Why all the hate for small rapides?

Here is the result of a test, where the failure point of the 3/16” National Hardware rapide was definitely above 4044 (+/-40) lbs. I pulled the rapide between two HMPE soft shackles. One soft shackle was attached to my jeep via a 3’ piece of 5/8” nylon rope, supposedly rated to at least 8000 lbs, but tied in figure-8s on bights for the connection. The other soft shackle was attached to a linescale 3 load cell on a fixed anchor. The rope broke at the knot, and the rapide was not even deformed. The stated WLL of this rapide is 1150 lbs. The breaking strength is presumably n*1150 lbs., where n is 3-5. Below are some “after” photos.

after4044

I’d done a similar test over a year before, also with a National Hardware rapide where I got to 3280 lbs, when the rope broke first, and the rapide was fine.

 

So why is there so much hate on the web for these small, non-PPE rapides? Partly, people don't realize they can be much stronger than other parts of their system. Then there is just the “lore” that unmarked rapides are always bad; and small rapides are much less likely to be marked. There is less area to put a numbers on a small rapide, the surface is highly curved, and the writing (whether by stamping or laser etching) adds cost. (Péguet does stamp 3/16" rapides, but the writing is barely readable, and is on the inner surface; US Stainless stamps just the steel type.) People take boxes of anonymous rapides they bought off eBay, search for the mankiest-looking piece, and then show break tests. Huh.


Bear in mind that your harness belay loop is rated to just 3500 lbs, and it is doubtful you would survive 3500 lbs on your hips without serious spinal damage.

 

I don’t mean to single anybody out; you will find similar opinions all over the web. But let’s look at comments from a clear canyoneering expert (whom I respect; I emphasize I respect him, but sometimes we disagree).

 

“What you are looking for are Rapid Links (aka Rapides, Quick Links, Maillon Rapide, Links). Available in many styles and sizes. The cheap ones are junk, meaning, they have no brand name on them, and their quality varies widely... and they are not really made for life-safety applications. These no-brand rapides are often referred to as "Pacific Rim", and are available at Home Deport, Harbor Freight, etc... I recommend against using non-brand-name hardware for life safety applications.”


The "life safety applications" is an interesting comment, because many of the more expensive rapides (including those sold by the quoted author) are not PPE-certified -- i.e., not certified for "life safety applications."

 

Do your own QA, dammit!

I agree that the sort of hardware store rapides that one used to find in a bin could vary a lot in quality; but I haven’t seen those bins in years. I’m also pretty sure the author of this comment did not judge the variation based on his own break-testing, and the idea probably spans observations of links from a myriad of sources. Today most stores sell plastic-closed cards with one rapide apiece, at about 2x the price of the old “bin” rapides, like so:

 

national_card

 

Here’s where you need to do your own QA/QC. No matter the brand, you need to examine the area that’s most like to have failures: the threads on/in the male and female sides of the nut. Partly open the rapide, and spray PTFE dry lube on the "male" threads and into the nut threads. You should be able to tighten the nut easily but firmly, so no thread is left showing. Do this several times.

Three years back, I bought 3/16" (5mm) rapides from three sources (Lowe’s National Hardware, Péguet, and US Stainless) in batches of ten, then subjected three from each batch to a pull to at least 1000 lbs (after the PTFE treatment above). After weighting, I was always able to unscrew the rapide with finger strength, or in one case (a Péguet rapide), light crescent wrench force.

 

 

Some thoughts QA/QC for those stamped rapides

  

Here's what the Péguet Maillon Rapide catalog says about their QA/QC: “Quality control throughout the manufacturing process is made through internal control procedures. Quality materials from our raw material suppliers is fundamental. A monitoring plan makes it possible to ensure a variety of controls on the entire production process. The final check allows us to offer top of the line products. In particular, our manufacturing of stainless steel is subject to unit control of the products.” OK, what exactly does that mean? 


Well, here's what Lowe's says: "In 2016, more than 13,000 product tests were conducted in 87 third-party labs around the world. The Quality Assurance team also works closely with third-party agencies to conduct preshipment product inspections prior to acceptance by Lowe’s. Products are inspected for proper labeling, functional operation and consistency across production samples to help ensure customer satisfaction. In 2016, factories were visited nearly 11,000 times to perform these preshipment product inspections."

 

Note that most Péguet rapides sold are NOT certified for PPE. And “certified for PPE” does not mean that occasional screwups won’t leak through:. Edelrid recalled PPE-certified rapides made by Péguet, because some looked like this:

edelrid_quicklink

 


As a long-time scientist for the gub’mint, I tend to have a skeptical view of QA programs. For chemical analysis, we had a choice of using our own QA/QC checks on the work of a colleague, or using a firm that had a gub’mint approved QA/QC plan. Use of the latter lab supposedly would not require us to make our own checks. Mixed in with our samples, we sent “ringers” of known composition for analysis, and our colleague got much more accurate results than what were quoted by the officially-approved lab.


(When I worked as a QA checker, I always had people pressuring me to sign off on QA. In one case, after I found errors, the originator of a spreadsheet had management assign him another, less-strict QA checker, who signed off. To no one's surprise, when  I confronted the originator about the mistakes, he responded "Doesn't matter, it was already signed off.")

 Your iPhone is made in China. Is there a Q/A Q/C program for it? Yes, through Foxconn. In fact, the exacting demands lead to strikes by the workers. And yet in 2016, it was reported that Android phones—which use far fewer QA workers – were less likely to fail.

Would 5mm rapides cause undue force for rope pulls?

Again, from the respected canyoneering site: “I sell Rapides made by a French company with excellent quality control called Maillon Rapide. They come in many sizes, and the 7mm seems to be a good balance of weight vs. strength and working size. Some people use the 6mm (especially for explorations), but they are small enough that the fattest ropes people might take into canyons might get stuck in the pull. The 8mm size is good for high-use canyons, but they are heavy. Larger rapides are not only quite heavy, but also may be big enough to let a biner block slip through.”

A minor note: “Maillon Rapide” is simply French for “Quick Link;” I’m not sure this is a brand. Péguet started using this term about 1940, but the name “quick link” has been used for the same product from other manufacturers for a long time. The “Maillon Rapides” that people respect are made by Péguet in France. Similarly, “Crescent Wrench” and “Channellock Pliers” are registered names, but we all use those terms for tools made by other companies.

 

Let’s look at the forces required to pull 8.3mm polyester and 11mm nylon (dynamic) ropes through 5mm and 8mm rapides:

8mm_vs_5mm

You can see the thickness of the rope made the greatest difference; the size of the rapide was a relatively minor effect. Possibly some of the extra force required for the 11mm dynamic rope, is taken up by the energy of stretching. The inefficiency of the pull over the 5mm link is a little greater – but comparable – to the inefficiency of pulling a rope over a carabiner. A 13mm rope (13mm is the inner diameter of curvature of a 5mm rapide) certainly will get stuck; but I don’t know any canyoneers who use ropes over 10mm.

 

How was the test done? I used a 44x2 lb AWS (CE-certified)  hanging scale, calibrated on small dumbbell weights. To record the forces, I took videos of the digital screen with my smartphone attached to the scale through a flexible tripod.

 

Each point on the plot represents 4 pull tests (hence the error bars). To get a value for each of the four tests at each point on the plot, ~20 times were used to define the “plateau" on a single pull test, and I averaged the 3 highest values from each plateau. I had the rapides hung from the top of my stairwell, and had ~20’ of rope up and through a rapide “pulley,” down the other side. The far side had a 5lb dumbbell, and I walked down the stairs on the other side, trying to maintain an even rate as I raised the dumbell. The force went up as the weight was lifted, plateaued, then went down at the end of the test.

 

The same canyoneering web site recommends an anchor tied with an overhand on a bight, hanging off one strand of webbing. Now, I use this type of anchor, and think is great for canyoneering-type raps. Assuming 1” climbspec tubular nylon webbing, the breaking force at the knot would be <2500 lbs. That’s plenty for most canyoneering rappels… but wouldn’t a 4000 lb-breaking-strength rapide be more than enough?

 

Do you think there are many rapide failures in canyons?

What are the “well-documented” cases of rapide failure in a canyon? I’ve searched hard on this; maybe I’m just using the wrong terms. I find just one case in the Blue Mountains of Australia, in a Péguet-made rapide. The failure was probably due to two factors: 1) to a minor extent, use of a delta versus an oval rapide; and 2) failure to close the gate. There are conflicting reports about the gate, possibly because no one wants to take the blame for screwing-up.  In canyons with lots of fine sediment, it is easy to get grit on a rapide threads, making the rapide hard to close.

 

I have personally seen a case where someone opened the rapide, then closed it over just a few threads. I had been there a few months before, and had left it completely closed; and I know one other party came by in the interim, and apparently opened and partly closed the rapide. Always check.