why did my workbench warp?

[Chris Skoglund]

I'm new to this and for my first large project I decided to make a workbench. This is the one that I made. (Originally I thought I'd do a Roubo bench but decided that I didn't have the resources or the skill to handle that at this point). http://www.handymanclub.com/Portals/0/uploadedfiles/Robust%20Bench.pdf

Unfortunately, as I got into the project I liked this plan less and less and now my tabletop is warped and there's a sizeable bow upwards in the middle of the bench running lengthwise. The plan calls for gluing several lengths of maple together and then mounting that assembly to a stack of two sheets of 3/4 MDF. The maple top is glued to the MDF along the front edge only. My thought is that this is were the problem is. Did that glued edge prevent the top from expanding? Also, I did not pay attention to the direction of the grain in the individual lengths when I glued them together, but as I've been reading up on this it seems that I should have to prevent cupping (?). Not sure if I understand that correctly. Perhaps what I'm seeing is cupping?

We just came out of a rainy spell and the weather will now become very, very dry as is custom for my area. Will the tabletop flatten out as moisture is pulled back from the wood into the air or will I face new problems? Will it crack?

Any insight is greatly appreciated as well as resources to turn to to understand this better. (I'm waiting on Understanding Wood as per several members recommendations).

At some point I may just build another top and mount it to the base and use this one as expensive firewood if necessary.

Thanks in advance.

[Jim Matthews]

It would be helpful to view a few pictures.

If your top is a lamination with MDF at the center, flip the works over and apply some weight for a few days.
My guess is that you can pull this down with a batten under the middle, spanning the rails that connect the trestle legs.

Still, it would be helpful to see the extent of the problem.

I wouldn't say that my creaky bench is dead flat, but nothing rolls off while I'm working.

[Tony Shea]

I have become very unhappy with any MDF laminated tables/benches. I personally in my life have made 3 different benches using laminated MDF. 2 were for me and the 3rd was one I helped a friend out with. My last attempt was with 4 layers of MDF and no matter how hard I tried I never ended up with a flat bench top using this method. There really is no good way to lay them up in lamination to obtain a nice flat surface. I will never go this method again and ended up making a my last bench top of wood. Now I can flatten the top any time I need/want. I mainly use hand tools so a flat bench is an absolute requirement to get flat stock from hand planing.

[Chris Skoglund]

Here are two pics. So, the top is laminated maple. There are two sheets of mdf beneath that and a sheet of 1/2 inch birch ply at the very bottom. At the edge closest to the camera, the maple top is glued along the front edge to the mdf beneath it (the plan called for this to keep the front edge flush to the front). At the other end of the bench there are wood screws driven through the ply and the mdf into the maple top. The warping is most pronounced at the front edge where the gluing is and less pronounced at the end where the screws are. I'm guessing that the problem was caused by the gluing as that would prevent the maple from expanding right?

The maple frame that you see surrounding the top isn't glued to the top, it's held in place by biscuits. The biscuits are only glued to the frame, not the top.

Thanks for taking the time to consider this. I'm looking to a) never make this mistake again, and ideally, b) make it better.

270685270686

[Jim Matthews]

Gotcha - I was thinking the bow was left to right rather than front to back.

Is there a set of top rails, also?
If so, a batten or two on either side of the bow (front to back),
with a couple of screws will pull this back to flat.

Just make sure that the screws "bite" into the hardwood on top without coming through.

If you're planning dog holes, care in placing the battens is in order.

[Chris Skoglund]

Ok thanks! Why do you think this happened? Is it because that end of the bench top is glued along the front edge? Trying to avoid having this problem again...

[Paul Murphy]

My opinion is you would have your wood movement with or without being glued along the front, so you don't need to worry about that part of the situation.


I believe the MDF is blocking the bottom of the wood layer from migrating moisture at the same rate as the top surface exposed to the air. Whenever the top and bottom of the wood layer are again at the original “flat” moisture content your wood layer should again be flat.


Taking into account the potential wood movement of nearly 1/4” per foot of width, you have the option of routing holes through the MDF layers, and screwing into the wood layer. For example, at a distance of 2' from the front edge if you took the worst case your hole would be a slot ~1” long with the screw placed at the center of the slot. The worst case protects you from false assumptions on existing and future moisture levels, and assumes future movement from center to either extreme of moisture content.


So, fasteners of ~2” length through fender washers. Space fasteners about every 6 inches, and now your MDF is acting as a batten on your wood layer. If your slots are not long enough, when the wood layer is shrinking the screw will hit the end of the slot and the wood layer will split, and when the wood layer is gaining moisture the screw will act as a fulcrum for a contained “bulge” where the arc radius is defined by the distance the wood is trying to expand. As you can see, the slots need to allow for full movement of the wood layer! I would space fasteners & slots ~every 6" to form a "grid batten", and tighten them only as much as the changing moisture will allow. A little patience will prevent stripped fastener holes.



Don't let this bother you, as woodworking is a learning experience, and most woodworkers build other benches as their needs evolve.

[Art Mann]

I just took a critical look at the link you posted. The base appears to be designed well enough but my opinion of the top is that it is an engineering disaster. There is nothing in the original design to hold the thin maple top flat other than a single screw. The difference it the coefficient of expansion of the maple and MDF is enough so that the two layers can't be tied together more securely. Problems such as you are seeing are almost predictable. I say shame on "Handyman Club" or whoever did this plan for publishing such a flawed design.

If you want a maple top, I would suggest you buy one that is thick enough not to warp without any MDF. You could also glue one up yourself if you have access to hard maple at an economical price. Where I am, the materials to build a solid maple top are more expensive than jut buying one.

If you can do without the maple, two layers of 3/4 MDF properly glued together and flattened during curing will make a pretty stiff and stable top. I chose to build my benchtop that way and then put a captured and replaceable sheet of 1/4 Masonite on the top. Whenever my bench top gets warn, I just replace the Masonite. Having said that, I must say I am planning to replace that top with a factory made solid made maple top.

[Jim Matthews]

Art, do you think a couple plywood battens across the center span would help him bring the top back to flat?

This same problem appeared in a solid wood chest that warped in the opposite direction (high on both ends).
It appears from this design that the maple top is little more than a veneer.

When I've made veneered panels, I put a piece in opposition so, the panel doesn't bow as the glue dries.
I'm guessing this is a similar effect, on a larger scale.

[peter gagliardi]

Above, Paul nailed it in paragraph #2 ! You can't leave solid wood with 1 exposed face, and cover the other with any panel product.
It is not so much differential expansion of different materials, but rather differential expansion due to moisture level differences between wood faces. Both faces need to always be treated the same for optimum results.

Peter

[Brian Holcombe]

My opinion is that the design lacks strong consideration for how much wood can move. Normally a solid wood workbench top can move 1/4" or more across the grain. This design joins the ends and the sides all while joining one of the sides to the top and putting a screw into the top from the bottom.

That leaves too many boards locked in with no where to go except up.

[Steve Baumgartner]

Alas, too many handyman magazines are published by people who are professional journalists with only a hobbyist understanding of wood! This design is an example of what can result. It was probably inspired by looking at typical laminate-surfaced kitchen cabinets and jumping from there to solid wood surface (instead of thin plastic laminate) with several layers of MDF and plywood to shoot for the "thick is strong and heavy" mantra of good workbenches - all without much understanding of the techniques and materials involved! Several very serious errors in the design resulted.

Others have pointed out the issue with seasonal expansion and contraction of the maple (which does not occur with plastic laminate) so I won't belabor that other than to note that any lamination should be symmetric about its middle else it is likely to warp. A layer of solid maple atop two layers of MDF and another of plywood is anything but symmetric! There's a reason why plywood is always made with an odd number of layers and why you should always veneer the back side of a panel if you veneer the front.

But here's another problem: both MDF and plywood are notorious for creeping under load unless they are restrained by fastening them to something rigid. The layers of MDF and plywood may be rigid enough to resist the bending of the maple for a while, but then they will creep and either bulge or sag. You need crosswise supports under the top and should screw (*not* glue) them in place. If deep enough, they should keep the MDF and plywood flat, though you still have potential for the joints in the maple top to open up in dry weather.

[Tony Shea]

If you put the same straight edge along the bottom face does it show the same bow in the opposite direction? Is this whole bench top bowed top and bottom or is it the maple? I assume it's the whole top but I suppose the maple may delaminate itself due to the complete difference in materials used (solid wood glued to mdf). If the whole bench is bowed then I agree that your only option is large battens glued underneath. Then you will need to flatten the top of the bench with handplanes as I doubt you will get the top completely flat with these battens. They should at least prevent it from moving too much more or at least keep the bowing down slightly. But you will probably be spending a lot of time flattening your top until your through the layer of maple into the MDF.

If all else fails then go to the lumber yard and glue up a lamination of 2x lumber. This will certainly be rugged enough for a bench top even if you have to use Doug Fir. I would probably not use spruce or pine 2x material but southern yellow pine or doug fir would be fine, depending on what's available where you live. This method will keep your cost down and give you a very nice top.


270856 270857

270858

sorry about all the spokeshaves and messy shop. Just showing you what a laminated doug fir 4x4 top looks like after about a year and a half of use. I've flattened it twice and the second time was surprisingly not that out of flat. I also don't really care for a finish on top of my bench but I use hand planes a lot so the rougher surface is welcomed.