A bit more info here: https://vertikal.net/en/news/story/35316/5000-tonn...

Apparently the hook block failed and what you see on the video is the crane jib flicking over backwards and given it wasn't designed for it collapsing at the mid point.

This photo shows it before and you can see the lack of rear flip capacity of the top half of the boom.



from here https://www.khl.com/international-cranes-and-speci...

The 5000 tonne capacity is clearly at minimum boom angle, but if it snaps then the whip back can't be avoided for high angle operation.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Good view here. Supposedly the hook was made by Ropeblock, I am curious to see how these are constructed especially how the hook portion is retained to that center pin. I was not able to find a good cross section.
https://www.youtube.com/watch?v=6BJTLRfWocM&fe...

What would "rear flip capacity" look like? Is it a strut or a cable system?

Edit:

1. Equipment owned by Cosco. Not good timing considering the mass of inventory they probably want to move in a short time.
2. Amazing that the 4 point hook just snapped off. It is hard to tell how it was secured to the shaft. Perhaps some weldment failed or someone forgot shear pins.
3. I'm wondering if the hook assembly had some type of fuseable link designed to prevent overloading. Even so I'd think it would have a safety factor of at least 2 or 5.
4. Sorry, one more comment. The uncertainty of the load has to be a prime concern for a ship mounted crane to avoid capsizing. How could they not consider the reaction due to loss of load?

It is a tubular member telescoping inside a larger tubular member, trombone style.
One part is pivoted from a fixed member, in the illustration midway on the top of the machinery housing, and the other part is pivoted to the boom.

The two back stops look like hydraulic cylinders.
They may stop wind events but given the violence of this event they would not have helped.
The tower or gantry would have acted as a back stop, but the upward reaction was so powerful that it bent the boom over the tower.
Conventional telescopic stops low on the boom would have been much less effective.

Bill
--------------------
"Why not the best?"
Jimmy Carter

At one time jibs were protected by a cable system.
A strut was pivoted near the pivot between the main boom and the jib and swung below the pivot.
Slack cables ran from a point on the main boom through the end of the strut and to the jib.
The cables came tight when the jib reached the maximum allowable upward position.
These systems seem to have fallen out of favour.
On a large crane they may do more damage to the jib than they prevent.

Bill
--------------------
"Why not the best?"
Jimmy Carter

Supposedly the failure was at around 2200t, so well below capacity

That makes sense.

Apparently this crane has no counterweights and uses water pumped into ballast tanks. There must be a huge amount of energy stored in the loaded system due to cables elongation, jib deflection, ship rotation, and the sudden out-of-balance condition of the ballast tanks. It seems like a sudden loss of load is simply a catastrophic, unrecoverable event.

Looks like a Din 15402-c Quad Forged Hooks. These hooks are forged. Unlike single and duplex hooks, the bottom shank is very challenging to perform NDE. There are not many forges capable of making a 5000T Quad Hook. Did you notice at the end of https://www.youtube.com/watch?v=6BJTLRfWocM&fe..., the front prop for the sheave assembly is bent. Most likely all of the structure above the machinery house will need to be removed before repairs can start. This is going to be challenging.
Fred

Must have been very exciting in the control room with the boom collapsing in the back and the hook block flying up within feet of the front window.

Quote:

the front prop for the sheave assembly is bent.

Did you mean the fron member of the gantry? that is bent.
Same thing, different names?

Quote:

Must have been very exciting in the control room

I wonder if some of the injuries were self inflicted when diving for cover?
I made more than one frantic run or dive working as a logger out west.
Finally sanity prevailed over the adrenaline rush and I quit and went back to school.

Bill
--------------------
"Why not the best?"
Jimmy Carter

The operator likely died from the horseshoes up his...

Dik

'Honey, you'll never believe what happened at work today'

Some more detail (including a schematic of the hook, with a possible failure plane drawn on) and pictures in this imgur album: https://imgur.com/a/gJCNHLE

Whoever drew the failure plane seems to think the center pin snapped.

Interesting photos. It looks like the shoulder for the nut is still part of the center pin, so it did not break where the red line went. It looks more like the nut failed, the threads failed, or the nut somehow came unscrewed.

Are we sure this huge stud was threaded? Seems more like some locking collar or something, hard to tell from the drawing or the photo...

waross You are correct the assembly which holds the luffing sheaves is called a Gantry by some and an A Frame by others
.
What I was trying to point out is the tension member which the boom is resting on, has taken the shape of the boom. In the first picture provided by Eufalconimorph, the member is straight. Once a large box section is bent it is usually less expensive to discard it and start over.

Well that is not a Din 15402-c Quad Forged Hooks. The red line is not correct. In this photo (screenshot from Eufalconimorph's set
) The bottom of the taper the cylindrical section, which is threaded is still attached to the shank.


Possibilities -
1) The nut cracked and the inside diameter increased until the threads no longer engaged each other.
2) Threads stripped.

I am out of guesses
Fred

Quote (Eufalconimorph (Computer) 6 May 20 00:18)

Some more detail (including a schematic of the hook, with a possible failure plane drawn on) and pictures in this imgur album: https://imgur.com/a/gJCNHLE

Whoever drew the failure plane seems to think the center pin snapped.

Granted, the following overlay of images from the quoted link is very crude, I would agree with the assesment that the failure was in the pin, at or above the seat of the nut. With this failure occurring at less than half the design load, it will be interesting to learn what metallurgical flaws were present and why they weren't discovered earlier in the fab process.

That is certainly the least expected place. Someone is definitely off to the x-ray vault to see what those films look like. And the signatures.

Agreed the overlay makes it clear, the shank fractured. Particularly unusual.

12, injured, two treated in hospital. All injured persons where in the control room (on the ship). There were a further 120 persons on the ship, who apparantly are all well. Guess this could have gone far worse.

I'm still just surprised that the design is basically a giant Jesus Nut.

The bearing areas on the nut from the hook block don't look very big to me.

And the hook looks like it is able to swivel around.

Any difference in hook load or rigging would seem to put a large bending moment on that thick, but reducing diameter shaft.

To break in what looks like pure tension is difficult to understand.

I'd love to see more drawings of the hook assembly though.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Quote (LittleInch)

To break in what looks like pure tension is difficult to understand.

I'm still trying to convince myself that it was loaded in pure tension. Looking at the rigging, the mooring lines would need to have been looped around to opposite sides of the hook in order to ensure uniform loading. Was this done? If not, it would be difficult to avoid putting a large moment on the shaft. I wonder what is the load rating if only one or two adjacent hooks are loaded?



Edit:
From this second view I'm gonna say there was almost no hope of ensuring a uniform load.

With this design, when load is applied, the pin will stretch and constrict. But still, at only less than half the design load, the moments had to have a flaw to exploit.

Added: To break in pure tension, implies necking, which would be visible.

Looking at that photo above, maybe it's the camera angle but if looks like the main lines are not vertical. The barge clearly can't move any closer but if it's trying to lift at an angle it won't help.

It looks like such a clean break that there must be some massive inclusion.

I hope the design didn't mix up lbs weight with kg....

But for five thousand tonnes that shaft looks mightily small to me. I reckon you would need a 1m diam shaft. Not easy to make in high strength steel.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

I interpret that drawing to show a bearing behind that nut that allows the hook to rotate. I would call that a slewing bearing if it was in one of my machines. Slewing bearings like this can be rated to loads in this range. The rating of the bearing for pure thrust is highest. If any moment is applied, the thrust capability of the bearing is greatly reduced. This is the load chart for a bearing of that type.



If the hook is free to rotate on that bearing, then the tapered area above the bearing is just a guide and not an interference fit, as I first assumed. I would look for fretting damage from rubbing on that area for crack initiation. We have seen this sort of failure in large piston rods in reciprocating compressors.

How is this area lubricated?

Johnny Pellin

I found a better shot of the rigging. This view makes me wonder if the mooring lines slipped off the reaction bollards causing an impact load on the shaft. That could explain the clean break. You can see that the bollards are not in line with the load and that the inside ears of the bollards do not have much of a rim to prevent a large diameter line from rolling off, especially when doubled as they were.

[img https://lh3.googleusercontent.com/LlIFK1K7gb13MQNN...]

I'm assuming that photo was after the incident.

But it's the design of the shaft and hook attachment I just don't get.

Why is it tapered?
How does that bearing work?
What are the strange tapping holes?
What's the plate on the bottom?
Nothing looks right.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Is that a hydraulic nut?

A.

I think the plate at the bottom retains the grease.

The plate at the bottom is probably there to lock the nut so it can't unscrew. The holes coming in radially in the outside diameter of the nut could be for the tooling used to install and remove the nut (spanner holes). The bearing looks like a cylindrical roller bearing that I would call a slewing bearing or slewing ring. We use them in coker drill stem rotary joints that operate with loads in the range of 100 tons while spinning at maybe 30 rpm. The drawing does not make it clear that this is a threaded nut. But, if it was hydraulic, I would expect to see clear indications of the o-rings and back-up rings in the fit and the hydraulic connections. I suspect it is most likely threaded. The pin is tapered just because it needs to transition from the small nut to the larger area at the other end where the horizontal pin passes through. It is probably better to make this transition with a gradual taper than large steps.

Johnny Pellin

Quote (charliealphabravo (Structural) 6 May 20 19:40)

... makes me wonder if the mooring lines slipped off the reaction bollards ...

It appears that the lifting points are canted longitudinally but not rotated towards the center of lift. This of course leaves all of the inside lines vulnerable to slipping off. I see some cinching cables, highlighted below,
but I struggle to understand this system.



Added: I see it now, when the hook is at the determined height, the canted planes pass through the center of lift.

The hook manufacturer chimes in. I think we already knew this though.
https://renews.biz/60152/orion-hook-designer-cites...

Brad Waybright

It's all okay as long as it's okay.

In short, the front fell off.

https://www.youtube.com/watch?v=3m5qxZm_JqM

With such components of existential importance to the system and without redundancy, and especially when lives are at risk:
Rules should require a pre-test in a safe working environment, step-by-step increase of load, thorough check for cracks prior and after.
But I don't know whether this is perhaps already the case with load hooks?

I don't know whether the enquiry (at least the technical part) shall be public, or being made public. But I'd very much like it to be made public, in all detail.
Such sudden death of a big size forging is quite unique to my professional experience. The more with the certificates and checking that must have undergone this particular one.
I have seen once a cluster of forgings that had been heat treated incorrectly, they just fell apart during transport, due to internal stress state.

Does anyone know whether there's a requirement to have a material sample going with the main hook components during heat treatment, in order to do a metallographical analysis?

Roland Heilmann

Another video;

https://www.linkedin.com/posts/tamlynlingham-a197b...

That's actually very interesting.

The hook falls off and the block doesn't move for several seconds.

It's the fact the vessel is ballasted on the side that causes the large motion and movement of the crane arm backwards.

Seems like a bit of a design flaw to me. If the load had to be suddenly lowered in operation then the ship clearly can't ballast level fast enough to prevent large motion of the vessel and crane arm.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

I'm thinking when you're lifting 5,000 tons, you don't "suddenly" do anything. I expect that would have pretty much been a disaster with a land-based crane too.

Yes, I was thinking the same thing, the tilting of the ship as quickly as it did probably added enough inertial momentum in the boom structure that this contributed to the failure. And as LittleInch noted, there was very little movement caused by the failure of the hook itself, as nothing happened until AFTER the ship tilted toward the wharf, that's when all the 'excitement' stated to take place.

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

J Stephen. The massive difference with something like this monster https://www.ale-heavylift.com/sk10000/ is that the counterweight doesn't move....

It's clear that is the motion of the vessel which causes the crane jib to flex back until it goes beyond the vertical.

If the job wasn't as vertical as it was it might have survived. I do wonder though if this scenario ( loosing the load very suddenly) had been properly analysed.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Was the loading really only 50% of rating? It seems to me that lifting something off water is harder than lifting the same weight off dry land due to surface tension.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! https://www.youtube.com/watch?v=BKorP55Aqvg
FAQ731-376: Eng-Tips.com Forum Policies forum1529: Translation Assistance for Engineers Entire Forum list http://www.eng-tips.com/forumlist.cfm

Actually, because the 'weight' was a barge, floating in water, as it was being lifted out of the water the suspended load would have increased as the effect of buoyancy was reduced. As for surface tension, I suspect that compared to the weight of the barge, that it was negligible.

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

"It seems to me that lifting something off water is harder than lifting the same weight off dry land due to surface tension."

I suggest you think that out a little bit more. You are probably thinking about suction effects that may occur when trying to pull something out of mud, but that too is not relevant here.

Well, I was looking at some Youtube videos to see what happens in similar cases. And, there aren't too many cases where the load is dropped like that, and even fewer with a crane that can go backwards, and where they show what happens to the crane.

In the first clip on this video, it looks like the crane turned over rearwards due to load release, but it's hard to tell for sure what was happening with it:
https://www.youtube.com/watch?v=3q8ylgL0Fy0

In this clip, there's a pretty signifcant load release and you can see the boom (and crane) bouncing around pretty good, but not flipping over backwards, either:
https://www.youtube.com/watch?v=YJq-RuJxA-A

I'm uploading that new linkedin video that shows the beginning of the failure to ET for posterity. At least I think that is what I did. It says jpg and png only but it still created the link for an mp4.

https://res.cloudinary.com/engineering-com/video/upload/v1589138202/tips/orion_linked_in_failure_video_furjwm.mp4

Edit:
I don't think that they ever intended for the barge to come clear of the water. Those lifting points seem too close together for such a long load. At any rate, the real-time load would be metered/monitored by the crane.

Dynamic control of counterweight to maintain crane stability - like an automatic ballasting system - or moving counterweights used in some tower cranes (https://patents.google.com/patent/EP2360113A1) creates a situation where counterweight position (or moment) is critical to a safe lift - and sometimes structural integrity of the crane.
In this case in addition to the obvious situation with the hook, there is (to me) a question hull stability design.
So on a barge or ship mounted cranes is the backwards boom stability with maximum dynamic hull rotation in a loss of load event a required consideration when designing the load chart?

Quote (FacengrPE)

So on a barge or ship mounted cranes is the backwards boom stability with maximum dynamic hull rotation in a loss of load event a required consideration when designing the load chart?

Well I think it will be now!

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Trying to think of somewhere in Europe that could do a static pull on a 5000 ton test piece. Never mind geometric issues.

850 tons was the max I knew of 20 years ago.

They used to test the offshore cranes in Aberdeen harbour with a similar setup but a lot smaller barge. They were though the recertification checks.

I think this will be covered by DNV who were always pretty hot on logging of heat treatment and NDT.

This test could only have had one acceptable outcome, and that was successful completion. Perhaps there was some monitoring of power and stress point metrics to verify calculations, but that catastrophic failure could have only had one result.

Brad Waybright

It's all okay as long as it's okay.

It's curious though as to why the barge has eight strong points, but only 4 were in use. Was the hook supposed to be higher to make the rope loads more "vertical". See the picture from CAB - 6 May 20 17:54

But once that boom went past vertical then the result was not in doubt.

I guess there were very good reasons why the crane is on one side of the vessel, but the ballasting arrangement to allow any load to be moved on or off the vessel would be significant and involve a lot of water movement between ballast tanks. Fundamentally this is the key to this failure.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

I think its a test barge that they use in that port for all of the crane certifications and recertifications.

Its not the first time its been used and it won't be the last time either.

They don't lift off the actual deck they lift off barges and then position. The ones I have seen so far building off shore windfarms have all used jack up rigs next to the work site.

But they are wanting now to have one vessel putting foundations and piles in then build the complete mast plus turbine and blades on the shore or maybe on the deck then lift the whole lot into position in one lift. Before it would be piles then mast, then turbine on top then fit the blades. And fitting the blades apparently was extremely weather dependant.

It seems to me that it wouldn't be too hard (in a relative sense) to test that hook. You could come up with hydraulic jacking arrangements, you wouldn't need to actually do it on a crane.
In the US, smaller lifting hardware typically uses a factor of safety of 5 or so. They may lower that in a case like that, especially if they can analyze the heck out of it. But still, it had to have failed at just a fraction of its intended strength.
I'll bet barge has a pretty good ding in the deck now.

Quote (JStephen)

But still, it had to have failed at just a fraction of its intended strength.

Is that certain at this point? Firstly, the crane is only good for the rated load under narrow constraints which surely include restrictions on uniform loading of the hooks. Secondly, I understand that safety factors might be as low as 1.3 for these heavy lift systems. Thirdly, the rigging system doesn't give much confidence of a uniform load application. I can see a rather large moment being applied to the hook which might quickly put you well past the stresses expected at the 50% load reported by the crane sensors. Finally, I can't really see them getting this far in the process without testing and certification of the hook prior to installation.

JStephen-
You're right in that of all of the component parts of the crane, the hook would probably have been one of the easiest to conduct a static load as a separate part. In fact, I guess we don't really know whether such a test wasn't performed prior to shipping. We also don't yet know the failure mode, so it's possible that it might have passed in a different configuration. Ropeblock might be having similar thoughts in hindsight as I'm sure their liability is probably into seven figures.
BTW, from reading the press releases It seems to me that Liebherr was quick to note that the hook came from a 3rd party vendor and I get the feeling they're trying to throw them under the bus. I would say that unless that equipment was purchased/provided by somebody else like the customer, then it's part of Liebherr's equipment it and it's theirs to own. Ropeblock makes a similar notation that the failed part comes from an outside supplier, but they seem to defend that supplier's design, construction, materials, etc. a little more than Liebherr.

Brad Waybright

It's all okay as long as it's okay.

I suspect that the mindset of the people responsible was that they were NOT testing the hook. They were focused on the crane and its structures, and had made certain assumptions about the other components.

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

I'm curious to know how far into the testing program they were. These thoughts are easy to come by in retrospect, but perhaps some thought needs to be given ahead of time to the consequences of plausible failures. I'd imagine a "lesson learned" coming out of this is to evaluate the repercussions of a hook failure when the boom is at such a high angle, and that there should probably be progressive testing at lower angles (and necessarily lower loads) before getting to that point. Again, I'm not saying I would have thought that before seeing this, just thinking about what could have been done to limit damage if the hook were destined to fail.

Since the crane only controls 1/2 degree of freedom of the boom in an application where 1 full degree of freedom is a possibility; perhaps there will be some design change to add the other 1/2 degree of control with a second mast located below the boom and a smaller cable reel to restrain the boom in a potential recoil in case the load is lost for any reason. OTOH the same 1/2 degree of control is true for tower cranes - if they have counterbalance a heavy load I'm not sure that sudden loss of the load can be withstood by the tower bearing and not topple off, which seems to be a rare occurrence, no doubt due to the high standards for fabrication of components.

Quote (Pete K)

I'd imagine a "lesson learned" coming out of this is to evaluate the repercussions of a hook failure when the boom is at such a high angle, and that there should probably be progressive testing at lower angles (and necessarily lower loads) before getting to that point.

The system may not have the moment capacity to test the hook at a low angle. Although this is interesting to think about. The moments have to zero out. Is the system safe from the failure mode of flicking the boom over at a low angle? This may not be too hard a problem to solve with energy methods.

Quote (I'm uploading that new linkedin video that shows the beginning of the failure to ET for posterity. At least I think that is what I did. It says jpg and png only but it still created the link for an mp4.)

Thanks for the MP4 heads up...

Dik