LumberJocks

Are Most Benchtops Woefully Underclamped during Glueup? (probably not)

  • Advertise with us

« back to Woodworking Skill Share forum

Forum topic by Nicholas Hall posted 12-06-2012 05:46 PM 2331 views 2 times favorited 88 replies Add to Favorites Watch
View Nicholas Hall's profile

Nicholas Hall

348 posts in 762 days


12-06-2012 05:46 PM

Hi All,

In the course of researching my split-top Roubo workbench build, I came across an article “Get Serious About Clamping” in Nov/Dec 2007 issue of Fine Woodworking written by Roman Rabiej, who is an avid woodworker and a professor of industrial engineering at West Michigan University. This is the guy Titebond calls when they have a glue question. Roman spent an inordinate amount of time testing glue joints for failure using a engineering testing equipment to determine the recommended glue clamping pressure as a function of wood species and grain orientation.

Here is quote from the introduction of the article:

“Most woodworkers have only the vaguest idea of how much
clamping force to apply when gluing boards. Even those perfectionists
who rely on dial calipers and feeler gauges when cutting
and planing wood often judge clamping pressure simply by
the amount of glue that squeezes out. The results are occasional joint failures
and embarrassing gaps between boards on the ends of tabletops.
During my career in wood technology I’ve done scientific
studies of glue joints using different types of glue, different
clamping pressures, different species of wood, and even different grain
orientations.”

Obviously I can’t really post the whole article because that wouldn’t be right. If you subscribe to Fine Woodworking online access you can get the article, and its definitely worth a read. If you don’t, you should, it’s only $3.00 per month and you get access to every article written and tool review from the last 30 years! The tool reviews alone have paid for 10 years of subscriptions by helping me to get way better tools for much less money. Anyhow here is the table of clamping pressure:

I was actually quite surprised that the species of wood can vary the required clamping pressure by 1000% (in the case of flatsawn Ponderosa Pine vs Sugar Maple). I was also quite surprised to learn that the grain orientation can double the required clamping pressure. Now, let’s assume that someone who has tested tens of thousands of glue joints like Dr. Rabiej has learned a thing or two about gluing. What does this mean for the glue-up of a workbench top?

In my case I’m doing a 4” thick by 96” long top, so the total square inches of glue face is 384 sq inches. The boards I have are flatsawn, with the flatsawn faces in the glue joint. I’m using ash, which is most similar to red oak in density and grain structure, so we look at the chart and see that I need 900 Pounds Per Square Inch (PSI) in order to get the recommend clamping preassure. That means I need to exert 345,000 pounds of force in order to get the recommended 900psi distributed across the glue joint (900psi * 384 sq inches). Further in the article, the author shows that the average woodworker can extert 1050 lbs of force with a 3/4 inch pipe clamp (which is what I’m using). Luckily I have 10 pipe clamps already, so that gets me 10,500 pounds of force. Now I just need to find 335 more sets of pipe clamps and devise a way to get them mounted on the wood. Thats probably not going to happen…

Now I know that most people use 10-15 pipe clamps for their workbench top glue-ups because I’ve seen the pictures. I only have 10 pipe clamps myself. What I’m wondering is whether or not I should bother to get more clamps or if I should just go with the 10. After all, it’s not just the workbench tabletop that gets glued together. I do conference room tables as well that are 1.5” thick. I also often make table legs by gluing together two 2.5”x2.5”x30” pieces of 4/4. I’ve found that no matter how I prepare my glue joint (jointer + smoothing plane) I’m rarely satisfied with the glueline. I’m wondering if the problem isn’t pressure, I’ve obviously been seriously underclamping just about everything.

I can accept that I’ll never come close to the recommended pressure for the workbench top, there’s too many square inches. But for a $5,000 conference room tabletop, I really want a perfect glueline. I’m wondering if its worth it to rig a glue table up with bottle-jacks, which is the only way I can think of to generate 100,000 pounds of clamping pressure. At any rate, it’s my lunch break and this is the sort of thing I think about. I thought I’d see what you folks think over your own lunch/coffee breaks… :)

-- Outside of a dog, a book is a man's best friend. Inside of a dog, it's too dark to read. -Groucho Marx


88 replies so far

View Monte Pittman's profile

Monte Pittman

14197 posts in 993 days


#1 posted 12-06-2012 06:09 PM

Part of my concern would be the pressure needed to crush the wood. I find it easy to comprehend that different woods require different pressures or that grain orientation makes a difference. But I don’t want to ruin the wood trying to create the perfect glue joint. Just a thought.

-- Mother Nature created it, I just assemble it.

View crank49's profile

crank49

3434 posts in 1626 days


#2 posted 12-06-2012 06:16 PM

Air bags.
A 12” diameter air bag exerts 12,400 lb of force at 110 psi.
A row of 8 bags could be arranged to push a big caul along one side of a bench/table top glue-up.
That would be 99,525 lbs of force applied over an 8 ft length.
And the units could all be piped together so you would only need one valve to control it.

May be a little spendy; the bags are between $200 and $800 each.
The structural frame will be a couple grand at least.

But hey, quality costs.

-- Michael :-{| “If you tell a big enough lie and tell it frequently enough, it will be believed.” ― A H

View David Craig's profile

David Craig

2135 posts in 1764 days


#3 posted 12-06-2012 06:30 PM

I think there is a little confusion here on how this article translates. From the math, I get the feeling that there is an assumption that the clamping pressure of one clamp is only providing force for one square inch. The clamp is providing pressure across the whole board but the concentration of pressure is within a smaller radius. Keep in mind that sometimes too much pressure causes failed glue-ups. Some people exert so much force that the joint is starved for glue.

-- There is little that is simple when it comes to making a simple box.

View Nicholas Hall's profile

Nicholas Hall

348 posts in 762 days


#4 posted 12-06-2012 06:42 PM

@ Monte: It’s actually not that much more force than I typically use on a kitchen table-top. For a kitchen tabletop with a finished thickness of 1” x 6 ft long, made out of cherry, with the quartersawn grain in the glue joint, the table above would give us 250psi*72 square inches or 18000 total pounds of force exerted. That means I need 17 clamps to do a 6 ft long tabletop. I currently use 10, just because thats how many I have.

The trouble comes when you get to cherry and oak, because these have more demanding force requirements for an optimal glue joint. But by the same token, they are much stronger woods that are more resistant to the force so no wood is going to be damaged, particularly with when using cauls.

-- Outside of a dog, a book is a man's best friend. Inside of a dog, it's too dark to read. -Groucho Marx

View crank49's profile

crank49

3434 posts in 1626 days


#5 posted 12-06-2012 06:43 PM

Well, the chart in the article says pressure required per Sq.In. for oak is 900 Lbs for flat sawn face glued.

A bench top glue joint is 4” x 96” = 384 sq. in.

384 sq. in. x 900 lb required per sq.in. is 345,600 total lbs of force.

-- Michael :-{| “If you tell a big enough lie and tell it frequently enough, it will be believed.” ― A H

View Nicholas Hall's profile

Nicholas Hall

348 posts in 762 days


#6 posted 12-06-2012 06:52 PM

@crank: In my family business PRC Industrial Supply in Porland ME, we used airbags to generate pressure for the vulcanization of conveyor belts. The system used four 10”x6” H beams and an airbag. We would charge the bag to 100psi and it would exert 151,200 pounds of force on the splicing area heated to 300 degrees. That’s how you splice 6 foot wide industrial conveyor belts in paper mills. It works like a champ, but the splicing press is wee bit costly as you suggested!

@David Craig: It’s possible that I read the article wrong but I don’t think so. What he does say specifically in the article is that it is very nearly impossible to “starve a glue joint” with too much pressure, because most people have never seen enough clamps to accomplish this feat, let alone own enough clamps to do it themselves.

Pounds per square inch seems fairly unambiguous. If you have 384 square inches, and you need to generate 900 psi for a glue joint, you need to generate 345,000 pounds of force to accomplish this. At least that’s how I interpret it.

Are there any engineers out there with access to Fine Woodworking.com who can take a look? I’m really curious at this point.

-- Outside of a dog, a book is a man's best friend. Inside of a dog, it's too dark to read. -Groucho Marx

View David Craig's profile

David Craig

2135 posts in 1764 days


#7 posted 12-06-2012 06:58 PM

crank, I am not disputing your math in regards to total force, but the assumption that the clamp is only providing 1050 pounds of force. The clamp is providing 1050 pounds of force per square inch. Lets say you have 3 12 inch wide boards pressed together. The clamp optimum is in the center which is providing 1050 pounds per square inch of force. Wouldn’t you now have 37,800 (edited due to poor math) pounds of force applied?

-- There is little that is simple when it comes to making a simple box.

View Nicholas Hall's profile

Nicholas Hall

348 posts in 762 days


#8 posted 12-06-2012 07:13 PM

Here is a pic of an Almex sectional hot vulcanizing press like I used to use. This is a 6 beam unit, but the principle is the same. Those airbags put out some crazy force. Mind you, I’m not proposing doing this by any stretch of the imagination. These cost $10,000. What I’m wondering is if there is a cost effective way to generate 50,000 of clamping force, in order to maple conference room tables. It’s just a though experiment…

-- Outside of a dog, a book is a man's best friend. Inside of a dog, it's too dark to read. -Groucho Marx

View Nicholas Hall's profile

Nicholas Hall

348 posts in 762 days


#9 posted 12-06-2012 07:22 PM

@ crank: I here what your saying; it is possible that you are right. I’m no engineer, and I’ve misunderstood more than a few things in my life:). If you want I can email you the article. PM me if you’re interested.

The method for calculating the total force for the glue-up comes straight from the article though. The article’s author also references “Understanding Wood” by Hoadley, and Hoadley comes to similar conclusions regarding the calculation methods for determining total force required for a glue up.

Is there an engineer out there who might shed some light on this?

-- Outside of a dog, a book is a man's best friend. Inside of a dog, it's too dark to read. -Groucho Marx

View PurpLev's profile

PurpLev

8476 posts in 2304 days


#10 posted 12-06-2012 07:36 PM

a few things:

1. while many articles discuss not applying enough pressure, I have never seen any reference (with supporting evidence) that a joint can be starved of glue – or over-clamped.

2. there are Optimal numbers, and then there are acceptable numbers, I think what the article refers to are optimal numbers meaning BEST CASE SCENARIO. does not mean than anything below those numbers will result in joint failure. From the wording of some of those sentences one would think that 98% of all woodworking projects should have failing joints, but in reality this is just not the case.

to sum up the above, this is why you should use whatever clamps you have, and clamp those table tops as much as you can – as much as your clamps will allow you to. it’s “the best you can do”. If you mill your wood properly, the friction between the wood fibers will also reduce the chance of those joints failing, the glue is really only an additive.

don’t over think it. History suggests that it’ll be OK ;)

-- ㊍ When in doubt - There is no doubt - Go the safer route.

View bondogaposis's profile

bondogaposis

2528 posts in 1006 days


#11 posted 12-06-2012 08:05 PM

Interesting to be sure, and it supports the old adage that you never can have too many clamps. Even if you do have enough clamps you can’t generate enough clamping pressure even if you space them side by side w/ no space between them! 345,000 lbs of force?, you would have to run into it with a pickup truck to generate that much force. I think there is something lost between the research and practical application of the theory.

-- Bondo Gaposis

View DannyB's profile

DannyB

46 posts in 2077 days


#12 posted 12-06-2012 08:09 PM

Stop trying to think in terms of how many individual clamps you need.

If the goal is to exert 900 PSI evenly across the piece, you aren’t going to need 335 clamps.

You need just a few cauls and clamps to hold them

The cauls will distribute pressure evenly up to some amount
For example: http://www.woodpeck.com/ottclampingcauls.html

These say they will deliver up to 225 lbs evenly.
So you’d need exactly 4 pairs of cauls (8 clamps) to get to 900psi.

Now, that may not be possible to fit, but it’s certainly better than 160 clamps

View Nicholas Hall's profile

Nicholas Hall

348 posts in 762 days


#13 posted 12-06-2012 08:34 PM

@ PurpleLev: I wholeheartedly agree that there is an important distinction between optimal numbers and acceptable numbers. I’m not suggesting that any joint below optimal will fail. I do however have a cherry kitchen table built by my father-in-law 15 years ago that has 18” glue-line failure. The table is perfect otherwise, there is just the one glueline failure. My guess is he didn’t use enough pressure. I’ve worked with him in the shop so I know for sure he tons of glue (it gets messy). Every time I look at that table, I stare at that glue-line. It’s awful. Non-woodworking friends comment on it. It’s a $1000 table that I only keep around because he made it.

However, if I’m going to sell someone a 2” thick maple conference table that’s 8 feet long, and I know that the optimal force for the glue joints is 230,000 lbs based on the chart above, I’m not sure I’m comfortable using 10,500 lbs using my 10 little pipe clamps. That only gets me 5% of the actual pressure I need.

A good analogy is a car tire. If I look at the tire and see that the optimal pressure is 35psi, but only have a pump that can produce 2psi, should I call it good (that’s about 5% of 35psi). My guess is that even at 2 psi, the car will probably drive acceptably. Just the same, it seems like I’d be cutting it a bit close to the wire.

With all that said, I think you are right. The bench will be fine. I’m going to buy 10 more Jorgenson pipe clamps at $14.00 apiece and call it good. You can never have too many clamps. My experience using high pressure vulcanizing clamps just got me thinking. It’s unfortunately my nature to tinker and think about wierd stuff like this.

Thanks for the reality check though; I need one every once in awhile :)

-- Outside of a dog, a book is a man's best friend. Inside of a dog, it's too dark to read. -Groucho Marx

View Nicholas Hall's profile

Nicholas Hall

348 posts in 762 days


#14 posted 12-06-2012 08:45 PM

@ DannyB: The unit is pounds per square inch. The more square inches you have, the more pounds you need to generate the same PSI.

Force radiates away from clamp in a 45 degree angle. The point of a caul is to increase the surface area over which a single caul is able to apply force. They don’t multiply the force, they divide it. Take a look at the diagram below to see what I mean.

-- Outside of a dog, a book is a man's best friend. Inside of a dog, it's too dark to read. -Groucho Marx

View derosa's profile

derosa

1556 posts in 1491 days


#15 posted 12-06-2012 10:05 PM

Get half the number of clamps but get ones with significantly more strength and use good cauls. I beam clamps can apply 3 times as much pressure as a pipe clamp; by using cauls to change the location of the pressure you can really crank down on the clamps.

-- --Rev. Russ in NY-- A posse ad esse

showing 1 through 15 of 88 replies

Have your say...

You must be signed in to reply.

DISCLAIMER: Any posts on LJ are posted by individuals acting in their own right and do not necessarily reflect the views of LJ. LJ will not be held liable for the actions of any user.

Latest Projects | Latest Blog Entries | Latest Forum Topics

HomeRefurbers.com

Latest Projects | Latest Blog Entries | Latest Forum Topics

GardenTenders.com :: gardening showcase