What size roof joists?

[les45]

I used 1X2's on 12" centers for the roof on my weekender build on a 5X8 NT trailer (60" roof span). With 1/4" ply skins that seemed to be plenty sturdy. On my new "pop-up to retro" project, I will have an 80" roof span (out to out). Does anyone have any experience building this wide? Will 1X2's be adequate or do I need to move up to 1X3?

[linuxmanxxx]

As long as your standing them up long ways they will work just fine. Double stack 3/4" blue or pink folks the gaps perfectly for insulation.

Sent from my Pixel 2 XL using Tapatalk

[Tigris99]

depending on what's available may not need to double stack as 1.5" is available in areas that get a winter.

As for the wood matter, 1x2 is plenty you shouldnt be running the length should be running the width and as long as your not planning in standing on the roof it will hold up no problem.

Sent from my SM-G950U using Tapatalk

[linuxmanxxx]

If you don't stand it the long side up and down it will sag I know this from experience. If you glue the sheer length of all the foam with a strong skin like aluminum and Luan you can stand on it. Total glue up adds crazy sheer strength if both sides are done and enclosed like a box.

Sent from my Pixel 2 XL using Tapatalk

[Andrew Herrick]

If you want to run some numbers, there are plenty of load span calculators on the Internet.

Here's the main thing to know: the bending strength of a beam has an approximately squared relationship to the height. To borrow an example from stick-built construction, a 2x6 is only 30% more wood than a 2x4, but it's more than twice as strong.

So, if you want your rafters to carry the load, then I'd make them bigger. Will 1x1-1/2's break when spanning 80 inches? No. But they'll sag eventually. And that doesn't look good and it increases your risk of roof leaks. When I build to 80 inches wide, I use 1x2-1/2's. You only lose an inch of headroom, but wow, is the roof stronger! And you get more room for insulation. If you want more space to use standard electrical boxes, go to 3 inches deep.

One poster mentioned using glue to bond the ceiling and roof skins to the roof rafters. This is a form of torsion box or stressed skin construction, and it transfers the load-carrying capacity of the assembly from the rafters to the skins. Nothing wrong with this, but for most home builders, it's better to size your rafters to carry the load, and then to glue the skins as a belt-and-suspenders approach.

[linuxmanxxx]

I did 1x2 standing vertical did about 12" spans and did cross pieces about same with stacked 3/4" foam stuck together with contact cement and wood all glued tightbond3 nailed and skinned both sides with Luan complete glue of both sides with contact cement with a vent Hood toward the front. Zero sag and very little flex when I walked on it 200 pounds. The glue was 3M 30NF water based loves foam and strong. Glue with that much surface would be stronger than traditional construction any day at half the weight or even less. 2018 materials and innovation collide with 20th century engineering they keep teaching in school.

Sent from my Pixel 2 XL using Tapatalk

[noseoil]

Remember, the strength of materials isn't just the size, it's the combined strength which counts. If you glue, nail & skin a structure from both sides, the resulting structural strength is amazing. It forms a "beam" if it's a single member (look at an "I" beam or a TJI made from wood), or a torsion box if it's an entire system.

Standard roof construction for residential building is normally based on 2' centers & 1/2" plywood (now OSB). The spans determine the depth of a member, 2x4, 2x6, 2x8, etc. Longer spans need a trussed roof system. Floors typically have heavier loads & centers are typically reduced from 24" to 19.2", 16" or even 12" for heavier loads, if depth can't be increased.

I'm just saying that an 80" span would be ok with 1x2's if glue & nails are used for skins on both top & bottom. An 1/8" plywood skin would be fine if centers were kept at 12" or less (10" might be better) & you aren't planning on walking around on the roof. Add a glued-in foam core & it would be really good. There is one other factor to keep in mind, the "creep" factor can come into play here. As wood ages it can sag over time, as was mentioned previously. The way around this is to add a bit of camber or "crown" to the shape. For residential roof construction, we added 1/8" of camber or "crown" to each 10' of truss along the bottom chord (the ceiling rafter), so it would settle back with a load to a flat ceiling. With an 80" span, I would add at least 1/16" to 1/8" of crown to the roof, prior to skinning & glue. Easiest way to do this is to find some 1x2's which are already slightly curved along the long edge & set them all so they have a uniform "crown" running across the width.

This is a long-winded way of saying that a light, strong roof can be built if attention to detail & some common sense are used to make it.

[Andrew Herrick]

In response to the commentary on torsion box/stressed-skin/I-beam structures:

While it's certainly true that a stressed-skin structure is a more efficient and lightweight use of material, I think it CAN BE deceptively difficult. I started building stick-built campers and transitioned to stressed-skin structures, myself.

What takes one sentence to write - "glue the roof to the foam" - MIGHT take a lot longer to put into reality. Can the glue bridge a gap? Are the thermal- and moisture-driven expansion rates of the glue, foam and wood compatible? Will the glue fail if you ever have a leaky roof? If the glue fails, what will the resulting strength of the roof be?

Now, thankfully, teardrop trailers are small and simple enough that Macgyver solutions can work surprisingly well. So don't misunderstand me; I'm not bashing on stressed-skin designs! It's by far the better way. But as one poster mentioned, it's "20th century engineering." If you don't have experience with stressed-skin design, I'd just recommend taking a little more time to learn, design, model and plan. You don't want your roof assembly to delaminate after a few months of use or after sitting all afternoon in direct sun - and yes, that's happened to quite a few people! As the old saying goes, "Measure (research) twice, cut once."

[Andrew Herrick]

Oh, and to add to the advice of crowning the rafters:

Solid advice - if you're culling through 1x2 S2S or S42S stock. If you're ripping boards to width, unless you hand-pick your stock, this will be very difficult. Plus, if you want a reliable bond between the roof skin and the core foam, you either need a strong gap-filling foam or some way to crown the foam, too.

Until I get a $10,000 CNC router, all my roof rafters are ripped on a tablesaw and planed if necessary, so they're straight as an arrow. If someone knows a reliable method to crown a teardrop roof for a stressed-skin design, man, I'd love to hear it!

[noseoil]

Easiest way to cut crowns is to use a router & a pattern. Just make the pattern you need & then cut the number of rafters needed for the build. Screw the pattern to the raw stock & use a flush-cut bit to make the trim. Bandsaw first to get close to the line, then use the router to make a finished cut.

This is the way I did the hatch ribs. It was about a year+ until it was assembled, but fit well enough when it was time to put it together. Same thing will work for the crowns. Plywood, 1X3, poplar, oak, whatever you need.



You can also make a simple glue jig to laminate the spars, but it takes extra work, clamps & thinner strips of wood for the lamination. They would be plenty strong this way (think miniature glu-lam beams) & would be uniform in section & strength. No need to worry about grain run-out with lamination, as cheaper wood can be used to give a strong beam.

[Andrew Herrick]

Easiest way to cut crowns is to use a router & a pattern. Just make the pattern you need & then cut the number of rafters needed for the build. Screw the pattern to the raw stock & use a flush-cut bit to make the trim. Bandsaw first to get close to the line, then use the router to make a finished cut.

This is the way I did the hatch ribs. It was about a year+ until it was assembled, but fit well enough when it was time to put it together. Same thing will work for the crowns. Plywood, 1X3, poplar, oak, whatever you need.



You can also make a simple glue jig to laminate the spars, but it takes extra work, clamps & thinner strips of wood for the lamination. They would be plenty strong this way (think miniature glu-lam beams) & would be uniform in section & strength. No need to worry about grain run-out with lamination, as cheaper wood can be used to give a strong beam.

Looks like you have time to burn!

However ... you could run into issues attempting to screw or nail into the bottom or top of the glue-lam rafter, which will be at least 50% end-grain. If anyone goes this route, follow SOP for fastening into end-grain: predrill and/or use confirmat screws and/or pot the fasteners with glue.

[aggie79]

For the OP, a roof camber or crown will work, but only for a "flat" roof. For a teardrop or canned ham, where the roof transitions to the front or rear walls, you won't be able to "bend" sheet goods around the camber/contour. I guess you could "feather out" the camber/crown to "flat" from the center of the roof to the point where the curve begins (PI/PT in surveying terms) but that too would be tricky.

[Andrew Herrick]

For the OP, a roof camber or crown will work, but only for a "flat" roof. For a teardrop or canned ham, where the roof transitions to the front or rear walls, you won't be able to "bend" sheet goods around the camber/contour. I guess you could "feather out" the camber/crown to "flat" from the center of the roof to the point where the curve begins (PI/PT in surveying terms) but that too would be tricky.

Theoretically, you can compound-bend plywood sheet goods a little ... what's the maximum camber you could use on a teardrop, you think? And more importantly, would it matter? I think a builder could bend a 1/8-inch plywood around a "flat" roof with a 1/4-inch crown ... but would that 1/4-inch crown (roughly 0.10 in/ft slope) be enough to matter for strength and water shedding? That would depend on your roof material, too.

[aggie79]

Theoretically, you can compound-bend plywood sheet goods a little ... what's the maximum camber you could use on a teardrop, you think? And more importantly, would it matter? I think a builder could bend a 1/8-inch plywood around a "flat" roof with a 1/4-inch crown ... but would that 1/4-inch crown (roughly 0.10 in/ft slope) be enough to matter for strength and water shedding? That would depend on your roof material, too.

Flat or with a camber, the Achille's heel of a roof is the roof edge to sidewall transition. Often, where the the edge of the roof material meets the edge of a sidewall material, there is trim that causes water to pond. If the gap between the roof and side wall materials isn't enough of a challenge to seal against water intrusion, the fasteners holding the trim to the substrate penetrate either the outer roof or side wall material causing another potential point of water entry.

[noseoil]

Compound curves with a two-dimensional material would be a bit much to deal with for me (not worth the trouble). I'm thinking a laminated build would have to be the way to go in order to make it work properly. No problem with laying a fiberglass skin on these curves, but plywood is a different matter. You would be building a strip plank type of shape, basically a boat or airplane!