Self rescue hauling

what do you mean it doesn't work if you have to clear a lip?

jdrago wrote:what do you mean it doesn't work if you have to clear a lip?

I think he/she means the lip of a roof, i.e. that the casualty is hanging in free space and there is an edge creating a lot of friction between the anchors and the casualty.

The system I posted deals with exactly this situation. (And most other situations.)

Tangential observation, but it might help non-technical readers follow things if you made a more clear distinction between stroke efficiency and force efficiency. These systems are typically named for one but their effectiveness is judged on the other, and the same notation is used for both.

I gotcha. I guess every situation is different but I had that same thing happen the other day but fortunately the roof wasn't to far down. I gave myself plenty of slack on my clove hitch and then where I can reach I placed the prussik(tractor) down below the lip of the roof therefore getting rid of the friction of a tensioned rope bent over an edge. That might not work in every situation but it helped in mine.

David Coley wrote: We now just await tests from Jim (please, please, please) so we have some numbers rather than my word for it!

1.51:1 for that one using 8mm dyneema slings and 12mm dia HMS karabiners. It´s quite an easy set-up though with a few (or one long sling and a few clove hitches). As you say some of the stroke is going to be lost in the progress capture rope stretch unless of course you just put it below the hauling Prussik, with an ascender and maybe a bit of weight it would be self-tending as well.

Jim Titt wrote: 1.51:1 for that one using 8mm dyneema slings and 12mm dia HMS karabiners. It´s quite an easy set-up though with a few (or one long sling and a few clove hitches). As you say some of the stroke is going to be lost in the progress capture rope stretch unless of course you just put it below the hauling Prussik, with an ascender and maybe a bit of weight it would be self-tending as well.

Jim, is that 1.51:1 for the 3:1? If so, what would it be with the same sling and carabiners, but the 2:1 version? or was the 1.51:1 for the 2:1?

Thanks.

That was for the 3:1 shown by MartyC, the right hand version with karabiners. Theory says it should have been better at maybe 1.9:1 but it wasn´t. I shall play with this a bit more today and see what went wrong.
Edit:- I set it up again as shown with an 8mm Dyneema Mammut Contact sling as the top one and a 12mm Dyneema hybrid as the lower through 12mm HMS krabs and hung myself off it, 1.38:1. (84kg lifted 116kg).
As a Z haul with the 12mm sling it was 0.74:1 (84kg lifted 62).

Jim, any chance you could use a single strand of 5mm cord as I'm been using for both the 3:1 and the 2:1? The 2:1 seems to work really well - in that I can haul someone heavier than me up a cliff. The 3:1 seems to give a better mechanical advantage, but is such a pain to use it would be good to know if it might be worth it.

Thanks, as always.

5mm cord as 2;1 (Z haul) 84kg gets 68kg so 0.81:1. As a 3:1 128kg so 1.52:1 but I had to use 4mm cord for the upper part, my 5mm isn´t that long.
The stretch is horrendous!

tibloc, pulley, microtraxion, done.

I have hauled someone using a reverso3 and a nylon pulley that slips over a caribiner. Fucked up my hands real bad, did not realize I was destroying my hands because I was pulling so hard. Stupidest part about it was that I had gloves hanging off of my harness.

Bring gloves, use gloves.

OK. I'll go straight to my main question. The other stuff makes plain how this question relates to this thread but I am afraid that the question will be lost in the mix if I start with preambles.

When pulling on the simplest 1:1 counterbalanced haul situation shown on the left-hand diagram, I get the distinct impression that I am having two separate effects: the obvious pulling up of the load, acting directly on lifting it, rather than indirectly through the counterbalanced pull, and the reaction against that pull that increases the counterweight force I apply with my own weight (my own body feeling much heavier and less comfortable inside the harness as I pull). So my main question is: is it correct/appropriate to take both of these forces into account in calculating approximate "real" mechanical advantages in this and other haul systems ? In other words, assuming that my pulling up is equivalent to lifting a 25kg weight, is it OK to add 25kg to the pull-up side and another 25kg to the counterbalanced pull-down side, as shown on the diagrams?

To complete the explanation of the diagrams: I rounded off my own weight to 70kg for the calculations. I also used 66% transmission efficiency when pulling a static 7mm cord through a round-stock HMS biner (for the diagrams illustrating David's cord 2:1 system) and 50% when it's the climbing rope going over this kind of biner (both numbers taken from David's post). The practical experience appears to support counting both forces: with the simple 1:1 in the "tree tests" I discuss below, I can indeed quite easily raise a 70kg load but am barely able (pulling up as hard as I can, so probably more than 25kg) to raise a 90kg one and only in small increments at that.

I’ve generally found that I don’t have difficulties figuring out the theoretical mechanical advantage of standard hauling/pulley systems. However, when you get into this kind of situation where you're pulling up against your own weight, it's not ones, twos and threes anymore. It becomes necessary to quantify the effect of each pull and counterpull to get an idea of the real mechanical advantage. I am well-aware that back-of-the-envelope calculations will not give precise numbers for the real mechanical advantages of different systems or a precise appreciation of how easy or difficult the haul is but, from what I've seen so far, these calculations should be able to provide a ranking of the systems that more or less follows the ranking derived from the subjective experience of real tests. As I explain later I now heartily second bearbreeder's oft-stated recommendation that climbers should run their own haul tests in realistic (read "less-than-ideal") situations. However, I am hoping to be able to provide a comparison of the two rankings for a selection of systems that may help others identify the systems potentially more likely to work for them in order to focus their testing on those. To do this, it's essential that I get the calculations straight, hence the "main question".
In the past, I have used a quick set-up of a rope & biners Z 3:1 in pulling-up configuration and an on-rope 2:1 to help a stuck second, so I know that these systems can be useful but the second was helping in both cases: still using the holds for the 3:1 and pulling on the anchored side of the 2:1. Over the years, I've also occasionally tried other hauling systems but always by themselves, not in comparisons and I made a number of assumptions about how they would perform in more difficult conditions. Bearbreeder keeps insisting we should try hauling in adverse conditions if we want to find out whether hauling can be a realistic solution. I'd been meaning for a while to do just that with different hauling systems, using biners instead of lightweight pulleys, a load heavier than myself, some friction on the rope, etc., thinking that, despite these difficulties, I ought to be able to find systems that work. David's thread provided the incentive I needed to get started with these tests, which led to more tests, etc.

In the first series of tests, I used a short 15m sport route with a 2-bolt anchor set below the lip and lacking a belay ledge. The test victim weighed 15% more than me. The loaded rope was clipped through 3 bolts with single lockers and followed a very slight zig-zag path (short dogbone QDs would have allowed the rope to run straight, through slack biners).

I tried David's 2:1 system on a 7mm static cord with a walk-up system on it and an on-rope Z 3:1 redirected down also with a walk-up system on it (using a 10.3 mm rope), both assembled with biners (no pulleys). When these didn't work, I converted David's system to a block tackle 4:1 and the Z 3:1 to a 9:1. These didn’t work either, even standing with my full weight on the pull side of the haul systems and pulling up on a grab loop connected to the primary haul prusik. Thanks, bearbreeder, that was an eye-opener!

Following these first tests, I decided to move the bulk of the testing to my backyard. If hauling is only going to be realistic if you've taken steps to remove friction as much as possible, I feel that hauling an inert load from a hanging belay position up in a large tree becomes a fairly representative test and it does not involve trips to the crag and overtaxing partners' patience. I also wanted to try different counterbalanced hauling approaches. I've now tried a bunch of systems but I have lingering doubts about the calculations and I'd really like to get some feedback on that before I report further on the tests. I tried to search the web for an answer to the main question without success and also tried to get the answer from various engineer relatives, friends and acquaintances, also without success. So here am I.

Despite this lingering uncertainty on the calculations, I should highlight the practical outcomes I got from trying David's system in the tree test since I am using it to illustrate this post.

Testing David's system without an upward-pulling grab loop, I was able to raise a load weighing about the same as me (70kg load against my own weight of 69kg with gear on) but was completely unable to raise a 90kg weight until I added the upward pull component to the system Adding this component made the 90kg raise fairly straightforward. This made me think that either David had this upward pull component in his system when he raised his 80kg partner with his own 60kg or he weighed somewhat more than 60kg during his test (with gear on) while his partner weighed somewhat less than the estimated 80kg. Hauling is really a numbers game and a few kgs one way or the other can make all the difference between a doable haul and an impossible one; which is why I really want to get the calculations straight and why I whole-heartedly support bearbreeder's recommendations for climbers to run their own tests in "realistically difficult" situations.

The middle and right-hand diagrams show David's 2:1 cord system with two different attachment points for the pulling up loop. Unless my calculations are way off, these diagrams also show that it's important to place the pull-up loop just on the other side of the rescuer's counterweight (middle diagram) as opposed to right on the loaded rope (right-hand diagram). In the middle diagram configuration, the mechanical advantage of the system works to amplify the pulling-up action, resulting in a more efficient system.

jktinst,

Great to see you helping out on this.

I may have had a couple of advantages over you.

1. I might have been using 4 or 5mm spectra cord in some tests. This is slippy and doesn't stretch. Also Jim Titt explained that the friction of a cord over a carabiner reduces as you reduce the diameter of the cord - so thin is good as long as it doesn't stretch.
2. I probably sat down hard at the start of each stroke. Jim has explained to me that the friction when the cord is not moving is much higher than once it is moving. So if you can lick start it you then just ride it down.

pulling up really helps and also kick starts, but I hadn't thought about pulling the other side of the "pulley". Walking sown the wall also helps.

The reason I kind of think it is a potentially good system is that it can be used as a near far end haul. i.e. you head down removing all the runners and hauling from the victim. And it isn't hard work. Or at least not too hard.

I'll let others comment on the math!

petzl.com/us/outdoor/single…

It can help...

The carabiner efficiency corresponding to Jim’s 2:1 test with 5mm cord is about 53%. With 12mm sling the efficiency goes down to about 49%. On the other hand, Jim’s figures given by David earlier in the thread for single carabiner efficiency give about 69% for 5mm cord, and David reports weighing 60 kg and hauling 80 kg, corresponding to a carabiner efficiency of about 76%. jktinst reports being unable to raise 90 kg with a body weight of 70 kg which would have required 74% efficiency, so we can assume he had less than that.

These calculations assume that the only source of friction comes from the carabiners, in which case the efficiency of the 2:1 system is (e^2+e):1, where e is the carabiner efficiency (as a decimal, not as a percent). Using this, the break-even carabiner efficiency for the 2:1 system is about 62%, i.e. with carabiners less efficient than that you won’t be able to haul your own body weight (I’m not yet considering jktinst’s extra pull tactics here).

The huge range of apparent carabiner efficiencies (53% to 76%) and the inconsistency of individual experience suggests that either the same things aren’t being tested in comparable ways, or that the weight estimates are not accurate, or that these systems are very sensitive to rope and carabiner type; probably at least two of these effects are present. Sadly, this means that the success of the system may be questionable for you even if someone else has managed to use it in similar conditions.

An additional possible difference in testing protocols is that David mentions getting his apparently low-friction system to work by “sitting down hard” and so taking better of advantage of the typically lower values for sliding friction over static friction. (It may be worth mentioning in this regard that if everything is clipped to the same anchor, then that anchor will be loaded with the sum of the rescuer-applied load and the raised load and accelerating the rescuer to get things going will put a high instantaneous load on the anchor.)

As far as I can tell, all the math in jktinst’s diagrams is correct. If we turn it into algebra, letting L be the (biggest) load the system will haul, W the weight of the hauler, h the amount of additional one-handed lift the hauler can generate, and e the carabiner efficiency, then jkinst’s middle diagram corresponds to the equation

(W+h)e^2 + (W+2h)e - (L-h)=0,

which can of course be rearranged to emphasize the role of any one variable in terms of the others, for example

L=W(e^2+e) + h(e+1)^2.

Setting h=0 gives the L=(e^2+e)W efficiency of the 2:1 system mentioned above. The h(e+1)^2 term tells you how much extra load you can haul with given pull strength and carabiner efficiency. So, for example, with the break-even carabiner efficiency of 62% and a pull strength of 25 kg per stroke, you could haul 66 more kg than your body weight, getting you close to the 2:1 theoretical frictionless efficiency if you are strong and light.

Here's a video of it in action hauling someone who weighs more than me. As you can see the action is very gentle and I could do it for 50m of hauling - and I could haul more weight if I put some effort in, i.e. pulled on the up-rope etc.

youtu.be/iTHI3YEt3LU

jktinst,

I just had another look at your drawings, and while they might all help to get the person moving, you won't really be able to use them to haul. This is because as the person hauling goes down, the lower prusik moves up. Most of the time the thing you show being pulled upwards will be above, not below, you. Please see the video link I added above.

Hence if you have to pull you end up pulling on the ROPE with both hands once they are moving. You kind of pull yourself down the rope arm over arm. As the rope is tight, I'm not sure you can grip as well as you might want to, but it works.

Does that make sense?

Yes but the added advantage at the beginning of the stroke the motion might be just what is needed to get past the static friction obstruction to motion.

The corresponding formula for pulling up on the main load rope is

L=W(e^2+e) + h(e^2+e+1).

The second term is perhaps most easily compared to the second term in the formula for jktinst's haul position by writing it as h[(e+1)^2-e), which shows that the hand force multiplier for the first method is reduced by the carabiner efficiency to get the hand force multiplier for the second method.

Applying this to the same parameters used in the example for the first method (which is to say still assuming the break-even carabiner efficiency of 62% and a hand pull of 25 kg), the system gets a 50 kg load-raising boost rather than the 66 kg boost from the jktinst haul position. This is still pretty substantial, although I'm not sure that the hauler can consistently pull that hard. But maybe so, because the hauler is using both hands. Maybe they can pull quite a bit harder than 25 kg...

I'm still interested in any explanations about all the inconsistent test results. Jim's 2:1 numbers for 5mm cord seem to say the David can't do what he's doing, for example.

Or actually not. On second viewing, the haulee's legs are braced against the wall and he's walking them up as he rises. This means less than his full weight has to be raised. Roughly measuring the rope angle (which however changes during the haul) and the leg angle and doing a very basic statics calculation, I'd estimate the hauler only has to lift 3/4 of the haulee's body weight in David's video.

The situation gets better if the wall isn't vertical. This could be a major reason why some people report successful hauls when it seems clear that the dead weight involved probably couldn't be raised.

David: this clip is only 14sec long and I can't make out the system. I take your word that it shows your cord 2:1 system and can see that you go down about twice as fast as the other guy goes up but that's about it. One thing that the clip does not show is the sequence of resetting the primary haul prusik lower down after coming to the end of the haul stroke, prusiking back up on a different system to re-stretch the haul cord to its full length and starting over with the next haul stroke.

I was expecting a clip showing the "sitting down hard" technique (and was having trouble reconciling that expectation with the gentle movement you mention ;)). Imagining this sitting-down-hard approach based on the photo of the system in your first post, I anticipated that the full haul stroke would consist essentially of the distance between standing up and crouching down. I expected that this system would tend to work mainly in very specific circumstances; ie with a wide enough ledge to allow performing the "stand up, take up the slack on the cord and sit down hard again" sequence but not so wide as to result in edge friction. I'm having a hard time imagining how the sitting down hard approach would work in a hanging situation, with a longer cord, etc.

Regarding the question of the cord diameter: yes thinner cords cause less friction over biners but, based on the diagram you showed, the difference among cords is really quite small, despite the scary-looking slope of the curve. The transmission efficiency over a biner seems to be about 66% for 7mm cord (which is what I used for my calculations), 68% for 6mm and 69% for 5mm, these differences will have a pretty small impact on the real mechanical advantage of the system. Could it be that the 5mm Spectra cord you used is way slicker and results in a much better transmission efficiency than the (presumably regular nylon) 5mm cord Jim used in his tests?

As for the problem that when using a pull-up loop, your pull-up stroke is quite short because you go down at the same time as the loop comes up, yes, that's exactly what happens but it doesn't follow that the system does not work for hauling. My tree test is obviously limited in height but with my own 70kg, I had no great difficulty raising a test load of up to 90kg 6-7m into the air in a matter of 10 minutes or so with these two pull-up versions of your cord 2:1. For efficiency, I used a fairly long pull cord so I could do several pull-up strokes for each overall haul stroke. Of course, after each pull-up stroke, the pull-up prusik had to be reset back down (easy part) and I had to prusik one step up the pull cord (a bit harder) but in any counterbalanced situation, you have to expect a lot of "going up" motion to reset the system, be it prusiking up a rope or cord or standing up again after sitting down hard.

rgold: thanks for the validation on the principle of the calculations. Although I feel relatively at ease with pulley systems, I'm definitely no math-head and must confess that you lost me there with the formulas.

rgold wrote:...The huge range of apparent carabiner efficiencies (53% to 76%) and the inconsistency of individual experience suggests that either the same things aren’t being tested in comparable ways, or that the weight estimates are not accurate, or that these systems are very sensitive to rope and carabiner type; probably at least two of these effects are present. Sadly, this means that the success of the system may be questionable for you even if someone else has managed to use it in similar conditions...

Indeed.
I don’t imagine that it would be an option to come up with an SOP for haul tests that the entire climbing community would agree on so that we could really compare each other's test results or expect to obtain the exact same result as reported previously the first time we try a system, provided we stick to using the exact same materials, procedure and environmental conditions as dictated by the SOP. This is why it’s essential to perform one's own tests in realistic circumstances, with one's own gear, rather than rely on blanket recommendations from manuals, forums, etc. This is also why I feel that it’s important to provide a panoply of systems from which the testers can select, for their own testing, one or two that may be more likely to work for the actual weight differential between the lighter "rescuer" and the heavier "victim".

I'm still working on completing the diagrams and calculations for the various systems I considered (including subjective experience reports for those I actually tested with different loads, which is not all of them, thank god). I hope to report back soon.

jktinst wrote: Although I feel relatively at ease with pulley systems, I'm definitely no math-head and must confess that you lost me there with the formulas.

Sorry about that. My calculations are exactly the same as yours with symbols replacing your numbers. But I just posted the end results, not the derivations, which I felt few would have the stomach for. Moreover, I assumed that those who do have the stomach know how to get the same results themselves. But if anyone wants a detailed account, pm me and I'll send something in a format that is more readable than what we can do here. (I actually tried making the math a little more readable with HTML markup. The message previewer rendered it, but when submitted the HTML tags I added were stripped and so I gave up on readability.)