Origami steel yacht construction

...so,it's a punch-through thing?and the pipe end itself could just as easily punch directly through pad and next,the hull?or do you expect a pad edge to shear through the hull making a crack?in fatigued (per those dynamic loads still undefined) plate ( I thought) or is it now the strut itself failing in shear?

Man,just when I think I'm getting it....

 

Assuming the pad is thick enough what happens is that the plate it's welded to flexes as load comes on & off. All the pad does is increase the area of flex. If it's still too small to stay below the point where fatigue cracking will take place, you've bought more time before failure but the structural problem still remains.

The pipe won't punch through the pad, it transfers its load to the pad and the pad transfers it to the hull plate which flexes. How many times can it flex before it starts cracking? That is the crucial question. Brent assumes that the answer is infinity but has no math to support his assumption. The engineers say it'll happen but without a lot more detail on loads etc can't predict when.

The design I'm building which Brent derides has this load fed directly to the keel which is a 7.6m length of 200x40 flat bar welded to 6mm deadwood plates forming a hollow box.

PDW

 

This is where “generalising” can only go so far, since it is not a “catch all statement”, but you’re on the right track.

The “punch through” is the direct shear load. In the example I gave, if the plate was 6mm thick, if the strut is welded directly to the hull the shear area available is 6 x 237 = 1422mm^2. With the pad it is 400x6 = 2400mm^2.

So this leads into your question/curiosity.

The example is a one off direct punch load. If that load on the shear area available is below the shear stress of steel, then hey presto it doesn’t break. BUT, this is a one off load.

What happens if the weld isn’t so good? …why does that matter, it is punching through the plate…well, the pad or strut is welded to the hull plate. If the weld is poor quality, the load carrying capacity is also poor. Why is this important?

The load goes from the strut, via the weld, to the plate (or pad) - the load path is eccentric too. So the weld transfers the load. BUT, if the weld is poor quality the amount of load it can transfer is reduced. Most basic welds, even if poor quality can usually transfer a load at least once. But, this is where I need to start generalising again.

If the weld is poor quality, when a load is transferred through it, any flaws that are now in this crappy weld becomes sites for crack initiation. A simple thought experiment can explain. If you take a rubber, the ones you use when drawing in pencil and need to rub out a mistake, bend it back and forth …no problems. Now, pierce the surface with your pencil, and repeat, what happens?....the rubber slowly starts to split from the location of where the pencil penetrated the rubber surface. This is a simple explanation of a flaw.

So, in poor quality welds, there are many such flaws. So each time the strut is loaded, even well below a load that would cause a static shear load failure, this causes the flaw to become a crack. Fatigue is best described as the failure under repeated or otherwise varying loads which never reaches the level to causes failure under a single application.

Thus, if the load is 1kg or 100kg or 100 tonne, it is still a load that is being applied. Doesn’t matter if the load is well below the yield stress limit. Once a crack has initiated, all it needs is a cyclic load to propagate. Being in the sea, every time a wave hits the load, is a cyclic load. So if your boat experiences a wave say every 3 seconds, in one hour that is 1200 cycles. In one day that is 28,000 cycles, in one month it is 892800 cycles…..and so on.

Once a crack has been initiated, you’re doomed. But these cracks are microscopic in nature. You’ll only know just before failure, by seeing a huge crack (which means the failure is imminent) or once it has failed.

There are also many other influences that affect fatigue strength, but I have somewhat generalised for simplicity. Since as PDW notes it is rather more complex, using SN curves, stress concentration factors etc etc, but can be done, with supporting facts and dimensions etc. And anwsers the "when" and "how"...

 

In the past people who bought your plans who asked for the lines and the weights and moments were never supplied them, even on the origami yahoo groups when people asked or even offered to calculate or make a 3D model but asked you for the lines you'd just fade quietly away strangely silent.
People who specifically asked for lines were told there were none available right now and they were of course not required to build.
Ditto for weights and moments calcs. Now suddenly it's ******** and All your plans come with the hull lines.

Would you like to say when you started doing this?

Now you can do a stability calc using your lines ( although it's better to use a computer and avoid the math and errors) but fist you'll have to find an accurate COG from weights and moments. Something else you cant provide.

 

PDWILEY,ADHOC, thank you for explaining at length! I was thinking it would most likely fail at the HAZ ( a new word learned from other earlier posts!) but I see now the flaw in a weld is yet another problem too,because it'll concentrate stresses locally-Question-even if the surrounding weld might hold(being thicker than plate) these stresses would still in turn tear at the hull?..Otherwise, I do think I understand that fatigue would be repeated loading surpassing the elasticity of the material to return to original "condition" (as in bending crushed beer cans!)...
and as you say,once the crack has started,it'll go fast!Yikes!Oh ,wait a second- wouldn't a cracked weld be INSIDE the boat and not through the actual hull?The pole end would slip out and and the deck/cabin corner would sag inward (given that the deck alone could not support it as it's pictured) and the mast would fall too but there's no water coming in....?Another question: people have said that they would add extra welds to nearby framing -is that just so that the shear acts against the entire length of the weld instead of across it(per the example as you describe?or is it something else.

And there's those "Cycles" and whether or not the accelerated mass of the mast+sail power-BOTH being accelerated by wave motion- will indeed be a threat....hmmm

 

It's worth reading David Lewis' 'Shapes on the Wind' if you haven't. He was a sailor but a mechanical ignoramus of the first water, as he freely admits. Read pages 241 & 242 for an example of what happens when someone with no structural 'feel' builds/modifies a structure like a mast. The mast sheared its base supports, punched a hole through the hull and the boat sank. Note that alongside, under a static load, no problems were evident. The mast came loose in a moderate sea and became uncontrollable when the swell picked up closing the land.

That particular hull was ferrocement. Brent goes on & on about how steel is massively strong in tension and it is. Ferrocement is massively strong in compression. Is this relevant? Well, not when the failure mode is shear, and this is the point that the engineers here have been making time and time again to Brent, and he just ignores it.

Would you be comfortable working underneath a steel pipe welded to 5mm thick plate overhead, supported by a few lengths of 25x25 angle iron intermittently welded to it, and having that pipe loaded/unloaded with both vertical and lateral load sufficient to cause its attachment point to flex every 3 seconds?

PDW

 

quote
B S

giving a boat that is often indistinguishable from a round bilged design.

 

QUOTE B S

Get your advice from those I have built boats for , people who have experience building and cruising in my boats, not from those who have a personal stake in keeping the price of boats high
UNQUOTE

Brent, this thread is not about price

 

To start answering these questions depends upon how much engineering knowledge you have really. Since the more answers we give, the more questions you’ll have….but as PDW says worth a good read! But here goes…

That’s a good question.

Firstly, cracks need to be classified into 2 groups. They can be in a ductile or brittle manner. Ductile fractures propagate into thinner sections or lower yield strength regions whereas brittle fractures propagate at stresses less than yield and their crack paths are dominated by brittle regions, residual and/or thermal stresses, they also tend to propagate from thin to thick regions.

These can also then be looked at as either transgranular or intergranular fractures…..why is all this mentioned and is it important?

The reason being is that structures that have no redundancy, ie more structure to pass the load to, and structures which have high residual stresses, ie compressed, all this affects crack growth, propagation and initiation. And since there is no structural redundancy in a BS boat, where does the crack from the weld go…well, only one place it can go…through the plate! Can’t go from thin to thicker regions…can’t go into thinner or lower stress regions and so on…..ie, problem!

Adding extra welds, I assume you mean extra structure/frames…since the welds must attach something!? This what is termed structural redundancy. Lets say you design the structure and all is well for all loads and all conditions. If you take away one of the members does it fail??...if no, all is still well, then you have more than you “technically” need. This is structural redundancy. Some may say you’re adding unnecessary weight. But here’s the thing….

Lets say that shear load has a value of 100. If you have one structural member passing into this, say a frame, this halves that load to 50. That is 50 either side. If you then add a longitudinal also at this location ie now having 4 members, you then have 100/4 ie 25 per structural member. So suddenly that 100 load becomes 25. You have added a factor of safety of 4, just by adding a frame and a longitudinal.

Remember the bit about cracks??...well, if you now have more frames/structure, this 100 load which is now 25….in other words, the stress is lower in this region. This means the ductile cracks that start to propagate once initiated, tend to propagate into these lower stress regions, ie into the 25 region. The beauty is, since it is lower stress, the crack growth slows down and then you can spot it, before it becomes a disaster. Also if the crack is brittle in nature, these extra members are now “thickening” up or 'stiffening' the region too because there is more material where they are welded, thus the cracks propagate here too. Hence, if the structure is well designed, cracks can propagate into “safe regions” and even arrest totally. Thus, it becomes a fatigue resistant structure. Again, a broad 'generalisation.

The waves slamming onto the boat is just one source of cycles, there are many on a boat, not just environmental. If your genny is running a lot, there’s another…your mast, flapping about is another…and so on.

Great quotes Peter

 

Whoa - I consider myself a good boilermaker and above average welder and can not pull such a feat off. Get back to reality - we had all seen photos of your work.

I once had a boat built to Xray procedure welding as per request of client and welded by qualified/certified coded welder. Root runs done with TIG and filler/cap with E7018 and Xrays/ NDT done by TUV. During the welding the welder himself also performed dye-pen testing and particle testing on all welds and all looked well.
However, at the Xray session still a few flaws showed up, but luckily within the percentage allowed for the random testing and passed.

Ad Hoc, thanks for your brilliant posts, even pro BS supporters enjoy to have that kind of quality information shared with them based on sound principle engineering facts presented instead of the demolition quantum theories by you know who based on foolishness.

 

Thank you for the informative and factual post, Ad Hoc.
I think while it is understandable that not everyone able to be as informative as you, sticking to facts and forgetting mockery could theoretically be learned from you by anyone, which I'd like to recommend.

So the bottom line: because lack of structural redundancy, Brent boats should have been scantled to smaller stresses and strains than a framed one. The question remains whether they are, i.e. whether the rules of thumb used in design give enough safety factor. This is again a question which can be ansvered by structural analysis.
Did I understand it right?

BTW I have found a very good source for engineering information: the indians have put whole university courses to the web bot in webpages and video.
Here are the most relevant ones (Ocean Engineering), but there are a lot on mechanics also.
http://nptel.iitm.ac.in/courses.php?branch=Ocean

 

Also a fairly complete Naval Architecture series on youtube.......

http://www.youtube.com/watch?v=W1HH5-QhdEQ&feature=channel

 

Holy smokes, Wynand, the roots were all GTAW(tig)? That must've taken a very long time!

When I was doing a job in The Bahamas - fuel tanker - I had quite a clash with an inspector, because he wanted me to go back into the bowels to put another partial cap pass on a weld. The cap that I had laid had to be ground flush & just as I finished grinding it - in a 1m x 1m x 1.5m(approx) confined space, with crap being blown everywhere(pos vent.), 38degC outside temp+ & only a flashlight - I nicked the plate with my grinder, right across the weld, in a couple of places. But, the inspector couldn't be sure that I hadn't removed a flaw, so, back I went, to lay beads of about 20mm each, then grind again. The trip, each way, required 2 (600ml) waterbottles. LOL, that was the job from hell! Working straight 12's(7am-7pm), in incredible heat, Romanian workers screwing with our machines because they were pissed that we were brought in & were paid a lot more, no chance to tour where I was(too drained), most of the hotel menu not available, because of a recent hurricane, cheap gear from China, because they told us not to bring ours, etc. . I couldn't even see the view when we flew out, because I had so much metal in my eyes that I could only see white light! Great fun thinking back on it, though.

No, no way Brent's done U.T. welding.

Mike

 

Ad Hoc Thanks again.Hey,thanks for the links too but I've got a coupla professors right here!

Now,PDWILEY if you're still about,I've a detail that I'm confirming
PDWILEY post#513

"Assuming the pad is thick enough what happens is that the plate it's welded to flexes as load comes on & off. All the pad does is increase the area of flex. If it's still too small to stay below the point where fatigue cracking will take place, you've bought more time before failure but the structural problem still remains.

The pipe won't punch through the pad, it transfers its load to the pad and the pad transfers it to the hull plate which flexes"

I get from the above that the pad MUST be thicker to restrain the pumping shear of the strut(a round hollow pipe on edge) -a match for it's push/pull strength in other words,
But then you say "All the pad does is increase the area of flex. If it's still too small to stay below the (fatigue)point...etc. "
So,you think a larger-area and thicker pad is needed.But it might not fit beside the chine or curve of the hull there.Putting the strut off centre on the pad gets a little more complicated?

Also,there are two pads,two struts and while they can be Pull/Push or Push/push or Pull/pull,they still work in unison so it's "DynamicShear/2..

 

Yeah. I agree. It's kind of like Sybil ... you never know who's talking. If there is a coherent response in some of those posts, I wouldn't know who's making it. Often sounds like one man arguing with himself.

I've got nothing against you Brent, I'm reading this thread with an open mind -- at least I started to. I'm still trying, but I'm losing the will to fight. Put your best qualities forward, proofread your responses, add some form as well as substance. I think we all acknowledge that a Volkswagen Beetle is not as safe as a Mercedes ... make your argument for "safe enough". If you're going to rely on anecdotal evidence, this thread might as well end now. There have been so many opportunities for you to either show some humility or show some respect, and it seems you have trouble moving in either direction. I'd love to be wrong -- show me.