Can I get a better definition of a torsion box?

Alright so I am ordering the new xcarve on Tuesday (first time xcarve owner) and I quickly realized that I will need a table for this sucker. So I did some research on here, found a nice thread full of table builds and kept getting the word torsion box thrown around. From what I can gather its just a table top that has extra bracing? I am looking for the simplest table build. I don’t have much experience woodworking so I want to keep it simple. My plan is to build a table top frame out of 2x4’s and use 2x4s as feet. If I got the torsion box correct, I will throw some braces on the table top. And use 3/4 mdf as the surface.

Here is another question. The nex Xcarve has the Xcontroller, and also has the option for the sideboard kit.

How many people here plan to use it? Or should I mount it below? I only ask so that I know if I should add the extra length to the table or not.

when you build a table top you will usually do a big slab surface of some type. Either get some thick, heavy stuff that won’t sag like thick lumber (like most workbench tops) or you use thinner stock and add braces under it to prevent sag and flex (like most counter tops where the cabinets act as braces). a torsion box is basically made of a thinner top and bottom pieces and has structural ribs in between. This enables the top to be both strong, inexpensive and light at the sacrifice of time. Ikea uses lots of torsion boxes in their designs because they can use exceedingly thin surfaces that are ecologically conscious, light and strong. Many of their side tables are basically a cardboard honeycomb with very thin veneer around it (just a few mm)

for my torsion box, I have top and bottom surfaces of 1/2" mdf that are around 48" square, I put a set of strips of 1/2" mdf sandwiched in the perimeter of the surfaces and then cut cross members that cross each other every few inches in between the perimeter pieces that create a grid inside. The thickness (height) of these middle pieces will determine how strong the box is overall, I eventually settled on around 2", though thicker is stronger.

Theoretically you could skip the bottom surface but in my experience it really adds lots of rigidity. Some people also fill the voids in a torsion box with foam. this only marginally improves rigidity but does wonders to reduce resonant frequencies of the box. I elected not to fill voids and have had no issues with resonances (they won’t impact carves, but just add to the noise)

this video should give you a good primer:

This thread should give you lots of ideas as well (including pictures of various torsion boxes in various stages of assembly): X-Carve Workspace Showcase

To your other question; if I had the space and hadn’t already built my station I would probably consider it. The design of my station has a rack under the x-carve and I’m just going to put the x-controller on a pull-out shelf when I get it… I’ve only been waiting since May…

1 Like


Depending on where you live and the climate that your machine is going to be stored in dictates how you proceed in your build. If your x-carve is going to live somewhere humid now is the time to plan ahead and avoid a lot of the aggravations I had.

I’m a woodworker, I’ve been doing it for 15 years now and I am starting to get good at it. I built my machine first and thought I was going to lift it from under a assembly table I have when I wanted to use it and boy was that a mistake. So I quickly figured out this bad boy was going to need a home of its own. If you look up my previous posts you will find what I wound up building for my machine.

Because of the way the stock carve is put together unless your machine is going to live in air conditioning full time or you live in the desert you are going to have problems with the spoil board not being level. The best setup I have seen so far on this forum and I’m sorry I don’t remember the name of the guy who built it but it wasn’t hard to find, involved no spoil board at all, the Y Axis plates were screwed to blocks that were in turn screwed straight to the torsion box.

The reasons I think this design was brilliant are:

When I was done building my torsion box table I checked it with a Rockler Aluminum straight edge and feeler gauges and at the lowest point I was 5000 thousand, which over 4’ is pretty darn good. I recently checked it again and it is still 5000 thousand at its lowest, hasn’t moved at all.

After 6 months of my stock x-carve being on top of the torsion box the MDF spoil board had warped as much as 1/8" in places prompting me to spend a whole bunch of money I didn’t need to upgrading my setup to get it back level again, I really wish I would have seen that guys setup first I would have thrown away the base completely.

I have followed Mark Spagnolo (the Wood Whisperer) since 2007 and he has never steered me wrong, my entire table setup is based loosely on his torsion box assembly table. Watch the video in Ryan’s post and mod to fit your needs. Mark has redone his a couple of times so you might want to watch all of the videos he has on it. You don’t need 2X, just MDF but LDF (light density fiberboard) would be better. The trick with torsion boxes is that they will sag if you don’t support them properly across the length on the bottom. Maybe its just my environment but you can’t just have a wide open center, there needs to be some cross support. If you have more questions on the table give me a shout and I will help you out.


P.S. Once you get your machine setup you are going to need to tram your setup, if you have a friend that is a machinist invite him over for drinks, he/she can bring their indicator and you can find out where you are hi and low.

Gah all this talk is really scaring me into purchasing my X-Carve, I was under the assumption that If I could build it and place it on a level surface I would be good to go. I will continue to do my research and see what I can find. Thank you for the info!

@KaseyHellawell don’t get put off by all the talk of the need for torsion boxes. There are many many folk who use the x carve with huge success on little more than a table.

Yes, a torsion box is more stable than a table. But unless your intending to make really accurate things and you also have the time, inclination, tools and the space to make one, I’d suggest not bothering.

Find an existing table that is reasonably solid. Screw or glue a nice thick piece of plywood to the top of it if you’ve any doubts about its stability and you’ll be good to go.

Yes, there are some very effective improvements that can be made to the “old” x carve design that will improve rigidity and if you’re looking for improvements to your machine, I’d be doing these before working on a fancy torsion box. If you’re buying one of the “new” machines, you’re already good to go.

Please note, I’m not criticising or dissing the work done by others - every bit of stability is worth the effort. But for many XCarvers, I suspect this level of accuracy isn’t really a good “bang for buck” decision. I know this might incite a storm of protests, but I suspect for those on a limited budget, spending a couple of hundred bucks on something like VCarve Pro and a few good quality bits, would possibly be a better investment.

My opinions only.

1 Like

Don’t be scared! Would you probably be fine with a flat, level table if you are doing basic things with your router that don’t require fine tolerances? Probably. But don’t let the torsion tabletop thing freak you out.

My definition (and anyone can chime in on this) is basically a sandwich. Flat material on the bottom and top and stiffeners in between. I used 3/4" MDF cut to 44"x44" for both top and bottom and used 1x’s trimmed to make them all equally sized as the stiffeners in between the MDF pieces. I then went an extra bit and filled the cavities with spray foam to minimize any hollow sound or vibration. A picture is worth a thousand words:


This photo is obviously before I screwed the top on. I then added doubled-up 2x4 legs, a bottom shelf, some pegboard sides, a laptop drawer, a 1/4" thick rubber mat surface, a drawer unit from IKEA and leveling casters to make it all pretty. There’s also a little hidden shelf under the back edge for the electronic components to keep them out of the way.

You can do it! And the support here in the forums is second to none.



That looks really good! And looks like something I could make! What did you use inside as braces? 1x2s? Also did the spray foam add much weight? Thank you guys so much for all of the detailed information! I am not as worried now after that.

yes sir thats all it is, i didn’t fill mine because i didn’t think i needed to

To me… any table made by wooden base will change condition in short period of time. It doesn’t matter torsion or not. That is a WOOD.
I welded steel square profile like other torsion boxes you can see on others pictures on perfectly flat surface without heating up too much. Attached legs and support bars to it. That is I call torsion Table.

I put layer of rubber seal for vibration problems…

Here is the result :slight_smile:


It was stolen from someone, might be yours. :innocent:

1 Like

Here is the table and internals I built for my machine last year.

There are three internal braces that goe from one side to the other and the rest is just smaller blocks fitted into the space. The intrnal bracing is glued and nailed together inside with an 18 gauge brad nailer and it has 1/2" ply on the bottom and a 3/4" MDF top was installed after this pic was taken. The bracing was also glued to the top and bottom panels. Solid as a rock, you lift one corner and the whole end comes up with no twist. then I covered the thing in two coats of shellac to seal it. It is a bit of overkill but I had the time, materials, and tools to do it so…

The enviroment I live in ranges in temp from 90+deg F in the summer with high humidity to -40 deg f in the winter and so dry that a desert would look humid by comparison. I have noticed no appretciable twist or warp in the table.

Unless you are making hyper-accurate parts for an oscillation overthruster out of unobtainium, somthing similar to this should more than suffice for your machine.


“over thruster out of unobtainium,”. :laughing:

Also what do you mean by you trimmed the 2x1s to make them the proper size? “used 1x’s trimmed to make them all equally sized”

I wasn’t lying when I said I haven’t done much woodworking.

Sorry, 5" of water in my (finished) basement has me scrambling.

I cut them all to equal, flat depths on a table saw. If you don’t have one, now is the time to make a new friend and use theirs / their assistance!

Back to bailing…

Store-bought dimensional lumber can vary in size, or worse yet, be warped. it’s almost never a bad idea to trim boards down a little to make them consistent with each other when using several boards for a project.

The reason for this is most of the lumber is either cut before it’s fully dry or it’s been rained on or seen moisture changes before you buy it. Sometimes the lumber is also sourced from different mills that may have a saw adjusted a little different. this is something I see often as finding something as simple as having a table saw, tape measure, calipers and rulers in the shop that all agreed on what an inch was can be a trying task. I ended up relegating all my stanley tape measures to rough work as they were too different from my other INRA and fastcap measuring tools.

I buy my lumber rough directly from a mill that kiln dries the lumber to below 15% moisture content (they aim for 10-12%) and I can say with confidence that even when buying that, letting it dry further for a year in the attic and then milling it, it will still warp both during the process and a couple days after as cutting it changes which internal stresses are dominant and the moisture in the wood still allows for some warping as it dries. when I need very precise lumber I have to use a several session milling process where I do some initial milling, wait a few days and check for warping, mill a little closer to final dimension, wait a few days and check for warping, and (hopefully) one final milling to final dimension. I almost always see some warping after the first milling, and it’s not uncommon after the second.

Oh… that’s not good… :frowning:

Can you use the Xcarve to build a boat? :slight_smile:

1 Like

I happen to have 2 pieces of plywood. One is maple the other is just a sheeting plywood. Would I be able to use these? Or is there a reason most use mdf?

The sheathing plywood should be ok for the underside of a torsion box, and the maple may or may not be suitable for the top.

It all comes down to how the panels are manufactured. plywood is a bunch of layers of veneer glued together under extreme pressure. a thick slice of veneer or a couple pices overlapping in he wrong way results in a hump in the panel. Sheathing panels are more susceptible to this issue and a bit more care is taken with the higher end materials.

MDF is made from fine wood fibers and an adhesive binder, again under pressure, resulting in a fairly uniform panel. It does have a tendancy to absorb moisture from the air quite readily resulting in a slight swelling, especially along the edges. Being made of fine fibers rather than layers of veneer, machining it flat is a straight forward affair.

Sorry to hear about the flood! I sure hope you got everything worked out! I was wondering if you could answer a few questions. The top board of MDF is it just glued to the 2x1s or is it also screwed on? I landed up getting 2x2;s would that make a negative difference or can I use those instead of 2x1s? Why doubled up 2x4s instead of 4x4s? Is there a benefit of doing it that way?

I can’t speak for @SusanHalla but 4x4’s are typically pressure treated which is not recommended for indoor use and can be quite misshapen. having 2 2x4’s connected together also allows for the different grains to interfere with each other to create a generally stronger leg that better resists warping. I can also say that working with 4x4’s is a pain in general, especially if you don’t have a 12" saw.