joining legs to a nested torsion box in an outdoor table...

[Matteo Lorenzo]

Hey all.

This seems like it should be so much more simple than I am making it. So perhaps you pros could lend an ear and point out the obvious to a novice? I would be very grateful for anything you can offer.
So, please bear with me. I am going to try and make this as short as I can and semi-intelligent.

I am in the last 30% or so of a build on a Parson's style, rustic farmhouse table intended for the outdoors in Los Angeles.
The kicker is, the table is 14'x33"x29". Add to that: two nesting ~11ft benches and i have myself a quite unwieldy job for a single set of hands. But I have managed so far.


here is the general plan in sketchup:
original size: http://i.imgur.com/Fr1bOeTh.png


http://i.imgur.com/Fr1bOeTl.png


I am using cypress, in the rough, from a local lumber yard out here. It will be weathered with vinegar/steel wool then given a latex paintwash.

On to the confusing bits:

In order to reduce sagging overtime and deflection from big dinners and elbows, I built a 12ft torsion box that will be nested up under the table top and enclosed somehow behind the apron and between the legs. The torsion box dimensions are as follows:


12ft long
21in wide
4.5in deep


constructed of 1/2in mdf skins and webbing and 3/4inch mdf sidewalls. it weighs somewhere in the neighborhood of 100-120lbs.

http://i.imgur.com/AxIJTqCl.jpg


standing up on edge just for the pic's sake

http://i.imgur.com/3OeXTP5l.jpg


I thought I could notch out a section of each leg to accept each of the 4 corners of the torsion box, then use 2 hangar bolts and some reinforcing hardwood (to be secured inside the torsion box on each corner) to secure the legs directly. BUT, because I plan on using haunched M/T to attach the apron members to the legs, I start to run into issues of clearance and usable meat to drive in the bolts. (The legs are 3.75x4.75, laminated cypress)


here is my attempt to visualize how that might work in sketchup:
orginal sizes:http://imgur.com/a/ez1aV


all angles are from above the leg/apron
http://i.imgur.com/Ex2GrYvl.png
http://i.imgur.com/IuJceoRl.png
http://i.imgur.com/fART2gHl.png


you may be scratching your head at the small tenon with stub dowels. That is because I came up with a setup that I think is probably overcomplicated and begging for disaster since the measurements/clearances dont have enough play to allow for mistakes.


here is a grab of the torsion box (represented by just a corner segment showing the inside hardwood pieces and bolts.
each bolt would clear the tenons and hanches/stub dowels by 1/4 in or less... which makes me really hesitant to call this idea a success.


original size:
http://i.imgur.com/f7s53bUh.png


http://i.imgur.com/f7s53bUl.png




SO.... I feel like I am painting myself into a disastrous corner and I am hoping that there is some very simple solution to all of this. Perhaps I can find some sort of hardware that wont be too obvious and stick out too much to attach directly to the legs without removing material from the stock? Maybe I can notch out the legs but not fasten the torsion box to them, instead allow it to "float" within the confines of the apron (assuming I build some sort of cage or shelf to hold it in place on the apron members. Maybe I can change the orientation of those hangar bolts, or use a single one either vertically into and along the length of the leg ... or counterbored diagonally into the leg member?


On top of all that, once I nest in the torsion box, how will i be able to attach the table top to the apron effectively? the torsion box will be in the way. I suppose I could fasten the long rails with pocketscrews on the inside, drop in the torsion box, attach the legs to the box and apron, then finish the short apron members with pocket screws on the outside... but that feels sloppy.


I have no clue if any of this makes sense. It confuses the hell out of me, so feel free to ask for clarification... if you made it this far, that is.


Thank you all very much for any and all help you can offer. I really appreciate it! and again, im sorry for the length and confusing description.

[Jamie Buxton]

You're putting MDF out in the rain? That's going to be a disaster.

[Doug Garson]

You're putting MDF out in the rain? That's going to be a disaster.
+1 on the disaster

[Jerry Miner]

You're putting MDF out in the rain? That's going to be a disaster.

My sentiments exactly. MDF would be a disaster outdoors.

[Jerry Miner]

As for the leg-to-apron connection, I think you are over-thinking it. Ignore the torsion box and connect with a standard M & T joint.

Then build a torsion box with exterior materials---or better, a steel frame--- and connect it to the aprons. You need this for anti-sag, but not so much for leg connection. The apron joint will do that.

If you DO use a torsion box, how does the moisture get out??

Another point: you show a 9" breadboard end. How is it supported? 8" tenons?

[Matteo Lorenzo]

You're putting MDF out in the rain? That's going to be a disaster.


+1 on the disaster


My sentiments exactly. MDF would be a disaster outdoors.

Thanks for the heads up on that, incredibly helpful. I was sure you could seal it against the elements enough to make it useable with varnish or paint, but I guess it would just drink it up and swell. No matter - I will chop it up into an assembly table (which i desperately need and an infeed/outfeed for my jointer, which i desperately need. And that means I can rethink the issue and build it smarter this time. Silver linings.

So - that in mind, is there a good alternative to making a light and structurally tough torsion box? I think i'd prefer to go lighter than that damned MDF. thoughts?

thanks

[Matteo Lorenzo]

As for the leg-to-apron connection, I think you are over-thinking it. Ignore the torsion box and connect with a standard M & T joint.

Then build a torsion box with exterior materials---or better, a steel frame--- and connect it to the aprons. You need this for anti-sag, but not so much for leg connection. The apron joint will do that.

If you DO use a torsion box, how does the moisture get out??

Another point: you show a 9" breadboard end. How is it supported? 8" tenons?

looks like ive got a lot to learn.

I dont even know where to begin on building a steel frame, which is why I opted for a torsion box.

if i build a torsion box with another material and seal it properly, would moisture be an issue? It is Los Angeles and dry as a bone, but there is this looming El Nino to consider.

as for the breadboard, I have been meaning to adjust the plans to rest the bulk of the breadboard over the apron. would i still need to be concerned about tenon length?
thanks for taking the time.

[Dave Richards]

Why not make just add a long stretcher down the center and some crossing ones with halving joints? Sort of build a ladder like this (https://flic.kr/p/Aq6PuU). Use M&T joints, dowels, loose tenons, etc. to join the stretchers to the inside of the aprons.

If it's not too late, I'd made those aprons thicker, too.

I'm curious about your haunched tenons, too. What is the benefit in this application to putting the haunch on the bottom? Typically the haunch would be on top if it's used at all. If you're going to put it on the bottom, why not just make the full length of the tenon longer and gain some additional gluing surface in the mortise?

[lowell holmes]

Quarter sawn white oak would be my choice for outdoors. It is one of the premier outdoor woods.

[Jamie Buxton]

Unless your patio is extraordinarily flat, a torsion box is a bad idea. One of the primary characteristics of a torsion box is that it does not twist. (That's why it is called a torsion box. It resists torsion.) If you put it on a non-flat surface, it will rock back and forth on the two highest spots underneath it. If you've ever sat at a teeter-tottering dining table, you know it is not comfortable. What you want for a big table is that it should be willing to twist just a bit, so that it gets all four legs contacting the floor. That is, while you're designing the structure of this table, you should be focusing on the sag issue, but not worry very much about the twist issue. To combat sag, I'd make the aprons thicker than your drawing -- 1 1/2" or so. I might even add a third "apron" running down the middle of the table. You can evaluate the sag of your design by mocking it up: support your proposed aprons at the ends, and see how far they sag when you load them.

[Matteo Lorenzo]

Why not make just add a long stretcher down the center and some crossing ones with halving joints? Sort of build a ladder like this (https://flic.kr/p/Aq6PuU). Use M&T joints, dowels, loose tenons, etc. to join the stretchers to the inside of the aprons.

If it's not too late, I'd made those aprons thicker, too.

I'm curious about your haunched tenons, too. What is the benefit in this application to putting the haunch on the bottom? Typically the haunch would be on top if it's used at all. If you're going to put it on the bottom, why not just make the full length of the tenon longer and gain some additional gluing surface in the mortise?

thats an interesting idea... its sort of making the table a torsion box without skins. its exactly how i constructed the benches, but without a center runner.

i have wondered if I shouldnt double up the apron members - so i think i will end up doing that and just go a full 2" thickness or close and make a double tenon on the ends.

as for the haunches, they were strictly for allowing the bolts to go through untouched leg meat. mmm

great ideas dave, thank you

[Matteo Lorenzo]

Unless your patio is extraordinarily flat, a torsion box is a bad idea.

worse! its out on the ground in the backyard. actually, in thinking about that, it could be better out there with a teeter totter, at least you coul shove a pinecone under there or something... but i digress ;)


To combat sag, I'd make the aprons thicker than your drawing -- 1 1/2" or so. I might even add a third "apron" running down the middle of the table. You can evaluate the sag of your design by mocking it up: support your proposed aprons at the ends, and see how far they sag when you load them.

yeah, I think that I have convinced myself that 4/4 stock is not beefy enough for rails on this beast. I will likely double them up and build something like Dave Richards mocked up in his post, sort of an open, skinless torsion box. that will solve the problem of allowing twist while combatting sag. and I probably can just do that all with cheaper fir sticks.

thanks for the insight!

[Don Orr]

Great advice so far-ditch the MDF, Cypress is excellent outdoors. My suggestion is in line with others to double ALL the apron thicknesses and add 2 stretchers in between the side aprons (double thick also) which could then extend through the end aprons to add some support to the breadboard ends of the top. You can add some design elements to the "tusk" tenons for appearance and clearance sake-like an ogee for example, or just a straight angle cut. Please let us know how this ends up (with pictures).

[Matteo Lorenzo]

Great advice so far....

yeah, im brand new to the forum and im floored at how knowledgeable and respectful you all are to an amateur like myself.
i sort of got my arse handed to me with the MDF torsion box thing, but its strangely relieving and not all is lost because now I dont have to build another one for my shop. I will be cannibalizing this one into a few small ones

great idea on the "tusks"! I think I could do that and make it look like it belongs with a simple angle or two.

I am going to work today on redesigning the table structure and I'll post it up later to see how it tests with the pros. I should also start thinking about learning to weld :)

[Doug Garson]

One more suggestion if it's going to sit on the grass. You need to make sure moisture from the ground doesn't wick up into the bottom of the legs and cause rot. One way to prevent this is to seal the bottom of the legs with epoxy. You could lay the table upside down before applying the finish, make a dam with tape around each of the legs, pour in say 1/2" of epoxy and let it soak in. You may end up with different thicknesses of epoxy in some legs depending on how porous the wood is, you can level this out with a file or sander after the epoxy has fully cured.

[Matteo Lorenzo]

One more suggestion if it's going to sit on the grass. You need to make sure moisture from the ground doesn't wick up into the bottom of the legs and cause rot. One way to prevent this is to seal the bottom of the legs with epoxy. You could lay the table upside down before applying the finish, make a dam with tape around each of the legs, pour in say 1/2" of epoxy and let it soak in. You may end up with different thicknesses of epoxy in some legs depending on how porous the wood is, you can level this out with a file or sander after the epoxy has fully cured.


this is LA, we dont have grass :)
thats a great suggestion. I coated the ends of the legs with a thick wax, but I figured that wouldnt last, so I will check into the epoxy solution. presumably, you are talking about an epoxy like this one, correct?: http://www.amazon.com/Great-Planes-30-Minute-Formula-GPMR6047/dp/B001BHI4JU/ref=sr_1_6?ie=UTF8&qid=1446661517&sr=8-6&keywords=epoxy

Are there other methods to coating table leg ends that is durable?

[Doug Garson]

Not familiar with that particular brand but yes that's the idea. I have used West System, System Three and Cold Cure epoxies. Don't think you will find a more durable coating. Another approach is to add a non porous cap or foot to the bottom of the leg. I've used an old uhmw plastic cutting board to make replacement feet for some aluminum patio chairs. They were attached with screws into a scrap of cedar wedged into the hollow aluminum extrusion. That could work but I think sealing with epoxy is a better solution. You might want to rough up the bottom of the leg or drill one or more holes in it so the epoxy can key into the wood (probably overkill but can't hurt).

[Matteo Lorenzo]

Why not make just add a long stretcher down the center and some crossing ones with halving joints? Sort of build a ladder like this (https://flic.kr/p/Aq6PuU). Use M&T joints, dowels, loose tenons, etc. to join the stretchers to the inside of the aprons.

If it's not too late, I'd made those aprons thicker, too.

I'm curious about your haunched tenons, too. What is the benefit in this application to putting the haunch on the bottom? Typically the haunch would be on top if it's used at all. If you're going to put it on the bottom, why not just make the full length of the tenon longer and gain some additional gluing surface in the mortise?


Unless your patio is extraordinarily flat, a torsion box is a bad idea. One of the primary characteristics of a torsion box is that it does not twist. (That's why it is called a torsion box. It resists torsion.) If you put it on a non-flat surface, it will rock back and forth on the two highest spots underneath it. If you've ever sat at a teeter-tottering dining table, you know it is not comfortable. What you want for a big table is that it should be willing to twist just a bit, so that it gets all four legs contacting the floor. That is, while you're designing the structure of this table, you should be focusing on the sag issue, but not worry very much about the twist issue. To combat sag, I'd make the aprons thicker than your drawing -- 1 1/2" or so. I might even add a third "apron" running down the middle of the table. You can evaluate the sag of your design by mocking it up: support your proposed aprons at the ends, and see how far they sag when you load them.


Great advice so far-ditch the MDF, Cypress is excellent outdoors. My suggestion is in line with others to double ALL the apron thicknesses and add 2 stretchers in between the side aprons (double thick also) which could then extend through the end aprons to add some support to the breadboard ends of the top. You can add some design elements to the "tusk" tenons for appearance and clearance sake-like an ogee for example, or just a straight angle cut. Please let us know how this ends up (with pictures).


One more suggestion if it's going to sit on the grass. You need to make sure moisture from the ground doesn't wick up into the bottom of the legs and cause rot. One way to prevent this is to seal the bottom of the legs with epoxy. You could lay the table upside down before applying the finish, make a dam with tape around each of the legs, pour in say 1/2" of epoxy and let it soak in. You may end up with different thicknesses of epoxy in some legs depending on how porous the wood is, you can level this out with a file or sander after the epoxy has fully cured.


Thanks for everyones input.

I have decided to go the manual approach and make an unskined torsion box out of cypress or fir with 2 stretchers at around 2in thick and cross members half-lapped down the lenght of the table. they will be connected by blind dowels that mate into the side apron members to the tune of about 1 inch. I havent quite figured out how to do the stretchers connection into the short apron members, but probably blind dowels since I dont want the apron to appear "broken through". I also havent decided on extending the stretchers out to support the breadboard yet.

as you can see the side/front/back apron members will each be around 2 in thick now with double tenons to mate into double mortices in the legs. it will be a lot of work, but the way I see it, its great practice.

any comments as to the complexity? am I going overboard? It feels like it will be a solid table but I really have no frame of reference. lastly - any reason or need to throw in corner braces between the front/side aprons to anchor into the legs?

large size:
http://i.imgur.com/qclFa04h.png
http://i.imgur.com/qclFa04l.png

larger:
http://i.imgur.com/Yv2hS8xh.jpg
http://i.imgur.com/Yv2hS8xl.jpg

[Dave Richards]

I expect it'll work but I also think you could park a car on it. :)

On another note, why are your tenons inside out?

[Matteo Lorenzo]

How do you mean? in the sketchup model? as in, flipped normals?

im not sure why its doing that different color.. but if thats what you mean, i promise I wont cut my tenons inside out

[Matteo Lorenzo]

unless you mean that it looks like a single tenon was been inverted... in which case - no... they are intended to be double tenons

like this
http://www.bois.com/media/214824/assemblage-double1.jpg

[Jerry Miner]

This looks like a big improvement over the original design. I think you could lose at least half of those cross-pieces. All you really need them to do is hold the aprons straight. The real structural issue is sag, and they don't help with that---and every half-lap joint carries a risk of a loose fit and increased sag. I certainly would load-test an apron. Twelve feet is a big span.

(p.s. there is no such thing as a "skinless torsion box"--- the skin is a critical component. But I still wouldn't put one under this table. I WOULD consider sandwiching a 1/4" steel plate between your apron halves. That would solve the sag issue, and you could eliminate the center aprons/stretchers

[Matteo Lorenzo]

This looks like a big improvement over the original design. I think you could lose at least half of those cross-pieces. All you really need them to do is hold the aprons straight. The real structural issue is sag, and they don't help with that---and every half-lap joint carries a risk of a loose fit and increased sag. I certainly would load-test an apron. Twelve feet is a big span.

(p.s. there is no such thing as a "skinless torsion box"--- the skin is a critical component. But I still wouldn't put one under this table. I WOULD consider sandwiching a 1/4" steel plate between your apron halves. That would solve the sag issue, and you could eliminate the center aprons/stretchers


Thanks Jerry.

I have simplified the cross members down to 5. I wouldnt have thought to sandwich steel between the apron halves, though it makes a lot of sense. I just dont know much about metal or where to source 12ft sections of it. I'll have a look and if it is cost effective and easy to get, ill take a stab. Otherwise I think I could live with doing the extra stretchers, assuming that it is okay at load. ill be sure to test after tomorrow when i laminate the main apron members.

is this something that is done commonly? a straight sandwiched layer of steel at the same width as its surrounding wooden counterparts? or are you talking routing a slot into both members and encasing a smaller flat of metal along the length and laminating the pieces together?

[Dave Richards]

I did just mean the reversed normals.

[Kent Adams]

I'm not much of an expert on anything, but why do a torsion box at all? The idea behind a torsion box is to keep the surface flat, but I don't see how that's possible if its outside. If you use any wood outside, its going to swell and shrink with the weather. I would think a torsion box would be too rigid to allow this movement and in short order, you'll have all sorts of problems with the expansion of the wood in the box. I can see it's fun and a challenge to build these boxes, but I'm not sure it matters which wood you use if its outside, it will all move. What about using something like Alaskan Yellow Cedar, 8/4 with 1 1/2" aprons? Or perhaps Ipe and ditch the whole torsion box. With the two woods I mentioned, you'll be long gone and dead before that thing rots. I'd put a coat of waterlox or something on the table. I've used waterlox on outside Versailles Orange planters and they still look new, years later. The wood I used was 12/4 locust, another rot resistant species. Good luck with whatever you build.

[Matteo Lorenzo]

So, I was thinking on Jerry's idea of sandwiching some steel plate between the long apron members to add rigidity and reduce the need for central stretchers.

I wasnt sure how I could cleanly put a flat bar inside the member, so i came up with this idea, maybe you guys can tell me if it makes sense...

I sourced (but not yet procured) some 20ft steel angle near my lumber provider that I think may work.

I figure that anything around 2x2 @ .25" thick up to 4x4@.25" thick would be sufficient to overkill. It isnt too expensive, but it will add a little weight (nowhere near my torsion box's addition!)

My thought is this:

instead of sandwiching the steel between the beams, I could buy a 20ft section, split it in half to 10ft and bolt it sufficiently to the inner-face of the inside apron member 4/4 slat PRIOR to laminating them to 8/4.
I can countersink or counterbore the holes to allow the bolt head to sit flush, allowing me to laminate the outer facing slat and have the angle permanently attached.

check it below and let me know if you think this is logical

http://i.imgur.com/tGW7Vjf.png

http://i.imgur.com/SisUSxe.png

[Jerry Miner]

...is this something that is done commonly?

Not too common in furniture (but then neither is a 14' table with no center support). Fairly common in building construction. Called a "flitch plate"---used to add load-bearing capacity to a beam.

Your angle-iron idea is a step in the right direction, but for the same amount of steel, you will get more anti-sag with a 1/4 x 4" flat bar. No welding required, just drill and screw through from the inside:

324798

[Matteo Lorenzo]

Not too common in furniture (but then neither is a 14' table with no center support). Fairly common in building construction. Called a "flitch plate"---used to add load-bearing capacity to a beam.

Your angle-iron idea is a step in the right direction, but for the same amount of steel, you will get more anti-sag with a 1/4 x 4" flat bar. No welding required, just drill and screw through from the inside:

324798

really cool, thanks for the illustration and info Jerry. ill read up on flitch plating for sure.

one or two quick things... I guess that I could conceal it by removing .25 with a couple passes over the dado stack, but am I correct in assuming that it would need some sort of adhesion? epoxy or something to make it a true lamination? or does that matter?

and since the facility i found locally that sells this stuff only sells in 20ft spans, would it be considered logical and safe to go with 10ft of reinforcement on either side instead of the full 12'1" that the beam spans?

[lowell holmes]

I would probably use a piece of 1/2" exterior plywood. We used to make long beams in residential construction with two 2x12 boards with 1/2" plywood sandwiched. That gave us 3 1/2" in width, which fits standard 4" board widths and it is strong.

[Howard Acheson]

>>>> Cypress is excellent outdoors.

That's true for old growth cypress which is almost impossible to find now days. New growth cam rot if subjected to constant out door elements.

To be safe, I suggest you figure out a way to keep it under cover when not being used or have a weather-proof cover made that will protect the table and benches.

The biggest entry point for water and moisture is through the end grain of the legs that are in contact with the ground.

[Jamie Buxton]

I would probably use a piece of 1/2" exterior plywood. We used to make long beams in residential construction with two 2x12 boards with 1/2" plywood sandwiched. That gave us 3 1/2" in width, which fits standard 4" board widths and it is strong.

You're missing the point, aren't you? The steel flitch plate is for strength, not thickness. Steel is way stiffer than plywood.

[Jamie Buxton]

...and since the facility i found locally that sells this stuff only sells in 20ft spans, would it be considered logical and safe to go with 10ft of reinforcement on either side instead of the full 12'1" that the beam spans?...

Yes, you should be okay. In fact, there might be a little advantage in holding the steel short of the beam ends. The steel may rust (if it ever rains in CA again). If the steel extends over the legs, rust stains might drip down the leg faces. But if the steel stops short of the legs, any rust drips would fall to the ground.

[lowell holmes]

You're missing the point, aren't you? The steel flitch plate is for strength, not thickness. Steel is way stiffer than plywood.

Really, I'm not missing the point. I just happen to think (from experience) that the beam I suggested would do the job without the corrosion of steel.

The steel will work and that's probably what he should use. Be prepared for rust though.

Has anyone calculated the loads on the beam in order to size it. That's where you should start. There is a book,by Harry Parker, SIMPLIFIED ENGINEERING FOR ARCHITECTS AND BUILDERS, available from Amazon. Spend a little time in that book, It is available from Amazon.

[Matteo Lorenzo]

I would probably use a piece of 1/2" exterior plywood. We used to make long beams in residential construction with two 2x12 boards with 1/2" plywood sandwiched. That gave us 3 1/2" in width, which fits standard 4" board widths and it is strong.


Yes, you should be okay. In fact, there might be a little advantage in holding the steel short of the beam ends. The steel may rust (if it ever rains in CA again). If the steel extends over the legs, rust stains might drip down the leg faces. But if the steel stops short of the legs, any rust drips would fall to the ground.


interesting thought. I was reading about how flitch-plating was less optimal these days due to the fact that its more cost effective to sandwich marine ply which gives a similar strength and costs less ( and you can drive nails in)
but for my application, I think steel is going to be the right route to take. thanks for the suggestion though! I did a cursory calculation on the sagulator and came up with an acceptable deflection result even without steel at a nominal load. I honestly think the steel is overkill, but I am a fan of overkilling it when underkilling it is a possibility.


And thanks Jamie, thats kind of what I was thinking ... and I honestly think that 10 feet is way more than sufficient, but ive never done anything like this so ... have to ask!

Wouldnt it be smart to coat the steel in some sort of krylon spray inhibitor or something prior to implanting it?

thanks guys

[Matteo Lorenzo]

Thanks howie - this is definitely a mixed bag of heart and sapwood - probably all newgrowth. I am going to coat the leg ends with epoxy as suggested in an earlier post, give the whole thing a few good coats of spar and I will also instruct the customer to keep it covered with a tarp. I still like to reiterate that this will be in dry dry Los Angeles - and while we do have a stormy winter ahead, its rare and I take some solace in that fact.

[Jamie Buxton]

...Wouldnt it be smart to coat the steel in some sort of krylon spray inhibitor or something prior to implanting it?...

Yeah, trying to coat the steel would be a good idea. There are anti-corrosion paints, like Rustoleum. There's also truck-bed coating. Ugly, but strong, and it sticks to steel very well. You can buy spray cans of the stuff.

[Jerry Miner]

Yes, coating would be good. Any good metal primer.

No adhesion necessary. Use plenty of screws. As a non-engineer, I would use something like a pair of screws every 12" or so.

10' length should be fine.

I'm surprised to hear that marine ply has a similar strength to plate steel. If you believe it, then use the ply.

[lowell holmes]

Yes, coating would be good. Any good metal primer.

I'm surprised to hear that marine ply has a similar strength to plate steel. If you believe it, then use the ply.

Obviously marine ply does not have the same strength as steel. However, double 2X12 or for that matter 2X10 with 1/2" plywood between the 2X's has been used a long time and the resulting beam is strong. The beam with steel is stronger, but if the wood filler is strong enough, it is easier to use. It will rot, but it does not rust.

It also is a beam design that has been around a long time.

[Matteo Lorenzo]

I just wanted to follow up and thank everyone involved in this thread for all of your insight and great ideas.

I managed to finish the table without incident, delivered and installed today to a happy customer.

For what its worth, I ended up going with the flitch-plated aprons that were doubled in thickness and connected to the legs with pegged double M/T. The connection is very rigid in the end and the table is really heavy and solid.

For the flitch-plated apron side members, I routed an 1/8th pocket in each 4/4 piece of cypress and sandwiched the steel into that slot prior to glue/screw. I also used a fair bit of epoxy to really seat it in there.
There is zero (noticeable) sag and I reckon that table could support a LOT of weight... the top will surely fail before the frame if there is a neighborhood dance party on it.

so, again, thanks so much for everyone's participation. it was absolutely priceless to me.

as promised, some snaps...

resident butterfly that checked my work daily
resting on a flitch-plated apron member after joining the two slats. 4in X .25in x 10ft plates embedded in both side apron members

325542

double M/T connections to leg stocks. you can see the steel plate here as well

325543

angle showing the joined apron from leg to leg and no sign of sag. this was loadless, but I have put quite a bit of weight on the top to test and was not able to create any sag.
it was probably overkill, in the end, but id rather that than underkill!
325544

and the finished piece at the new owners house

325545

325546

[Glen Flint]

Matteo... I've been losing sleep over how I'm going to build a torsion box table for a 10ft dining room table.... the usual method for a flat work surface in a workshop is not going to work. Nathan Day Design has accomplished this, but I'm not sure how they affixed the legs to the torsion box. In my internet searches, I stumbled apon this thread. I actually kind of like your first idea, with a nested nook for the torsion box to snugly reside. That was my working plan and I still might go with that.

[Lee Schierer]

Great advice so far-ditch the MDF, Cypress is excellent outdoors. My suggestion is in line with others to double ALL the apron thicknesses and add 2 stretchers in between the side aprons (double thick also) which could then extend through the end aprons to add some support to the breadboard ends of the top. You can add some design elements to the "tusk" tenons for appearance and clearance sake-like an ogee for example, or just a straight angle cut. Please let us know how this ends up (with pictures).

If you double the thickness of the stretchers, it has less effect than adding 1" to the height of the stretchers. The 1" of height will also add less weight to the table than doubling the width of the stretchers.

[Thomas McCurnin]

OK, since its already more or less built, I'll have a contrary opinion. It may fail, but oh well, I've never learned anything from doing it right, only by making mistakes, so let's carry this project through as is and make it work. Call me an optomist.

I'd just attach your legs in the corner in one of the hollow spaces. I'd attach them with construction screws from the aprons and shim them out to meet the cross pieces in the torsion box and add some more constructions screws from the interior. Then I'd brace the legs underneath with some more 3/4 wood, stupid steel L brackets or small shelf brackets. Glue the legs with epoxy, which probably won't help much, but what's the harm?

Lower mortise and tenon stretchers would be nice, but for dining that probably wouldn't work. Maybe an angled stretcher with an angled shoulder from the leg up to the bottom of the torsion box, or heck, go simple and add a 3/4 x 3 x 18" angled stretcher screwed into the leg then screwed into the underside of the torsion box apron. OK, the table structure is done.

Now let's figure out how to make the wood work. I'm in Los Angeles too (PM me if you want to get together) and I would seal the MDF both sides with multiple coats of a worktop varnish or polyurethane. Several coats, like minimum 4 (it will soak up the first two coats). The more the merrier. Then I would add a paint, go with farmhouse green, that would be nice, again several coats. You might have to use rattle cans for the underside of the torsion box for the varnish and paint.

Add your top, seal it with varnish, and seal the legs with multiple coats of varnish as well.

Buy a 20mil custom cover for this rascal and keep it covered.

My guess is you'll get a decade out of this.

[Jim Becker]

Matteo... I've been losing sleep over how I'm going to build a torsion box table for a 10ft dining room table.... the usual method for a flat work surface in a workshop is not going to work. Nathan Day Design has accomplished this, but I'm not sure how they affixed the legs to the torsion box. In my internet searches, I stumbled apon this thread. I actually kind of like your first idea, with a nested nook for the torsion box to snugly reside. That was my working plan and I still might go with that.
What you need to do is to plan for the legs in the design of the torsion box so you have structure available to provide the support needed. For a 10' table, where you put them is going to come into play, too as it will influence the internal structure so that it's in line with the legs for the same reason a "regular" table often employs aprons between the legs. If the legs have a simple rectangular/square top section, you can build in pockets for them to slip into snugly which means it can also be easy to knock down for moving in the future. Etc. Think it through.

[James Jayko]

I read 38% of the original post and my only thought is “this is too complicated by half,” and “outdoor MDF is a disaster.”

[Thomas McCurnin]

I'm not the OP, but one guess is that he wanted the look of a thick 3-4" slab without having to pay for the lumber or fuss with the weight. I have to admit that the torsion box is beautiful, but too complicated for my tastes and yeah MDF outside would be a problem.

I think I'd just put kiln dried straight grain Douglas Fir 2x4s face to face and skin that monster with a decent hardwood. Actually, for an outdoor table, there is nothing wrong with Douglas Fir and 4x4 legs, like an outdoor workbench. Heck, its just an outdoor table, not an heirloom piece.

But since its already built, rather than second guess his methods, I wanted to see if he could make it work as designed.