HR shackles can explode

[fxykty]

Originally Posted by Dockhead

You are getting correct and helpful responses which go to the heart of the matter. What happened to your gear has nothing to do with the properties of HR fittings.



Your running backs perform the same service mine do, but you are a PERFORMANCE CAT, which means that your rig is much higher loaded than a mono of similar size, which heels to relieve pressure in gusts.


You should listen to the engineers who have weighed in here -- they speak the truth. This gear didn't "wear out"; it was overloaded. There's a big difference.



The whole system should be upsized at least one size. The blocks and shackles will be heavier, but you can get the weight back by replacing the wire rope with Dyneema (something on my list to do to my own boat).



You should thoroughly inspect the whole rest of the rig and see if anything else is showing signs of being overloaded. God forbid you've got a toggle getting ready to give way.



You can, on the other hand, just replace the shackle (and block, and wire rope leader) like for like. But if the shackle was not up to the job, what part of the running back system will fail next? There is no guarantee that the next set will last 10 years like the first set did. The shackle in your photos was exposed to loads exceeding the yield strength, which means the whole system, if as you say it is all designed to the same WLL, has been overloaded, and will be overloaded again if you replace it like for like.


Running backs on a rig with a static backstay and/or aft swept spreaders may not be critical to keeping the rig up, which is probably why the designers felt they could undersize it, but the failure of any part of the rig in a dynamic situation can have unexpected consequences to the rest of it. You just don't want that.

You really think Atelier Outremer designed and built their signature boat and under spec’d it? Possible, but not likely. But our boat is due for a re-rig next year and as usual with any equipment replace we will be assessing the various components and replacing and upgrading as necessary.

I challenged the Harken tech on the size and working load of the running backstays system given the failure of the HR shackle in particular and he was definitive that the shackle showed age related wear that had weakened it. The adjoining block shows no wear or damage at all, even though same working load. He was confident to recommend replacing with like gear. I don’t believe he’s making his recommendations from ignorance.

We actually do have lower loads than a monohull of similar size - a friend with a Hanse 55 has 16mm shrouds while ours are 14mm, his primaries are 65 and ours 53, he has 50% more white sail sail area and weighs three times as much.

Performance cruising cat old style, not Gunboat style, means light weight, easily driven and relatively small rigs.

[Nicholson58]

Thanks for the heads up and reminder. We too have the occasional bits break including shackles like these. I’m a recently retired mechanical engineer, machine designer. I dealt with a lit of metallurgy and corrosion issues. Many of our parts are as old as 1984. Shackles run metal on metal causing localized wear and thinning at the contact points. This increases localized stresses and in turn, stress cracking and inter-granular corrosion. Many of our failed parts show the brown tea stain across the broken face. This is a regular issue with all of the SS we use on boats.

My opinion is that your components are well proper selections considering the performance and weight issues. You just need to keep in mind that the expected life under such conditions is shorter. In the ten years since your rig was selected, a lot has changed. There are many new options for performance rigging that could increase strength and reliability while even reducing weight.

Most of us on CF are not sailing performance machines. My boat is 40 tons and the main rig totals around 2000# with massive stays and hardware. We don’t expect stuff to break and mostly we don’t tolerate it.

[noelex 77]

That HR stainless steel shackle certainly failed in a different way to what I would expect from a conventional 316 shackle.

HR stainless steel is used by Whichard, Selden and others and this failure mode (if typical) is worth bearing in mind. On concern is the potential for injury from flying debris.

However, I agree with others that this indicates that the part was undersized. Personally, I do not have the faith in production boatbuilders that you hold. There is a great temptation to undersize components. There are very significant cost savings and with lightweight catamarans the weight savings are perhaps an even more attractive benefit.

For many of these components you can do the calculations yourself. This is worthwhile.

The Australian America’s Cup yacht designer Ben Lexcen had a famous saying along the lines of “Any component that does not break is too heavy”.

This is a fine philosophy for a state of the art day racing yacht crewed by professionals and carefully inspected after every race, but I suggest a variation in the saying for a cruising yacht: “Any component that breaks is too light”. I think this is applicable in this case.

[Boatguy30]

Thanks for posting the pics. Guess all these components need to be chucked in these applications at 10 years same as most insurers are now requiring for cat standing rigging.

[svHyLyte]

For the record, if one observes deformation of a metal part, stainless or otherwise, the part has been strained beyond its elastic limit. Strained in such manner repeatedly results in "strain hardening" or brittleness. Absent any residual elasticity, i.e. strain hardened, a shock load, which in a runner can be several orders of magnitude greater than imposed "steady" loads and is quite common, the result is often rupture or "breaking" such as you observed. It is not a failure of the part but a failure of the application of the part. Moreover, regardless of whomever may have designed the rig or specified the rigging, the failure of the part is conclusive evidence that impact loading was not sufficiently accounted for in the design. Engineers do make mistakes

Rather than debate the matter and accuse Harken or anyone else of anything, be grateful your rig didn't come down and up-size the gear. It's not that hard or expensive compared to replacing a downed rig, No?

[Hesti]

It might be interesting to see the surface of the actual breaks. Chances are, that the breaking mechanism was a crack growth mechanism (fatique), possibly together with crevice corrosion. There is something called dynamic corrosion crack growth or whatever it is called in english. Then, under a peak load, the rest tends to give way, with some deformation on this rest. Normally, this can be seen when looking at the surface of the break. With some higher yield strength material, the crack growth per load cycle can be quite high due to stress concentration on the tip of the crack. This could be called brittleness.



The thing is, that with many steels and other metals, most of the lifetime is in the crack growth phase. So, chances are, that you have had a slowly growing crack already for some time. And did not find it because you were not looking for it. Designing on the limit is a twitchy topic - IIRC, 10 % higher stress amplitude can reduce the lifetime of an item by a factor of 10.


So - assuming that the break was being prepared by a slowly growing crack - and you now know that your design is a bit on the limit - it would be good if you had your rigging inspected for crack growth every year. Expensive? Probably. Or you just replace the stuff every 2 or 3 years to give you some safety margin on fatique. In a way, it's how airplanes are kept flying.

[Dockhead]

Originally Posted by svHyLyte

For the record, if one observes deformation of a metal part, stainless or otherwise, the part has been strained beyond its elastic limit. Strained in such manner repeatedly results in "strain hardening" or brittleness. Absent any residual elasticity, i.e. strain hardened, a shock load, which in a runner can be several orders of magnitude greater than imposed "steady" loads and is quite common, the result is often rupture or "breaking" such as you observed. It is not a failure of the part but a failure of the application of the part. Moreover, regardless of whomever may have designed the rig or specified the rigging, the failure of the part is conclusive evidence that impact loading was not sufficiently accounted for in the design. Engineers do make mistakes

Rather than debate the matter and accuse Harken or anyone else of anything, be grateful your rig didn't come down and up-size the gear. It's not that hard or expensive compared to replacing a downed rig, No?

There have been a lot of really good posts in this thread, but I think this is the best -- explaining the very mechanism by which these parts fail.



And this is also exactly why such parts should be adequately specified in the first place -- if they are overloaded to such an extent that they start to deform, they lose their elasticity and become prone to sudden failure. They should be sized adequately to avoid this.


If you see any structural metal parts on your boat with deformation, you should immediately replace them and consider upsizing them. An anchor chain is another example.


By running backs are 18 years old and have been through hell including a lot of hard sailing unreefed in 30 knots and a full knockdown in the North Sea, and although the shackles show some rust streaks, they are fine. That is because they are big and strong enough for the job including a healthy margin of error for hard to calculate dynamic loads. That's how you want your rig to be.

[a64pilot]

What you have was in fact most likely correctly sized, as all that is considered in sizing is breaking strength, even General Aviation aircraft are designed with materials tolerance that only takes strength into account, there is no requirement to consider ageing.

What happened to your pieces parts is called high cycle fatigue, without serious testing one can’t determine how much fatigue material will experience, and even then you must first know what are the cyclical loads and their frequency, and that’s nearly impossible.

So you have a few choices.
1. Materials substitution, meaning a material that is less susceptable to fatigue which usually means much less strength so you need to go much bigger.
2. Stay with the material you have but upsize it so that it’s outside of the fatigue cycle.
Let me explain that, yes it’s true that any steel will experience fatigue if a force is applied repeatedly, but it’s possible to simply go so big that the force applied doesn’t fatigue a part.
I attended a symposium in California by an Ag aviation group that talked to fatigue on aircraft structures, specifically wing spars, the FAA expert attended and he of course spoke that fatigue is going to happen all you can do is determine the life limit and change out parts before then. Now I knew that is what he was going to say and it aligned with our competitors design, but we had gone another way and built a wing that was strong enough to be outside of the fatigue cycle. I stated that, he responded that wasn’t possible. So I asked him if what he was saying was that if he took a coat hanger wire and bent it back and forth by hand that it would eventually break, he said yes pretty smugly. So I handed him a Rail Road spike and asked him how long it would take for him to break that spike by hand, he didn’t answer and I responded with see, it’s possible to build strong enough so that fatigue isn’t a factor.
3. Determine a life limit, which you already have done and change parts before failure.

So the designer did their job, they specified a part that was strong enough to not break under the loads experienced by sailing, but they can’t take fatigue into account because they don’t know what the frequency of the cycles are nor the amplitude of the forces.
In aircraft that’s called the mission profile.

If it were me, I’d upsize the part significantly or determine a life limit. If it were an aircraft you would use what is called a scatter factor of seven, and if it lasted seven years, you would change parts every year.

By the way, I feel sure fatigue is why we change our rigging at a predetermined interval, and by that I mean the rigging wiring. It’s the simplest way, another way is called damage tolerance, that’s where you do frequent inspections and change a part when x amount of damage has been done, but that’s real tough because you have to determine how much damage can be tolerated, but more importantly how quickly the damage occurs so that you can have inspection intervals less than the time it takes to have excess damage occur.

Actually I’m changing my mind, I would both oversize and set the life limit to less than the time it took the parts to break, that’s just safer and these aren’t expensive parts.

[a64pilot]

Are Titanium parts available?

[meirriba]

There may be a different scenario that caused the damage.
The holes elongation can also appear when a rig is loose and the mast is 'dancing' applying periodical alternating high and low tension on the stay and connecting hardware.
I am not saying this is the case for sure but it a situation that shows similar results.

[Dockhead]

Originally Posted by a64pilot

What you have was in fact most likely correctly sized, as all that is considered in sizing is breaking strength, even General Aviation aircraft are designed with materials tolerance that only takes strength into account, there is no requirement to consider ageing.

What happened to your pieces parts is called high cycle fatigue, without serious testing one can’t determine how much fatigue material will experience, and even then you must first know what are the cyclical loads and their frequency, and that’s nearly impossible.

So you have a few choices.
1. Materials substitution, meaning a material that is less susceptable to fatigue which usually means much less strength so you need to go much bigger.
2. Stay with the material you have but upsize it so that it’s outside of the fatigue cycle.
Let me explain that, yes it’s true that any steel will experience fatigue if a force is applied repeatedly, but it’s possible to simply go so big that the force applied doesn’t fatigue a part.
I attended a symposium in California by an Ag aviation group that talked to fatigue on aircraft structures, specifically wing spars, the FAA expert attended and he of course spoke that fatigue is going to happen all you can do is determine the life limit and change out parts before then. Now I knew that is what he was going to say and it aligned with our competitors design, but we had gone another way and built a wing that was strong enough to be outside of the fatigue cycle. I stated that, he responded that wasn’t possible. So I asked him if what he was saying was that if he took a coat hanger wire and bent it back and forth by hand that it would eventually break, he said yes pretty smugly. So I handed him a Rail Road spike and asked him how long it would take for him to break that spike by hand, he didn’t answer and I responded with see, it’s possible to build strong enough so that fatigue isn’t a factor.
3. Determine a life limit, which you already have done and change parts before failure.

So the designer did their job, they specified a part that was strong enough to not break under the loads experienced by sailing, but they can’t take fatigue into account because they don’t know what the frequency of the cycles are nor the amplitude of the forces.
In aircraft that’s called the mission profile.

If it were me, I’d upsize the part significantly or determine a life limit. If it were an aircraft you would use what is called a scatter factor of seven, and if it lasted seven years, you would change parts every year.

By the way, I feel sure fatigue is why we change our rigging at a predetermined interval, and by that I mean the rigging wiring. It’s the simplest way, another way is called damage tolerance, that’s where you do frequent inspections and change a part when x amount of damage has been done, but that’s real tough because you have to determine how much damage can be tolerated, but more importantly how quickly the damage occurs so that you can have inspection intervals less than the time it takes to have excess damage occur.

Indeed.


And in aircraft design, because of the extreme significance of weight, you have to push this as far as you can. Therefore, you get concepts like "scatter factor". And even according to your aircraft concepts, the OP's running backs are sized to be replace every 1.42 years, but on any boat other than an extreme racing boat, all the rigging should be sized to be "outside of the fatigue cycle", as you put it. Because even a performance cat is not that sensitive to weight that you would put up with a 1.42 useful life to save 2kg of weight in the running backs.



And certainly, once you get deformation like that of any rigging part, under any circumstances and at any age, there is just no question that it is not anywhere near sized right.


Deformation like that might not even be from cyclical loading; it could be from a one-time overload beyond the yield strength of the metal.

Originally Posted by a64pilot

Actually I’m changing my mind, I would both oversize and set the life limit to less than the time it took the parts to break, that’s just safer and these aren’t expensive parts.

But that's why we change the whole rig every X years; for example I have 6 years on my standing rigging and will change it again if I hit 10 years before selling this boat. It isn't reasonable to manage different useful lives of individual parts of the rig; everything in the rig should comfortably last that long with comfortable margins of error.

[Dockhead]

Originally Posted by meirriba

There may be a different scenario that caused the damage.
The holes elongation can also appear when a rig is loose and the mast is 'dancing' applying periodical alternating high and low tension on the stay and connecting hardware.
I am not saying this is the case for sure but it a situation that shows similar results.

That's a very good point, and people forget how much a rig can be damaged at the dock.



But these are running backs which would not be set up at the dock, so not applicable here.

[Mal Reynolds]

My HR shackle exploded in 9 knots of breeze, smooth water. I don't know the age of the shackle, but could have been up to 17 years as the boat was relatively new to me. In the photo, you can see severe crevice corrosion that resulted in the failure.

[a64pilot]

No, if our rigs were outside of the fatigue cycle we wouldn’t be replacing it on a calendar basis, we would be replacing them based on condition, meaning when they fail inspection, most likely from corrosion. But if they weren’t accumulating fatigue, they would last essentially until they corrode enough to be weakened.

Of course that may well make a rig a whole lot heavier than we want, a lot of weight aloft, so they are sized so that they accumulate fatigue, and are replaced on a schedule.
Rigging is definitely a life limited item.

But if left up indefinitely on a performance boat that is sailed in a manner to use that performance, it will one day come down.
That doesn’t mean it was a poorly designed rig, nor does it mean that it was made from junk materials, it just means that you exceeded its life limit.

Same for these shackles, they failed from fatigue, so I’d upsize them, and replace on some interval that I felt safe with, but it doesn’t mean that it poorly designed or made from junk.
I’d also take a hard look at the rest of the system that the shackles were connected to, it may be nearing its life limit too, assuming it was designed with the same loads the shackles were chosen for.

[Suricat46]

fxykty deserves many thanks for his report on failed HR parts.
Rather than guessing a solution to find a replacement part let's focus on the causes of the problem.
After more than 50 years in aircraft maintenance and engineering I got a few ideas on material fatigue and failure. Of course, aircraft design benefited from several thousands of inputs from similar failures on similar parts under certain circumstances. Each failure was analysed by the designer/manufacturer and analysis was supervises by the Authority responsible for the design. Over the years and according to experience, this resulted initially in "hard time limitation", certain parts being limited in working hours, working cycles or even working days.
The process was then tuned according to inputs and investigations to retain hard time for "critical parts" while others were tracked according to "condition monitoring" and replaced according to wear, cracks, corrosion, etc.
Designers have now gathered enough experience to move to "predictive maintenance" and design parts or systems that should not wear, crack, corrode, etc. (What you try to achieve when replacing a steel shackle by Dynema).
Unfortunately, sailcraft production will never reach aircraft production and feed back data. However, sharing experience in clubs, forums... will certainly help each of us to think differently maintenance and utilisat