Roof Spar Joint - Alternatives to Butt Joint

[Andrew Herrick]

That's an interesting idea ... framing the roof independently of the sidewalls.

So far, it seems the best solution for a solid plywood wall would be to route either a mortise pocket or an entire rabbet into the interior lip of the wall. That gives the spar something to a) rest on and b) prevent it from twisting. Using two screws per end, rather than one, would also add strength.

I understand the apparent appeal of sandwich wall construction, but again, I promised not to high-jack my own thread, so I'll skip that :p

I also like the idea of stapling the aluminum siding to the framing before attaching the aluminum trim. Sounds like a good fail safe in case the screws in the aluminum trim ever weakened and pulled out.

[QueticoBill]

Not criticising, as it's just a different way, but it doesn't seem that noseoil nor aggie rely on the foam and bonding of the foam to the skins for the structure, but just for insulation. Maybe because of having been introduced to the concept of composite foam and ply skins over 40 years ago, it seems like a natural. I'm sure I can omit the roof spars except maybe at ply joints, and just contemplating on how much of the frame I can omit in the walls - basically just edges and blocking - and does it have to be more than an inch or 2 wide. Perhaps I need to add a motto like "In glue I trust" or something similar, but if I didn't , I wouldn't get in an airplane or probably do a lot of other things, like use plywood. The problem will be what do I do with all the tube sand after its built.

[noseoil]

For a foamie, just the "hard points" need blocking to provide anchors for fastening. A drink-holder or a dome light should have something to screw into for a permanent attachment.

[KCStudly]

I originally planned to have minimal spars just at the roof vent; and to have my cabinet face frames be an integral part of the roof support system using foam and the outer skin in a structural fashion. Unfortunately, even with a slightly crowned roof/side profile, the weight of the ceiling skin would not support itself during installation. The options that I was comfortable with at the time were to build a temporary support system (even if it was just some creepers and props) or add spars to the top of the ceiling skin. I reasoned that it would cost me more in temporary wood and take just as much, if not more effort struggling with the temporary arrangement getting in the way inside of the cabin, than it would just to go ahead and glue the spars onto the ceiling skin before installing it.

If I new then what I know now (Ooh La La - The Faces) I might have tried it by applying some of the foam w/o the spars, just to the main area of the roof with the least radius; but that would have involved some extra risk (not being able to bend it after the fact) or extra testing.

Other than the spars, my roof is 5mm marine ply ceiling (sitting on top of foam walls with the same 5mm ply inner skins), two layers of 3/4 inch thk foam laminated with PL300, and two plies of 6oz glass cloth/epoxy. No corner blocking, no trim, and no rim board (except along the galley portion of the side wall and far enough forward to tie into the hatch hinge spar). The glass wraps the wall to roof joint with a nice 3/4 inch radius, no seams to be sealed, no fastener penetrations and no trim to buy, form or fit.

Not traditional, not everyone's 'cup of tea', but it suits my design criteria.

[aggie79]

Not criticising, as it's just a different way, but it doesn't seem that noseoil nor aggie rely on the foam and bonding of the foam to the skins for the structure, but just for insulation. Maybe because of having been introduced to the concept of composite foam and ply skins over 40 years ago, it seems like a natural. I'm sure I can omit the roof spars except maybe at ply joints, and just contemplating on how much of the frame I can omit in the walls - basically just edges and blocking - and does it have to be more than an inch or 2 wide. Perhaps I need to add a motto like "In glue I trust" or something similar, but if I didn't , I wouldn't get in an airplane or probably do a lot of other things, like use plywood. The problem will be what do I do with all the tube sand after its built.

I can't speak for noseoil but I chose my method of construction as a balance of ease of construction, strength, lightness, and insulating value. In composite or more correctly stressed skin construction the bond of the foam to the skins adds very little strength. Plus, the foams typically used in teardrop construction are non-structural foams. Any strength developed from bonding the foam to the skins is lost once the structure starts to flex and the foam shears below the glueline. The dilemmas with stressed skin construction are how to treat edges and openings, how to construct attachment hardpoints, and how to address impact resistance for a thinness of the skins. Once these are solved, I am sure that a stressed skin teardrop can be built much lighter than the way I constructed my teardrop although I question if the difficulty in construction is worth the effort and time.

[KCStudly]

...I question if the difficulty in construction is worth the effort and time.

This^. It may not be.

But then, sometimes it is the journey as much as it is the destination. Good thing I'm not in this for any sort of monetary profit!

[QueticoBill]

In composite or more correctly stressed skin construction the bond of the foam to the skins adds very little strength.

When the foam is the web of the structure? There are stressed skin panels that rely on the webs or spars in TD speak, but there are also panels that rely on just the foam or honeycomb or other web. That bond is essential to the strength. What would plywood be without the bond between layers? Or a SIP in building design and construction?

[yrock87]

Not criticising, as it's just a different way, but it doesn't seem that noseoil nor aggie rely on the foam and bonding of the foam to the skins for the structure, but just for insulation. Maybe because of having been introduced to the concept of composite foam and ply skins over 40 years ago, it seems like a natural. I'm sure I can omit the roof spars except maybe at ply joints, and just contemplating on how much of the frame I can omit in the walls - basically just edges and blocking - and does it have to be more than an inch or 2 wide. Perhaps I need to add a motto like "In glue I trust" or something similar, but if I didn't , I wouldn't get in an airplane or probably do a lot of other things, like use plywood. The problem will be what do I do with all the tube sand after its built.

I basicly did as you are describing. My roof has 3 total spars, plus 2 on the front wall and one at the hinge. The Spars are all either at plywood joints or where i would attach something from the inside (front cabinets, roof vent and rear cabinets) the hinge spar is the only one that has any real strength being 1x2 pine laminated with 1x2 oak. the remainder are all 1x1 or 1x1.5 pine. even the roof vent only has one spar at the front edge, no framing on the back or sides of the opening. and there is no "roof" framing on the edges of my ceiling. the wall framing meets that requiem so why have twice as much wood that high up?

front of trailer, 2 spars due to cabinets and plywood seam

a shot of majority of the roof. a single spar in a 7 foot distance from front to back. it was where the interior ply joint and roof vent were. you can see how I did not add spars where the foam joints were. in hind sight, I should have cut a piece of 3mm ply to fit on edge in that gap. It would have cost me a few ounces but made a significant increase in strength. like a honeycomb door (hypothetically anyways)
not to hijack the thread, but you can see how minimal my wall framing is here. I only put wood framing where there would be something that needed to be attached. (galley wall, galley counter, plywood joints and door. the top of the wall framing was included to allow the roof ply and roof spars to be glued/screwed down onto the wall.

[yrock87]

In composite or more correctly stressed skin construction the bond of the foam to the skins adds very little strength.

When the foam is the web of the structure? There are stressed skin panels that rely on the webs or spars in TD speak, but there are also panels that rely on just the foam or honeycomb or other web. That bond is essential to the strength. What would plywood be without the bond between layers? Or a SIP in building design and construction?

One more and I'm done, promise! :-)

just check my profile pick for how much strength the foam adds. I am standing on a 14x14inch foam and ply sandwich, no spars there at all. the glue bond between the three layers is what is allowing the three to act as one and distribute my 180lbs over a 12 inch gap.

[QueticoBill]

In the past several months I've read 100s if not 1000s of posts here and probably 20+ build journals and forgot how similar what you did is in the same direction I've been thinking. Looking at the wall framing, I have to wonder if 3/8" skins would allow skipping all internal framing.

Now I need to find a discussion of adhesive. You used 3m Fastbond 30; I found 3M 78; and my acquaintances who are involved in the manufacture of stressed skin panels for a living suggested Ashlands Iso Grip SP 4005 D, but not sure I can get that as a consumer. I found a lot of foam adhesive discussion on a surfboard builders bulletin board.


Thanks!

[yrock87]

In the past several months I've read 100s if not 1000s of posts here and probably 20+ build journals and forgot how similar what you did is in the same direction I've been thinking. Looking at the wall framing, I have to wonder if 3/8" skins would allow skipping all internal framing.

Now I need to find a discussion of adhesive. You used 3m Fastbond 30; I found 3M 78; and my acquaintances who are involved in the manufacture of stressed skin panels for a living suggested Ashlands Iso Grip SP 4005 D, but not sure I can get that as a consumer. I found a lot of foam adhesive discussion on a surfboard builders bulletin board.


Thanks!

I don't know where the ply gets thick enough to skip the framing. I guess it depends on what you are attaching. places where I need to add strength and where I neglected to add framing blocks I used "external" gusset planks to spread the fastner load.

here is where I had to compensate for not including internal framing for the hatch hinge spar. since I didn't know where I would want to hinge, I put it off till it was too late. I made a 1/2in gusset on either side that I glued to the wall to spread the load out, then screwed to that gusset.

To play devils advocate with you, why stop at 3/8 inch thickness? why not just make it 5/8 and skip the foam entirely? I would argue that using 3/8in skins REMOVES utility because it makes for a wall that is twice as heavy as one made with 1/8 inch ply skins.


as far as foam-wood glue. after mediocre results with the walls using 3m Fastbond. I switched to using gorilla glue and/or PL premium trowled out flat like tile grout. both of those seemed to make at least as good of bonds as the 3M fastbond and where easier to use, and arguably cheaper. both also have other uses beyond contact cementing flat panels.

[QueticoBill]

Yes, leaving out internal framing doesn't justify moving from 1/4 to 3/8 ply. I'll figure out how to attach something without framing when it happens. I still think fewer spars is probably desirable and a good alternative to more.

[KCStudly]

My 1-1/2 thk foam walls with 5mm (3/16 actual) ply inner skins, with 2x2 cedar sill and door surround, let in pine blocking for numerous features, and similarly constructed door cores (less FG/epoxy outer skin) weighed about the same as solid 3/4 ply; 45 lbs. For all of the extra effort I get sound deadening, insulation value, and no worry about splitting plies due to driving screws thru the edge of plywood. I'm sure there are other benefits that aren't coming to mind at the moment, such as compatibility with thicker windows and door latches, etc.

Plywood is fast. Foam hybrid is better, IMO.

[Andrew Herrick]

The only thing in my mind about using 5/8 ply for the roof ... is that you'd have to kerf the plywood, right? So unless you kerf it just right and fill in the kerfs with the right filler, you're not getting the full 5/8 of thickness. And ... you just added at least a 100 pounds to your roof!

[working on it]

cabin roof bracing, no spars.jpg (29.37 KiB) Viewed 1298 times

I don't know where the ply gets thick enough to skip the framing...To play devils advocate with you, why stop at 3/8 inch thickness? why not just make it 5/8...

Or 3/4". I used straight-cut flat 3/4" plywood on my un-insulated squareback TTT. Joined together with steel connectors in all stress points, glued with PL Premium adhesive in/out of every seam. I only used one sparlike wood brace, only on the forward 45 degree slope front of the trailer, as a precaution (unfounded worry) against wind-pressure. I used that spar to mount an overhead stereo/ventilation fan system anyway. On the roof, there are no supports, save for side bracing around the doors (also bracing the front slope and shelf), and steel corner braces for 90 degree joints, and Simpson Strong-ties gusseted angle braces around the perimeter of the roof (same system used on the floor, as well). The walls are mounted on top and within the outlines of the floor; the roof and front slope are laying on top of the walls. Four and a half years after the structure was completed, there has been no twisting or sagging of any part. Granted, the largest part of the roof is only about 1825 square inches, there is no sign of sagging nor pulling apart from the walls. However, I intend to retrofit a couple of internal beams to the underside of the roof plate, to fend off any future sags and add strength in order to add a roof rack, awning bases externally, and grab-handles inside (with a bad back, I have trouble rising up and out of the trailer!). I also plan to retrofit some insulation inside, to prevent minor wall-sweating, but not foam. Sorry about that, foamies (just kidding, to do it all over, I would've done it with foam-faced 1/2" plywood- w/1/8" luan interior walls).

• front slope shelf, single spar/brace
• 1/2" oak screwed/glued to walls as doorframe & slope brace
• roof plate mounted on top of walls with steel corner brace; later glue and angle braces

corner braces,hatch roof,bulkhead supported.jpg (44.39 KiB) Viewed 1298 times

cabin roof bracing, no spars.jpg (29.37 KiB) Viewed 1298 times