gluing up workbench top

Re: gluing up workbench top

Paul M. in San Diego

>Hi Kelby,

Vacuum is an extremely powerfull tool. But whatever you use as a "platen" *must* be structurally much stronger than the 3" maple you are planning to use or else the platen will warp to match the shape of the maple.

I think you would be better off staggering your glue-up in stages without the vacuum. Glue a few boards together while this is on a known flat surface. Then glue a few more to this assembly the next day. Keep going until you get the width you want. If there is some movement, then you can slightly stagger the boards so that they follow the reference surface as you glue up.

No matter how careful you are, you are going to have to flatten this surface. Scrub planes are cheap, and they are your friends. Get yourself a long straight edge, and two long winding sticks, and you will be done in less time than it will take to set up this vacuum system.

I've done some vacuum bending and veneering, but the thought of vacuum assembly of a giant benchtop is mindblowing. If you do go that route, please take lots of pictures, as I would love to see this!

Re: gluing up workbench top

Kelby

>Paul,

Thanks for your thoughts and suggestions.

It makes sense that if the maple is warped, the platen will warp to match it, not the other way around. And that is what made me consider the possibility of jointing the maple edges first.

What got me thinking about this was that, if the maple edges are jointed to start with, then if the glued-up top is warped or twisted it could only be because the individual boards slid around during glue up --- in other words, some of the boards slid up and others slid down, such that the edges that constitute the top are not all in the same plane. But a vacuum press and a platen should eliminate that problem -- the edges of the maple boards should all be in the same plane (as would the platen) once the press is on. The only issue is whether that plane is going to be a flat one or a warped one. If the edges of the maple are all jointed, then perhaps it will be a flat one. But then again, perhaps not if there are other reasons for the maple not to lie flat.

I agree that it wouldn't be that difficult to laminate the boards a couple at a time as you suggest and then clean up with hand planes, and that's probably the route I'll go.

However, if no one has ever tried this before, maybe I'll give it a shot just to see if it works. The worse thing that can happen is that I waste a little time and effort and end up with the scrub plane at the end anyway. If, on the other hand, it works, I will have invented "The Kelby Method" of gluing up a benchtop. ;) If I do it, I will take some pictures and post them here so that Ellis' site can be the official site of The Kelby Method.

Thanks again for your thoughts.

Re: gluing up workbench top

Adam Cherubini, NJ

>Hi Kelby,

Here's my idea, and I would appreciate it if anyone could tell me why this might not work.

1) Wood is not aluminum, Kelby. You can't machine it accurately and if you could, it won't stay that way. Those boards won't be perfectly twist free. We've all tried to align the boards in our glue up, but it never quite works out that way.

Initially, I was worried about there being enough clamping pressure to laminate all those maple boards, but then I did the math.

2) Well you did your math wrong

Assuming a benchtop that is 5' long, 24" wide, and 3" thick, that would mean over 2000 pounds of even pressure clamping the boards together.

No that's 2600 pounds divided by the area (60x3") which equals 14.7 lbs/in2 (1 atmosphere). Now that would be about 1/10 of what you need if you had only two boards. But you have what, 8? You'd need a nastran model to know for sure what you were getting, but rest assured your vacuum bag won't cut it for PVA. You'd need autoclave pressure for that. And you probably couldn't glue up all those boards at that.

Admittedly, I would probably get a little more pressure than that by using six or eight pipe clamps

Now you're talking. Do one glue joint each night for two weeks. It won't be perfect and you'll have to flatten like everybody else. Don't use a scrub plane though. Get your self a long long try plane and work +/- 45 degrees

Vertically, the vacuum press should give me over 16,000 pounds of even pressure pressing the jointed edges flat against the platen.

Yeah and your body is supporting 12,000 pounds too! Look your vacuum bag is good for what its good for (veneers for example). It only allows 14.7. PVA wants 250psi.

As if I haven't rained enough on yoru parade, your bench is about 4 feet too short. You're going to run out of space even for jewelry boxes. But that's another math problem.

One joint per night. Use a long long plane to flatten and you'll be okay. That's my advice. Good Luck.

Adam

another reason - open/working time of glues

Tim of Dublin

>Most glues won't provide enough open/working time for you to apply glue to that many surfaces, get them assembled into the bag and apply pressure.

I also see problems when I or others get too ambitious when making more aesthetically demanding panels in too few steps.....even a light skin on the glue shows as a glue line in a panel.

For your workbench, you'll want as strong a joint as possible and that means getting the joint under pressure as soon as reasonably possible. Do one or 2 joints at a time....if you have enough clamps you can obviously glue up more subassemblies in a night.

Tim

Re: gluing up workbench top

Kelby

>Hi Adam,

Thank you for your thoughts.

I would respectfully disagree with you about the ability to machine wood accurately; although its true that I might not be able to get within .001" as I could with aluminum and a milling machine, I have never had a problem machining wood within acceptable tolerances, flat and square. So I'm not concerned about how well the boards can be milled before laminating.

But your points about acceptable clamping pressure are interesting, as well as Tim's comment below about open working time. I will give it some thought. There are ways to increase clamping pressure when using a vacuum press, so I'll think about practical ways to do that.

On a side note, I realize the label on PVA glue usually says 175-250 psi, but I have laminated many things using PVA in the vacuum press at 14 psi and never had a problem, even though the vacuum press is less than 1/10th the recommended pressure. (I used to wonder about using PVA in the vacuum press until I saw David Marks do it a few times. If it's good enough for David Marks, it's good enough for me. ;))

Also, as to the fact that there are multiple boards being glued at once, that should not alter the amount of pressure at each joint. That is why it is no problem to use the vacuum press to do bent laminations with eight or more laminations at a time.

As for your comment about my bench being 4' too short, I'm not sure where you're coming from. Are you saying that any bench less than 9' long is too short? I think most folks would disagree. Personally, I would love a 9' bench, but I do have some space constraints to deal with. So 5' it is.

So I guess I disagree with you about a few points, but that doesn't mean you aren't right. I do appreciate your thoughts.

Re: another reason - open/working time of glues

Kelby

>Those are great points, Tim. If I do go the vacuum route, I should probably do it a couple boards at a time just as I would if I were using ordinary clamps.

Putting off the inevitable

Christopher Schwarz

>Tim,

Even if your procedure works and you end up with a flat top on day one, you're going to have to flatten it some day. And judging by your attention to detail shown in your message, it will have to be real soon.

I've never encountered a workbench that didn't need flattening on a regular basis, though I've not been into space, where it might be possible.

I flatten mine every six months and when I do, I'm scolding myself for not tending to it earlier.

Glue up your top as best you can with clamps and orient the grain of the boards so they all run uphill in the same manner. If you're worried about PVA's open time, some hide glue can give you an hour of working time.

Get a long plane. Work diagonally both ways. Come back and clean up the diagonal marks going with the grain. The first time you do it this should take an hour. It is not hard. It is easy.

Good luck on your journey.

Chris

Re: Putting off the inevitable

Kelby

>Chris,

Thanks for your thoughts. (The original post was from me, not Tim.)

It certainly occurs to me that this could all create a lot more work than it could save, even if it did work.

The part I like best about your post is the suggestion to get a new plane. I already have a good jointer, but maybe this is a good excuse to get another one. :)

Re: gluing up workbench top

Frank D. in Montreal

>It's a good idea as others have suggested to glue up in stages. I would glue up 3-wide at a time, which would give you around 5-1/4" with jointed 8/4. Then joint the edges of the sections and do a final glue up. If you're so worried about pieces sliding around, use biscuits to line up the pieces (don't tell anyone here that I told you to do this). It makes monster glue ups like a bench or countertop a LOT easier. I made a laminated maple counter-top five feet wide by over 8 feet long last summer and with biscuits to line up the sections I flattened it with a scraper plane, that's all. Don't clamp too tight or it'll warp for sure; once it's clamped up check with a straightedge and adjust clamp pressure--you can get a really flat surface just by tweaking the clamps.

Frank

Re: gluing up workbench top

Frank D. in Montreal

>Lost the pic when I previewed--here it is

Sorry

Christopher Schwarz

>Don't know what I was smoking last night to botch your name on my reply. It must have been the salmon.

Finding excuses to buy more tools is a time-honored tradition in the craft. It might be the one thing we all agree on. Good luck with that top.

Chris

this worked for me

bill tindall

>What Adam has recommended is what I did for a very wide bench and it worked just fine- glue one at a time. Also he is right on about pressure, at least according to the adhesive manufacturers recommendations for edge glueing. For laminations, the recommended pressure is much less, which is why the vacuum works for laminations. In a furniture factory they would put 2000 pounds on each clamp, spaced 12", for this edge glue-up situation.

If you are using PVA glue and dry lumber you can glue one board per hour and be ok, assuming your joints are machined ok.

Re: gluing up workbench top

Adam Cherubini

>Hi Kelby,

I think my post was way too strongly worded. So I apologize for that. If you'll indulge me, I'll clarify a few things I wrote. Now keep in mind, I see these discussion groups as idea exchanges, not group brainwashing. You get to build whatever, however, and you and I get to be the best of friends. Okay?

I would respectfully disagree with you about the ability to machine wood accurately.

Well in this case you have a couple things working against you I think you need to take into account. Wood is like a bundle of rubberbands. Remove some from this side and a minute later, the wood moves to compensate. When you talk about .001 tolerance, (I know you weren't talking about wood) you're thinking in terms of a thickness. The problem you face is twist, and cast or flatness and you haven't a tool in your shop that produce a tight tolerance on any of those. Its a stiffness problem. How this manifests is when you go to glue up two boards that aren't perfectly flat. Inevitably you're forcing one or both to assume a new shape. That laminated shape is pretty stiff, will resist clamp pressure, and will result in a non-flat surface.

The idea of aligning the boards carefully is a good one and it will help. But its not going to get you out of planing. I guess I'm writing because I'm afraid you'll be dissappointed or discouraged if your plan doesn't worrk (I don't think it will). But clearly this is all a matter of degrees and expectations.

There are ways to increase clamping pressure when using a vacuum press, so I'll think about practical ways to do that.

You know something I don't. The only way I know to increase pressure is the use of an autoclave.

On a side note, I realize the label on PVA glue usually says 175-250 psi, but I have laminated many things using PVA in the vacuum press at 14 psi and never had a problem, even though the vacuum press is less than 1/10th the recommended pressure.

I've actually spoken to Franklin's engineers as well as composites experts about this very subject. The manufacturer was unwilling to share (or didn't have) test results for PVA glued low pressure joints. Their engineer said PVA had very low (poor) properties in low pressure joints. The manufacturer's recommended pressure is the minimum required to develop (its considerable) full bond strength.

As this applies to your work bench, low strength joints may hurt you down the road in several ways in which you may be unaccustomed or not expecting:

1) You may find joints or cracks opening at the ends of the bench depending on how you treat your end grain. But Maple is pretty stable.

2) Depending on your leg design integration, you could have either a weakness or a lack of stiffness or tightness in the leg joints.

These things may not be a big deal and there's a certain unrealistic academia in what I'm saying. But like I said in the beginning, now you know what I know and you can decide if your assembly is worth whatever risks you perceive.

Also, as to the fact that there are multiple boards being glued at once, that should not alter the amount of pressure at each joint.

That's only true if the boards and glue joints are infinitely rigid (stiff) and they aren't. Thus the need for a nastran model (everything is really a spring afterall).

As for your comment about my bench being 4' too short, I'm not sure where you're coming from.

If you look back in our archives, you'll see many discussions on this subject. Suffice it to say, longer is not just better. There are operations you simply won't be able to do with a short bench. A 5' bench places limitations on you that wouldn't exist with an 8 or 9 footer.

I thought my bench needed only be as long as my longest piece of furniture. I was wrong wrong wrong. The molding around a chest 40" wide 20" deep is 8' long. (I was thinking it would be no longer than 4'). Its important to be able to plane stock before cross cutting. And that's just one example.

Even if you can squeeze another foot or two it'll be worth it. Length is power in a woodworking bench.

Have I brainwashed you yet? :) Watch the trammel point. See it swing back and forth. You are getting very sleepy. You need more planes. Your chisels are crap. You need expensive chisels. Infills can pay for themselves in the fine furniture you can make.

Adam

Thanks Adam

Kelby

>I very much appreciate your thoughts. You make a couple of points that are persuading me to do it the old-fashioned way. First, your comments on clamping pressure are very interesting. Although I have not had any problems with PVA and vacuums in the past, it is true that those projects were not subjected to the kind of beatings that a workbench will take, and there is certainly a risk that a joint could fail down the road at an inopportune time. Second, the old fashioned way guarantees that I have a good excuse to pick up a new plane.

Thanks again for all the time you put into your respones.

Thanks to everyone

Kelby

>Thanks to all for your input on this. I have probably put more time into thinking about how I would use a vacuum to do this than what it would take for me to plane the top flat.

I suspect I will end up doing this the old fashioned way. I'll post some pics when the bench is done.

Thanks again.

Adam, why is it important

Frank Mutchler in Colorado Springs

>to be able to plane stock before cross cutting? Why not cut the pieces a little long, plane the desired profile, and finish cut?

I know there must be a very good reason and I can't wait to hear it!! ;>)

Re: Adam, why is it important

David Hammond, in Powder Springs GA

>I'm not Adam, but he told me the answer, so hopefully he'll forgive me for talking in his stead: If you're planing something, especialy a moulding made with a hand plane, it's far more accurate to plane an entire stick, and wrap the moulding around you're peice of furniture - this way the profile matches spot on. Same with table legs - plane all of your footage first, then cut.

Set me straight if I'm wrong Adam.

David

Terminology alert: what's nastran?

Bill Houghton, Sebastopol, CA, the rain's here

>Is that the federal agency that keeps throwing spaceships up there, or some new cold remedy?

Don't forget the "rubbed joint"

Ed DeLauter

>that doesn't rely on clamping pressure at all. When it locks with suction it stays. But not intended for

your bench top application.

bingo

Adam Cherubini

>Couldn't have said it better myself.

I do the same with carcass sides. I like to plane first, since the ends are never quite as nice to plane as the middle. Likewise, I prefer to edge straighten before cross cutting. If you are working a carcass side and top together, working on the long board means you don't have to match plane the two boards together later. I like doing it this way because teh grain will wrap around. If i could work 4 sides of a horizontal carcass at once I would- like a small bible box or something. That's why an 5' bench might be too small for an 18" x 14" writing desk or something similar.

Taken to the extreme, you might ask why a big case doens't require a 14' bench. Hello? That's how long the Dominy benches are. I don't get 14' foot stock though. I buy 8' stock so a 9 foot bench is the right length.

BTW- My current bench is 6' long

Adam

Off to Williamsburg for the conference. See you guys in a week. And you locals better not get any ideas, my wife is staying home to guard my tools!

Re: Terminology alert: what's nastran?

Adam Cherubini

>You've got the right idea. Pretty sure Nasa developed it. Its a finite element modeler/analysis package FEM or FEA. It considers the stiffnesses of individual components in an assembly to predict where the load is going and its approx magnitude. Its used by engineers in many industries. I'd be surprised if LV didn't have a FEM of their planes.

Adam

Clamps versus Vacuum Bagging for Bench

Todd Stock

>Just a comment on clamping and desired pressure for gluing with PVA.

Looking at the main portion of the bench top under construction pic, there are four clamps - each capable of applying about 1000 lbs or so of clamping pressure. Assuming the cross-sectional area of the joint is 60" x 2.5", that's 4000 lbs applied by K-body or pipe clamps to a 150in^2, or 26 psi. Using HD I-beam clamps, that pressure can be up'd to over 100 psi (assuming a 5000 lb max figure for the clamp - a good reason to grab a few I-beams, which are cheaper than the K-bodys)

Another consideration is the distance between clamps. Assuming you are making your bench-top up with 8/4 stock planed to 1.75" width, you'll want to have a clamp every 3.5" to get uniform pressure for the outermost lamination on each side. The rule of thumb to apply here to achieve uniform pressure across the joint is to space clamps no more than twice the width of the outer laminations (for engineers - assume St. Venant's principle)

On the vacuum clamping plan:

- Lots of pressure holding the laminations to the table (14 psi x length x width)

- Very little clamping pressure where it counts (the joint) - only 14 psi, versus 32 psi with four clamps or 64 psi with eight.

So...vacuum clamping is a pretty good method to achieve pressure for veneering, etc., because the pressure is uniformly distributed (not the case with most veneer presses) and probably higher than can be reasonably achieved using weights or a home-shop veneer press (can you find 4 tons of weight around the house for that 2x2 table top? Not a problem with a vacuum bag set-up).

Vacuum clamping is not a very good method for achieving adequate clamping pressure on something like a bench top glue-up, where the area of the joint is small - clamps will do a better job of applying pressure (the vacuum bag will probably do a good job of keeping things flat until the bag comes off, though).

If the job is to avoid having to do the heavy surfacing work with scrub and jack to get a flat bench top, go to the noisy side for instruction on cheap, fast, and accurate router sleds. Most of the folks here, though, consider building a bench to be a great excuse to pull out the heavy iron (or wood). The skill set developed will help when that long stick of 12/4 needs an 1/8" of twist removed.

Re: Clamps versus Vacuum Bagging for Bench

Pam Niedermayer - Austin, TX

>Another consideration is the distance between clamps. Assuming you are making your bench-top up with 8/4 stock planed to 1.75" width, you'll want to have a clamp every 3.5" to get uniform pressure for the outermost lamination on each side. The rule of thumb to apply here to achieve uniform pressure across the joint is to space clamps no more than twice the width of the outer laminations (for engineers - assume St. Venant's principle)

Todd, this is very interesting. Let's say the first joint is being glued. Is it possible that the wood itself, being so thick, would work as a caul, thus lessening the requirement? Or that using another board as caul would also lessen that?

And in turn, on gluing the second joint could the first assembly work as a caul?

Pam

Why not?

Pam Niedermayer - Austin, TX

>