Kalymnos Anchor Failure

My preliminary report on the recent anchor failure at St Savvas on Kalymnos is now up on Crag Chemistry. This one was something different.

Really appreciate your work David.

https://www.gofundme.com/f/s8w3fp-crag-chemistry-independent-bolt-failure-testing-fund/cl/d?utm_campaign=pd_ss_icons&utm_content=amp20_t1&utm_medium=customer&utm_source=copy_link&lang=en_GB

Thank you.

The conclusion, for those that don’t need the methodological details:

“Until proven otherwise, all installed 10mm Petzl Goujon expansion bolts of vintage, say 2000 to 2005, should be treated as dangerous. It should be emphasized that this condition arises from the nature of the bolt itself, and is independent of its installed environment.”

Wow.  Great work!

Curious...is hydrogen assisted cracking the same as hydrogen embrittlement?

Should austenitic stainless usually be more resistant to HAC?  

Wild stuff!   

Thanks!!

Brian in SLC wrote:

Wow.  Great work!

Curious...is hydrogen assisted cracking the same as hydrogen embrittlement?

Should austenitic stainless usually be more resistant to HAC?  

Wild stuff!   

Thanks!!

Maybe HAC is best thought of as a special case of hydrogen embrittlement. I have seen the term H-SCC used. Although not strictly correct because it is mechanical rather than chemical, it gives a better idea of the supposed mechanism. As you say, fully austenitic SS is pretty much impervious to atomic hydrogen, and so by analogy to Cl-SCC, the crack is progressed by diffusion of the attacking species (in this case H) along the crack to the tip, where it reduces the local fracture toughness.

Intellectually it is a great concept, but in practice it means coming across a nice stainless bolt doesn't bring the same cosey glow it once did.   

I guess that leads me into the question...

If your buying/placing bolts made from 316, how would a person know they weren't "good"?  Is there an inspection or test one could do to maybe sniff out a mis manufactured bolt?  Liquid dye pen or some such?

Thanks!

Brian in SLC wrote:

I guess that leads me into the question...

If your buying/placing bolts made from 316, how would a person know they weren't "good"?  Is there an inspection or test one could do to maybe sniff out a mis manufactured bolt?  Liquid dye pen or some such?

Thanks!

Depends on what you call mis-manufactured, back in my distant past I worked for a company that overhauled F1 engines and we had a spate of main-cap bolt failures which meant we had to reduce the engine run hours, as the specific version we worked on was used for endurance (Le Mans) this was kinda a hassle. The actual bolts were made by an aircraft engine supplier but in the end the original engine builder had to have it's own set of rolling dies made and these were personally carried to the bolt company and only used while the tech personally watched. And then they all went under the microscope. The dimensional standard allowed under the DIN norm for rolled threads is reasonably slack as is the surface finish, it's mass production and that's what you get.

It's worth noting that we as climbers are using this type of anchor off-standard, no building code in the world is working on one bolt with that type of loading to provide 100% safety especiaĺly without routine testing so mis-manufactured isn't really the correct term.

Right.  But...here's a few of my takeaways from David's (excellent) report (and given that its preliminary):

but likely prime cause is hydrogen assisted cracking (HAC): extreme cold-working of the rolled threads has caused a transition from chloride SCC to hydrogen‑assisted SCC.

We are certainly looking at a progressive cracking phenomenon that is being driven by the stress field of the preload applied during installation.

And here we come to another important observation. The metal, even at the centre of the bolt, has been so work-hardened, presumably by the final dimensional reduction of the bar stock, that it is hard to discern actual grains.

The hypothesis here is that severe cold work at thread roots creates conditions in which cathodic hydrogen‑assisted cracking becomes the dominant failure mechanism, rather than classic anodic slip‑dissolution SCC.

Cold-rolling, as a properly controlled process, leaves the thread root with a compressive residual stress. This is considered beneficial for SCC resistance. However, when incorrectly controlled, the very opposite of that which is desired, ensues. The material of the root and the thread flanks not only becomes extremely hardened, but a high residual tension is imparted. And this, even before we consider the impact of this treatment at the microstructural level.

Ok...so...what can folks do to mitigate the above issues?  Back off installation torque to reduce preload?  Inspect the rolled threads and look for...smearing?  Or?  Is there a more discreet dimensional check that might flush any issue with the rolled threads?  Visual check?  Non-destructive test?

Maybe, with regard to 316 bolts, cut threads might be better??  

Edit to add:  maybe glue in's are the (only) way?

You hate to think that any bolt installation could become a time bomb when the bolts aren't "mis manufactured" or botched during installation.  We get what we get and there will always be a risk?  I'd have a hard time swallowing that...

Brian in SLC wrote:

Ok...so...what can folks do to mitigate the above issues?  

You could reduce the risk by using non-identical bolts for the anchor. Different manufacturer would be best, or different batch, different year, different diameter.

Brian in SLC wrote:

Right.  But...here's a few of my takeaways from David's (excellent) report (and given that its preliminary):

Ok...so...what can folks do to mitigate the above issues?  Back off installation torque to reduce preload?  Inspect the rolled threads and look for...smearing?  Or?  Is there a more discreet dimensional check that might flush any issue with the rolled threads?  Visual check?  Non-destructive test?

Threads are the failure point, so, for mechanical bolt designs, those that have threads deeper in the hole, like Powers 5-piece, are at a slight advantage. 

Consider, if threads are near the surface, the failure mode described here will cause immediate, catastrophic bolt failure. If threads are deeper most of the bolt remains in the hole, giving, hopefully, time for the issue to be noticed and addressed.

EDIT: Thank you, David Reeve, for your detailed, careful, and illuminating report.

So, what can we do? I think Jim Titt has hit the nail on the head when he points out that this is mass production, and yet perversely we are using the product in a life-critical application. I like the idea of enforcing true redundancy at the anchor, but this will be a hard sell. When stoke is high, corners get cut - that is why so much crap ends up on the cliff. Yet common sense says having two different sources for the anchor bolts must work. Maybe this is something for the rebolt funds? In my experience, it is the rebolt funds that bring a more engineering-oriented view of the task in hand.

Short of looking at a polished section under a microscope, I don't know how one can guard against this. And even if you did that, how would you interpret the result when you have no control identifier like batch number?

The UIAA have done the right thing in finally insisting that manufacturers have incoming goods checks on material composition. You never know, a day may come when something marketed as 316 is actually of correct composition. However, that only takes us halfway on the path of quality control. Of late, I am becoming increasingly aware that correct composition is not enough to ensure the safety we need. The production process from the mill to the bar stock the manufacturer receives is fraught with mismanagement. In the case we are looking at now, I am very suspicious that the bar stock received by the factory was in an excessively hardened state, and that is why the thread-rolling went wrong. I believe that only by performing a full engineering stress-strain analysis on every bar entering into production can we be sure the toughness required by a climbing anchor application is properly met. OK, maybe just checking the hardness of each bar would suffice.

David Reeve wrote:

So, what can we do? I think Jim Titt has hit the nail on the head when he points out that this is mass production, and yet perversely we are using the product in a life-critical application. I like the idea of enforcing true redundancy at the anchor, but this will be a hard sell. When stoke is high, corners get cut - that is why so much crap ends up on the cliff. Yet common sense says having two different sources for the anchor bolts must work. Maybe this is something for the rebolt funds? In my experience, it is the rebolt funds that bring a more engineering-oriented view of the task in hand.

Short of looking at a polished section under a microscope, I don't know how one can guard against this. And even if you did that, how would you interpret the result when you have no control identifier like batch number?

The UIAA have done the right thing in finally insisting that manufacturers have incoming goods checks on material composition. You never know, a day may come when something marketed as 316 is actually of correct composition. However, that only takes us halfway on the path of quality control. Of late, I am becoming increasingly aware that correct composition is not enough to ensure the safety we need. The production process from the mill to the bar stock the manufacturer receives is fraught with mismanagement. In the case we are looking at now, I am very suspicious that the bar stock received by the factory was in an excessively hardened state, and that is why the thread-rolling went wrong. I believe that only by performing a full engineering stress-strain analysis on every bar entering into production can we be sure the toughness required by a climbing anchor application is properly met. OK, maybe just checking the hardness of each bar would suffice.

Hardness varies considerably through batches, I can tell by how hard each length is to bend. The actual annealing standard is extremely wide as it is fundamentally a crude process and that isn't going to change soon, so long as the metallurgical structure is correct it's good enough. In one delivery batches may come from up to three different sources, we get the countries of melt on the invoice.

The works certificate gives a full engineering analysis of the pour but there's a hell of a lot of bars in one pour and as I wrote the anneal can vary a lot. Like it or not climbers have to find bolting protocols to work around what is supplied, sure as hell they aren't willing to pay for aircraft quality products!

Steve Bartlett wrote:

Threads are the failure point, so, for mechanical bolt designs, those that have threads deeper in the hole, like Powers 5-piece, are at a slight advantage. 

Consider, if threads are near the surface, the failure mode described here will cause immediate, catastrophic bolt failure. If threads are deeper most of the bolt remains in the hole, giving, hopefully, time for the issue to be noticed and addressed.

EDIT: Thank you, David Reeve, for your detailed, careful, and illuminating report.

Steve I think this is a reasonable idea. The only problem is the 5 piece are 304ss which would be not at all appropriate for this location. 

The hownot2 folks are selling 316ss sleeve bolts. I have some and may put at least one on new anchors.