How many clamps???

[Greg Becker]

According to an article in FineWoodworking by Prof Roman Rabiej, Cherry being glued on a quarter sawn face needs 250psi. My table top is 78" x 1.75 along the seem which calculates to 34,000 psi required.

Using all my large clamps (12 pipe and 5 K Body) and giving a maximum pressure (due to my hand strength) that totals out to 20,000 psi.

Do I really need to go out and purchase 10 more pipe clamps? I am almost ready to glue up the dining table top that I have hand planed 3 boards to prepare. The final top will be 82 x 42 (including breadboard ends) and is my most important project to date. This is beautiful cherry that deserves to be done right.

Any suggestions? Below are photos showing my clamp base with built in panel clamps and a frame to support the pipe clamps. 10 clamps will be from underneath and 7 will be applied from the top.

[george wilson]

I haven't heard that figure. I think that a properly fitting seam should NOT have to be clamped together with great force to make it fit. If it does,it is going to pop open in the future. just make sure neither seam is slightly convex. It should rather be slightly concave,so the ends of the seam do not have a tendency to come open later.

It looks like you have enough clamps if the joint fits properly.

[Gary Max]

We have over a 100 clamps and still buying more----I hate it when I have to stop because I don't have enough clamps to get the job done.
You might as well go ahead and get them.

[Thomas love]

Seems like you have plenty of clamps, I like Georges advice.
tom

[Andrew Gibson]

People use math when it comes to glue up?
I clamp it up untill I get nice even squeeze out along the joint and call it a day.

[george wilson]

The math makes no sense to me. You just need to make sure a joint fits properly. Attempts to CRUSH it together will only come back to bite you later.

Some old Spanish guitar makers only used twine and tightening sticks to tie guitar tops and backs together,and even to tie the body of the guitar together when gluing. They often were too poor to have clamps. They still made some very fine instruments that way,though.

I did the same thing when working in public in Williamsburg. A form the shape of the guitar with nails driven all around the edges was used to lash down the vaulted backs of early guitars. The inlaid guitar I made in the FAQ section here was glued together that way. It was actually better than regular clamps because you could damage the vaulted shaped back with clamps.

Clever ways to clamp things can always be devised. Extreme pressure isn't needed.

Guys who sit around figuring out esoteric math formulae for things like clamping wood aren't getting much work done. Besides,WHO has the equipment to even measure clamp pressure anyway? It's a bit ridiculous.

[David Keller NC]

According to an article in FineWoodworking by Prof Roman Rabiej, Cherry being glued on a quarter sawn face needs 250psi. My table top is 78" x 1.75 along the seem which calculates to 34,000 psi required. Any suggestions? Below are photos showing my clamp base with built in panel clamps and a frame to support the pipe clamps. 10 clamps will be from underneath and 7 will be applied from the top.

Take what you find in FWW with a grain of salt. I don't know Roman, but I can tell you that the new crew running FWW lacks a lot in the "right" department, particularly when they're designing experiments (the "which joint is stronger" comparison from last summer is a glaring case in point).

It may well be that Roman is correct from an industry side of things - more than likely much development work has been done by engineers/scientists to determine the least amount of glue and clamping pressure possible to produce an acceptable joint. However, those results aren't really applicable to home shops and custom built furniture.

In particular, you are very unlikely to subject the joints in your table top to any appreciable force, particularly if you properly allow for expansion/contraction of the top with seasonal changes in humidity. What you should be more concerned with is what George mentioned - how straight are your edges, are they either match-planed or dang near exactly 90 degrees to the face, and are your panel clamps properly set-up to correctly align the boards.

On that last point, your panel clamp bars should not be perfectly straight when they're not under compression. Instead, you should have about 1/16" - 1/8" of convex curve across their length. The reason for this is that the wood will flex when they're put into compression, and if there's no convexity on their edges, most of the force you will be putting on the glue-up will be at the tabeltop edges, which defeats the purpose of the panel clamp.

A good way to introduce this convexity is with what David Charlesworth calls "stop shavings", where you "land" the plane a copule of feet or so from the end and take a shaving all the way off of the edge. Then repeat the process except "land" the plane closer to the end, etc... The final shaving should be the entire length of the panel clamp edge with a short plane (such as a smoother), which will take out any arrises left from the stop shaving process.

No, you don't need any more clamps - you've got plenty.

[David Christopher]

with 78" of lumber and 17 clamps, thats a clamp every 4 1/2 "....that will be more than enough if your edges fit

[James Taglienti]

The FWW clamping pressure article from a few years ago is a complete joke. There are a number of instances where the recommended number of clamps wouldn't even fit on the workpiece.

I'm also utterly unimpressed with their entire issue this month and am considering cancelling my subscription.

[Greg Crawford]

Here's what Titebond recommends for Titebond III;

http://www.titebond.com/IntroPageTB.ASP?UserType=1&ProdSel=ProductCategoryTB.asp?prodcat=1

Much different that the specs you mention. One problem many people introduce is starving the joint of glue with too much pressure. I think the previous advice is very sound.

[Tom Henderson2]

Greg-

I couldn't find clamping pressure recommendations in the site you linked... could you point me in the right direction?

-TH



Here's what Titebond recommends for Titebond III;

http://www.titebond.com/IntroPageTB.ASP?UserType=1&ProdSel=ProductCategoryTB.asp?prodcat=1

Much different that the specs you mention. One problem many people introduce is starving the joint of glue with too much pressure. I think the previous advice is very sound.

[Sean Hughto]

Good grief, you have plenty of clamps for the job. At just over 6 feet long, and having 17 clamps, you'll have a clamp every 4 or 5 inches, if my math is right. That's more than enough to yield a perfect joint (as with any joint - assuming you've prepared your edges well).

[Frank Drew]

Since Dr. Roman Rabiej has spent the last several decades studying and teaching wood technology, particularly as applied to furniture manufacture, I have to think that he's not a complete idiot and that something is being lost here between his clamping pressure recommendations quoted in FWW (250psi) and what that actually means in terms of type and number of clamps used.

[Tony Shea]

That's a first for me of ever hearing any such calculation. Way too much thought is being put into simple procedures such as glue-ups with articles like the one referenced to blame. Like has been stated above, most of your energy and concern should be focused on your joint and how well they come together without clamps. It's strange they come up with this insane calculation yet in a couple articles back you see a guy writing an article about using tape and rubber bands for clamps. How much pressure can one achieve with tape? Just do what you feel is right and concentrate the energy on your joints.

[glenn bradley]

I gave up crushing things together in favor of good fitting joinery some time ago. As others have said; if your joints fit dry, adequate glue and good clamping will result in a joint that is stronger than the wood in many cases.

[Jim Koepke]

I think for the job at hand you have plenty of clamping.

As far as the having to have this or that pressure on the clamping, the article misses by a mile. Not all glues need clamping, some will not work without clamping. As others have said, a joint can be over clamped and become glue starved.

Now, back to what my thoughts ran to when the question was first seen, "how many clamps?" There can never be too many in the shop. If 17 are being used to glue up one piece, what are you going to do if you want to glue up something else?

jim

[george wilson]

I do have to think that 250 psi would starve the joint for glue. Think about it.

[Greg Becker]

Here's what Titebond recommends for Titebond III;

http://www.titebond.com/IntroPageTB.ASP?UserType=1&ProdSel=ProductCategoryTB.asp?prodcat=1

Much different that the specs you mention. One problem many people introduce is starving the joint of glue with too much pressure. I think the previous advice is very sound.

My joints are as close to perfect as I can get. Running a straightedge along the length gives me a little more "rub" on both ends than in the middle - thus indicating slight concavity (but not to the point of visible light under the straightedge). They are also 90degrees to the "top" surface of the board.

As to the panel clamps, Mine are jointed straight and parallel to each other. I will compensate with very thin cauls in the center.

Thanks for the link to Tightbond III. They call for 150 to 200psi for cherry so I am there in terms of pressure (even though I am spaced closer than their 8 to 12 inches specified in the same Product & Application Technical Guide.)

Thanks to everyone for the sound comments and advice. I will do a dry run to check the panel clamps and visual alignment.

Now, any comments or suggestions from anyone on gluing top side up (to see what's going on with alignment) or top side down (so the "as close to perfect as I can get" flat surfaces are resting on the flat clamping surface)?

As always, thanks to everyone for the time and advice. This is a great forum.

[george wilson]

Not sure what you are asking.To keep the top flat,I'd put clamps on both sides of the top,so it stays flat and doesn't open up the seams.

[James Taglienti]

I read later that his main motivation for writing that article was to prove that it is nearly impossible for "joe woodworker" to starve a joint of glue, no matter how many mechanical clamps he used.

[Frank Drew]

Thanks for the followup information, James.

Greg, clamps above and below as George suggests, and use your straightedge to check for flat once the work is in the clamps.

[Kent A Bathurst]

with 78" of lumber and 17 clamps, thats a clamp every 4 1/2 "....that will be more than enough if your edges fit

Exactly, David - I'd have to go measure to see if the "throw" on the pipe clamp cranks will clear each other. That's my "math" - how many of them suckers can I toss on there and still rotate the cranks.

[Mark Wyatt]

The FWW clamping pressure article from a few years ago is a complete joke. There are a number of instances where the recommended number of clamps wouldn't even fit on the workpiece.

I'm also utterly unimpressed with their entire issue this month and am considering cancelling my subscription.

Glad to hear it wasn't just me. On the other hand I've been borrowing my father-in-law's copies of Popular Woodworking and enjoying them.

[Larry Marshall]

I'm with George and most of the others. Coming up with a magic pressure number doesn't reflect the variables of joinery, if nothing else. What kind of glue? How well do the pieces fit together? What about if you put cauls to spread the pressure between clamps? In the "good old days" hardwoods were often put together creating well-fitting edges, hot hide glue and rub joints. If clamps were used at all they were simply to hold things in place while the glue set up. The notion of pressuring joints together wouldn't make sense and evidence of that are the simple wooden clamps that hang on the walls at Williamsburg and other 'vintage' shops.

Cheers --- Larry

[george wilson]

We made some of those wooden clamps in Williamsburg. Rub joints are often used on things like drawer bottoms,where the wood is thin. Hot hide glue is used.

[David Weaver]

so, at 250 psi, and the comment from the fww article that bessey clamps can only exert 300 pounds, one needs to be every...

300/250 x 4/3 = 1.6 inches on a 3/4" thickness glue-up.

I think both numbers are goofy, though I've never found a scale to measure the bessey clamps.

What are you supposed to do if you glue up 8/4 boards, put bessey clamps on top and bottom?

Too much worrying -way too much.

[Gene Crain]

Give the pipe clamps to your neighbor...clamp rack is the way to go.

plantasymaderas dot com
Fine Custom Woodworks Since 1997

[Prashun Patel]

For gluing panels, I've become enamored with using cambered clamping cauls. You can get a couple Bowclamps for a no-brainer solution, or you can mill yrself a few out of construction lumber.

They really distribute pressure evenly and make the glue up process quick and less sweaty.

As for all dem fangled PSI calcumations, I use some more sophistated math:

Small bead of glue squeeze out along length of joint = Good Enough.

[john brenton]

To dovetail on the luthier's post (who is an awesome luthier by the way, great posts),"rubbed joints", which use no clamps, have been used for centuries. That table is going to have no where near as much pressure on it as a harps top gets when it is tuned up and it is very thin, most likely "clamped" up with nails bent over to push the sides together.

This reminds of the kind of "facts" that the government puts out. It sounds like this professor worked all of this up on a computer model or something...how do we know that he didn't forget to "carry a 1" somewhere?




According to an article in FineWoodworking by Prof Roman Rabiej, Cherry being glued on a quarter sawn face needs 250psi. My table top is 78" x 1.75 along the seem which calculates to 34,000 psi required.

Using all my large clamps (12 pipe and 5 K Body) and giving a maximum pressure (due to my hand strength) that totals out to 20,000 psi.

Do I really need to go out and purchase 10 more pipe clamps? I am almost ready to glue up the dining table top that I have hand planed 3 boards to prepare. The final top will be 82 x 42 (including breadboard ends) and is my most important project to date. This is beautiful cherry that deserves to be done right.

Any suggestions? Below are photos showing my clamp base with built in panel clamps and a frame to support the pipe clamps. 10 clamps will be from underneath and 7 will be applied from the top.

[Adam Cherubini]

PVA requires high pressure to develop full mechanical properties. 100-300psi is the range depending on the wood. If you don't produce the full recommended pressure, Franklin can't (or won't) tell you what your properties are. Rest assured they aren't nothing. But you should be shooting (no pun intended) for high pressure and that means a lot of very tight clamps.

The problem in your logic is the supposition that:
a) you need 250psi across the entire joint - you don't
b) that any number of clamps will give you that (as a consistent pressure)- it won't

What is really going to happen is the wood will deflect around the clamps such that you'll get higher pressures where the clamps are and lower pressure between them.

My recommendation is to spring long joints. In this case, you get almost no pressure in the center and a ramp of increasing pressure toward the ends. The shape of the ramp is roughly linear except for the addition of added clamps.

Failing that, I recommend you simulate a sprung joint by concentrating the clamps you have on the ends of the assembly in hopes that you'll develop properties where you need it most.

Wood shrinks as it dries and that shrinkage happens first at the ends where end grain is exposed. You can try to limit the about of exposure with finishes or breadboard ends. But invariably, you may see the joints open at the ends.

For sprung joints, as the boards shrink, they must first overcome the preload in that area. Only after that preload is used up will the joint open.

In your case, you won't have preload. But at least you'll have decent tensile properties. In reality, the tensile properties probably won't be good enough, but they'll be better than nothing (which is a little less than you'd get with a PVA rub joint).

Note that PVA is unlike hide glue or epoxy, neither of which like high clamp pressures. In industrial applications, they include material to prevent glue starvation. I believe this is why folks in the past thought roughened (like rough sawn) surfaces produced superior glue joints. This is not the case with PVA glue.

Good luck with your table and don't worry too much. Wood moves and joints open up sometimes regardless of what we do. God didn't make trees so we could build stuff out of them. Wood is a very imperfect construction material.

Adam

[Gaz Palmer]

This reminds of the kind of "facts" that the government puts out. It sounds like this professor worked all of this up on a computer model or something...how do we know that he didn't forget to "carry a 1" somewhere?

I wonder how on earth we all managed before "scientists" became involved in woodworking and cabinetry. :p

[harry strasil]

old fashioned doweled joints, tongue and groove and those really long biscuits called splines work for me, and I make the grain in my splines go 90 to the grain of the pieces I am glueing together as well as common sense in applying clamps and the amount of pressure with at least 2 cambered caul clamps. just my 2¢

[Rich Dill]

I wonder how on earth we all managed before "scientists" became involved in woodworking and cabinetry. :p



We were all doing it "wrong":rolleyes: