LumberJocks

Torsion Box Assembly Table/Workbench?

  • Advertise with us

« back to Focus on the Workspace forum

Forum topic by clin posted 05-23-2016 01:57 AM 869 views 0 times favorited 14 replies Add to Favorites Watch
View clin's profile

clin

510 posts in 457 days


05-23-2016 01:57 AM

I’m planning to build a 72” x 36” assembly table and general workbench. I’m thinking of making a torsion box for the top. Probably make it with 3/4” MDF skins and webs. Heavy duty for a torsion box, but this box will most likely take some hammer and mallet blows as well as some clamping pressure. And while heavy sucks while building, the heavier a workbench the better when using it.

I have a pretty good understanding of the physics involved, but wanted to run this by LJ in case someone can explain why I have the following wrong:

I’ve read about many complicated techniques, but it seems to me cutting the internal webs to full length and using half-laps where they cross would be quick and easy. And I’ve seen this is a popular technique.

But, people seem to be gluing these half-laps, and I can’t see why. The only joint that matters is the one between the web and the skin. I don’t even think it is necessary that the webs touch each other where they cross, though of course large spaces would begin to defeat the purposes of the webs.

Does anyone know of a reason the webs need to be firmly attached where they cross? I’m 99.9% certain it doesn’t matter for the completed torsion box, but does it serve a purpose, for example, to ensure a web doesn’t pop up out of position when securing the first skin.

-- Clin


14 replies so far

View squazo's profile

squazo

28 posts in 1106 days


#1 posted 05-23-2016 02:32 AM

i used all screws to make mine offsetting the ribs by 50 percent (only the ones running in the short dimension) so i could get a drill in there to work. I would not recommend mdf for the ribs. too weak lots of sag.

you have worded this very confusingly which is why you probably dont have any responses yet. maybe reword it a little.

also fill it with concrete if you want more weight.

View waho6o9's profile (online now)

waho6o9

7171 posts in 2038 days


#2 posted 05-23-2016 02:45 AM

View crank49's profile

crank49

3980 posts in 2432 days


#3 posted 05-23-2016 02:48 AM

I can’t see having one table serve as assembly and also as a workbench.
Assembly tables are usually short; like 20 to 28 inches.
An assembly table is usually light and often is designed to be mobile.
Torsion box construction makes sense for an assembly table.

Work benches are usually taller; like 32 to 36 inches.
Hand tool work would usually be near the low end of this range where power tool work would be near the top.
Work benches need to be heavy, solid and stationary. Especially for work like hand planing.
Torsion box construction is, by its nature, a light weight way to produce a rigid surface. That is in conflict with the properties desired for a workbench.

I have a bench made of 4 layers of 3/4” plywood topped with a replacable 1/4” hardboard surface. This sucker is heavy and flat and rigid. Works great, and was a lot easier to construct than a torsion box would have been. My assembly table was only 48×36 and two layers of 3/8 mdf with a spacer (grid) made from latice. It was surprisingly rigid and light and only 20” tall. I could fold the legs down and stand it on edge for storage when not being used. Had to leave the asembly table when I sold the house and moved. Plan to build another one like it, though.

-- Michael: Hillary has a long list of accomplishments, though most DAs would refer to them as felonies.

View newwoodbutcher's profile

newwoodbutcher

552 posts in 2311 days


#4 posted 05-23-2016 02:50 AM

I made this 6 years ago 3/4” MDF top bottom and web. I made all short pieces of the web for one direction and full length of the table for the rest I used glue and a pin nailer to put it all together. The legs, aprons and edging all Birch from the big box store. Still perfectly flat and level after a lot of use and abuse. About six months ago I added a 1/4” hard board to the top to cover all the stains and gouges.

-- Ken

View Aj2's profile

Aj2

687 posts in 1259 days


#5 posted 05-23-2016 03:12 AM

I’ve made several assembly tables two were almost a full sheet.I used a nailer to toe nail the ribs.And I think I did use glue.Mostly for my design I needed it to be ridged.I had lots of over hang on the base.
The other thing that’s important is to cut all the ribs on the same set up off the saw.Cut extras.And you might have to cut them twice.If you get lots of crowning ones don’t expect the skins to flatin them out.
The bottom always came out falter then the top skin so be prepared to use the flattest side.

Good luck it’s a great project .

View clin's profile

clin

510 posts in 457 days


#6 posted 05-23-2016 03:16 AM


you have worded this very confusingly which is why you probably dont have any responses yet. maybe reword it a little.

- squazo

What exactly is confusing, other than the 1000 words, instead of a picture? If you can point to something specific I’ll see what I can reword.


I can t see having one table serve as assembly and also as a workbench.

- crank49

I don’t have the luxury of having a table only for assembly. But as I said, it is meant as a general workbench, not specifically a woodworking bench or dedicated assembly table. My interest in a torsion box is for flatness, not light weight.

Again, my question is concerning gluing or not gluing webs to each other when using half laps.

-- Clin

View brtech's profile

brtech

893 posts in 2384 days


#7 posted 05-23-2016 07:53 PM

You are probably right, the glue isn’t important.

I think you are over-engineering the web. You can probably use 1/2” material, and I think ply might be better than MDF for the webs. Hammer blows on a 3/4” top supported by a 1/2” web will fail the MDF top before the web I think.
The reason ply is probably better for the web is that it won’t bow as easily because of the stranding. Solid pine is probably better yet, but more costly and more work.

Torsion boxes are STRONG. The crummiest hollow core door uses cardboard and 1/4” Luan, and they are amazingly tough.

View Rentvent's profile

Rentvent

148 posts in 310 days


#8 posted 05-23-2016 08:18 PM

I found this project on another site and I am planning on doing something similar when I have some time:

The top and bottoms are 1/2 plywood. The sides are 3/4 plywood. The top and sides are joined with pocket screws. The top is laminated with formica.

IMHO, MDF would be too heavy and flimsy.

View clin's profile

clin

510 posts in 457 days


#9 posted 05-23-2016 10:12 PM



You are probably right, the glue isn t important.

I think you are over-engineering the web. You can probably use 1/2” material, and I think ply might be better than MDF for the webs. Hammer blows on a 3/4” top supported by a 1/2” web will fail the MDF top before the web I think.
The reason ply is probably better for the web is that it won t bow as easily because of the stranding. Solid pine is probably better yet, but more costly and more work.

Torsion boxes are STRONG. The crummiest hollow core door uses cardboard and 1/4” Luan, and they are amazingly tough.

- brtech

I agree that 3/4” is much more than needed for the webs. and likely more than needed for the top. It’s just that you can make the top and bottom and all the webs from 2 sheets of material (some webs from both sheets). So having all material the same makes for the best use. May just go with 1/2” all around.

Though heavy is only an issue while building. Once done, heavy is an advantage. Though a 1/2” top will still be a bit over 100 lbs, and I think heavy enough for my purposes.

MDF is more than strong enough for all aspects of this. It’s more dimensionally stable than plywood and certainly more so than solid wood. No question plywood is stronger than MDF, but for webs, the loading is pretty low. Could probably be balsa wood and would work well.

“Cardboard” webbing is not just for cheap stuff, very high-tech composite panels use resin impregnated “cardboard” for webbing.


I found this project on another site and I am planning on doing something similar when I have some time:

The top and bottoms are 1/2 plywood. The sides are 3/4 plywood. The top and sides are joined with pocket screws. The top is laminated with formica.

IMHO, MDF would be too heavy and flimsy.

- Rentvent

That looks to be a variation of the Paulk workbench. Those are specifically design for portability. Not what I’m interested in now, but it is a great design.

MDF is certainly heavy, but it is not too heavy. While I won’t be using this as a traditional wood working bench (have that on the RADAR as well), I do want this bench solid and massive enough it will allow me to do anything reasonable I’d want to do.

As for MDF being flimsy, it obviously doesn’t have the strength of plywood, but in this application it works well. It’s a very popular material for torsion boxes due to its stability and flatness. That’s the surprising thing about torsion boxes, they provide excellent rigidity, from materials that don’t seem that strong.

Similar is airplane wing construction. I build RC model planes as a hobby. A simple balsa wood frame is very floppy (especial in twisting). Once you cover that frame with even something as light as tissue paper, it gets very rigid. It’s a bit different principle than a torsion box, since the skin only supports tension, but it is amazing how well engineered things that must be light weight, can be quite stiff with rather flimsy looking structure and materials.

-- Clin

View Ger21's profile

Ger21

1047 posts in 2592 days


#10 posted 05-23-2016 10:21 PM


and I think ply might be better than MDF for the webs. Hammer blows on a 3/4” top supported by a 1/2” web will fail the MDF top before the web I think.
The reason ply is probably better for the web is that it won t bow as easily because of the stranding. Solid pine is probably better yet, but more costly and more work.

Torsion boxes are STRONG. The crummiest hollow core door uses cardboard and 1/4” Luan, and they are amazingly tough.

It doesn’t really matter what the ribs are made of, as their only job is to hold the top and bottom panels.
The important thing is the glue bond between the ribs and the faces. This is what carries the load, not the ribs themselves. As long as the ribs remain glued to the faces, they cant sag.

-- Gerry, http://www.thecncwoodworker.com/index.html http://www.jointcam.com

View bbasiaga's profile

bbasiaga

754 posts in 1456 days


#11 posted 05-23-2016 11:48 PM

I would plan to put a piece of laminate (formica) on top of the MDF. It soaks up any little glue, oil, finish, water…anything, and it never comes out. Sometimes it swells. It also doesn’t like being banged on. The paper layers will start to delaminate. I had an MDF work bench before, and these were my main complaints. A layer of laminate over the top, and they went away.

-Brian

-- Part of engineering is to know when to put your calculator down and pick up your tools.

View clin's profile

clin

510 posts in 457 days


#12 posted 05-24-2016 05:10 PM


I would plan to put a piece of laminate (formica) on top of the MDF. It soaks up any little glue, oil, finish, water…anything, and it never comes out. Sometimes it swells. It also doesn t like being banged on. The paper layers will start to delaminate. I had an MDF work bench before, and these were my main complaints. A layer of laminate over the top, and they went away.

-Brian

- bbasiaga

I plan to put a piece of hardboard on the top as a replaceable wear surface. Of course arguably you could just apply another layer of laminate on the top once the first wears. Not sure how well contact cement sticks to the finish surface of laminate.

Also, not sure how much variations in glue thickness might affect flatness. Then again, no idea how consistent the thickness of hardboard is either. Probably all these things don’t matter as far as a surface for wood working.

FYI: (and geek warning)

In reading up on torsion boxes, I have seen in several places that claim that the “strength” of the torsion box goes up with the cube of thickness. First off, what they mean by strength is not defined, but I assume they mean resistance to deflection rather than it structurally failing.

In any case, the claim of the strength going up as the cube appears to be wrong. This is true for solid beams, but with structures like torsion boxes and I-beams, this value is between the square and cube of the thickness. In the case of typical torsion boxes that I’ve seen for wood working tops, the proportions put it a lot closer to the square of the thickness.

I ran some calculations for a 4 inch total thickness, using 1/2” material with 6” between webs, and compared this to an 8” thick top using the same materials and spacing. The 8” top was about 5X stiffer (5X larger moment of inertia). This compares to 8X stiffer if in fact it varied with the cube of thickness.

Deflection is proportional to the inverse of the moment of inertia (1 over).

Regardless, even a square of the thickness relationship strongly supports the idea of thicker is better.

Just throwing this out there for the interested geeks.

-- Clin

View HokieKen's profile

HokieKen

1742 posts in 600 days


#13 posted 05-24-2016 06:00 PM

You’re right that the webbing does not need to be glued to itself. However, I think (intuitively, without pondering it for a while) that “loose” half-laps will weaken the resistance to deflection.

Torsion boxes rely on the webbing being rigid and fixed to both the top and bottom skins. If you half-lap one of the web members, you introduce a point at which it’s no longer rigid. Now if you half-lap the adjoining cross piece and fit them together loosely, you still have a gap which is still a point where the member can flex. On the other hand, if the joint is tight, the gaps are filled completely and the weak point is eliminated.

Again, just my initial intuitive thoughts…

-- Kenny, SW VA, Go Hokies!!!

View clin's profile

clin

510 posts in 457 days


#14 posted 05-24-2016 07:11 PM



You re right that the webbing does not need to be glued to itself. However, I think (intuitively, without pondering it for a while) that “loose” half-laps will weaken the resistance to deflection.

Torsion boxes rely on the webbing being rigid and fixed to both the top and bottom skins. If you half-lap one of the web members, you introduce a point at which it s no longer rigid. Now if you half-lap the adjoining cross piece and fit them together loosely, you still have a gap which is still a point where the member can flex. On the other hand, if the joint is tight, the gaps are filled completely and the weak point is eliminated.

Again, just my initial intuitive thoughts…

- HokieKen

Look at versions of these made with alternating web positions, where one row of short webs are placed between the position of the webs in the other row. Here’s an example I found on the internet:

The structure doesn’t depend on continuous spans of the webs.

The webs keep the skins parallel to each other. The main force on the webs is a shearing force, parallel to the skins. This is why the web to skin glue joint is critical. You could randomly place short webs throughout the structure and it would still work well.

Now, to some extent, the web provides basic support just like a thin, solid beam. So if it is not continuous, you will not get this extra support, where there is not web. If we had extremely thin webs, too thin to contribute to the moment of inertia, the total moment of inertia is about 5% less. But you could just as well argue reducing the spacing between webs makes it stronger.

I’m not starting this build for a while, but at this point, I’m going to assemble the core webs dry unless I get information to the contrary based either on structural theory or someone’s direct experience.

As I mentioned in my OP, structurally I don’t think it matters, but there may be some advantage while building it. Though to me, anything like gluing or nailing the web joints, just introduces the possibility of building in an offset preventing the webs from sitting flush against the skin.

But who knows, once I have it dry fit, I may find a reason to add some glue or shoot some brads here and there.

-- Clin

Have your say...

You must be signed in to reply.

DISCLAIMER: Any posts on LJ are posted by individuals acting in their own right and do not necessarily reflect the views of LJ. LJ will not be held liable for the actions of any user.

Latest Projects | Latest Blog Entries | Latest Forum Topics

HomeRefurbers.com