Naval Architect Needed (paid?) - I Cut My Houseboat In Half

That is really good advise to avoid the epoxy if you can't take it back it can be painted on as an excellent waterproofing layer that will need paint over it, avoid sanding it and use water if you have to.. The stringers look about half size before the hull extension.... taller stringers are better and doubling the number as well will get back to something like safe. The hogging is lack of transverse strength so double that too . . Surfboard manufacturers are often not that careful with skin contamination and fumes, sometimes for decades, they use polyester it's way more forgiving than epoxy. Foam stringers with limber holes and good long lay ups are likely the easiest way to mate and increase the sizes of the originals, shape them so there's no corners that need filling, mix in manageable batches .. Disclaimer no training except home made stuff and reading a bit..ha. All the longitudinal structure including cabin and gunwales could use some extra strengthening,, The boat is now longer and skinnier so be extra aware of weight above the COG. I'll probably delete this in a couple of days as my station isn't qualified. ha cheers.

 

It makes sense to get the long fibers in contact with the existing skin. But this brings up another issue and highlights the need for engineering. Finding an engineer is the challenge as the risk/reward for them is a bit off.

I'm tired and it is late, but I think you need a zero degree uni in this layup. That complicates things because it will also require overlaps around the hull, but my intuition tells me 1708 is not the only glass needed because a zero uni would be the intuitive opposite of a worthless 90 degree uni. A composites engineer may agree you need some zero degree tows. Whether it is preferred all at the end to keep the layup cleaner, I can't say. But my intuition tells me 20 layers of 1708 is missing something important.

 

I find it interesting that it seems to be American practice to add the large pieces of glass first,here in the UK it is precisely the opposite.We always begin with the smaller piece and build up from there and I just ran through my mind the number of composite engineers and designers whose specifications I have worked with and the total was eight composite engineers and four designers that had consulted composite engineers.They all increased the size of additional laminates as the item was laminated whether the item was being made from wet laid material or prepregs.

Looking again at the pictures of the boat sections I can see that there are some large gaps evident and this is unfortunate as nothing will bridge those distances nicely.The choice is to trim some adjoining edges for a better fit,or if the overall length is important,to create an outer surface.The simplest way to create an outer surface to laminate against is to use something like 1/4" plywood and melamine faced is even better.It can be in several pieces if necessary and after dry fitting,it can be removed for waxing-after which it needs to be handled carefully so that bonding surfaces don't get contaminated with wax.If it means 50-60 self tappers need to be used,go ahead as it is a much simpler matter to fill screw holes than to fair and finish several square feet of surface.

I'm reassured to see that others agree with using polyester or vinylester and I would harbour a suspicion that the original boat may have been laid up with a chopper gun for speed of manufacture.This type of boat tends not to be used for any purpose where performance is a guiding criterion for the designers and lots of material applied speedily and inexpensively tends to be the norm.That being so,I would tend to use CSM for the repair rather than introducing a fibre with different characteristics,but as mentioned in an earlier post I would use one or two plies of sturdy unidirectional tape along the top of any new stringers with mat below,between and above them.

 

I'd avoid waxes for precisely the reason cited. The form can be covered with ship tape. Some prep is needed; you cannot form over gelcoat, for example.

On the matter of which piece first; the quality of the layup is key. If you are laying 20 sections of glass and start woth the narrowest; each subsequent layer may entrain air at the overlap of the prior piece. Additionally, no way can he layup 20 pieces or the laminate will cook itself; so he will need to sand a/o peelply in between. Sanding the bond with smaller first runs the risk of sanding through the longest layer which is defeating the most vital bond.

I'm a bit sad my finger sketch shows the opposite, but I agree with longest first. Since I'm not a composites engineer, I can honestly say, I would probably have filled the scrarf smallest first to the edge of the taper to get the hull dimensions the same and then laid longest pieces over hull and scarf to smaller if I were flying by the seat of my pants.

I cannot imagine only using csm for the layup. If me, and no help, I'd use 1708 in the taper and then put the uni over the existing hull and scarf. But all guesses here. He needs an engineer.

 

Nobody would dream of laying up 20,plies at once I would hope.I haven't seen problems with entrained air at the overlaps of a well consolidated laminate and if using CSM I often use a torn edge in preference to a cut edge to eliminate the appearance of ridges.Again,a well consolidated laminate shouldn't need sanding between plies unless you were unwise enough to use epoxy with no peelply and on this project I don't see a need for epoxy.Were it a racing boat constructed of epoxy and in need of a new section of hull,it would be different.

 

Thanks guys, I have a couple answers and a couple questions.

• The boat luckily is not a chopper gun boat, it’s good ole 1970s thick woven roving mat for the hull. It's anywhere from 4 to 8 layers of the stuff and must be ~0.080 thick for each layer.
• The boat sections have a big gap between them right now just to allow for easier scarfing of the edges. The sections are on a rolling cradle so they can be butted right up against each other before glassing.
• Curious why the preference for poly or vinyl? When I was researching for that decision I read some articles about epoxy resins superior secondary bonding characteristics vs other resins. One's like this (WEST SYSTEM Epoxy vs. Polyester - Epoxyworks https://www.epoxyworks.com/index.php/west-system-epoxy-vs-polyester/) showed a test repair of a polyester fiberglass specimen using poly for the repair and then epoxy for the repair with a single sided 12:1 scarf. The poly being ~70% of the original original strength and the epoxy being 80% of the original strength. Articles like this are what led me down the path of epoxy being the strongest for a big job like this. Of course I know a lot of these articles are from West System so I understand they would definitely be biased trying to sell their epoxy.
• The big side vs. small side first is really throwing me off. I know there have been some older threads on this forum discussing it. When I look at a lot of boat repair books it shows the biggest piece first, and Boeing/Aircraft/NASA composite repairs always show the smallest piece first. Curious why the differences in layup schemes on repair joints and if strength changes with either way.
• I have started searching the internet Googling for "freelance naval architect" and coming up with some sites like Upwork.com with people on there offering their services. Not sure if this is the best route to find a qualified engineer, but I'm actively searching! If anyone knows of someone on here who might be interested let me know!

I appreciate everyone's ideas and feedback, and will definitely enlist an NA (once I can find one) to help validate some of these ideas.

Thanks!

 

My advice, visit the reasonably local boatyards and ask who is their go to guy for engineering and naval arhitecture. It doesn't matter if the yard is doing steel fishing boats or leisure fiberglass, they all must know someone.
After you find someone you don't "have your ideas validated" you follow whatever the guy prescribed, even if you don't understand why.

There is no worldwide consensus on how to install patches when scarfing, you can stop obsessing.

 

Peel ply between epoxy layups, yes! I forget about using it when I am not thinking in terms of vacuum bagging. The peel ply will leave a bondable texture and remove amine blush without sanding as it is pulled off the laminate.
So most of the patching is in the 1/2' (13mm) thick sections between the various longitudinal features.
It would be worth asking the naval architect if some of the stitched biaxial layers should be 0/90 and some 45/45 to the lines of the hull. In the last repair on my boat it made sense to re-orientate some of the layers.

 

Are you sure it is not C-Flex ?

 

Epoxy vs esters.

Yes, epoxy provides stronger bonding.

Epoxy is not ideal for the gelcoat finishes.

The epoxy bonded sections will vary in strength from the areas nearby. And this repair is not simply about making the repair area super strong, but making the strength of the repair similar to adjoining areas to avoid high variations in local areas. As a builder, I understand the basic concept, but the NA and engineer would appreciate it with precision.

Conceptually, consider an aluminum popcan glued to a sheet of paperboard. Then for a moment consider a bond that is not the weakest. Now, apply a force to the aluminum; the failure is on the paper. The strength differences are too great. Same, too, for your boat; this amazing epoxy repair next to half inch polyester hull may result in the hull failing. But honestly, I'd be more worried about the quality of the work and so, you need an engineer to decide what is best. And this is alsothe reason I think you need longitudinals, and many; not a couple. The longitudinals will help alleviate local stress I'd say. As mentioned earlier, when the boat is in slings; you'll forever worry about cracking the repair if done poorly. Not fun.

 

I hired two freelance Naval Architects to help come up with the calculations for the houseboat connections. One came back with 77,000 psi max shear stress, the other came back with 1.0184e+007 Pa which is 1,567 psi max shear stress.

So obviously with one being 49 times higher, I'm a little confused. Also from reading other articles and old West System tests, it looks like even old polyester fiberglass usually breaks with a tensile stress of ~26,000 psi, so I don't know how a boat could ever be rated for 77,000 psi. I think even new epoxy fiberglass composites seem to be rated for 250-300 MPa which is 36,000 - 45,000 psi.

So needless to say I'm a little more confused than when I started. Any thoughts or commentary are welcome again as always!

Attached documents from these two NAs are attached if anyone wants some light reading.

 

Hi Kyle,

Im coming late to this party....
The first issue you really need to address, is the overall global strength. If the vessel ended up hogging, it wont be "just" because of the stringers.
The stringers, whilst they contribute to long.t strength, the main "box" of the hull does that!
Therefore I would surmise that the hogging occurred over the years simply because, like all boats, over the years more "stuff" is added. Therefore the design allowable load has slowly been exceeded.
There may have been on occasions, a large overload (big birthday party with many guests - for example),...and when these larger than anticipated loads are added, this very slowly compromises the structural integrity.
Thus a concomitant effect over the years leading to the hogging.

Therefore I would suggest a review of the existing structure, the main box, to ascertain if this meets your desired "loads" over the next 5-10years.
Then add the 8 foot section, and review again to see the difference.

If the main global strength is satisfied, then your scarfing issue become minor.
It is only an issue, if the global overall hull girder loads are close to their max possible.

And, for reference, these hulls are not driven by the max/min stress, but by deflection!
Composites are a low modulus material, which means it bends more for the same applied load than a material with a higher modulus. Think difference between rubber and steel.

So, in doing your calculations, 1st check the stress, but the most important check, is the defection.
Since I would guarantee you can get the stress calc's to be ok...but your deflection, not!

 

I agree with the analysis because, from what has been explained to us, the problem seems to be due more to the lack of longitudinal strength of the hull, global loads, than to local loads. That said, I would like to know the opinion of the experts, and what they base on when considering an acceptable or too high deflection.

(One of my conclusions : FEA, poorly applied, is useless, except to have "very technical" images.)

 

Probably a bit off topic and I don't have very much knowledge of FEA and especially no knowledge with Ansys:

The mesh in the pictures looks strange to me with Trimesh aft and Quadmesh mid and fwd, and it looks like there are some disstorted areas within the mesh.. anybody an idea about this?

The stress peaks appear in the aftmost toppart of the hull. Is this normal?

 

Agreed.
There is a mismatch with node compatibility at this interface: