Hardlinking (Joining) of Multiple Karabiners

[lutan1]

I've had numerous discussions with people in relation to hardlinking or joining multiple karabiners together with arguments on each side of the fence- yes you can do it, no you shouldn't do it.

What does everyone else think and why/why not?:confused:

To see the differences of opinion, I've put this same topic up on the Australiab Techrescue website- check it out at http://www.techrescue.org/smforum/index.php?topic=13450.0

[FIRELT03]

We have been taught never to go carabiner to carabiner, people say it places a torsional load on the carabiners at the point of contact and presents a possible point of failure. Not sure if any official testing has been done on this, would like to see it if it has been done though. Would not hurt to place a short soft link in between the components of your system. Just my thoughts on the subject.

[resqtek]

I have also been taught to avoid this with little or no science to support the opinion. I was told that when things want to twist, this can place a torsional force on the carabiner. How is it any different than clipping a carabiner to a harness d-ring, descent control device, hard eyed anchor strap, belay device, pulley, etc???

[lutan1]

Would not hurt to place a short soft link in between the components of your system. Just my thoughts on the subject.

Do we really want another link in the chain so to speak? It's just another thing we woudl have to do, it's just another thing that has the potential to fail, etc:confused:

[lutan1]

I have also been taught to avoid this with little or no science to support the opinion. I was told that when things want to twist, this can place a torsional force on the carabiner. How is it any different than clipping a carabiner to a harness d-ring, descent control device, hard eyed anchor strap, belay device, pulley, etc???

Your final statement is the cruncher- why is hardlinking any different to all of these connections?

Give me evidence/research- someone? Please?:confused:

[DCFDRescue2]

I have also been taught to avoid this with little or no science to support the opinion. I was told that when things want to twist, this can place a torsional force on the carabiner. How is it any different than clipping a carabiner to a harness d-ring, descent control device, hard eyed anchor strap, belay device, pulley, etc???


I think that the difference is whether or not there is anything in place to absorb the torsional load. For instance, if a carabiner and pulley is attached to an eye bolt in the floor ( on a roof, etc..) for a change of direction, there is nothing there to absorb the twisting motion created if the load is creating that motion.

However, I have no problem hooking a block and tackle to a hard attachment point at the top of a tripod because the rope in the system between the pulleys will absorb the torsion.

With the examples you gave, there is something in the system that allows for absorbing the torsion. On a biner to harness attachment, there is all of the rope between your attachment and the anchor to absorb that torsion.

When I teach, I make sure that people know the difference between hard attachments that will not absorb twisting and those that will.

[lutan1]

I think that the difference is.....

All good points DCFD, however your opening statement still makes me wonder about the facts to support it- I THINK.....:confused:

[DCFDRescue2]

Agreed, I too like empirical evidence. However, since there isn't any regarding this, that I have seen, we're all just in the "I think" category. Not a great way to operate, but better than not thinking about it at all.

Does anybody have any ideas as to how to test this or tests they would like to see? I'm sure given enough time I could do some tests. I'm not the most scientific guy in the world and I'm sure there would be variables I wouldn't think of accounting for, but I have a access to a dynamometer and could give it a go. Heck, even if I do nothing but break stuff, I've still had a good time:D .

[MEDIC0372]

It has to do with side loading the gate...when the biners twist.

Connect two biners and unlock one of the gates (or both). Now twist them while keeping pressure on them in the op direction.

If you did the test right the biners will come apart...(some people call this "rollout")

For that reason I will never connect two biners together...

[Bones42]

Connect two biners and unlock one of the gates (or both). Now twist them while keeping pressure on them in the op direction.

If you did the test right the biners will come apart...(some people call this "rollout")
So, can't that be overcome simply by locking both gates?

[Halligan84]

Don't simplify things here Bones, we're talkin rope

My 2 cents on the subject.. similar to DCFD, make sure that the load can absorb that torsional energy. As a rule I try not to link 3 hard points together, we did some backyard tests at work years ago and 2 never seemed to be an issue, but you could maneuver things around with 3 to put some stress on that middle biner.

[MEDIC0372]

So, can't that be overcome simply by locking both gates?

Yes my friend you should have both biners locked...no kidding.

Do they always stay locked...NO.
Plus this is an example of the stress that is put on the gate. If it was locked for your little test you would not see what happens under a true Rescue load. I may not be the greatest writer in the world and I may not be explaining this to you the best way possible but...DO NOT LOAD TWO BINERS TOGETHER. The standard locking biner WILL come unlocked and open when twisted (maybe not every time).

This is the reason not to lock two biners together. It is not safe.

Do a search on "Carabiner rollout" also search the OSHA area.

Also call some manufacturers and ask them. The NEWER carabineres are much better at preventing rollout. If you do rope rescue or anything with ropes you should know about "rollout".

Here is a small sample of what you may find if you do a search. Much info about snaphooks but biners work the same...Remember the older style bines do come unlocked (I have seen this many times!!!)

http://www.capitalsafety.com/portals/0/MISC004.pdf

http://www.islandnet.com/cgi-bin/postit?login=sarbc&topic=rope&order=thread&article=5079

Please be carefull...
MEDIC-0372

[ResCap661]

I was always taught not to.
My question is, why would you have to?
You would be using more equipment, One more spot for something to go wrong.

Keep in simple.

[Rippin73]

I worked with an experienced climber. He explained that "roll out" was also known as "The American Cross." He stated if you have no choice to go biner to biner then you double them up (a total of four) with the gates opposite and opposed. If you can avoid hard linking, if you cant "I think" this is the way to go.

[resqtek]

Carabiner roll out can happen when a carabiner or snap hook is attached to any d-ring, o-ring, carabiner, or any other attachment. The probability of roll out increases when the object diameter decreases. Roll out is most likely when a system is subjected to loading and unloading which may allow carabiners to rotate during the slack moments.If roll out is the reason you are avoiding hard linking carabiners, you may have a false sense of security.

As for being protected by locking gates....... recent tests done by rigging for rescue have shown that a locking gate on a 40kn NFPA carabiner can be broken open by a latteral inward impact on the gate of as little as 2-3kn. It is important to pay attention to the orientation of biners before loading. Once they are oriented and loaded, roll out is a non-issue.

[Techresq]

In rescue work we use the "D" shaped Carabiner because of its strength. The strength of the Carabiner is in the spine. One of the main purposes to the D shape is to help hold the load to the spine. When cross loading and/or side loading the Carabiner the load transfers from the spine to the frame of the Carabiner. Once this happens the weight is not only transferred to the frame but also to the gate. The pin in the gate that goes over the nose of the Carabiner has the weight transferred onto it also. From some of the articles I have read it is usually the pin in the gate what fails. In these tests they found that the pin would fail somewhere in the neighborhood of 600 to 1000 pounds. I cannot remember where exactly I read this but it was in one of the trade magazines possibly the one Resqtek is referencing.

[Riversong]

This is just one more example of the "mythology" or "dogma" of the rope rescue community.

Are there real issues here? Yes. But those issues are hidden rather than revealed by blanket "rules" about such things as hardlinking carabiners.

Most of the issues have been addressed here, with the exception of the point loading and metal hardening potential (also speculative and not empirically tested, to my knowledge).

The real issue is not so much hardlinking - as has been pointed out, we routinely hardlink carabiners to harness D-rings, rigging plates, pulley beckets, etc - and not even so much torsional loading, since it is rare that there will be no softlinks in the chain to absorb moderate torsion.

The potentially fatal issue, as has also been addressed, is rollout. But this is far more of a problem when linking a descender to a carabiner and failing to make sure they're properly aligned before weighting. It is devices with a small eye that tend to cause gate locks to break and then roll out of the carabiner with potentially deadly result. The major culprits have been 8-plates and rappel racks.

This has been documented in the UIAA journal in 3/2000
http://www.lancs.ac.uk/socs/lumc/uiaa_equipment.pdf and it's been tested by Peter Poppal, director of technical information at Petzl, demonstrating that less than 1kn of force (basically body weight) can break a carabiner gate when levered by an eye.

- Robert

[jmatthe2]

Interesting article.

I noticed in the figures that the carabiners are oriented with screw gate needing to be threaded upward. Do you think orientation of the carabiner would make a difference in the outcome of case studies in the article?

[Riversong]

Actually, in one set of pictures the gate is facing away from the climber and in the other towards the climber. The gate is typically oriented away from the harness in rock climbing for ease of access, which makes it down when belaying a second from above and up when belaying a leader or when rapelling.

I can imagine scenarios in which levering the gate open could happen in either orientation, particularly when the carabiner is somewhat captive in the harness and can't easily reorient itself.