Crack Simulation on Alex Thomson Hugo Boss 2020 Vendee Globe Boat Failure - Imoca Tech

[@salmiakki5638]

@Peak Torque i know your working off the little that is known and lot of assumptions, and therefore your analysis is to be taken with more than a pinch of Salt. But how it can be even remotely accurate when you don't know the size of the core, the ply number and arrangements, geometry of the plies etc etc..., regardless of the loads assumed? Calculating stressed on cfrp is really difficult.. besides what failure criteria are you using? Anyway, really appreciated the Video and the discussion, thanks a lot!

[@PeakTorque]

Of course, but I did this over a lunch hour for the sim and one evening for the video production. This is just a passion for me, so trying to find the time for a more detailed analysis is tough sometimes. The point of the video is two-fold; 1. When you get back to basics, flexural moduli, laminate thicknesses, cores or no cores, layup information, it all becomes academic if the basic design rules are broken in terms of stress raisers, and smooth load transitions between members. For this simple educational case, modelling the panels as solid orthotropic materials (albeit with estimated flexural modulus of 80GPa) was sufficient for the graphic demonstration. 2. To educate non-technical, but interested parties on how these failures can just 'happen' and how tough a hull design actually is because of the hull constraints and shifting load cases. If you want a full report, send me the drawing pack of the hull, rig, and full composite layup schedule, and a full weather and sea state history since leaving LSDO, and I'll send you back a very, very large quote 😉

[@PeakTorque]

If you don't mind I will pin this comment, because others have commented the same. I am grateful for your interest and comments. All of them. 👍

[@stephenclark9771]

I am a boatbuilder with quite a bit of experience and knowledge of advanced composite processing and fabrication. I also personally know some of the structural engineers who work on there projects. In fact I have worked with Gurit, the firm which did the laminate design for Hugo Boss, and found their analysis of structures to be pretty conservative and safe. As you know, Carbon laminates are very stiff and so fail abruptly. They are either fine, or broken because there is little flexibility in the composite. All it takes is a single load spike and things go from fine to broken. These guys use all the best FEA tools and aren’t bone brains but the bows of high performance ocean racing boats has been an ever moving goal post as speeds and slamming loads have increased. The last two generations of foil assisted IMOCA60s have significantly increased all the loads on the vessel, and not in a linear way. All that being said, Alex Thompson is a premier racer in this discipline, and Hugo Boss was built to be the fastest possible boat sailed by the most expert skipper so the factors of safety in the areas which most effected performance were probably minimized.

[@HughSheehy]

@@stephenclark9771 The last time I worked on any of this was decades ago but the physics doesn't change. As Peak Torque says in the video, it's the load cases that are hard to define. If (as you say) slamming causes a load higher than the load cases then the skill of the design engineers is almost immaterial (though they do seem to have some tight curves in the cutouts). I do wonder whether the boats are instrumented and whether they share load data. It would make the fleet safer.

[@stephenclark9771]

@@HughSheehy These boats have fiber optic load cells throughout many with alarms to inform the sailors when they are approaching maximum design load. There probably aren’t load cells in the bow, but vertical acceleration is monitored with alarms to warn if the boat is being chucked around too much. I think this data is shared on the Hugo Boss Hub site along with Alex’s biomeds. As a result of some rule specific constraints, it is entirely too easy for these hydrofoil boats to break themselves. For example the boats are too stable for the design of the mast. The mast design and construction is defined by class rule and cannot be changed. So load cells and alarms are vital. As I stated earlier, the designers use Velocity Prediction Programs to evaluate the consequences of all design choices. It is an accepted fact that adding weight to the bow slows any sailing boat, so this is reflected in the VPP. The designers would have some theoretical answer to the speed and time lost for each kilo forward of the mast expressed in seconds per mile. On a 20 mile race this is trivial, on a 24000 mile race you can believe these things will cost you a day or two. This creates a situation where the most vulnerable portion of the hull is subject to the greatest effort to reduce weight and pare down structure. Because this generation of IMOCA60s are quite a bit faster than their predecessors and the way the meet waves is very different, the previous best practice probably isn’t good enough.

[@kevindakin3847]

P is for Port, you can see in one of his previous videos that he has 4S marked on the fwd starboard crash bulkhead

[@CliffordDive]

As an old engineer who's got hooked on this year's Vendée, I'm enjoying your techie chats. Great stuff!

[@nathanchan82]

As a student engineer also hooked on this year's vendee, really enjoying learning things about engineering and analysis that I probably won't learn at uni!

[@santiagobenites]

Dude, you are everywhere! First I enjoyed the great analysis of the Hambini bottom bracket, and now it's on to Alex Thompson's 'barge'. Nicely done!

[@davidtydeman1434]

Thank you for your analysis which has helped this non engineer understand this issue. It is my understanding that a number of the French boats strengthened this area in their boats after failures during test sailing. I hope that Alex has not been compromised by his reduced preparation caused by COVID lockdown. He didn’t participate in Vendee Artic. Good luck and safe return for all the sailors

[@STANLIZ4]

Given the scant information available, all your comments are very interesting/relevant to my untrained engineering mind. This is the 2nd build so surprised there was no evidence from previous, but the French boats had similar problems in trialing, I understand. The boat surely has to be durable to withstand its speed capability. Having to back off because the hull cant withstand it would seem an odd approach. Surprised there was no water ingress considering the distortion needed to create these cracks. You didnt mention making the beam thicker as well as using larger radii, given the strength benefit to wt when carbon is so light for its immense strength. I hope H Boss dont stop providing details of any future problems, they are so interesting to contemplate cause and solution. Thanks again for your video, I found it riveting ( excuse the pun!)

[@andrewpease3688]

I think that the beams are numbered from the front and P is port side. The fore foot is usually monolithic and the frame itself is almost certainly cored with a slab of uni capping. The diagonal seems to be taking the forestay load in tension. The boat has had some secret modifications up front,which will change the loads a bit, although I doubt that this is the cause. Any mods are done with full design support these days. Alex has almost certainly gone through most of his repair materials now and these problems are not rare. Alex described himself as not a boat builder but he is very experienced and will have seen the pros at work so he has a very good idea.

[@VictorMarinus]

Wow! This is real prime content on the Vendee. No babling with previous skippers and dramatic words from a host like on the French daily Vendee show. But a real analysis of major problems. Even with a lot of assumptions and probably errors this is very interesting. Many thanks. you just won one subscriber. :-)

[@PeakTorque]

Thanks for the kind words, Victor

[@jonathanwilde1634]

Really good explanation of the engineering fundamentals, judgements and compromises that the designers are having to make. Really appreciated this as an engineer. In particular cheers for noting the caviats and lack of information you're working with - it's obviously much easier to find design problems in hindsight and designing and building something like this is exceptionally challenging. Great video with lots of information, explanations of your opinions, and acknowledgements of the limitations/judgements. Thanks.

[@PeakTorque]

Hi Jonathan, appreciate your comments. Cheers.

[@chapmag6578]

Excellent analysis and discussion. Let’s hope the repairs do not introduce additional stress concentrator. Thank you for doing this.

[@MrWightHD]

Excellent analysis and have subscribed on this basis. I'm a sailor, not an engineer but you have pretty much hit the nail on the stress gradient head

[@sheilamorrison1954]

Ah. A nice reminder that KZbin isnt just full of face vlogging drama in Beverly Hills. Fantastic work, thanks for taking the time. Everyone is learning something here.

[@PeakTorque]

Can you explain face vlogging in Beverley Hills? 😅 You mean semi-beautiful rich people walking about telling you about their day and Starbucks? Luckily my feed doesn't show me that. If you press 'subscribe' on my channel the KZbin algorithm will actually help you stop seeing that stuff 😁😁

[@VidarrKerr]

You must watch a lot of dumb videos. The algorithm is what you watch.

[@bpetersson5024]

great video and my concerns exactly; they compromised the integrity of the boat in exchange for weight gains. It seems like all the foiling boats have the same problem and the mast setup is also sketchy. The problem with carbon fiber is that once it cracks, the whole piece is compromised and considering the numbers of cracks in different places, the fix he did seems just cosmetic...

[@Man-go-Everywhere]

It should have Kevlar in the weave. I know what slamming our Glassfibre legend hull takes I would expect a carbon only to be too stiff in this place

[@superior451]

I’ve watched similar repairs on fast sailboats carried out by design/repair teams who explained that laminating plies is calculable abs how these panels are designed in the first place. However when done on the go with limited materials they introduce more variables along their periphery.

[@simonpuxley7374]

Thank you. Loads to learn there. In particular the tough job presented to the design team. Your analysis of the ultimate factor of the behaviour of the skipper is cracking too. Pardon the pun!!

[@AlexanderPanzeri]

I just found it thanks to suggestion by KZbin. Very well done! Also it confirms my initial “thesis” that the structure has been hyperstatic and not “elastic” enough. Without offend the designers looks like they forget a bit the basis of cells structure (we called in Italian).

[@pauloriain]

My thoughts exactly..... the broken are is too stiff, boat has to flex....

[@AlexanderPanzeri]

@@pauloriain I would say, you can make it rigid, but must survive... and because it’s real stuff, so think smooth...

[@whiplashe13]

great video i have been contemplating the same things... i read in one article they have load sensors over the whole boat optic fibre and the like and alex can see those in the cockpit. so the boat has to be sailed with in load tolerances which have been calculated assuming taking all the data available into account.... i was at a rope manufacturer a while back and some of the loads on these high performance boats on halyards and standing rigging etc can be upwards of 20 tonnes at times ... the tech is really quite astounding ... i hope he gets the spare rudder on and gets going though :)

[@PeakTorque]

Having a boat packed with load cells and alarms does not constitute good engineering let's remember. Particularly when the skipper's psyche (utterly sleep deprived and tormented) can so easily override it. Good engineering is also taking a proven design and refining it, with many many sea hours, a la the older gen Imocas.

[@s7ev0]

Great video and thanks for making the time to do it. I am a boat builder. When he drilled through it to make his repair it did look like it has a core the way it drilled my guess is honeycomb. It also looks like the carbon used in the panel was too thin. The repair panel took a lot of drilling where the boat drilled through quickly. The lay of the carbon looks like it was made in a vacuum bag so I would rule out any air voids. I would say design error as the boat should be sailed as fast as possible where Alex can handle the slamming he should not be slowing down for the boat

[@PeakTorque]

Thanks for the comment, all good points. However vaccuum bagging does not rule out voids and total de-gassing. Its only 1 atm of pressure! Very plausible to still have planar voids between plies. Obviously these things are too big for the autoclave process, but vaccum is not always reliable with only 14 psi!

[@donnahiebert8887]

@@PeakTorque interesting to consider, in light of your comment regarding original panel thickness, that, aside from the design, the interpretation of the design as it was being built is also a consideration - these are custom, really hand-built structures.

[@grahamsnyder762]

My gut say's you're right about the failures cascading once the first crack forms. Would be super interesting to see how your FEA predicts the truss would respond if you model a fracture at 2P where your model predicts it

[@barrydavies2977]

Interesting analysis. Thank you. The interesting, and worrying, thing is, he has not been pushing the boat too hard. He's been deliberately keeping his foot off the gas so as not to over-stress her. She's capable of 40 knots. He hardly ever done much over half that. I'm hoping that with the additional bracing he has put in, it will reduce fatigue stresses.

[@in4merATP]

Also remember there's that bloody great side/rear load along the deck plane where the headsail attaches, which likely creates a huge torsional moment about the bow box. I imagine the sail loading also plays a huge load. Those sheets (headsail lines) are huge for a reason, and only some of that is resultant thrust; the rest is balanced from the headsail attach on the bow.

[@jakobmathiszig-lee8046]

Super interesting analysis, watching how hard the vendee globe boats get pushed in rough sea's the dynamic loads on them must be enourmous. Hearing about the various repairs sailors are having to do is part of the fun of the race, i remember a few years ago someone had to keep their boat heeled over to allow the repairs to one of the rudders to cure. Keep up the videos!

[@clausmayr3204]

I think, similar happend now to PRB where the bow broke and pointed 90° upwards within seconds. Broke in two. I think the bow section from this new kind of foiling boats plus the stress by flying into waves permanently can't be simulated. So... i assume Hugo Boss und PRB won't be the only one and other foilers will follow during the race.... with same kind of cracks

[@490957]

I think this is a great background and suspect that your analysis is correct, failure was a result of a combination of all of the factors mentioned, sailing into tropical storm Theta wouldn't have helped. I understand that Corum had to have extensive stiffening of the bow section after trials but then the mast breaks! There is no real substitute for testing in representative trials of any prototype.

[@allgoodfun3843]

Good on you, you’ve done as much as you can from the information available to you, but also there’s huge diagonal leadings applied when the boats slam down at roughly 35/40 degrees when coming down onto a wave/or off the foils, and due to the roundish bow shape I think it may have been more as a result of the repairs not being up to it in the southern ocean rather than a rudder damage.

[@PeakTorque]

Yep, i agree. I did mention those loads as a combination of the vertical and horizontal. When its heeled the slamming loads are somewhere in between. I'm yet to see the broken rudder.

[@franks8327]

Well sailing nerds unite! Saw a lot of good comments on here including someone mention a video where Alex said: “the thickest carbon layup on the entire boat was either 2 or 2.5mm.” Also noteworthy is his history of catastrophe but then again at least he’s out there doing it. The team has also pushed there design concepts, both in 2008, and 2020 away from the French teams. I so want this thing to win with the enclosed cockpit.Then it’s interesting where the boat was built. UK? Think some French boats were built in NewZealand. Wicked Pissah video.

[@milusocal5325]

Thank you for the time and effort on this matter, very intresting! 👍🏻🍻

[@grahamm2015]

Boris Hermann has been putting out some videos on sail changes and his sail locker only seems to have a single beam. I realise it is an older generation of boat but still a long time at sea.

[@PeakTorque]

Boris' boat to me has a much more preferable design - Multiple floor and roof stringers, evenly and regularly spaced. May not be ultimately stiffer than HB, and arguably heavier, BUT no stress raisers. Big fan of Boris.

[@MrSmithToday]

One load that people overlook is the possibility of the forstay being over tightened. Would explain the bow on one of the Vendee Globe boat pointing up 90 degree. Snap. I was on an Island packet yacht that some doors would not open or close due to over tightened forstay. Reforming the hull.

[@billhanna8838]

I believe the major stress comes about when the rigging arms dip into the water at 20+ knots then the strain on the forard section (Twist) becomes to much for the forard section of the hull. Great representation .

[@johnswimcat]

Very interesting. I wonder how these foiling boats will fare in the southern ocean. My feeling as an ex designer is that they are not really strong enough but, as you say, that depends a lot on how they are sailed

[@jeffmorris9893]

If I remember correctly, the caps on aircraft wing ribs have gaps. They're not an uninterrupted circuit of the rib.

[@matthewwillis4892]

Great video, thanks, longitudinal bulkhead and ring frames is what we call them here on the west coast of USA.

[@mhaddadi]

If I was the engineer , I will implant a lot of stress sensors inside the structure, that can be recharged wirelessly, and a control station that will monitor the stress of all sensors wirelessly, that will give a clear picture of weak links, and allow to prevent a crack by reinforcing weak spots.

[@PeakTorque]

To do that you would need strain gauges absolutely everywhere, a shit load of wiring, amplification, signal conditioning, data acquisition...etc etc. Just not plausible unfortunately. Maybe for rnd purposes, but not on race day. Essentially this race boat is the rnd process, so maybe you do have a point! All depends on the legacy of the team and how much they want to invest in learning.

[@EngineerK]

@@PeakTorque You would think they would be doing this on a few critical components. They already know their forestay loads (see the link in my earlier post) so this strain gauge information would could help them in quantifying hydrodynamic forces - I would think they have a very good understanding of the loads on the structure just from the sail rigging. The delta between this and the measured strain give indication to the hydrodynamic force.

[@PeakTorque]

@@EngineerK yep would be good to see that delta in real figures. You can see on corum footage when it's slamming into big swell, serious detentioning of the forestay as the bow is slammed upwards.

[@EngineerK]

@@PeakTorquefixed the link in my post above with Seawolf live tracking check it out very cool

[@reedarcher8367]

Great video. According to the details presented in the quote below the hull only accounts for about 25% of the boat weight. I would think that increasing the safety factors in the hull design wouldn’t account for a huge weight penalty but could ensure that the boat is strong enough. They seem to discount the effects of having to stop and make repairs or getting rescued. “Thomson says Hugo Boss weighs 7.6 tonnes, of which the hull structure itself is just 2 tonnes (the keel accounts for around 4 tonnes, with everything else - mast, engine, winches - another 2). Some of the cleverest details of the yacht are almost invisible.Nov 14, 2019”

[@zbeekerm]

Great video! I wonder if the current race leader’s understanding of his boat (Charlie Dallin is a naval architect) and his close work with its designer has led him to know how to push his boat safely! Also, I’m mostly a CFD guy, but I agree that a cored central truss/web/stringer seems like it would have been a good idea. Nomex core would be hard to fix/replace but they could carry balsa coring for emergency repairs.

[@PeakTorque]

As an engineer Charlie will naturally have more feel and mechanical sympathy, and when you're pushing something with such fragility you really need to understand mechanical sympathy. You're right i think. Based on what I can deduce from the bullishness of AT he seems pretty gung-ho. I may be wrong. And this attitude has certainly gained him more fame, PR, sponsors, and arguably personal wealth than anyone else.

[@freddy01577]

Hi Sherlock! A good analysis and (probably) the first of many - you have raised a number of issues which I had arrived at and your concluding remarks are on the button ie adding structure/stiffening only moves the problem further up the chain - when I first saw Alex's video showing the damage it was pretty obvious the longitudinal beam was (in my books) a bit of a bodge. CF is very brittle and does not lend itself well to high stress areas - the loads exerted are dynamic and so the structure should have some sort of stress absorption - I don't know how you achieve this but I would be thinking of some sort of shock absorbing system. I'm reminded of very tall buildings having hydraulic pendulums (500+ tons) in them to counteract the effects of high winds which allows the building to 'move' and prevent the windows falling out. I fear the weight issue would kill this idea. Stay well

[@TomLeg]

Interesting to discuss engineering challenges or modern technology. Arecibo radio observatory in Puerto Rico has self-destructed as a result of earlier problems. Might Alex's cracks be related to PRB's splitting in two?

[@frankstocker5475]

The designers knew this boat was for Alex, Hello. If this structure had been made stronger would the arrival time be less than the time he had to stop to repair it? It will be interesting to see what modification they make when he gets back.

[@stephenclark9771]

To answer the load cell question. Yes these boats have load cells and alarms on every critical structure. They also have accelerometers and alarms to warn if they are being thrown about too violently. The hardest thing to determine in structural engineering is the load case, so not only are the sailors being given guidance on how hard to push ( like a tach and temp on a race car) but the designers are recording data for refining their FEA models. Hugo Boss/ Alex Thompson Racing is as close to Formula One style design as any in sailing. The builder Jason Carrington is one of the best and the boat has more than 50,000 man hours invested in its construction. So whatever has gone wrong should not easily dismissed as transparently stupid or shoddy workmanship and design. It is more than likely that if the boat hadn’t lost it’s keel striking an unidentified floating object in the TransAt Jacque Varbre, this might have been wrung out in more extensive sea trials.

[@PeakTorque]

I understand your comment and thanks. I hope I was not dismissing the designers. Of course its a fine line when you're pushing a the very edge. However I also don't believe packing a boat with load cells and fibre optics qualifies as good engineering. Good engineering is also building a solid boat, and refining it year on year with empirical testing. Look at how many of the last gen boats are going so well and not having problems. There is just no comparison to sea hours.

[@stephenclark9771]

@@PeakTorque You should read the rules which are available from the IMOCA site. There are things you can and cannot do, and some of the things that you point out as illogical are on the list of things that cannot be changed. I personally know one of the principals of VPLP and can assure you that they are very thorough and thoughtful designers with years and years of success in fast off shore racing boats. Alex Thompson Racing is not a slipshod group either. This was pretty close to good as it gets. One point however, these boats are now running around the oceans at speeds equivalent or faster than most commercial shipping. Suffice it to say that a container ship is not troubled by the UFO that vaporizes an IMOCA.

[@michaelphillips9979]

Excellent analysis PT. It appears to be Repetitive Flexural Stress Failure. Very similar in failure pattern to that which brought down the composite-constructed Commet jet airliners. The rectangular windows on that jet aircraft had very tight radius corners and the flexural stress from repeated pressurisation and then decompression (not to mention in-flight external buffeting) led to metal fatigue of the aircraft skin. Cracks formed between the window corners and several aircraft broke-up in flight. Let's hopethat Alex and the |HB team have strengthened it enough to resist the huge forces of the Southern Ocean.

[@sheadford]

OK, so, not an engineer but done plenty of offshore miles. As we now know Hugo Boss is retiring with (what must be significant) rudder damage. Rotten luck all round. My take on the structural problems at the bow just by piecing together the scant information available and maybe making some outrageous assumptions. He did broach badly in TS Theta. My impression is that these latest Imocas designs are actually quite hard to broach. Hypothesis - pushing hard as he does, and in the worst of the wind and seas, Alex buries the bow in a wave and the rear of the boat comes right out of the water, the rudders lose their grip of course and she's laid over on her beam ends. Imagine the forces at those speeds at that moment and the resulting stresses transmitted through the rig/hull/mast, particularly in the fore-stay triangle. I'm thinking of that frame in the bow as a basic parallelogram with the fwd vertical edge forced upward in that exact moment. There wasn't much happening after then that might've caused such structural damage so I reckon he was carrying it from TS Theta.

[@RobHealey]

Thanks for the vid. I haven't studied engineering, but I have a lot of experience (69 years) with things breaking and I've sailed a lot of my life. I posted this 3 days ago on one of Alex's video's and it seems you feel the same, that the repairs have now set up potential stress points where they end - "Brilliant work. BUT I hope that the repairs don't stop and miss the low curved part of the longitudinal brace because the edges of the forward and rear repairs will create the weakest point and it will fracture there next. Obviously the bow section is twisting and bending,. A design and/ or building fault for sure."

[@marcofrancioni1155]

you are missing the loading from the multiple stays holding the mast up, there should be a vertical load and a bending moment trough the front bulkhead and a vertical load applied above the oblique truss.

[@PeakTorque]

Yep, i mentioned i didn't apply the rig/stay tensions. Tried to keep it simple. I'm not a marine engineer so the force magnitudes were all guesses.

[@felixl4572]

I think the cut outs in the bulkhead are propably well thought of and designed using topology optimization as shown in this presentation: www.altair.com/resource/hyperworks-helps-to-develop-race-winning-yacht-for-volvo-ocean-race-2009 So Failure is more likely because of incorrect loads, because the dynamic behavior in waves is hard to predict.

[@scottl1989]

I hope you’re part of building the next one! We’ll see how right you’ve got things :)

[@carlamerritt490]

Just watched this video. I think Hugo the most aggressive skipper in the feild. I suspect you knew that but was being polite. 🙂 But just found this channel and love it. Biking and sailing my favs.

[@skydiverbob1]

The question is... Will the repair be good enough for the boat to remain safe for the rest of the journey or fill the boat fall apart completely in freezing waters when all the seams in the CF contract? I sure hope Alex survives this catastrophe!!!

[@tonyaustin4629]

I am not an engineer but I find it more than plausible that Alex Thompson needs his boats over-designed for durability since he has a history of damaging boats.

[@VUO4E]

He's not the only one. One of them even won the VG twice...

[@2adamast]

Meanwhile he got the lightest boat of the fleet, because that's also what he needs.

[@maxgover2990]

THANKYOU FOR COVERING THIS ❤️❤️❤️

[@johnpaulson996]

At 2:21 you can see the honeycomb core in the panel.

[@grahamm2015]

I guess you have another simulation to set up, hopefully not too frightening

[@PeakTorque]

Terrible news. No simulation needed! Poor chap. Back to AT... I still haven't seen the broken rudders. PR cover for the bow....?

[@grahamm2015]

I watched the Le Cam interview and one with the race director this morning. It shows how lucky they are and it reminded me of when Pete Goss turned around to aid Raphael Dinelli in 1997 I wonder if there will be added inspections of HB and Corum to see if they are showing hidden issues. If only to inform the other teams of possible issues.

[@jean-pierredeclemy7032]

Would strain gauges with downloadable memory be of any use to the designers in situations of ongoing development that are occurring here?

[@solowinterwolf]

Hindsight is 20-20, right? Just standing back and looking at that beam, my sense is that the lightening measures are not well conceived. There looks to be too little material remaining after the cutouts, and their shape seems incongruous with the application. A couple of diagonal ovals might have been better, or perhaps circles. Hard use would be the test --- very hard to simulate it graphically.

[@PeakTorque]

Just big central circles equispaced just along the neutral axis would have been better, surely. Currently there seems too little material, too far from the neutral axis with those polygonal cutouts.

[@al35mm]

And then once he fixed it, Alex's starboard rudder got damaged beyond repair and sadly he had to retire from the race. The bottom line is that the Vendee Globe is the ultimate, single-handed, round the world yacht race where the big players build new boats every 4 years to compete for the surprisingly small prize money. These boats including Alex's are well tested before the race as they have to complete qualifying races as well as testing. The engineering is similar to that in F1. They push the envelope of what is possible for boat and man. None of these boats ever cross the finish line without some damage. These boats are all about strength and lightness, hydrodynamics and aerodynamics. Then you have to have some mad bastard like Alex to sail the thing without any tactical, navigational help from anyone else including his team. Alex only had about 4 hours of sleep during the first week of the race. It is incredibly hardcore, Alex's was not the first failure in the race and won't be the last. It really is very sad that he is out but it is the nature of the game and hopefully all the skippers out there will stay safe whether they complete the race or not.

[@ff-he8rg]

Excellent analysis. Even though in modern design tech, we may not be able to find them? So I just worry about air plane design.

[@TheAntoine191]

The fact the crack propagate like this in multiple place mean 2 things in my opinion : This part was smartly designed with load properly shared along the structure. It must have been a very nasty load. You can tell it was strong. Large fillet, top crasftmanship, huge sections. But still it exploded. Everybody talk about the slamming which obviously must have been huge but the rigging also must put tremendous load on those axial/vertical structure.

[@pauljrcarty9314]

Sounds like your analysis is inline with mine. Understanding sailing without ever being on a sailboat. For the foils I'm inverting exotic car spoilers and scaling them up 12 to 14 times.

[@peterculverwell2030]

Great video and excellent analysis, thanks for sharing. What I have not heard much about on any posts or vids is the fact that the stress alarms did not warn of this problem, or maybe they did? I recall In one of the vids, Alex going in to gets detail about the boat being able to alarm him in the event of an over stress situation which seems really great. I guess like anything, if not set with the correct parameters, not much use?

[@PeakTorque]

Which maybe suggests the cracks originated from momentary impact loads, or fatigue loads, rather than steady state static forces which would trigger the so-called alarms. But just a guess.

[@christophereyers8480]

Last time I saw and S shape failure such as the one at the middle plate was in a glass fin being tested, it was then very clearly lateral torsional buckling...thin plates have a habit of failing in such a way under bending and compression, I imagine the kinetic compression loads must be massive on the plate. Also explains the angle stiffeners added.

[@michaelwaters9532]

I am a retired NA, (and sailor) used to designing such structures for large ships etc. Seeing the Hugo Boss fractures, I immediately asked myself, why use such a construction that gives so many shear loads, when a simpler truss could handle things better? The structure also needs to be able accept the high impacts and still remain inside the elastic limits of the material and chosen joint. So I'd like to see even some diagonal cables up forward, that could help absorb the punishing slamming and forestay load. I fear that if Alex insists to smash away at high speed that the whole bow will fail.! A true truss with material working more in tension might avoid such a problem, especially if a fiber with more elasticity were used. These High modulus fibers might work in the lab, but are treacherously brittle at sea, where even steel structures can be totally demolished Just hope the weather gods are kind to Alex Mike Waters na

[@Rod2001able]

Interesting that Alex was the only foiler skipper to take on tropical low Theta - maybe the others were concerned about failures on this sort.

[@harryadams5651]

Good to start design reconsideration. Maybe distance between ring frames should be smaller, maybe another longitudinal frame from angle of chine to deck head.because a lot of the dynamic hull slamming and rig loads static and dynamic are 30deg from vertical. I wonder if forstay loads in recoli from deceleration have been considered

[@PeakTorque]

I like Boris' bow. Many and evenly spaced floor and ceiling stringers, maybe 2-300mm high. Arguably heavier and less stiff but no stress raisers there and smoother load paths.

[@richardcannon7327]

Nice job and I agree with everything you said. One final thought would it be possible for a faulty rig rake system to overload the boat during a rake change? Alex is cranking on the rake and bang the combination of this and a wave slam overloads in bending. It definitely looks a shear failure of this panel. I was also no served where the web caps had deboned from the bottom skin and had been t lamed over that the resulting joint will not have the peel strength required. I had thought about modelling it up like you and found a set of plans for the boat. But work got in the way. Would be interstellar to add the splits and repair and apply your load case. Nice job

[@PeakTorque]

Cheers. Yeh this was a rough sketch and lunch hour(s) job. Would be interesting to see the actual plans.

[@richardcannon7327]

@@PeakTorque I assume we are both working in Mortorsport I've not been involved in boats for the last 25 years or so. I think that the build of the boat looks beautiful. Glad to see not nomex in the hull of the boat. which looked to be PVC core from what I saw which is much more sensible. I did some interesting work on battle damage of Cored panels a long time ago and Bi or tri directional cloth and pricked PCV core was amazing. and even when delam started it just kept on going. Where as nomex or Rhoacell just kept delaminating. PVC panel with a load of scrapnel through it would still be usable for non structural areas of the aircraft. Bit of Ali tape and fly again.

[@hjortureir]

What about stress coming from the rig? You seem to be omitting that alltogether. But the job you have done so far is fantastic! I think that this may have something to do with pull forces from the rig. Please look into that if you can.

[@PeakTorque]

Yes i mentioned i omitted the forestay tension and rig loads, although for this basic simulation i lumped them as part of the vertical load acting on the bow. Cheers

[@ponemark]

You know he's going to be going a lot easier until he gets past cape horn. Great tech info we learn more from mistakes. Is carbon just to stiff ? Or epoxy ?

[@pratwurschtgulasch6662]

what you say makes perfect sense. hugo boss should have hired you or someone like you. i feel bad for alex, he's been trying so hard but the boat keeps breaking apart :(

[@davidburgess741]

The sea is unpredictable. The loads the structure will encounter are not predictable. If it hasn't failed it must be over built, right? Eventually any timbers will be shivered!

[@normanboyes4983]

Great analysis. The nub of the issue is designing for complex indeterminate dynamic loads without sufficient empirical (real world) testing . Edit - Probably the only sure way to move on understanding is to to mount strain gauges at the ‘key points’ in the structure and go storm hunting and do 30+ knot runs. By key I mean those identified by FSA as the stress concentrations under different loads and assumptions.

[@normanboyes4983]

@Chris Woods Well that it is true but only based on that assumption. It is perfectly true that there are sensors fitted to the boat but the fact he only discovered the cracks during his ‘rounds of the boat’ suggest there were no sensors fitted in this specific structural member, indeed in the albeit limited video footage there was no evidence of any strain gauges at that location. My suggestion was strain gauges for design understanding during development and trials rather than permanent fitment and monitoring during service.

[@skynineUK]

PQ I am not an engineer but a questioning retired pilot but I am fascinated to know why the designers of “the Boss” would have wanted to cut those large openings that lead to uncertain or complex load paths into what is 2 sheets of carbon fibre with a nomex or balsa core. The weight saving must have been minimal, the loss of rigidity huge, and the ability to isolate the port and starboard section of the bow would have been a fantastic safety measure. Whether such a panel would be the best way of increasing the rigidity and stability of the bow section is something that should be looked at. My feeling is that structure should have manufactured as more of an integral part of the design so it spread the loads better but that is for people like you to question. The next question is should the underwater bow section be more rounded or a deep V to dissipate the pounding it experienced when falling off or driving into a swell? You can’t have a reverse camber in the IMOCA rules so it would have to be a basic profile. The 3rd question is why would IMOCA allow foils when they don’t allow any foiling at the stern and is anything that the designers could do within the rules to mitigate what in reality has all the stability of a tail dragger aircraft like the old DC3.

[@PeakTorque]

Hi David, yes I'm still trying to get my head around why that compartment is so scantily clad. The way these boats are dropping like flies is an absolute embarrassment, in my opinion. Serious questions would be asked if I was the sponsor. Goodness gracious. Bow shapes - few different philosophies there. A v shape will slap less but have less bouyancy, meaning possibility for more water on the decks (more tonnage and drag) from wave piercing. Finally you're dead right about the lifting t foil rudder. These things are like flying a plane with no elevator. They have very rudimentary means of pitch control, almost zero. They can rake the foils by a few degree but realistically the CL will never line up with the CG. A lifting rudder foil would ironically make these boats safer, and i think they're considering it for the next gen, but it'll need some really well thought out auto trim closed sensing loops if it's to work with the AP. Cheers

[@aardvarkmindshank]

Very cool analysis. Interesting. Thanks.

[@grahamm2015]

Thanks for the modelling of the structure and analysis. I can recommend watching Vittorio d'Albertas recent video on the same thing, unfortunately it is in Italian we can only hope they put out an English translation. From experience of delivering a 36' boat upwind for 5 days not only is it mighty uncomfortable but when we got back we could not lift the floor boards due to the continuing slamming of hitting waves. Back in 2005 El Corte Ingles had a system of hydraulic rams to counter the rig loads pull the bow and stern up like a banana. Keep them coming, you might end up with an audience of one, but I will be watching.

[@PeakTorque]

Cheers Graham. Feel free to share it with friends/sailing associates.

[@nickmalein]

Very interesting analysis. I am concerned that he is still going very slowly........further problems?

[@barnybrogan7850]

Damaged rudder they say. I wonder if people are trying to save face. I reckon the boats falling apart. Those rudders are obviously very susceptible so I'm sure they would have impact damage weak points and surely carry 2 or more spare ones?

[@johnpaulson996]

It is a shear failure. A sailboat is like a big balloon. There is a large ring of tension from the forestay, backstay/main leech load, and the bottom of the hull. It is being pushed apart by the mast and the deck. That is the primary load. (there is another ring around the chainplates/sidestays etc. but not part of the failure.)The bulkheads push the hull and deck apart. There are secondary (slamming) and tertiary loads as well. If you load that forward bulkhead up and do a shear calculation you will see the stress. Carbon is wonderful in tension and compression, but very weak in shear.

[@michaelwaters9532]

I would like to add this to my earlier post. Is this hull not the same one that experienced a massive underwater collision that ripped off the canting keel just about 12 months zgo? At that time it was reported "that no further hull damage was apparent'. I suggest that 'perhaps' the more recent interior damage to the central girder, might well have grown from that event, especially as Alex was unable to associate the current damage with any collision. Was there no ultrasound done on this girder? All 'water under the bridge' now, since yet another collision with underwater obstacles has now destroyed a rudder, taking unlucky Alex out of the race, though perhaps 'a blessing in disguise', considering the unhealthy state of the bow structure. Its now up to Samantha to come through with a strong showing to give the Brits someone to cheer for. Go Sam Go!

[@tommypetraglia4688]

From what I read in all accounts, they descended on it like a horde and picked it apart from stem to stern, paying special attention to not only the damaged keel but the mast and everything forward. I thought I read somewhere all the structure in the bow is new work I have quite a few links on it but not right handy. Give it a search

[@failureroom9817]

Thank you for your explanations and efforts. When I saw the first video of the damage, first thing came to my mind is that the cracks happened due to extreme rigidity of the materials. Smaller parts cracked becouse smaller parts has even less elasticity. I think repairing with the same stiffness is not a solution. Maybe it is better to leave these parts not repaired as they will create a flexibilty. Most probably I am totally wrong but I think if the boat will be pushed, same or worse will happen again after the repair

[@rexwolfe9916]

"Load cases are linked to psychology." It will be interesting to see how many breakages occur when the IMOCA 60s are used in The Ocean Race, which is fully crewed. I understand The Ocean Race crews push their boats a lot harder than the Vendee Globe solo skippers.

[@alexandermenck6609]

Whatever the correctness of this all is the key question is that VPLP should know all this and design the boat accordingly. It cannot be a question of weight saving taking into account that Alex then loaded the boat with all the spare parts now used for the repair. It makes much more sense to put some additional weight in a structure that holds than to put that weight in spare parts while designing such that it may not hold.

[@imboredofthisnow]

Great analysis cheers for looking into it. The shear web is like a vierendeel truss.

[@para1324]

Interesting analysis well done.

[@sailawayteam]

As a boat builder I have always thought that cutting holes is not a good way to save weight, better use uniform but light structures. If this bulkhead really was cored, I doubt much saving in the weight really was achieved due to extra cappings. Especially the lower half diagonal seems completely redundant and might actually make the whole bulkhead so weak that we can doubt if it really contributes much to the boat strength.

[@MrWillow900]

Pretty sure it's a cored panel. If you watch the video where Alex was drilling a hole through it seems to be cored.

[@PeakTorque]

Originally i thought that too. But the other episodes you can clearly see his fingers go around the capping and underneath it leading me to believe its plate not core.

[@debdoyle119]

These cracks in the structural bulkheads really concern me since he still has Cape Horn and he Southern Ocean ahead of him.

[@SimonJones-jy8ly]

Interesting! Wasn’t it this hull that had the keel dropped out in the Jaques Vabre two handed race? Whilst the original design might have been sound, the structure may have been stressed in that previous race incident. I believe the boat has water ballast capability too. That must put enormous stress on the shape.

[@xeels2708]

It is the same, but the keel didn't drop out, they had to cut it because it broke after impact with a submerged object It does have water ballast too, so indeed great loads

[@VUO4E]

IMO, this didn't just happen at once. The different cracks don't seem "related" so this is something that's been going on for a couple of days before being identified (taking into account where he was when he discovered it). Looking at the panel and the cutouts, I would have mounted it upside down compared to how it's mounted but that's 100% non-scientific :)

[@bobblakley8749]

Based on your excellent video of your analysis of the longitudinal beam structure, I have a possible solution to share. Briefly, I believe the repair needs to bond an area of reinforcement around the openings in the longitudinal beam, and perhaps also overlay the caps. What we know is that the existing structure had enough strength to last this far into the race. So conceptually, if that strength can be restored, Alex can logically continue racing, perhaps to the finish before the cracks come back. ASME code has an extensive treatment on the reinforcement of openings, to maintain the stresses in the base membrane, so the idea is to install reinforcement pads on both sides of the openings to regain the strength. As to the bonding, I would use the 12:1 scarf joint thinking; if the panel is 6mm thk and the pads are each 3mm thk, then the width of the reinforcing pad would be 36 mm. What I cannot do, as you correctly pointed out below regarding your analysis, is determine how thk to make the pad, because this approach is more along the lines of a constant strength rather than absolute basis meaning we are trying to get back to what we started with. Do you have any idea how we could run this by the folks at VPLP?

[@michaelgreenland4682]

When the first Comet was flown back in the fifties they fitted 'oval' windows for a better passenger view; they were changed to 'round' after cracks appeared in the framework around the windows! Hence if you wish to lighten a structure have round holes! I've met Peter Hobson who had an involvement with this flying machine when redesigning the Osprey dinghy, really brilliant innovated designer but sometimes old school (me) rules. Many thanks for you’re in depth report and view. I didn't sleep that night when I saw his boat at 3 knots pointing towards Argentina and felt it was over but, they have given him the tools and advice to carry on albeit not doesn’t seem to be going fast as the others at present.

[@damientelle]

I was sure one guy will take the challenge to simulate this structural fail ! thank fo this ! i am not a naval architect, but sailor competitor and aeronautical engineer (but my structural and stress simulation knowledge are as old as my diplom...) Nevertheless having seen nearly all the main frame of the Airbus family aircraft i can say that cut-out are allways, circular or elliptic. I can also say that the main static load along the boat central beam is the forestay load (on the imoca only 2 of the 3 forestay are structural for the mast integrity) the capability to preload the forestay is a key factor to control the shape of the foresail. even in light wind condition we may stiff the backstay to max load in order to preload the forestay and keep a flat foresail profil. This pre load is so high that at a first assumption : static load = forestay pre load and all others loads = dynamic

[@PeakTorque]

Great insight thank you. I'm very new to sailing. I can picture all of the loads, but the magnitudes are unknown to me. Exactly right, the cutouts should be circular, not polygonal as I said in the video. The weight penalty would be minimal, but the stress concentration effects would be far less.

[@stephenclark9771]

The centerline bulkhead is cored. Probably 35-50 mm thick with no ex core. The “p” would be Port or left hand side. When designing these boats a key tool is a Velocity Prediction Program ( VPP) which predicts the advantage or penalty of various design compromises. Racing boats want to be light in the bow, and the VPP will penalize the design that is he a year in the bow by some seconds per mile. On a race as long as the Vendee this adds up. More than the margin between first and second in the last race. The structural designers shave the structure to the minimum on the basis of this analysis. Slamming and G lol ads are, as you suggest , very hard to predict. 10 kg of preprogrammed preprogrammed would have improved the factor of safety by lots, but woul have been resisted by the sailor and designer.

[@markomarkomarko]

Excellent analysis. A solid panel would have been very prone to buckling. What was designed is more of a truss with beams following the load paths. So the main diagonal member transfers the reaction from the forestay (which also holds up the mast) back into the hull. A solid panel would not have been as effective and would need to be quite a bit thicker than these beams in a truss. The biggest issue is that you want minimum weight in the bow for maximum performance. You will note that there is nothing stored in this compartment to keep the weight down. Hence why they tried to optimise this truss for minimum weight! Clearly they've underdesigned it and the point re stress concentrations and low radii stands. The entire truss should have been beefier. Very good point about the loading being uncertain, but isn't this the case with all structures? Boats are very tricky in terms of loading and constraints. I've tried to analyse the mast for my boat and it was very difficult to even conceptualise loads and supports! It is very unusual in that the mast bending moment actually depends on the amount of the righting moment available and is not necessarily proportional to the wind load! The fact that the boat now also lost a rudder and is out of the race does suggest that this boat was structurally inadequate!

[@batchint]

my assumption from your analysis is.. does this barge.. require two longitudaul members... as a virtue in design processing

[@gbr562]

I'm a boat builder and based on structures I make and repair I very much doubt that the structure is solid, More likely a 4 to 5mm skin either side of 25mm nomex honeycombe. The capping is there to add a bit of sideways stiffness and also to keep each side of the laminate in column and stop it from delaminating from the core. My guess is the cracks are due to compression when pounding. Oh and the music is not helping.

[@Wysiwig99]

It will for sure not be solid laminate but cored carbon skins

[@abdust]

i was hoping to save a bunch on my crack bill on a simulator but it turns out that's not what this is

[@PeakTorque]

😅

[@garstanforddasilva9663]

Thanks PT, all sounds v plausible. Surely the design engineers knew who would be skipper and built in enough risk assurance for aggression or bullish sailing tactics? Like the modelling supposition. Chak, chak,chow!

[@christinehomans5108]

Too Sade for Alex ; courage Alex ; congratulations for your humility

[@psystealth]

Your analysis is pretty impressive, ....i tend to agree the skippers mindset creates the voyage.....@21:38 hit the nail on the head. This downwind Atlantic sail has not stressed the boats at all, the Southern ocean will do that with ease. LinkedOut has broken some part of his foil (UFO?)? do we see more failures in the upcoming weeks? For Alex I hope for his sake his boat comes through the weather ahead.

[@jackshea6937]

Excellent analysis!

[@Wellorep]

I thought maybe you might have played with their design a little to see if you could alleviate some of the stress.. Thanks for the cool video though, I appreciate it.

[@shahrulhanuarmdaris6105]

The cut out composite should be more thicker - the cracks were due to stress made by the tremendous head on waves

[@simonjohnson1]

How can he consider running repairs? Surely the structure is too compromised and the boat needs major work to make it safe.

[@PeakTorque]

The boat has failed now and is out of the race, apparently due to a broken rudder. But I'd not be surprised if the real reason was the big structural damage discussed here

[@pauljrcarty9314]

Because the boats heel over. But when I look at them they're symmetrical. It appears that the boats should be level to me. I have to dump that to understand the swinging keel

[@EricRidesDirt]

What software are you using for the simulation?