it - the general gist seemed to be for low force falls it reduced peak force, but for high force falls (assuming the pulley actually rotates, which is debatable) it increases the peak force due to the 2:1 pulley effect.

I’m probably being blind but I’m not seeing where this (2:1) mechanical advantage is coming from in the case when the climber is not also the belayer.

EDIT:
I think, if my search found what you were referring to, your SuperTopo person was looking at the situation when the leader’s top runner is also the belayer’s belay (ie on a multipitch route). I think the person posting then forgets this when re-posting it on another thread.

http://www.supertopo.com/climbing/thread.php?topic_id=731822&msg=732669#msg732669

http://www.supertopo.com/climbers-forum/859036/Physics-question-DMM-Revolver

[X]

Ok re the revolver, I think it's back-of-envelope physics time.

I think your para 3 is accurate (e see also https://m.petzl.com/US/en/Sport/Fall-factor-and-impact-force---theory) but the other paras I’m largely unconvinced by.

I think the following is probably a better basis to start from: https://4sport.ua/_upl/2/1404/StandardEqn.pdf

[i_a_coops]

I think the '2:1' is correct if the belayer is fixed to the ground, then the climber puts X kN of force into the rope, the belayer experiences X kN of force through the rope, and the gear gets 2X kN.

If the belayer jumps into the catch it's a whole different 

[X]

Hmm. I’ve never heard Newton’s Third law expressed in this way before.

In any case, as we’ve discussed it’s not entirely clear what overall impact (pun intended) having a pulley of greater efficiency actually has due to the other implications.

[duncan]

I'm also curious what people's reasoning is for all the gear being on one rope is. I was also under the impression that truly equalising things was really hard.

I was also under this impression. 

Another consideration is that ropes do some weird things when a piece blows. The rope is stretched before the piece pulls which can reduce its elasticity and so can increase the impact force (?) on the next piece that comes under tension.

High speed film after a piece breaks show the rope snaking around. I’ve twice experience a rope unclipping from the piece below the one that pulled in this situation, possibly because of the way the rope behaved.

Because of this I would tend to choose two thin ropes on separate pieces rather than attempt to equalise unless I felt there was a strong case for a single rope.

[X]

I'm also curious what people's reasoning is for all the gear being on one rope is. I was also under the impression that truly equalising things was really hard.

I was also under this impression. 

I think you’re right that equalising it perfectly is impossible in practice but the contract is with splitting over two ropes and I don’t think it is clear cut which is superior due to the previously mentioned issues. I’d be interested to hear about JB’s tests.

...
Because of this I would tend to choose two thin ropes on separate pieces rather than attempt to equalise unless I felt there was a strong case for a single rope.

I think this is a misunderstanding. I (others?) simply meant have the entire equalised top cluster on one rope (a skinny half), next lower gear on another rope.

(EDIT: expanded quote to make it clearer)

[Paul B]

I certainly wasn't advocating for a single!

[cheque]

My experience of falling on rollers is you go absolutely miles!

  As a layman it doesn’t seem like such a big modification but when you think about it (and/ or start using them) you realise how different the properties of a pulley and the bar of a standard carabiner really are!

I used to climb with someone who had aspirations of being a climbing instructor. They created what they thought would be the perfect gear for rigging top ropes on sport routes. These things were quickdraw tapes with a normal locking D carabiner at one end and a locking Revolver at the other, the idea being that you put one on each bolt of the anchors (with the gates opposed of course, safety first) and had a foolproof and ultra smooth toprope system.

I only saw them in use once but it was quite something 

[Johnny Brown]

(Obviously this is physics rather than engineering, I wouldn't be completely shocked if the variability of the efficiency of the pulley at different loads and/or weird boinging somehow meant that the peak force was actually higher with a revolver. I just think the first order approximation suggests otherwise    )

I'm no physicist but my empirical experience from doing a lot of testing is that it might do either depending on the situation. You can also get an amazing amount of variation test-to-test, even on a professional rail-based drop rig with no apparently obvious variables, whereas youtube/ blogs almost always do one 'representative' test for each variable and then the world treats it as gospel. My old back-up device research below is a great illustration - look at the Shunt



I could arm wave as to why the Shunt shows such variability, but why bother. In a climbing context some of the variation is due to the way knots tighten - a big part of the energy absorption especially on small falls - but more significantly, I suspect, due to the fact you are often creating something akin to a double pendulum - i.e. a complex system that creates chaotic results. When I spoke to an experienced rigger close to Dan Osman he had similar thoughts on why the rope broke on what was, after all, a Factor 1 fall. And that's before you factor in behaviour like Duncan mentions.

That's not to say you can't sometimes get fairly consistent results in tests, but I would always urge caution when applying them to real scenarios which are almost always more complex.

So if you have a specific scenario, copy it as closely as you can and then test it repeatedly. It does get expensive with devices like screamers - the above graph had government funding...

[Johnny Brown]

I think you’re right that equalising it perfectly is impossible in practice but the contract is with splitting over two ropes and I don’t think it is clear cut which is superior due to the previously mentioned issues. I’d be interested to hear about JB’s tests.

I haven't tested the exact scenario. I have tested the efficacy of various ways of equalising anchors. With more than two anchors over short distances like a belay it's impossible - the shorter strand always takes the bulk of the load. Even with two anchors it's very hard to do effectively - make the two strands identical length and a slight change of the angle of the applied load negates the equalisation. Whereas you can do it by doing something similar to a lead fall on double ropes - i.e. long distance and small angle between the two connections, letting the rope stretch do the work of equalising.

[X]

Whereas you can do it by doing something similar to a lead fall on double ropes - i.e. long distance and small angle between the two connections, letting the rope stretch do the work of equalising.

I'm interested in what this setup does to peak impact force on the two pieces/clusters compared to the other setup. While, I appreciate from what you write that, essentially in the "equalised" setup often the peak impact force will be on only one runner, that Peak impact force may(?!) be lower for the hazy reasons I previously mentioned

[Steve R]

I think this is a misunderstanding. I (others?) simply meant have the entire equalised top cluster on one rope (a skinny half), next lower gear on another rope.

That's what I meant too.  (Obviously not achievable most of the time if onsighting.)  My rationale was simply that if* the load could be spread between the 3 pieces effectively, then one rope gives a longer deceleration window due to more stretch and, intuitively, a lower impact force.

On the equalising perfectly thing, I'd agree it's hard for >2 pieces to line things up well - use slack knots(!?)  For 2 pieces, can think of a couple of ways to equalise perfectly - thinking similar set up to how the sling on a totem cam auto equalises between the two sets of lobes.  Granted it seems hard to avoid a bit of a potential shock load if one piece fails with that sort of set up though.

*seems like a very big if from this discussion

[Johnny Brown]

essentially in the "equalised" setup often the peak impact force will be on only one runner

Yeah that's what I'd expect.

that Peak impact force may(?!) be lower for the hazy reasons

I can't see it myself, but note that the impact force on the climber might be higher on double ropes, but lower on the individual gear as better shared. I'd also expect the result of a piece blowing to be more predictable, because you haven't got a sudden change in the direction the force is applied.

On the other hand, if I wasn't confident in the top cluster holding, yes I'd want them all on one rope and the other rope on the next piece lower. But that does increase the load on the top cluster and increase the chance of it failing, so it depends on where the next piece is and how much room you've got to fall into...

thinking similar set up to how the sling on a totem cam auto equalises between the two sets of lobes.  Granted it seems hard to avoid a bit of a potential shock load if one piece fails with that sort of set up though.

Ah yeah I was thinking predominantly of static equalisation. On two a sliding X or similar works pretty well, unless one blows at which point I think one per rope would be preferable.

Perhaps we can get DMM to make a Pivot for 3 x 7mms?