Pulley jammer warning

[Bob Mehew]

Like andrewmc says, it is complicated and frankly I can't explain it even in simple terms.  Whilst it is not quite a replica of the drop test, https://www.youtube.com/watch?v=GIkeGBXqWW0 has a number of similarities.  The slinky shows you can get variable forces along the length of the slinky since from Newtons 2nd law (F = m * a), some parts of the slinky are accelerating and other parts are not.  I suggest our drop test is similar to a speeded up version of the slinky.  My expectation is the peak force seen by any part of the set up is the same, it just occurs at slightly different times.  (Though the time frame will be different in rope & steel since the speed of sound is different.)

mikem - Would you like to come to the Bradford to witness an experiment where we locate a second load cell within the system?

The use of low stretch (semi static / SRT) ropes in place of dynamic ropes risks giving the climber a bigger peak force.  But as I indicated, if the climber is being belayed by another person using the classic belay stance, then that will reduce the peak force seen by the climber.  That will be less true if one uses one of the other belay systems.

The report states "Discussions with representatives of Petzl have clarified that the manufacturer does not consider the Petzl STOP to be a belay device of any form.  Furthermore, Petzl have never advocated the use of the STOP part threaded. So use of the STOP in this manner would be out with manufacturer?s guidance."  BCA have urged "cavers to only use devices ... in line with current manufacturer?s instructions."

 

[mikem]

I think these tests are applicable in theory to the pulley jammer scenario (the anchor impact forces are very similar), the climber creates the force on the top anchor, but experiences a much smaller force themself. The force on the jammer is what deforms the pulley (obviously the jammer is not as forgiving as a belayer on a long length of rope, so climber will experience a greater percentage of the total than shown here, & until the pulley deforms will experience almost all the force): https://www.petzl.com/US/en/Sport/Forces-at-work-in-a-real-fall

On Mendip laddering is still the norm, so belaying from fixed anchors at the top is commonplace, therefore it would be useful to know if a short length of dynamic will make a significant difference to using semi-static when someone comes off the top of the pitch with some slack in the belay...

 

[MarkS]

On Mendip laddering is still the norm, so belaying from fixed anchors at the top is commonplace, therefore it would be useful to know if a short length of dynamic will make a significant difference to using semi-static when someone comes off the top of the pitch with some slack in the belay...

I suspect whether a difference is significant depends on all manner of things, such as your definition of significant, the type of knot(s) in the system, whether they have been pre-tightened, how much slack is in the belay, whether the belay is directly above the climber, how high the belay is above the climber, what the belay device is, what sort of harness (or belt?) the climber is wearing...

Ultimately, all belaying should be done with kit and methods falling within manufacturers' guidelines, e.g. a grigri + dynamic rope. There are so many variables that I suspect advice on what else may be OK is going to be subject to so many caveats as to make it worthless.

 

[andrewmcleod]

On Mendip laddering is still the norm, so belaying from fixed anchors at the top is commonplace, therefore it would be useful to know if a short length of dynamic will make a significant difference to using semi-static when someone comes off the top of the pitch with some slack in the belay...

a) don't do this (the bit in bold) - obviously it can still happen, but if it's a risk, put in a traverse line and use your cowstails; if you can't be bothered, then don't fall off!
b) an appropriate belay device will limit the load e.g. from memory of something Jim Titt said years ago on UKClimbing I think a standard climbing belay plate will slip at 2 kN or so and an Italian hitch at about 3 kN, a Rescuecender (not supposed to be used for belaying) is designed to slip at 4 kN (I think), and something like an I'D or Rig (probably similar for a GriGri) will slip at 6-8 kN I think (depending on the rope - skinny rope, more slippage which can be up to a metre or more...). And Lyon Equipment found a Stop cut/damaged the rope on large dynamic falls (due to the edges on the plates I think).

I'm fairly confident that the small amount of dynamic will make bugger all difference. See the difference in forces between static and dynamic cowstails - about 1-2 kN I think. You are better off with cowstails with knots in static rope than sewn cowstails in dynamic rope, for example.

Swildons has a thread and a P-bolt for a safe approach traverse line. GB ladder dig has a thoroughly bizarre and not entirely safe method of access, but anchors well back from the pitch once you are up there. Can't remember about Sludge Pit. Eastwater 35ft pitch has dodgy old anchors but plenty of backups and you can run a traverse/anchor backup line from the chamber before. If you can't get on/off a ladder pitch safely, then I would suggest it is not well bolted/conceived...

'Bottom-roping' is obviously safe with a semi-static rope (possibly safer as there is less stretch making it harder to deck out on falls near the ground) but so is 'top' top-roping, done 'correctly' i.e. the climber does not climb above the anchor, or if they do they don't fall off, or if they do they fall off slowly (e.g. sliding off a ledge and THEN falling)!.

 

[Ed W]

I am probably being a bit thick, but is the high peak force more a factor of the high Fall Factors (1.4 to 1.7) rather than being due to the jammer?  It should be no surprise that a semi-static rope will generate high peak forces in such conditions.  Isn't it also a little unrealistic for caving lifeline situations, where the worst result of a FF1 is the impact with the floor of the pitch.

I'm not advocating the use of jammers in a lifeline system, more a defence of semi-static rope for lifelining as long as slack is minimised to keep FF as low as possible.

 

[Chocolate fireguard]

The only question I have is where was the load cell located? As force on climber will not be the same as force on pulley (the easiest place to locate the cell).

If the belayer is still pulling on the rope then that force will register on the load cell at the top as well as the force on the climber, so the load cell will register a bigger force than the force on the climber - by the few 100N a person is capable of in such a situation.

If there is no such pulling force then the force on the climber will be the same as the force on the load cell (to within a fraction of a Newton - I will explain this caveat to anybody who is interested, but not now).

Neglecting the weight of the rope:
The upwards force on the climber is the same size as, but opposite in direction to, the downwards force on the rope (Newton's 3rd law).
The downwards force on the load cell is the same size as, but opposite in direction to, the upwards force on the rope (ditto).
If the force on the climber was less than the load cell reading there would be a net upwards force on the rope.
Bearing in mind that the mass of a few metres of rope will be less than 1kg then a difference of even 10N would result in the rope accelerating upwards at more than 1g.
This means that in the course of arresting the fall (taking typically a few 100ms) the rope (represented by its centre of mass) would travel UPWARDS a distance of a few 10s of cm.

Someone would have noticed this by now.

 

[Oceanrower]

GB ladder dig has a thoroughly bizarre and not entirely safe method of access, but anchors well back from the pitch once you are up there.

It is, however, vastly amusing when you get there with someone who's not done it before and watch them trying to work it out...

 

[andrewmcleod]

In a class bottom-roping situation, i.e. a karabiner at the top, a person on the ground belaying, and a person falling off who hasn't hit the deck, the forces on the anchor, the climber and the belayer are all different.

The force on the anchor is largest as it has to hold the downwards force from both the climber AND the belayer.
The force on the climber is the next largest.
The force on the belayer is the smallest, as friction reduces the peak force from the climber (which is why belaying over a pulley is silly and you will never see a climber doing it).

In the steady-state case (i.e. someone hanging on the rope), the force on the climber and belayer will be equal and the force on the anchor will be twice this.
In the dynamic case, the peak forces can be different - but how different will heavily depend on the situation e.g. a pulley will reduce friction and thus the difference in force.

When directly belaying off an anchor, there is not a belayer hanging down from the anchor, so forces on the climber and anchor will be more similar.

All of this has bugger all to do with toothed ascenders which should be nowhere near any belaying situation.
It also has bugger all to do with the testing carried out, which shows damage to the pulley, the ascender and the rope in this situation. The actual forces reached are irrelevant.

 

[mikem]

Sludge Pit does have traverse bolts before the pitch head. SRT ain't friendly there though as lots of rough edges close to the line.

 

[mikem]

The numbers are important as, if we understand what is going on in the system, we can use the results to model what will happen with other belay devices.

If others are correct about no force being taken by the pulley jammer, and if Grigri, I'D or Rig won't slip until 6 to 8kN, then using them to belay from fixed anchors at the top will also result in similar (high) forces being experienced by the climber if they fall off with slack on a short rope - thus suggesting they are probably not the safest items to be using in that position...

Obviously in the real situation the knots in the rigging & belay rope will provide most of the shock absorption, along with the way the body of the climber flexes (the only one of these that is in the test is the knot attached to the climber). The energy absorbed by the core & sheath of the rope (due to strectching of short lengths & squashing around device) & deformation of bolts / belay device will be miniscule.

 

[mikem]

Actually I've worked out where my error was - the load cell isn't measuring the force on the pulley, but the force on the whole system (which will be what the climber experiences). The pulley has been subjected to a much greater force, but this hasn't been recorded as it resulted in the pulley deforming, however, this did reduce the force on the climber as for a 1.2 fall factor onto 0.44m of rope attached to just a jammer the peak force was 6.6kN, whilst a 1.4 FF onto 0.45m rope running through a pulley jammer was only 5.4kN (82% - but higher FF, so probably even bigger reduction).

Other considerations are that somewhere like Swildons 20 isn't suitable for a handled belay device as the climber falling off will pull the device against the rocks, potentially either releasing the handle or obstructing it so can't be used.

Looks like Bob needs to do the same tests on a Grigri (or similar)...

 

[mikem]

There are a couple of discrepancies between the notice (weight given as 100kg):

that the rope suffered varying degrees of serious sheath damage.

& the main report (weight given as 85kg):

There was no visible damage to the rope in these three tests.

 

[andrewmcleod]

Numbers!

https://www.petzl.com/INT/en/Professional/Static-and-dynamic-tests-on-the-RIG?ProductName=RIG

My figures for the Rig are probably a bit high for likely situations.

 

[mikem]

So, yes a fall factor 1 drop of 1m on 10-11mm rope with a Rig does give 6-7kN force.

 

[Bob Mehew]

There are a couple of discrepancies between the notice (weight given as 100kg):

that the rope suffered varying degrees of serious sheath damage.

& the main report (weight given as 85kg):

There was no visible damage to the rope in these three tests.

Apologies for the discrepancies in the Note.  The correct weight for the test mass was 85kg, though this is considered to be equivalent to a 100kg human body when allowance is made for the energy absorption properties of the human body.  I can confirm that there was no visual damage to the rope in the three drops of the pulley / jammer set up.  I suspect the quote was lifted from the jammer alone set up where rope damage was observed, as has been by other persons.

 

[Bob Mehew]

Actually I've worked out where my error was - the load cell isn't measuring the force on the pulley, but the force on the whole system (which will be what the climber experiences). The pulley has been subjected to a much greater force, but this hasn't been recorded as it resulted in the pulley deforming, however, this did reduce the force on the climber as for a 1.2 fall factor onto 0.44m of rope attached to just a jammer the peak force was 6.6kN, whilst a 1.4 FF onto 0.45m rope running through a pulley jammer was only 5.4kN (82% - but higher FF, so probably even bigger reduction).

<SNIP>

Looks like Bob needs to do the same tests on a Grigri (or similar)...

It appears as if you refuse to accept the load cell measured the peak force in the whole system but I can't conclusively demonstrate that by simple argument.  I note that the pulley's safe working load was only 4kN so high forces are not needed to damage it.  I guess we will have to do some experiments to try and convince you.

Using Petzl data based on a different system to compare with our work is somewhat doggy. 

You suggest testing a Grigri which is built to meet EN 15151 which covers braking devices.  Unfortunately that standard does not place a limit on peak forces for a given drop and also is only valid when used with dynamic ropes - a most significant point I suggest.  I also note the Grigri instructions include a climber and a belayer, so you have two human bodies to absorb the load as well as being used with a dynamic rope. 

I hesitate to suggest an alternative as the EN standards are rather complex.  But EN12841 covers rope adjustment devices which are designed to ?? link a seat harness to a working line ? to allow access ? to work positions, to give support and protection against falls? and includes a test for peak loads as well as permitting the use of low stretch (semi static / SRT) ropes.  andrewmc's link to the Rig is a device which meets EN12841. But as the data indicates, if you switch from dynamic to low stretch  (semi static / SRT) rope, peak forces go up.  (Note Petzl state they use a lanyard of dynamic rope to meet EN 12841.  One has to read the fine print!)

I can only suggest one gets the instructions and work out what device fits the set up you wish to use it for.  And take great care to work out what the symbols mean (Petzl appear to use a weight symbol as a 'dead' load as opposed to a person or a harness as a 'live' load.)  I freely admit I remain uncertain about what can be used for what.

 

[mikem]

It appears as if you refuse to accept the load cell measured the peak force in the whole system but I can't conclusively demonstrate that by simple argument.  I note that the pulley's safe working load was only 4kN so high forces are not needed to damage it.  I guess we will have to do some experiments to try and convince you.

Using Petzl data based on a different system to compare with our work is somewhat doggy. 

The pulley should see at least the same, if not more force than a single jammer, but the load cell recorded less...

Quoting from the report (page 3 of 13, although the last is 16 of 15!):
"A simple theoretical analysis indicates that the pulley sees near double the force that is recorded on the load cell due to the mechanical advantage created by this particular set up, see Annex 1."

Yes, I know the Petzl data isn't compatible, but it is ball park of the expected figure. Some of the ropes give much lower values (mostly smaller diameter, but not all).

 

[glyders]

(Note Petzl state they use a lanyard of dynamic rope to meet EN 12841.  One has to read the fine print!)

Indeed is does, but that is a rope positioning standard. The ones more relevant to the situation in discussion is EN341 (and to a lesser extent EN15151).These are tests where the device is directly attached to an anchor in belay mode and a load is dropped. EN241 is tested on 10.5mm and 11mm semi-static rope. When a 120kg mass was dropped 60cm onto 4m of paid out rope, the RIG gave an impact force of 5600-5800N and a slippage of 2cm.

 

[andrewmcleod]

It doesn't really matter what peak force you reach in a drop test (as long as it is below 6kN anyway).

What matters is that if you reach a peak force of 2/3 kN or more on an ascender, I will get very nervy. When you get a peak force of 4kN or more on an ascender, you are into rope damage and failure territory.

You might get a higher impact force on a non-toothed device (be it a Rig or a Grigri or a standard belay plate or whatever) but what you _won't_ get in normal circumstances is rope damage - you will get slippage instead. Any device used for belaying should fail by slipping, not stripping.

 

[Bob Mehew]

"A simple theoretical analysis indicates that the pulley sees near double the force that is recorded on the load cell due to the mechanical advantage created by this particular set up, see Annex 1."

Apologies, the penny has finally dropped.  But the reason for the pulley seeing double the force is due to the jammer which in arresting the rope generates a force within the system, a highly unusual set up.  My prediction is that the force seen at the bottom end of the rope (and hence by the person) will be the same as that seen by the load cell.

The ones more relevant to the situation in discussion is EN341 (and to a lesser extent EN15151).

EN341 like EN15151, does not specify a peak force limit (except for once only use devices).  (I presume 241 is a typo since EN241 is about petrol.)  Could you specify which clause you are citing from which the 120kg mass comes?  My take is EN341 and EN15151 are only concerned that the device does not break apart under reasonable shock loads and not about limiting the load which could be incurred by the person using the device.