Cheeki Rafiki loss report

[Polux]

Originally Posted by europaflyer

The elephant in the room seems to me to be that the hull liner ('Matrix') is just glued in to the hull. A vast amount of evidence shows that there is no reliable way of gluing units in to a hull which have to take structural loading. It is done purely because it is quick and cheap. Combined with a single line of keel bolts, the entire bending moment of the keel translates in to a pure shear on the bonding paste.

There is a reason that more respected manufacturers leave cut-outs in the structural grid between the beams, which then allow the grid to be glassed to the hull. It's better. Beneteau don't, people die, but it's their fault for grounding it of course...

....

Edit: I may have been mistaken in the single line of keel bolts. The annexes to the report (https://assets.digital.cabinet-offic...eekiRafiki.pdf) show that the centre three bolts are doubled. I stand by the rest.

The problem is not using bonding agents but the way it is done. Some top brands and racing boats use bonding agents that can have a superior resistance compared with laminated. Bonding agents are used in several industry that requires superior bonding of parts, for example on architectural building (Opera house of Sydney was one of the first) and sports car chassis (Lotus and other top brands use it).

[Polux]

Originally Posted by transmitterdan

This was not discussed by the MAIB report but if we are allowed to speculate as they did then I think there is another possible issue. It is well known that compressing FRP laminates with bolts and plain flat washers is a poor idea. The pressure will be too high under the washers and quickly drops to nothing very close to the washer. This pressure gradient does bad things. Also, if you tighten the bolts enough to compress the FRP without crushing it then the temperature caused expansion/contraction of the FRP will eventually either crush the FRP (rising temp) or the joint will loosen (lowering temp). FRP is not a spring so either way the keel bolts inevitably loosen.

The only way I know to bolt a high compression load to a FRP laminate is to spread that load over a relatively large area with the bolts fitted with Belleville type washers (Stacking Disc Springs) to absorb the expansion/contraction. Then the bolts can be tightened to the required preload so as to not crush the FRP. The spring washers will keep the pressure more or less constant as the joint contracts and expands. If CR had a proper backing plate inside the hull it would still be afloat IMO. Just look at the pictures and you can see the hull damage is relatively minor.

How we can get thousands of boats in the water using stupid fender washers against comparatively thin FRP laminates borders on criminal IMO. The steel (or AL) backing plate should be measured in square meters not square mm. If there was a proper backing system then the liner/hull adhesive would most likely never be over stressed and this whole inspection/repair debate would be moot.

Yes I confess never understood that, but you should not generalize because not all use little washers and some use the keel boat to a carbon or steel matrix (talking about performance boats).

That is the keel attachment of the First 40.7. Not a good example I would say but not very different from many other boats.



This is the keel of the Bavaria Match 42 (one of them lost a keel too) but this one has the reinforcements that Bavaria added to all the existing Match 42.
I would say that is a good example how it should be done and that the Mach 42 has now one of the more solid keel attachment among this type of boats on the market

[transmitterdan]

Polux,

The Bavaria C-channel load spreaders are much better because they spread the horizontal loading out over a wider area of the hull. Still the main damage that happens in a grounding is fore/aft stress which the Bavaria could address better IMO.

The little plates used by the First 40.7 are a joke IMO.

And neither of these designs do anything about the keel bolts becoming loose over time, temperature excursions and hard sailing. CR clearly had loose keel bolts many hours before she inverted. It is a shame we have no information on that part of the boat probably because the skipper (22 years old) was looking at through hulls and engine cooling hoses for a leak and not the keel area. Again, more speculation, but if they had been able to tighten the keel bolts they may have prevented the sudden loss of the keel. And when 3-4 of the keel bolts just turned and turned because they were broken off they may have been more interested in getting the life raft ready.

[Polux]

Originally Posted by transmitterdan

Polux,

The Bavaria C-channel load spreaders are much better because they spread the horizontal loading out over a wider area of the hull. Still the main damage that happens in a grounding is fore/aft stress which the Bavaria could address better IMO.

......

And neither of these designs do anything about the keel bolts becoming loose over time, temperature excursions and hard sailing. CR clearly had loose keel bolts many hours before she inverted. It is a shame we have no information on that part of the boat probably because the skipper (22 years old) was looking at through hulls and engine cooling hoses for a leak and not the keel area. Again, more speculation, but if they had been able to tighten the keel bolts they may have prevented the sudden loss of the keel. And when 3-4 of the keel bolts just turned and turned because they were broken off they may have been more interested in getting the life raft ready.

Yes I agree. I read somewhere that there were evidence that the keel bolts on the First were lose from the beginning and that may be on the origin of the accident since the lose bolts would make the keel move around increasing dramatically stress on the fiberglass.

I guess that one recommendation that misses on that report regards having the keel boats verified and tightened regularly and certainly before any major voyage, like an Atlantic crossing.

[neilpride]

Originally Posted by Polux

Your post is a bit confusing since it is not clear what are your opinion and what are opinions of shipyards that have actually repaired the boat and Beneteau statements.

Regarding the first 40.7 to be a "bad joke" I don't think so. Several have made several full racing offshore seasons in very though conditions (including several Hobarts) without the boat experiencing keel or other problems. One has been sailed on a full several year circumnavigation with high latitude sailing (Antarctic) facing severe storms without any significant problem.

The size and the thickness of the baking plate did comply with the rule when the boat was designed however (much against what you have been saying) the RCD has been improving all the time the safety standards and today the boat would have to be built stronger on that area.

I believe the fact that most accidents relating with groundings and matrix detachment have been more frequent on this boat because it is simply the one of that type that has been built in bigger numbers. The First 40.7 was a very popular boat and 550 were built.

The boats are not all the same and a lighter performance cruiser in what regards groundings cannot be built to have the same resistance has an heavy cruiser. If a grounding occurs the boat should be properly surveyed and given the way it is built a proper inspection is not easy.

What seems more worrisome in all those that inspected the boat and in what Beneteau says is that a clear protocol is not established in what regards procedures in case of grounding and posterior inspection. They refer also that on the the PDF document that analyses the Cheeki Rafiti case.

Beneteau should make more recommendations regarding that, establishing a maximum safety protocol regarding inspections and also provide information and the needed pieces to upgrade the boat to the new safety standards.

Neil, most of the time I don't disagree with you (except in what regards new technologies and the improvements they bring), just with the exaggerated and inflammatory way you tend to deal with the subjects and that is in my opinion the case when you say about the First 40.7 : "the CAT A decal in this boat is to me a bad ass joke without further modifications".

Ok, Beneteau statement say clear how to repair the isue, save the flanges, remove the loose bonding plexus stuff , insert new bonding stuff and glass the flanges to the bays, now the most dificult part, for those who never attempt to fix properly a grid liner i have bad news for you, is imposible!!! imposible to fix it properly , you can destroy the whole salóon and rebuild part of the keel structure, yes , but in the case of the first 40.7 this task call for a really expensive bill and with a doubtful result... the only way to get a excellent job is to rebuild the whole thing following profesional advice , and forget Beneteau , is going to tell you the same thing , keep the grid liner OEM as posible, and this is the focal point, is a defective enginering design , so try to fix **** and you get a shiny piece of **** ,, it happen everyday in the boatyard...

Several sail the Hobart and other tough races and they cross oceans to, question is how many have modifications repairs in the keel área? Grid liner? i think a bunch of First have some kind of repair there, ....


The whole keel meet the ISO estándar but the bolt washers need to be 3 mm thick and 3 mm wider to meet the estándar so obviously those washers dont meet the estándar....

And you mention grounding but MAIB mention liner damage by hull slamming or pounding to, so is not just grounding related....


No protocol because inspect liner damage is like Black Magic, almost imposible, ask any surveyor...

And to end ,I would like to see in the 40,7 owner manual something like if you ran aground, please drop the keel and make a extensive inspection,,, every 5 years a grid liner inspection is necessary....if you can LoL.

[boatman61]

Originally Posted by scottmills

I'm so sorry to all involved and the loss the families must be feeling.
I must say now, as before - myself and many of my professional captain peers will simply not deliver a 'Bendytoy' anywhere offshore.
This is not the first, nor will it be the last time a keel detaches from a hull on this kind of boat. Not seaworthy sums it up.
If you want to cross oceans, buy an older pre 80's boat. If you have a bendytoy, stay within reach of shore, or use the ICW whenever possible.
Or sell it. I personally have had nothing but bad experiences on these lightweight, mass produced boats.
Please take extreme care, and beware.

Sorry but I must disagree.. Having owned 2 Bending Toys both of which I bought in the Caribe and solo'd to the UK without problems your talking out a year ass.
But then WTF do I know being an unprofessional delivery skipper with a big mouth.
.

[Polux]

Originally Posted by neilpride

Ok, Beneteau statement say clear how to repair the isue, save the flanges, remove the loose bonding plexus stuff , insert new bonding stuff and glass the flanges to the bays, now the most dificult part, for those who never attempt to fix properly a grid liner i have bad news for you, is imposible!!! imposible to fix it properly , you can destroy the whole salóon and rebuild part of the keel structure, yes , but in the case of the first 40.7 this task call for a really expensive bill and with a doubtful result... ....

Several sail the Hobart and other tough races and they cross oceans to, question is how many have modifications repairs in the keel área? Grid liner? i think a bunch of First have some kind of repair there, ....

And you mention grounding but MAIB mention liner damage by hull slamming or pounding to, so is not just grounding related....

No protocol because inspect liner damage is like Black Magic, almost imposible, ask any surveyor...

You call liners to everything including a structural grid so let's call it what they call it: Matrix and the problem is not with the matrix itself but with the bonding to the hull.

It could look like black magic to you but to verify if a matrix is bonded to the hull, without an expensive job the only thing you need is modern technology and devices that can read through the hull and show it to you. When my boat was rammed by a 55t motorboat I had done 1500nm to be sure that the hull was inspected that way by a top professional. No magic involved, just technology.

Yes I agree that in the rare event of having a matrix with bonding problems the repair is going to be costly. That building system that is used by Jeanneau and Beneteau with a "contre moule" bonded to the hull has advantages and disadvantages. A costly repair if there is problems with the bond is certainly one of the disadvantages. Not a very frequent problem dough.

Originally Posted by neilpride

The whole keel meet the ISO estándar but the bolt washers need to be 3 mm thick and 3 mm wider to meet the estándar so obviously those washers dont meet the estándar....

It seems what you did not understood what I said. The boat complied with the RCD standards at the date it was built. The RCD standards are higher now and the boat does not comply anymore.

If they demanded that all the boats complied with the current RCD standards most of older boats would have to be modified, some in a so extensive way that they would be scrap. Older boats do not need to comply with actual standards.

Originally Posted by neilpride

And to end ,I would like to see in the 40,7 owner manual something like if you ran aground, please drop the keel and make a extensive inspection,,, every 5 years a grid liner inspection is necessary....if you can LoL.

No, but they should say that in case of grounding the bond between the matrix and the hull should be checked by a surveyor with the help of an electronic device able to determine if the bond is consistent. If the bond is not consistent then the keel should be doped off and a repair should be made, according to their recommendations.

They should say also that on the case of extensive heavy use of the boat in hard conditions (offshore racing) the bond between the matrix and the hull should be checked with the help of such device each 5 years or so.

[neilpride]

Paulo , go ahead and educate yourself in Grid liners , if you mean termal or cameras inside of a hollow beam , good luck with that, the bonding agent under compresión is a cheater , and the repair is almost imposible without destroy half boat in some cases ,


You V hundreds of surveyors...

[europaflyer]

Originally Posted by transmitterdan

Polux,

The Bavaria C-channel load spreaders are much better because they spread the horizontal loading out over a wider area of the hull. Still the main damage that happens in a grounding is fore/aft stress which the Bavaria could address better IMO.

The little plates used by the First 40.7 are a joke IMO.

And neither of these designs do anything about the keel bolts becoming loose over time, temperature excursions and hard sailing. CR clearly had loose keel bolts many hours before she inverted. It is a shame we have no information on that part of the boat probably because the skipper (22 years old) was looking at through hulls and engine cooling hoses for a leak and not the keel area. Again, more speculation, but if they had been able to tighten the keel bolts they may have prevented the sudden loss of the keel. And when 3-4 of the keel bolts just turned and turned because they were broken off they may have been more interested in getting the life raft ready.

I'm not so sure about the small backing plate theory. If that was the problem, we would have seen the bolts pull through the hull with their plates, leaving holes in way of the bolts. What I see is:

1) first two bolts (single) sheared cleanly
2) next three bolts (double) pulled out through hull with backing plates, removing a substantial amount of both the grid and hull laminate
3) aftermost bolt (single) sheared, possible corrosion



Exactly what caused the initial movement we will never know. The order in which the bolts failed or started to work we won't know. However, the final failure mode seems relatively apparent from this: of the nine bolts, three are sheared and presumably their backing plates are still inside the hull, and six are still attached to the keel, along with the failed laminate section.

The initial cause may well be debonding of the grid and the hull following grounding. This shouldn't have happened in the first place, and that we can argue about for a long time, but it did. This is almost irrelevant. What matters is that the majority of the keelbolts and all the backing plates are intact, but it was laminate failure caused probably by repeated flexing which allowed the final separation.

Even with two, perhaps three of the nine bolts seriously compromised, even with the hull and grid to some extent debonded, if the hull and liner had been sufficiently strong it would have held. This is a critical point. There is no secret on how to make these things strong - more fibreglass. For literally no more than £50 worth of fibreglass and resin, and half an hour of labour, that whole area could have been twice as strong, or more, and it would have held. And that is the fundamental problem with Beneteau and others like them - for the sake of a tiny cost saving, they will happily compromise critical areas of the boat as long as the standards they are working to allow it. And that is why these cheap, lightweight boats will never be as safe for offshore sailing as what people here call 'proper bluewater boats', whatever some diehard enthusiasts of mass-market production on here say. A boat which is so damn flimsy that the keel, with most of the bolts intact, can rip out a whole section of the hull laminate, is not a seaworthy boat.

[tankersteve]

I found it interesting that the long email sent by the management company, after discussing thru-hulls, specifically addressed inspecting the keelbolts.

With a crew of four, and a bit of motivation, even in some fairly rough and cold seas, inspecting the thru-hulls couldn't take too long.

Worth having a look at the keel bolts and make sure there is no cracking around them.

Yet no apparent mention in the subsequent, albeit limited, conversation.

Tankersteve

[Polux]

Originally Posted by neilpride

Paulo , go ahead and educate yourself in Grid liners , if you mean termal or cameras inside of a hollow beam , good luck with that, the bonding agent under compresión is a cheater , and the repair is almost imposible without destroy half boat in some cases ,


You V hundreds of surveyors...

Yes, it is hard to repair. No it is not thermal cameras but devices that were developed from the medical hardware: Some work with ultrasounds other with magnetism. I am not an expert but the guy from the shipyard were I have the boat is. He patented recently one. I have seen amazing images of interior bonds and cores.

I guess that for each high tech surveyor there are hundreds of low tech ones that bang on the hull to see if it sounds right

[Polux]

Originally Posted by europaflyer

I'm not so sure about the small backing plate theory. If that was the problem, we would have seen the bolts pull through the hull with their plates, leaving holes in way of the bolts. What I see is:

1) first two bolts (single) sheared cleanly
2) next three bolts (double) pulled out through hull with backing plates, removing a substantial amount of both the grid and hull laminate
3) aftermost bolt (single) sheared, possible corrosion



Exactly what caused the initial movement we will never know. The order in which the bolts failed or started to work we won't know. However, the final failure mode seems relatively apparent from this: of the nine bolts, three are sheared and presumably their backing plates are still inside the hull, and six are still attached to the keel, along with the failed laminate section.

...

As I said your conclusion had already be made by others that found evidence on the hull that the bolts were not tight and the keel lose. With the bad conditions the keel movements increased and lead to a fail on that laminate section. Have it been more thick (the fiberglass) and with a lose keel it would only be a matter of more time to the delamination to happen.

The real problem on this case seems to be sailing for a transat without verifying if the keel bolts were tight...and they weren't.

[minaret]

Originally Posted by europaflyer

The elephant in the room seems to me to be that the hull liner ('Matrix') is just glued in to the hull. A vast amount of evidence shows that there is no reliable way of gluing units in to a hull which have to take structural loading. It is done purely because it is quick and cheap. Combined with a single line of keel bolts, the entire bending moment of the keel translates in to a pure shear on the bonding paste.

There is a reason that more respected manufacturers leave cut-outs in the structural grid between the beams, which then allow the grid to be glassed to the hull. It's better. Beneteau don't, people die, but it's their fault for grounding it of course...

Minaret is the expert here I think, and I would be very interested to hear his opinion.

Edit: I may have been mistaken in the single line of keel bolts. The annexes to the report (https://assets.digital.cabinet-offic...eekiRafiki.pdf) show that the centre three bolts are doubled. I stand by the rest.

I do believe I've made my opinion on this very clear many times before. Won't waste breath doing so yet again. I don't have the available time of many of the amateurs here, I'm too busy running one of the busiest boatyards on the west coast. Suffice it to say, my opinion has not changed, nor will it. It is directly in line with the MAIB report and supported by the opinion of Neil and every single other pro I know. That is a large number of professionals, I rarely miss an Ibex event. I doubt anyone else here has ever attended one, despite its international nature. In short, everyone who actually knows anything about the subject is in agreement. Hull liners are madness. The world will be a better place when they have gone the way of the dodo.

[transmitterdan]

Originally Posted by europaflyer

Even with two, perhaps three of the nine bolts seriously compromised, even with the hull and grid to some extent debonded, if the hull and liner had been sufficiently strong it would have held. This is a critical point.

It isn't practical to design a structure such as a boat hull so that it maintains its integrity even though it has suffered massive structural failure. Nor is it necessary. The trick is to design it for the conditions with enough margin that it doesn't fail. And if/when it fails the failure should be readily apparent.

Originally Posted by europaflyer

There is no secret on how to make these things strong - more fibreglass. For literally no more than £50 worth of fibreglass and resin, and half an hour of labour, that whole area could have been twice as strong, or more, and it would have held. And that is the fundamental problem with Beneteau and others like them - for the sake of a tiny cost saving, they will happily compromise critical areas of the boat as long as the standards they are working to allow it. And that is why these cheap, lightweight boats will never be as safe for offshore sailing as what people here call 'proper bluewater boats', whatever some diehard enthusiasts of mass-market production on here say. A boat which is so damn flimsy that the keel, with most of the bolts intact, can rip out a whole section of the hull laminate, is not a seaworthy boat.

I agree but don't agree with the solution. It isn't simply a matter of too little fiberglass. Too much fiberglass can lead to keel bolt loosening for reasons I pointed out earlier. It is a matter of not enough engineering attention to the actual loading and the proper use of materials. FRP fails based on pressure (either stress or strain) which is measured in kPA or PSI. There are two ways to prevent over pressure failure. Make it "stronger" which means increasing the failure pressure. The other way is to reduce the load pressure by spreading the load over a larger area. Either way is perfectly valid if all the conditions are considered properly. But one way or another it has to done else the structure is more likely to fail.

[transmitterdan]

One other thing is obliquely mentioned in the MAIB report. The CR master searched over the course of many hours for a leak and apparently never considered that the water could be coming through the keel bolt holes. This is another danger that comes from repeating the mantra that thousands of these type boats make it across every year. Repeated enough times it tends to reduce the probability of keel failure in the mind of a tired crew. The master initially concluded that the water was from a fresh water tank. That could be a tired mind at work but could also be because keel failure is way down the mental check list of most of us.

Owners and crew need to be alerted to potential failures in the right order of probability. The head in the sand approach some take contribute to losses such as this IMO. If the crew were sensitized to keel failure as a real possibility they might have been able to save the boat by acting early. I think we should acknowledge the possibility of keel failure and when a cursory check of thru hulls doesn't turn up anything then the keel needs to be checked early and often. But that doesn't happen because the proponents of light matrix built boats don't want to put the keel (i.e. liner/hull attachment) at the top of the flooding check list. They tend to put it last which is not where it should be. Someone mentioned FMECA and this is a complicated topic. But even if the probability of keel failure is less than 1% it should be high on the list of things to check because the effect of such a failure is 100% loss of boat and crew. That's why FMECA teaches to put low probability but terrible consequence events high on the checklist. But marketing people hate FMECA because it makes it seem as if the product is no good. They should learn to embrace it and so should owners and crew.

The MAIB report hints that RYA instructors should not let students ground these type boats as an educational tool. Or worse, as a way to automatically fail a student they don't think deserves to pass. If an instructor thinks a student doesn't deserve to pass it is a dumb thing IMO to let them ground the boat just to avoid giving a true appraisal of their skill. Maybe instructors do this because they don't think any damage will happen to the boat in a "soft grounding". Owners and masters need to change their way of thinking and protect and maintain the keel as if their life depends on it. Because it does.