Stitch & glue questions

[johnw]

I'm not pretty seriously thinking about building a 9 1/2 foot sailing dinghy, stitch and glue. I think I can get the patterns for the panels printed out full sized. So, I thought I might glue them to door skins, cut them out, and router the shapes in the wood I actually build the boat with. That way, if I decide to build another, it's dead easy.

Then it occurred to me that this would create a much smoother edge than cutting it with a jig saw, and that might not be a good thing, because a rougher edge would presumably give the epoxy a better grip. Is this a real problem, or should I go the router route?

I'm also hoping to come up with the least unpleasant way to deal with sanding down the fillets. I don't fancy being covered with little glass daggers, and I sure don't want to breath them. For such a small boat, is wet sanding a viable option?

This is the boat I'm thinking of building. It would carry an El Toro rig, which I've already got.



I'm thinking of building it in 1/8 inch plywood, which would make it a bit fragile, but quite light. The plan is to have a boat I can carry down to the beach and toss the rig in and go sailing with little rigging time, and it needs to be narrow enough to fit in the back of my pickup truck and big enough to carry two people without much trouble.

I've only built one boat (well, except for a plywood boat I helped my dad build,) and I've not built stitch & glue before, so any advice would be helpful.

[Yeadon]

At an 1/8th of an inch, you could almost just score them a few times with a knife and break them apart. But that'd be a pain. I'd probably just use a jigsaw, then smooth the edges with a block plan and/or a foam block wrapped with sandpaper. My concern would be tearout on the underside of the plywood. But then, I'd just mask that damage with the fillet.

[johnw]

At an 1/8th of an inch, you could almost just score them a few times with a knife and break them apart. But that'd be a pain. I'd probably just use a jigsaw, then smooth the edges with a block plan and/or a foam block wrapped with sandpaper. My concern would be tearout on the underside of the plywood. But then, I'd just mask that damage with the fillet.

Good point, I'll need to make sure the tearout is all on the inside. But of course, that's the beauty of the router plan, much less likelihood of tearout.

[Breakaway]

The filets and tape are "fastening" the panels, not so much the edge-joins. Even if they were at 1/8, there is so little purchase relative to the total panel area I doubt there's a measurable difference in adhesion cut versus routered.

And of course, as Yeadon noted, most cut edges are cleaned up with a plane, so they are smooth anyway.

kevin

[johnw]

Thank you, not having built this way, I'm a newbie. I recall Sam Devlin saying you didn't want too much precision in the joint, because it didn't allow enough of the material that holds the boat together. Didn't know quite how to interpret it.

[RodB]

The best way to sand down the fillets is to make them smooth and unifom in the first place and then... while they are only cured about an hour ... perhaps a bit more ... use a chip brush to epoxy coat the fillet carefully and you save time by avoiding the sanding a cured fillet and applying epoxy later on. I also many times save even more time by adding fiberglass tape to joints (that do not require biaxial tape) once I have applied this coat of epoxy to a very uniform fillet.

After all, in many cases with stitch and glue construction, once you form a good uniform fillet, you let it cure, sand off the high points or sand it fair, then apply fiberglass tape to really strengthen the joint. Important structural joints are taped with biaxial tape.

In joints that do not require biaxial tape, if yor are able to produce clean uniform fillets, you can just wait a little while and carefully epoxy coat the fillet, then carefully lay out a pre cut piece of fiberglass tape or cloth on the wet epoxy (with gloves) to finish the joint (when required). If your joint does not require taping then you can just use the application of a coat of epoxy to a "hour and a half old" fillet to smooth it imensely and get rid of all the pourous surface caused by using wood flour and get a very nice smooth finish. The fresh epoxy softens the surface of the fillet and allows you to smooth it nicely and save steps in the construction. Sanding a smooth surface takes less time than an bumpy porous surface.

On joints... you can bevel each side of an edge and it lays nicely along the surface to be joined to... apply an appropriate fillet to each side and you have a proper joint. The epoxy fillet material just covers plenty of surface area on both of the panels to be joined. With thicker stock than you are talking about, it is common to round over the corner of the panel that will sit against the adjoined panel and that way you get a uniform joint... with only a narrow part of the radiused edge laying on the panel it will join. Now you apply a proper fillet, and you have some uniform joints that look great.

Quality work in stitch and glue should involve reasonably fitted edges laying along the surface to be joined to.

RodB

[johnw]

Thanks, Rod, that's very helpful.

[wizbang 13]

John, By now you have seen what I have done with Wizbang and Bingo. 403 filets and no tape.
If I can bang out 3000 miles @ 25- 45 mph, NO 5 knot sailing dingy will come apart that is built the same.
Anyway, my trick for sanding filets is to cut down an old inch n a half thick softpad and glue sandpaper that is an inch oversize in diameter. The paper folds over nicely @ 1500 rpm with a polisher .freakin dream. I never have seen anyone else do it that way.
I had gaps , half inch some places. They didn't let go. Even hitting the rocks @ 25 K .
I am not suggesting others build 40 mph speed boats with out tape, like I do ,but for wee dingys, ya.

[johnw]

You're an inspiration, Bruce!

I was thinking I'd just tape the outside of the chines, for abrasion resistance. I might do that skinny little bottom panel as well.

[johnno]

John, I basically didn't sand down any fillets and I certainly didn't bevel any edges. My sides come down against the edges of the bottom, which makes for a little 'V' on the inside chine which gives the epoxy a nice bit of thickness for strength.

The method suggested to me for the filleting was to tack fillet first. These are not large fillets and sit below the thickness of the final fillet. Then remove the ties when set. I then ran masking tape either side of the seam about two and a half inches away. Once done, brush on a coat of epoxy out to each piece of tape.

Then squeeze out a long fillet of epoxy from a plastic bag (like cake icing) and wipe once or twice with a rounded scraper. My first attempt at this I did the wrong way, I kept trying to tidy up the little running beads which get squeezed out either side of the main fillet.

Just forget about them and keep going till the bag is empty. Because the bag collects the epoxy into a big lump, it will go off faster so it's much better to get it out of the bag and onto the seam. Then you can come along and scrape the little edge beads away from the fillet with a flat scraper (the masking tape keeps all this neat), collect that stuff and add a bit more length to your fillet with it.

The reason I didn't sand any fillets was that once I had done one seam this way, I then placed fibreglass tape (biaxial in my case) over the fillet and brushed epoxy into it till it was fully wet out.

This method leaves a slightly raised surface along the edge of the fibreglass, but I quite like that look. It just shows the strength of the seam to my eyes. The whole interior needed perhaps ten minutes at the most of very light sanding just to take any tiny high points off that the brush left. The only real sanding I had to do was of the glassed bottom.

By the time I had finished the job (this was a first time for me as you know), I was so much better, neater, and faster at making fillets that it wasn't funny. The trick to this was getting a whole mix of epoxy bead laid out before going back to clean it up, rather than tidying as I went. Once the bead is out and onto the boat, you have ages to clean it up as much as you want, and always scrape the waste outwards, not back into the fillet.

If I didn't need the biaxial as part of the structure, then I could now make a perfectly good fillet Wiz-ztyle without the tape, and it would only need the lightest sanding..very quick.

Re-reading this, I should have said that the brushed on coat of epoxy between the lines of masking tape is for the fibreglass tape which comes later, and nothing to do with the fillet. If you don't use fibreglass tape, then you would not need the epoxy coat obviously, and you might also dispense with the lines of masking tape, or bring them in a bit closer to the fillet.

[johnw]

Thanks, Johnno, that helps. I'm quite a bit more nervous about attempting this than I was about Black Swan, because I understand wood a lot better than glass and glue.

How about the glue escaping the seam on the outside? Did you just make your joints tight enough that this wasn't a problem?

[johnno]

The joints need to come up tight. Assuming the panel shapes are okay, you pull them up as tight as you can. Perhaps as important though is that you do all the adjustment necessary to the shape before you start putting on the tacking fillets, because once that's done your shape is set.

I remember Ross was very interested to see how well the panels fitted together, as the dimensions came off his design programme. They don't always come off right I suppose.

I'm guessing that if you are scaling off you might need to spend a little more time getting that adjustment right. Not hard because you have plenty of time to play with this. It's so easy to mark up the panels, cable tie them together, and take apart again if you need to. I remember your design model for Black Swan so I'm sure you'll be doing one here too. A little larger scale than normal will solve a lot of the dimensional issues in cardboard or balsa.

[johnno]

Also John, no need for a strongback, just a flat surface and away you go. My simple little frame gave me this, and also put the boat at quite a good working height for the build. You could do it on a couple of saw horses if you wanted to.

[johnw]

Yes, I've got to get my sawhorses back. The friend who let me build Black Swan in his garage has still got them. I'm still in the stage of working with Vashon printing on getting the patterns printed out. Think I've got a handle on that.

[johnw]

The joints need to come up tight. Assuming the panel shapes are okay, you pull them up as tight as you can. Perhaps as important though is that you do all the adjustment necessary to the shape before you start putting on the tacking fillets, because once that's done your shape is set.

I remember Ross was very interested to see how well the panels fitted together, as the dimensions came off his design programme. They don't always come off right I suppose.

I'm guessing that if you are scaling off you might need to spend a little more time getting that adjustment right. Not hard because you have plenty of time to play with this. It's so easy to mark up the panels, cable tie them together, and take apart again if you need to. I remember your design model for Black Swan so I'm sure you'll be doing one here too. A little larger scale than normal will solve a lot of the dimensional issues in cardboard or balsa.

I did a little paper model, which is not a very precise way of doing things, and discovered that the boat was too flat aft, then adjusted the lines and did another paper model, which looks much better. That's at 1/12 scale, cutting the bits out just as they came out of the printer. I figure if that shape came out right, the ones that come out of the plotter likely will as well.

[lumberjack_jeff]

We cut our panels on a CNC machine, but when doing it manually, we will use a skilsaw - a jig saw is too hard to make a fair curve. On 1/8" ply, I'd cut both panels out at the same time.

We chamfer the inside edges of the panels half their thickness with a block plane or a bottom bearing router bit.

We use 4 or 5 wire stitches on the high stress ends of the panels and T50 staples in between (1/4" or thinner material) or 7/16 crown construction staples (for thicker material).

The fillets don't need to be big at all. On 1/8" thick ply, the fillet should be no more than the diameter of a tongue depressor.

Tidy up your fillets meticulously. It's better to have excess epoxy cure in the cup than to have to sand it off.

Epoxy doesn't care how smooth the wood is - it'll hold just fine.

One last thing: bear in mind that your truck is 48" between the wheelwells, but wider above them.

One other, last thing: this boat is comparable in size to a devlin 5x10 skiff, which is built from 1/4" ply and does not strike me as overbuilt.

[johnw]

Thanks, Jeff. The donor boat for the rig is a beat-up fiberglass El Toro. The pickup is a '97 Nissan Hardbody with a canopy, so the opening of the canopy is the critical measurement. The wheel wells are less than 48" apart, but the boat can sit on top of them. The El Toro, with a beam of 3' 11" at the rail, is a tight fit. The molded beam of the new hull is planned at 3'8", so with the rub rail should just fit nicely in the bed. This will save me about $80 every time I take it on the ferry, compared to my 17' sharpie on a trailer.

[johnno]

John, hope I don't seem like I'm telling you how to suck eggs, but the other important thing with this method is to put in the correct temporary spacers to get the beam at deck level correct. I used a combination of the permanent bulkheads and temporary spacers, as you probably remember. You might work this out on the model too. Then you're 100% sure of the shape and panel dimensions coming out right. Check carefully for any twist in the whole boat before epoxying it all up.

[johnw]

John, hope I don't seem like I'm telling you how to suck eggs, but the other important thing with this method is to put in the correct temporary spacers to get the beam at deck level correct. I used a combination of the permanent bulkheads and temporary spacers, as you probably remember. You might work this out on the model too. Then you're 100% sure of the shape and panel dimensions coming out right. Check carefully for any twist in the whole boat before epoxying it all up.

Hey, any advice is welcome. I was thinking that I'd put a thwart in at the point of widest beam, which is about 3'9" from the transom. I suppose if I mark the center of that and the center of the bottom panel and hang a plumb bob there, it will give me a good check looking forward and aft with it in place, and a line running the length would give me a check as well. Thanks for getting me thinking of this. I'm also thinking I'll have a couple feet of deck forward to act as breasthook (not as kinky as it sounds) and mast partners. A bulkhead up there would give me some buoyancy as well as helping keep the boat straight.

By the way, a Russian woman taught my father to suck eggs when he was still a kid on the farm. She was the girlfriend of one of his less reputable relatives. For some reason, she thought either that he needed to know this, or that he would think better of her for knowing she possessed this knowledge.

The trick is to put a small hole in each end. There, now I"ve told you how to suck eggs.

[johnno]

Haha, to be honest I'd never thought about sucking eggs in reality. I think they also used to do it to preserve the egg shells, in the days before the world became more PC?

Yep, sounds like a good plan to get those decks, bulkheads and watertight compartments sorted about now. The bulkheads can go in as stitch and glue at the same time as you're doing the hull panels. The thwart would probably be made and fitted much later, so a temporary spacer screwed to the ply sides from outside will keep the shape of the hull right while you do the first stitch 'n glue steps, and then the real thwart can go in when it's time.

My order was bottom, sides, bulkheads (and temp spacers) for first stitch and glue-up. Second off the rank was internal frames, centreboard, and in your case, permanent thwart if you have one. Also at this time, decking frames, and then gunwale.

Lastly, deck went on top, just glued down.

[Boater14]

This is all great info. Could be a book. Small batches and complete organization. Accurate mixing. There's a YouTube called EPOXY posted by zeitgeistsurfer. Shows a guy coating full 4x8 sheets with a coat of epoxy using a squeegee, roller and one of those yellow plastic bondo spreaders. It's time lapse but is almost real time. It's a hoot but a good first step if you plan that level of encapsulation. Be stingy within reason as the sanding is a real pain. Drop cloth or you'll be hearing about the condition of the garage floor long after the boat is history. No need for any fast setting epoxy.

[johnw]

And by the way, Jeff, 1/4 inch ply is also standard for El Toros. I have a bad back and want to be able to carry this boat, which is why I'm thinking of 1/8 inch. I might still go with 1/4 inch.

[wizbang 13]

4mm okume, I used it for Bingo's deck. Edensaw right in Tacoma. Light, tougher than doorskin.

[johnw]

4mm okume, I used it for Bingo's deck. Edensaw right in Tacoma. Light, tougher than doorskin.

Thanks, I was thinking okume. One thing is, there's a lot of twist in the bilge panel, and I'm sure that will be easier in 4 mm. Do you think that twist is doable in 6mm? Thing is, there's more skin area than on an El Toro, and I want it light enough to be easy to carry.

[JimD]

How about the glue escaping the seam on the outside?

Masking tape. Peel it off after the epoxy cures. A few years ago I built a stitch and glue kayak without a single stitch. Masking taped the whole thing together on the outside, filleted the inside. After cure remove the masking tape and do the outside. There's really no need to drill and twist all those stitches in a boat made from such thin and flexible plywood. You might want to try just putting a few stitches at strategically important points and masking tape everywhere you can.

PS, you might save yourself a little grief if you round that really hard corner where the stem meets the bottom planks. The plywood is not going to like what you've drawn. Good luck, John. That's a nice looking boat.

[johnw]

Masking tape. Peel it off after the epoxy cures. A few years ago I built a stitch and glue kayak without a single stitch. Masking taped the whole thing together on the outside, filleted the inside. After cure remove the masking tape and do the outside. There's really no need to drill and twist all those stitches in a boat made from such thin and flexible plywood. You might want to try just putting a few stitches at strategically important points and masking tape everywhere you can.

PS, you might save yourself a little grief if you round that really hard corner where the stem meets the bottom planks. The plywood is not going to like what you've drawn. Good luck, John. That's a nice looking boat.

Thank you, masking tape it is. I may end up rounding the corner if it proves hard to plank. The panels develop with good precision as to area, but that spot has a good deal of stress.

[JimD]

Thank you, masking tape it is. I may end up rounding the corner if it proves hard to plank. The panels develop with good precision as to area, but that spot has a good deal of stress.

Here's a pic of the kayak going together. The panels sit in a craddle so they don't flop around while taping it together and I had to employ a clamp or two. :



And putting the deck on:



I haven't done it here (but have on other projects) but you can also run the masking tape the whole length of the seams to stop glue from pushing out the other side.

[johnw]

I'm gonna need more clamps.

[JimD]

Even the fancy combing fit was done without wires:

[johnw]

Vashon Printing just printed out full-sized versions of the panels. Doubt I'll be able to do much before the Seattle Wooden Boat Festival is over, but progress begins. I'll start putting together the materials and trying to find the time.

[James McMullen]

Stitch and glue is easy and very forgiving. You won't have many problems with it, I'm sure.

I think you should fix that hard, sharp chin on your bow though. There's a reason why nobody does it that way: not only will it suck for beaching your dinghy or encountering floating line or driftwood, and possibly root and broach you if you stick your boat into a wave, but also on a purely practical structural level you will be able to get the panels to take the twist ever so much easier if you rake and curve that line. Look at all the other stitch and glue designs of kayaks and rowboats and whatnot to see the shape that works better in real life than the approximation on your computer screen. And then build a scale model to check it. You don't want to torture your ply panel as much as that shape will demand.

[johnw]

Actually, this is a flat bottomed design, and that sort of chin at the bow is usual for flat bottomed designs. And keep in mind, that's the waterline at about 300 lb. which is with me and a smaller companion. I figure as the load increases, the bow goes down, which is better than having the transom in the water. Normally I'll be sailing the boat about 100 lb. lighter, with my weight aft of amidships. When I'm sailing single handed, which will be most of the time given the boat's purpose, the trim will be more like this:



and with the intended kick-up centerboard and rudder, it should beach rather nicely.

Anyway, that's the theory. We'll see.

[johnw]

Oh, and when it's trimmed like that, the center of flotation is right where the seat is.

[James McMullen]

John, you've got an enormous amount of twist in your garboard panels. Look at how the Nutshell pram, the Shellback dinghy or the Bolger Cartopper--all proven boats with similar five-panel construction--do it. It was hard enough for me to get the Cartopper's panels wired together smoothly, and it's the easiest curve of the lot.

I think you really ought to at least mock it up with utility ply to see how much stress you're building in. I think you're exceeding the practical limits with this shape. With that much tweak, you may need to add a layer of glass to reinforce the tearing fibers on the tension side--and that's where the weight and the costs start to really pile on.

[johnw]

John, you've got an enormous amount of twist in your garboard panels. Look at how the Nutshell pram, the Shellback dinghy or the Bolger Cartopper--all proven boats with similar five-panel construction--do it. It was hard enough for me to get the Cartopper's panels wired together smoothly, and it's the easiest curve of the lot.

I think you really ought to at least mock it up with utility ply to see how much stress you're building in. I think you're exceeding the practical limits with this shape. With that much tweak, you may need to add a layer of glass to reinforce the tearing fibers on the tension side--and that's where the weight and the costs start to really pile on.

Here's what Delftship says about how that panel develops:



Now, when I was designing a boat for this purpose, I ran up a single-chined hull of about the same weight, beam and displacement:



It has the raking stem and the rounded forefoot. Here's what Delftship says about the equivalent panel:



The faceted bottom allows me to develop a shape with less stress in it than a straight V-bottom shape. The model I'm planning on building has about a third the stress my best effort in a conventional single-chined boat with a sharp forefoot has. It's still a fair amount of stress, though, a mock-up might be wise.

Thanks for the input. Not having built this way before, I'm not at all certain what shapes are possible.

[wizbang 13]

mattboard from an art store or framing shop. 2 inches to the foot. big model . It behaves just like ply at that scale.

[johnw]

Thanks, Bruce!

[JimD]

... Not having built this way before, I'm not at all certain what shapes are possible.

As soon as you start it will become very obvious why that corner is a bad idea. Fortunately, its a very easy thing to correct. You could even shape the curve 'in situ' with a belt sander. The more you curve it the easier those panels with mate. Stop anytime it feels about right.

[johnw]

I've tried a couple of approaches to raking the bow, but I keep getting more stress rather than less. And while a straight v-bottom design takes a natural curve like the one in post 35, on this one it doesn't seem to want to do that. Even though the flat bottom is very narrow, it seems to change the dynamic quite a bit.

[johnw]

Okay, I figured out how to add a control point so that I could induce some curve. I like the look better, but it has three times as much stress.

[JimD]

I think perhaps the design app you're using has a funny notion of what stress is and what is actually being measured as stress.

[johnw]

I think perhaps the design app you're using has a funny notion of what stress is and what is actually being measured as stress.

That had occurred to me, as well.

[Lewisboater]

One way to play with it visually is to set your perspective view to developability check, set your Precision to medium and then start moving the points around in your 3 other panes. You will see the red move...sometimes increasing and sometimes decreasing. Move your points to minimize the red while at the same time getting a shape you can live with. Then go check your model's stresses and see if you can live with that. Go back and forth until you get what you want or toss the computer out the window.

FS and DS are designed with Steel panels in mind...plywood will do the torture thing with a lot less complaining than will 1/2" steel plate...at least without heat and hammers.

[johnw]

One way to play with it visually is to set your perspective view to developability check, set your Precision to medium and then start moving the points around in your 3 other panes. You will see the red move...sometimes increasing and sometimes decreasing. Move your points to minimize the red while at the same time getting a shape you can live with. Then go check your model's stresses and see if you can live with that. Go back and forth until you get what you want or toss the computer out the window.

FS and DS are designed with Steel panels in mind...plywood will do the torture thing with a lot less complaining than will 1/2" steel plate...at least without heat and hammers.

That's pretty much how I got where I am now. I put one panel in the mode where red turns to green, and moved things around until all was green, then set about to reduce the error level from there. The version with the rounded forefoot looks more boaty, but I can't get the error or the stress to be lower than it is with the sharp chin.

[JimD]

John, have you been following flo-mo's tape and glue build? http://forum.woodenboat.com/showthre...ness-continues

Stitchless stitch and glue:

[johnw]

John, have you been following flo-mo's tape and glue build? http://forum.woodenboat.com/showthre...ness-continues

Stitchless stitch and glue:

yes, I've done some lurking on that thread. You can see how the v bottom goes into a natural curve at the bow. The narrow, flat bottom on mine seems to make it act differently.

[Lewisboater]

FS-DS is like someone from the flat earth society. It absolutely believes that you can't do that kind of bow but as we all know...you can. However...it also shows the panels at the bow with convexity...this is iffy. The plywood will want to take a conically correct shape which would be convex (look carefully at FloMo's boat's picture and you can see the subtle outward curvature) which is why the program is being cranky. However...we all know that you can torture the plywood to an extent and what you show is not that bad so it should make the grade. Depends on the thickness of the plywood, number of plies and quality overall. 3 mm 1088 occume will make that bend easy...up to 6 mm will not be much harder. Meranti will be a tougher sell but I give it a Probable. Other crappier stuff with fewer layers may do it fairly easy but it's long term acceptance of the situation might be questionable.

[johnw]

I'm planning on 4 mm okume, as Bruce advised. There's definitely something funky with the software, because it stops showing the surface deveolping properly when I add a control point, even if I haven't changed the shape.

[Lewisboater]

Of course...think of your drawing as a net...pulled into shape by the points. If you add another point the net will be distorted at that point. The best way to look at your model is with interior edges turned on...this actually represents how the program is thinking and representing your shape. Anything that distorts the mesh will affect the developability. However...if you can create another 2-3 points and form a 4 pointed shape it will like that much better than a triangle or 5 pointed shape. You must also subdivide the net with lines...so a 6 pointed shape would be much better as 2 4 pointed/sided shapes with a common line between 2 of the 6 points.

[johnw]

I made a rather crude model with mat board, as Bruce suggested. It works, as long as the bottom can be kept where it spreads the garboards, but it sure looks to me like it wants to be rounded.