Workbench: Slabs or boards?

[Zach Dillinger]

Joe, I wouldn't be surprised if others have had creep issues with laminated Roubos, but I suppose it could be written off as seasonal movement / wear and not considered glue creep. I don't have top-of-the-leg stretchers on my bench. If I had them, I suspect that the creep would have been minimized / avoided. On my next bench, I'll probably use hot hide glue, which to my knowledge doesn't creep like PVA glue, to glue the two slabs together. I'm not planning on a top stretcher, but I may make provisions for one just in case.

[Jim Neeley]

Zach,

I took your post back to the structural engineering team and they said:

From the front to the back of the bench, if you are seeing "steps" between the boards, that would be glue creep. If not, it's wood creep. Any sagging between the ends of the board (end to end) would also be wood creep.

Which are you seeing?

According to them, wood creep is a real phenomena and varies significantly between species... I'd never thought it would be measureable but apparently it is. I'm also not trying to tell you what you have.. but am curious to learn!

Jim

Jim,

I agree that creep would be negligible if you have cross supports. But for a Roubo, without such supports, I can assure you its quite real and measurable. Lacking proper slabs, I built my Roubo by laminating a large number of scrap birch drawer sides (cost me $45 to buy a pallet load, which was more than enough). They are 3/4" thick, 4" wide and 16 1/2" long. I lface-laminated enough of these together to make a 24" wide, 4" thick, 8 foot long bench. It took me the better part of a summer and a gallon or so of Titebond. In the years since I've made it, the center of the bench (between the legs where there is no support) has crept downwards about 3/16". I've flattened it out but the problem has made me consider putting a secondary brace in place to shore up the center. This is an extreme example, of course, given my choice of material, but any large glue-up without support will suffer a similar fate. It's just too heavy, with too little support. Its like an overloaded bookshelf. Nothing can be done unless the shelf is stiffer or you add a central support.

My next bench will be rather more rustic in composition and will be made from two large slabs of 16/4 beech that were still dead green when I made my current Roubo.

[Zach Dillinger]

Jim,

My bench has a slope towards the middle and I could definitely see some "steps" between the boards, especially in the middle few boards where there is the least support from the legs. I always assumed it was glue creep, given the sheer number of glue joints in the top. Of course, I've flattened it out a couple of times so I don't have pics of what I'm talking about.

Thanks for the info! I find this fascinating, as you do.

[Sean Richards]

Hi Rick,

Did you make a decision? Slabs or a glued up top?

[Rick Prosser]

Gonna go with glue up top. Probably 4x4-ish glue ups. I cut the boards 4" thick on the sawmill. Once they are jointed and thru the planer, they will vary in thickness from board to board - but I will saw into 4" strips to have consistent top thickness with varying lamination thickness glued together.
Been very busy with no time for shop. I have the yellow pine out of my kiln now and ready to start - waiting for some free time...

[Scott T Smith]

Remember that almost all glue joints, excepting hot hide glue, will have "creep". So, while your carefully selected quartersawn boards might not warp, they will slide around relative to each other.

Zach, I'm not following the logic about creep (and would like to learn more), could you please elaborate about this?

The reason that I'm not following is that quartersawn boards glued together should have consistent expansion and contraction in all three directions; ie width, thickness and length (presuming that they are the same species and from the same log). What is the cause of the creep?

Thanks.

Scott

[Mel Fulks]

The only commonly used glue I know of that creeps is "white glue".Before yellow glue it was widely used for good made to order furniture. A perfectly surfaced mortise and tenon joint would sometimes shift a few thousandths over night and have to be sanded again. Usually when I see something about creep these days it is in an old book. Anyone seeing creeps (other than in old teen movies)? After the white glue cured or dried a little more it was fine. Incidentily, we had a policy in shops around here that the finisher did no sanding on anything but finish. A small defect in the wood surface was sanded by the guy who made it.Yes , that meant the piece was moved two more times. Today many employers will let woodworkers export their duties to the finishing room.

[Joe A Faulkner]

Rick,

Check out this split top from an English benchmaker - Mark McGuire. Pretty nice looking bench.
http://www.youtube.com/watch?v=TiqG8w_ktXs

His web site is at: http://www.maguireworkbenches.com/

[Christopher Charles]

Hello Rick,

Having milled some slabs for a workbench with a friend's mill, I'd recommend going the 4x4 route- a good middle road, and for the flatsawn slabs, you'll get a reasonable number of "quartersawn" pieces once cut to width.

Good luck in finding time, a familiar problem!

Cheers,
Chris C.

[Scott T Smith]

Russell, one trick that timberframers use for planing large beams is to set the beams across a couple of saw horses, and run a lunchbox planer across the beam (instead of running the beam through the planer). The way this works is that you leave about 3' of beam sticking out past a sawhorse, and start the planer from one end. As it gets closer to the sawhorse, lift up the end of the beam and then move the sawhorse to the other side of the planer.

Works like a champ!

[Scott T Smith]

All - thanks for the education regarding glue creep. Wouldn't this problem be prevented by using dowels, biscuits or domino’s between the slabs?

[Mel Fulks]

Wood expands and contracts. Heavy spans can sag .Certainly I do not deny his bench has changed. I am skeptical of glue creep except when white glue is used. Anybody have any current test results or related information?

[Charlie Stanford]

Remember that almost all glue joints, excepting hot hide glue, will have "creep". So, while your carefully selected quartersawn boards might not warp, they will slide around relative to each other. Since the front and back will be supported by the legs, the middle will slide down relative to the other boards, giving you a cupped surface that is no different than you'll get with a slab. I vote slab, since you'll have to flatten your benchtop either way and the non-existent glue joints will not creep.

A face-laminated bench top moves through its thickness and not its width so the formation of 'steps' is not necessarily indicative of creep. PVAs creep, especially when glued edge-to-edge, but when gluing boards face-to-face there is a huge amount of gluing surface and step formation is most likely just the boards expanding in width (remember how they're oriented here) and not the PVA creeping.

[Zach Dillinger]

A face-laminated bench top moves through its thickness and not its width so the formation of 'steps' is not necessarily indicative of creep. PVAs creep, especially when glued edge-to-edge, but when gluing boards face-to-face there is a huge amount of gluing surface and step formation is most likely just the boards expanding in width (remember how they're oriented here) and not the PVA creeping.

Each of the boards I used is quartersawn, dry, stable birch drawer-side scraps used in a factory cabinet shop. There are a couple of steps, but the main thrust of the problem is that the center of the bench has slid down about 3/16" or so since the glue-up, which just so happens to be toward the unsupported area of the bench. I sincerely doubt that it is the boards changing in width. But I could be wrong, wouldn't be the first time and it won't be the last.

It won't matter soon, anway. I scored two 4" thick, 18" wide, 10 foot long oak timbers from an old falling down barn (my cost, $0.00). According to the barn's 70+ year old owner, they've been in that barn for as long as he can remember. So those will soon be my new top, and I can escape the creeping birch top.

[Rick Prosser]

boards.jpg
So, here are my rough boards (SYP). Should I joint and plane before or after ripping to 4 inches?

EDIT: Oh - I cut them 2"+ on the sawmill - not 4. (I was planning 4", but I wanted to be sure they dried completely - and quicker)