flat glue-ups

[Curtis Niedermier]

Let's say you're gluing up a panel with two 3/4-inch thick boards that are each 9 inches wide. For argument's sake, the boards are perfect flat, edges square, etc. Is it still possible to end up with a cupped glue-up do to offset clamping pressure? What about twist? Can you get twist in a glueup if everything is square? I can't figure out if I'm imagining things or if some of my boards aren't true prior to clamping.

I have been using cheap bar clamps, and it seems as if I get some cupping if I seat the panels too far down into the jaws of the clamps (too close to the bar). This even happens if I match the panels and joint the edges together prior to gluing so that they're at complimentary angles.

I thought about locking in the panel by clamping straight-edge boards on the faces and sort of sandwiching the panel flat.

I'm probably making too much of this, but I'm curious what you all think.

[Jim Koepke]

I thought about locking in the panel by clamping straight-edge boards on the faces and sort of sandwiching the panel flat.

That is the only way I have found to prevent the problem.

jtk

[Sam Murdoch]

Using Bessey K clams I find that I can very often simply stand the assembly on the clamps - in other words the glue up is vertical rather than horizontal. I alternate the clamps on either side of the glue up equally. Considering your "for argument's sake" qualifiers - you only need light pressure. I clean up the glue before I set the panel aside to dry. I do also check that the assembly is flat and will change the pressure and/or move the clamp in or out as needed to correct any shape I might have induced. I will add a caul if needed but that is often unnecessary.

With cheap bar clamps - you will do best to use top and bottom cauls as you describe. I add a film of packing tape to the cauls to eliminate the need to worry about them sticking to the panel. There are other ways to achieve the same result. Packing tape seems easy to use and to clean up or replace.

[Judson Green]

Its a good practice to alternate your clamps. I also try to get the screw of the clamp centered though the panel. Regarding the twisting you may want to have a better look at surface (sawhorses, work bench) that your doing the glueing up on. Need to confirm that its flat.

Oh) Don't know what cheep clamps you have but on some pipe clamps I've noticed that the faces are somewhat crowned.

[Ryan Baker]

It's not hard at all to get those sorts of problems when clamping ... especially with cheaper clamps. Check with a straightedge across the surface to see if you are getting any cupping. Adjusting clamps on alternating sides and avoiding too much pressure helps a lot, as the others have said.

[Tony Wilkins]

As an interested observer / hijacker I have a couple of questions:

1) what are the right clamps?

2) how would using hot hide glue change the answer to the original question?

[Jim Matthews]

If you're getting a curved panel after glue up,
your joint isn't actually square, or the wood isn't completely dry.

I match plane all boards I want to lay flat, that cancels planing errors.

[Judson Green]

As an interested observer / hijacker I have a couple of questions:

1) what are the right clamps?

2) how would using hot hide glue change the answer to the original question?

In the commercial shops I've worked in pipe clamps were typically used for panels. Guess a ideal clamp for panels would be stout but not have long arms on it. If you think of a clamp as an (F) than short horizontal bits would be good. I have Bessy K-bodyes and I turn them ~20° to the face of the panel glue up.

I haven't any experience with hot hide glue but can't imagine there would be much if any difference.

[Tom McMahon]

With hot hide glue, I use a rub joint, no clamps at all to distort the panel.

[Josh Doran]

I've been pondering the same issue for a while too. What I had tried was using cauls across the joint to flatten the boards. As it turned out, the cauls were actually causing the cupping since both sides were not bending equally to get to flat. I've come up with an alternative but haven't tried it yet. Use the flat surface of my bench and clamp the boards to it. Then use bowclamps (not necessary but quick and useful) to clamp the joint. Of coarse laying a protective layer between the joint and the bench. As far as hide glue goes, its viscosity would cause less sliding than regular glue.

[Brian Holcombe]

I try to save the pipe clamps for a last resort (overly large panel, ect). Bessey clamps apply pressure more consistently and less often bow out a panel.

[Judson Green]

I've been pondering the same issue for a while too. What I had tried was using cauls across the joint to flatten the boards. As it turned out, the cauls were actually causing the cupping since both sides were not bending equally to get to flat. I've come up with an alternative but haven't tried it yet. Use the flat surface of my bench and clamp the boards to it. Then use bowclamps (not necessary but quick and useful) to clamp the joint. Of coarse laying a protective layer between the joint and the bench. As far as hide glue goes, its viscosity would cause less sliding than regular glue.

Clamping the glue up boards to you bench can create another problem... your somewhat freshly planned wood not able to breath equally on both sides. Could induce cupping.

To help prevent sliding one of the guys I worked with would use little finishing nails, toe nailed at ends of the board, though the end grain, let just about 1/16 protrude out of the edge to be glued. I've heard of other things to like a few grains of kosher salt or sugar in the raw at the ends. I've never bothered always glued up over sized and just be mindful.

[Judson Green]

I try to save the pipe clamps for a last resort (overly large panel, ect). Bessey clamps apply pressure more consistently and less often bow out a panel.

Hear ya on that! I hate pipe clamps. Don't/won't own any. Seemed to always give me metal slivers. K-bodies are way better. Unless you get them all shnozzed up with glue.. those little serrations.

[Chad Bender]

My only large clamps are pipe clamps, so alternating top/bottom to balance out the clamping load is required. I've also gotten in the habit of using a couple pair of cauls with every panel glue up. The extra 30 seconds it takes to set them up more than compensates for the time saved in cleaning up the glue joint later.

[Sam Stephens]

i'd offer 3 comments: 1) the joints need to be well matched and should require virtually no pressure to close. I vertically stack the panel to check joints for large panels -very easy to tell the good joints from the bad. Sprung joints work well too. 2) minimal clamping pressure -basically till it's hand tight and then another 1/4 -1/2 turn. Definitely alternate clamps between top and bottom. It's easy to induce twist and cupping w/ uneven and too much pressure. Cauls can be really helpful, but can also cause problems as others have mentioned. 3) ensure after glueup, adequate airflow around both faces of the panel i.e. don't just lie it flat on the benchtop b/c it will warp. Sticker it and keep it that way. And sometimes, it just happens despite our best efforts so I try and leave some extra thickness just in case.

[Jim Matthews]

Clamping the glue up boards to you bench can create another problem... your somewhat freshly planned wood not able to breath equally on both sides. Could induce cupping.

I hadn't thought of that, and it's a really good point.

[david charlesworth]

There are two issues which I like to address.

1. Clamp bars bend. Alternating sides is a crude attempt to fix this.
My solution is to isolate the work from the clamp bars by setting it on wooden bearers which are about half an inch higher than the bars. It is also convenient to make feet for the clamps, so that they don't fall over.

2. We want force through the center of the board (thickness). Now most clamp feet are not square to the bar to begin with and are even worse when some force has been applied. I use "Pyramid" blocks to address this issue. They should be carefully made, in hardwood. They are more of a rectangle section, with heavy bevels on the clamp foot side. We now have a better idea of the line of applied force, and can move the blocks slightly up or down to maintain flatness.

This stuff is in my second book.

I once read an article which employed similar thinking. The person used half dowel rod, instead of my blocks.

Best wishes,
David Charlesworth

Glue ups are a ten minute opportunity to ruin weeks of work!

[Curt Putnam]

Here's what I've done = works for me & my circumstances. YMMV

I took two pieces of 48" 2 x 4 and cut notches every 6" to hold the bars of my clamps. Then I ripped a 2 x 4 down the middle for cauls, jointed the edges and waxed the heck out of them. The bottom halves sit at the 1/3 points of the panel length on the notched 2 x 4s. When the boards to be glued up are laid down they sit on the two cauls - which minimizes issues with the flatness of the table it's all sitting on. Even if I've power jointed, I run my LV BU jointer with fence over the edge until I get a clean full length shaving. I use a pair of F clamps on the ends of the boards to be glued, cauls in the middle and judicious tapping as each clamp gets tightened.

Seems to work well for the 3/4 - 7/8 stock and 36" long panels that I'm doing. I only glue two boards at one time. Since I'm aiming at a panel that is 24 x 36" I tend to try and glue in sections that are <= 12" in width so that I can run them 2 sections through the power planer, as needed, before the final glue up into a ~ 24 x 36 panel. I use a 2 handed scraper to rough clean the glue line. I've learned, the hard way, to sticker & clamp panels until they are glued into the final assembly.

[Mel Fulks]

Useing either pipe "bar " clamps or corrugated bar clamps ,I space two of them on bench or saw horses so that the board ends will be about 2 to 2 and 1/2 inches from clamp jaws. One wood Jorgensen type clamp is put on the the crank end of the bar clamps from underneath to keep bar clamp upright . Boards are put in clamps and spring clamps are used to bridge adjacent pieces to keep ends in alignment . Clamps are tightened lightly and hammer and block used to seat boards into contact with clamp bars , then tightened more. Other bar clamps added from top and bottom .Oak ,or any wood that turns dark with iron and glue contact ,gets a piece of paper between clamp and wood. I use sprung joints (cut on elec. jointer) which allow me to align adjacent boards by pressing down in middle of panel length while tightening clamps. I never attempt to do any aligning after glue has been squeezed out. I have never seen any permanent bowing in panels result from contact with clamps that have been slightly bent from over tightening .I'm sure a good job can be done with some variation from what I describe, but this is the way I was taught decades ago in a commercial shop and I have never had a joint fail . Glued as many as 200 ,varying in size ,for a single job. Occasionaly there will be one rogue piece of wood that does something weird and has to be replaced. Sprung joints are dismissed by some these days but they do work. I have glued up panels out of air dried wood that in spite of trying to get wood dry and warm as possible shrank so much overnight that clamps fell off the work left leaning against the wall , leaving a tangled mess. But the joints held. I remember well hearing an old timer laugh at me for thinking someone had "vandalized my work".PS the "rogue board" reference above means a piece that makes its weirdness known while being worked,usually something with oddball grain,nothing has ever had to be replaced that was sent to job.

[Andrew Pitonyak]

Go here:

http://www.sawmillcreek.org/showthread.php?126343-Tutorial-Shop-Made-Cauls

First post is by Mike Henderson, and he has a link to his site on how to make your own clamping cauls.

I always use them; always!

[Ryan Baker]

It sounds like a lot of folks are just using too much clamp pressure. With good joints, very little clamp pressure is needed, which will help with any cupping caused by the clamps. There are a lot of other good suggestions above. F-type bar clamps bow a lot when clamped. Pipe clamps should be avoided if at all possible. K-bodies (or similar) are a lot better if you have them. Obviously, the joints have to be good before you start.

[Chris Fournier]

Look at the glue squeeze for adequate pressure and keep the clamping pressure through the centre of the glue up. If you do this and you have squar edges you will be fine even with pipe clamps.

[Jerry Thompson]

I use the cauls Mike Henderson shows on his site and I have never had a problem providing the limber was acclimated long enough to the area it would end up in.

[Michael Ray Smith]

Anyone ever try these?

http://www.woodcraft.com/Product/2000321/1858/WoodRiver-Clamping-System.aspx

[Shawn Pixley]

I clamp the panels flat (loosely) and then use parallel jaw clamps to pull the edges together. Check for flatness and then tighten cauls. I typically spring joint my glue-ups. I then clean up any squeeze-out. I have glued up directly on my bench/assembly table with wax paper over the bench top. My bench is pretty primative compared to many you others have shown.

[Mike Holbrook]

Great link to Mike Henderson's tutorial on making clamping cauls. Michael Ray's link seems like an interesting alternative too. I have been thinking about buying a couple or three sets of Bowclamps from Highland Woodworking for a year or two now. I find the concept of "the perfect curve" interesting. I would make my own as Mike Henderson does but I am not so sure I would wind up with "perfect" curves. Although one may argue that the whole "perfect curve" concept is a little anal, we are attempting to spread pressure evenly over an entire surface. I just wonder how perfect the curves on wooden Bowclamps are, with changes in humidity, "normal" wood movement... In his DVD Build the 21st-Century Workbench, Bob Lang addresses the problem when he glues up the top pieces for his bench. He makes plywood i-beams to solve the issue of a "flat" bench to make a bench on. I think I see something similar in Mike's tutorial. In this case, other clamps are apparently used to force the pieces being glued together flat. Maybe there is a total solution among those ideas somewhere?

Does anyone own Bowclamps that can weigh in on how well they work and how long they retain "the perfect" curve?

[Mel Fulks]

Since we have a lot of commercial shops here , the Taylor clamp racks work well. I worked for a long time before getting a chance to try one. I used to wonder how they could work well with all clamps on one side . But they do! Only occassionally do you have to use a wood hand screw clamp or C clamp on each end. Biggest problem is the guy gluing up one panel on the TOP of the rack. Then you have to insist he help you pick the danged thing up and move it somewhere. THEN you can empty the rack and actually use the whole thing.

[dan sherman]

The attached photos show how i do it, the key step is book match jointing the edges. unless I really space out, a glue-up requires nothing more than a few swipes from a card scraper to clean up squeeze out and fix any minor misalignments.