LumberJocks

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

  • Advertise with us
View Nicholas Hall's profile

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

by Nicholas Hall
posted 12-06-2012 05:46 PM


« prev 1 2
88 replies

88 replies so far

View Nicholas Hall's profile

Nicholas Hall

348 posts in 857 days


#51 posted 12-07-2012 10:50 PM

@ lumberjoe: I’m not suggesting you lag bolt the workbench top. Don’t do it, it is’t necessary. No-one else should either. If all you used was a half a dozen pipe clamps you would probably still be fine. The surface area of the glue-up is so large, it’s own weight should be enough if you set it on edge. You will be planing a 20 pound board on a workbench built to his hold a 20,000 lb load. I’m not suggesting that reaching theoretically optimal clamping pressure is necessary. I’m saying that “I wonder how it might be accomplished?”.

-- 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 HorizontalMike's profile

HorizontalMike

6968 posts in 1665 days


#52 posted 12-07-2012 10:57 PM

”...So that equals 2100 lb of pressure PER CLAMP and NOT for the board’s length! And NOT PSI for the board’s area either (unless only one clamp is used)....”

I screwed up. The above is NOT true. The second sentence should be eliminated. Sorry.

-- HorizontalMike -- "Woodpeckers understand..."

View patcollins's profile

patcollins

1004 posts in 1616 days


#53 posted 12-07-2012 11:02 PM

I’ve been thinking about the mechanism of what happens when you clamp two perfectly fitting boards together. Glue is probably an incompressible fluid, like every liquid I know about, so when you clamp the boards together either the glue squeezes out, creates a pocket in the board by actually deforming the board (it has to go somewhere), or it is forced into the pores of the wood. Because the pores of the wood are pretty small and that glue is pretty viscous I dont think there is time for much to be forced into the pores of the wood. So what would matter would the the thickness of the layer of glue between the boards, as far as I know the properties of the glue do not change with pressure (its not that high tech), and the coverage of the glue. Personally I have never been one to clamp the hell out of a glue up because the harder I clamp the more likely the boards are to shift, hydroplane like a tire on a water covered road if you will.

I’ve been thinking about glueing two boards together using moderate pressure, measuring the outside dimension with a good set of calipers then really cranking the pressure up and measuring again. Unless clamping harder makes the layer of glue thinner I can’t possibly see how more than just moderate pressure increases the joint stremgth of properly faced boards

View Grumpymike's profile

Grumpymike

1187 posts in 1066 days


#54 posted 12-07-2012 11:29 PM

Pat, I think that you hit it on the head with your last statement. I truley believe that moderate pressyre till the glue squeezes is the right answer.
I also think that Nicholas has invented a new clamp (to sell to us) with a pressure guage so that we can have joints superior to the modern glues that we use.
I have pried apart joints that I have used modreate clamping pressuer on and destroyed the wood not the glue joint.
But they are entitled to their opinion as well as I.

-- Grumpy old guy, and lookin' good Doin' it. ... Surprise Az.

View BobLang's profile

BobLang

104 posts in 2151 days


#55 posted 12-08-2012 01:24 AM

I’ll never forget the day five years ago when that article was first published. The sound alone of the hundreds of thousands of glueups, inadequately clamped over the last several hundred years coming apart was deafening, not to mention the collateral damage as workbenches, panels and table tops around the world suddenly fell apart. Glad I had my boots on and don’t believe everything I read.

-- Bob Lang, http://360woodworking.com/

View Don W's profile

Don W

15559 posts in 1318 days


#56 posted 12-08-2012 01:58 AM

Take 2 pieces of wood, plane the joint, glue it, stick it together, go back tomorrow and try to break it apart. The we’ll talk science.

-- Master hand plane hoarder. - http://timetestedtools.com

View Dan Krager's profile

Dan Krager

1742 posts in 985 days


#57 posted 12-08-2012 02:05 AM

DonW, don’t you read carefully? It’s dangerous to generate that much force (to break a glued joint) in a home work shop! (he says, tongue in cheek :)
DanK

-- Dan Krager, Olney IL http://www.kragerwoodworking.weebly.com

View Nicholas Hall's profile

Nicholas Hall

348 posts in 857 days


#58 posted 12-08-2012 04:42 AM

It’s odd that Titebond overestimates the pressure that PVA glues work best under, just like the glue joint researcher did. Neither or them must have much experience with glue. Where do they come up with these crazy numbers?

All sarcasm aside, I agree with you guys that just using the pipe clamps that you have around the shop is a perfectly acceptable solution. I don’t think that there is anything wrong with it, or that a bench will fail because you only got 10 pipe clamps on it. That said, I got into this hobby because I like to build to the highest standard of quality that I can. I could use glues and screws for all my joinery, but I choose mortise and tenon because I believe that it’s the best I’m able to do with my own two hands. If I wanted adequate, I could go to IKEA and just buy a cardboard Morris Chair. I’m interested in glue pressure for the same reason I’m interested in mortise and tenon.

- The scientist says 1200 psi is optimal
- The glue company says 250 psi is optimal
- My 10 clamps gives me 26 psi…

I can do better than 26psi, and I’m trying to think creatively about how that might possible, practical, safe, and affordable. Most likely, I’m just going to go to the store tomorrow and buy a half dozen pipe clamps. Not because it’s necessary; just because you can never have too many clamps…

-Nick

-- 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 857 days


#59 posted 12-08-2012 04:52 AM

.

-- 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 jcwalleye's profile

jcwalleye

295 posts in 1824 days


#60 posted 12-08-2012 04:58 AM

Ok, here’s the definitive number of clamps needed for a 3”x84” hard maple face grain joint.

Let’s see that’s, err, 22 clamps of different make and model.

Glad to have resolved this.

-- Trees, a wonderful gift --Joe--

View derosa's profile

derosa

1557 posts in 1586 days


#61 posted 12-08-2012 06:08 AM

I’ve use Don’t method a number of times, just apply the glue, wiggle the boards together till there is just a tiny film on each surface and walk away. I’ve usually done this with 1/4” board glue ups where clamping causes too many issues with movement and bowing the wood. Glue lines virtually disappear and the wood snaps long before the glue line, gravity does all the work and that is more then enough clamping pressure.

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

View Al Killian's profile

Al Killian

85 posts in 1376 days


#62 posted 12-08-2012 07:46 AM

I always glue up table tops and stuff like that with one clamp 4 inches in from the edge then every 7”. Never had a problem with bad joints. I have several Bow clamps ranging from 2’ to 8’.

-- Owner of custom millwork shop

View RonInOhio's profile

RonInOhio

720 posts in 1615 days


#63 posted 12-08-2012 07:46 AM

Interesting thread but that glue guru sounds a bit condescending . I would think that using too much clamp pressure would force too much glue from the joint, thus starving the joint of glue.

Again, interesting thread but, I don’t think clamping glue joints should be made out to be brain surgery.

I think the old addage, “you never have enough clamps” pertains to gluing up multiple things at once .Or needing specialty clamps and jigs for different glue set-ups.
Not because you need to apply an ungodly amount of clamps to one glue-up project.

View HalDougherty's profile

HalDougherty

1820 posts in 1988 days


#64 posted 12-08-2012 11:53 AM

Nick,

When I started making laminated gunstocks, I talked to an engineer at Franklin glue. (tightbond) He advised me to clamp using 250 psi for laminated hardwoods. So, I built a hydraulic press that will clamp a board 55” X 8” and 4” thick. That’s 440 square inches of surface area. To get 250 psi per square inch it takes 110000 lbs of pressure distributed over the surface of the board. So, I built a press using bottle jacks for pressure. I make sure the boards I use to laminate are flat and smooth before I build my laminate and I drill a 1/4” hole and use a dowel shorter than the boards to allow for compression and to keep them from moving while I’m applying pressure. It takes about 15 minutes to apply the glue, put the assembly in the press and pump up the pressure. No, I don’t have any way of measuring the pressure applied, but the engineer assured me, I’d never have a glue starved joint. If I was using epoxy, glue joint starvation might be an issue. Epoxy is the only glue that fills voids and still makes a strong joint if the surfaces are not perfect. I built my temporary press using recycled wood from a construction project and it worked great. I’m still using it and I’ve not taken the time to build a more finished press out of better materials. My total cost for the press was $60. The commercial presses I found that would do the job were from $3500 to $10,000…

-- Hal, Tennessee http://www.first285.com

View Nicholas Hall's profile

Nicholas Hall

348 posts in 857 days


#65 posted 12-08-2012 02:19 PM

Hal,

That’s pretty awesome. It would be really great if you could post a picture of the press. It sounds like a really cool idea, particularly in a repetitive operation like custom gunstock making where the rough dimensions of the glue-up are relatively consistent. It sounds like you saved a pile of money vs a commercial press!

I’m building a rough model in Sketchup of an adjustable frame that consist of two 2” thick x 48” wide x 72” long frames with 5/8” acme threaded rods every 3 feet or so. The idea would be to snug up the frame to the panel being clamped with the acme rod, ensure that the frame was parallel, and then use the bottle jack to exert final pressure. Suffice to say it needs a bit of work; I’m guessing the press you built is probably a much better thought out solution.

Once again, for the folks who think this is ridiculous: I’m not suggesting that this level of pressure in necessary for a quality woodworking project anymore than I think that a mortise and tenon joint is a necessity given that a lag bolt or wood screws have equal holding power. It’s just something I think is really interesting. I’m a tinkerer.

-- 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 Dan Krager's profile

Dan Krager

1742 posts in 985 days


#66 posted 12-08-2012 02:29 PM

@Nickolas…the kid is learning! :)
DanK

-- Dan Krager, Olney IL http://www.kragerwoodworking.weebly.com

View Don W's profile

Don W

15559 posts in 1318 days


#67 posted 12-08-2012 02:44 PM

@ Nickolas
Planing on gluing up a bunch of that lumber?

Its cool to tinker with this kind of stuff. You wind up learning a lot about wood, and the fact that some of these scientist write stuff to sell magazines.

I’ve had some glue joints fail. I can explain every one and most have been trying to glue dirty joints because I was in a hurry or didn’t care. I’ve also let the glue freeze in my unheated shop before it had a chance to dry, but it had to be REALLY cold. Since we’re in the same climate, it could happen.

Keep us posted. I like to know how much of the article is really BS and how much actually has some merit.

-- Master hand plane hoarder. - http://timetestedtools.com

View Nicholas Hall's profile

Nicholas Hall

348 posts in 857 days


#68 posted 12-08-2012 03:13 PM

Dan: Story of my life… I’m a slow learner but I get there eventually :)

-- 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 Manitario's profile

Manitario

2378 posts in 1634 days


#69 posted 12-08-2012 03:21 PM

Nicholas: Thank you for posting this thought-provoking article! The assumption you make in your analysis is that the pressure induced by a single clamp only acts on the wood surface directly underneath the clamp pad. Following this assumption, total clamping pressure would be more important than even clamping pressure, ie. if theoretically you could have one massive clamp induce 384000lbs of force at one point along the joint that would fulfill Rabiej’s requirements. If force radiates out from the clamp pad (as you illustrated above) and a single clamp can induce 1050 psi of pressure, ensuring proper clamping pressure would involve placing the clamps so that their lines of force intersect.

-- Sometimes the creative process requires foul language. -- Charles Neil

View patcollins's profile

patcollins

1004 posts in 1616 days


#70 posted 12-08-2012 03:39 PM

Im wondering how the figure of 1050 lbs of force by the average wood worker was determined, did he grab a sample of woodworkers and average them or did he just call himself an average woodworker?

I would also like to see a chart of the strength of a glue line vs pressure applied. If the chart was a very sharp parabola or mostly a straight line with a small crown on it makes a whole lot of difference here. I’m guessing that it was a very wide rounded bump.

View vipond33's profile

vipond33

1405 posts in 1248 days


#71 posted 12-08-2012 03:56 PM

+1 on patcollins’ post #53. Well put.
PVA works on a molecular level, not mechanical. When the wood molecules and the adhesive molecules are cross linked then the object is to get those two areas as close together as possible. What you are doing is effectively linking the wood molecules from each board together. The adhesive is just the facilitator. It’s integral strength is pathetic by comparison, hence heavy glue lines are weak as well as ugly.
Two things are necessary. Wet both surfaces (which must be accurate in relation to each other), and bring them together quickly with no film layer in between. This is easily achieved with only modest pressure, witness the long term success of hand rubbed joints that rely only on suction for their pressure. In the tens of thousands of joints that I’ve made in my working life, any failure that I’ve ever had was due to my failure to observe these points.

An interesting picture on the rear of the CD “Passion: Music for The Last Temptation of Christ” by Peter Gabriel.
This is an electron microscope photograph of PVA.

-- gene@toronto.ontario.canada : dovetail free since '53, critiques always welcome.

View Manitario's profile

Manitario

2378 posts in 1634 days


#72 posted 12-08-2012 04:13 PM

Gene: Peter Gabriel + PVA = awesome

-- Sometimes the creative process requires foul language. -- Charles Neil

View patcollins's profile

patcollins

1004 posts in 1616 days


#73 posted 12-08-2012 06:17 PM

I don’t know why I didn’t mention this earlier but the thing you have working against you with the clamping pressure is working for you for the strength of the glue up…..the large surface area.

Titebond claims about 3600 psi strength on the glue, so lets say that you only get half of that 1800 psi. Now you have 384 sq inches of surface that means that the yield to pull apart those two boards would be 691,000 lbs, twice that if the full strength of the glue is realized.

View exelectrician's profile

exelectrician

1760 posts in 1178 days


#74 posted 12-08-2012 09:01 PM

Okay – To answer the question ” are most bench tops woefully under clamped during glue up?”

My definitave answer – No they are not!

-- Love thy neighbour as thyself

View Kelby's profile

Kelby

133 posts in 1162 days


#75 posted 12-09-2012 06:21 PM

Clamping pressure is critical if you do not have perfectly jointed boards. If the joints are perfect, it is not an issue. I often do glue ups without clamps at all. I’ve never had a failure doing so—the wood sometimes breaks, but not the joint.

-- Kelby

View Sylvain's profile

Sylvain

590 posts in 1250 days


#76 posted 12-09-2012 07:06 PM

I understand to make
Glued laminated beam to be used in building you would want to have the optimal pressure.

Just a simple question :

how do you apply clamping pressure in a tenon and mortise joint?

-- Sylvain, Brussels, Belgium, Europe - The more I learn, the more there is to learn

View patcollins's profile

patcollins

1004 posts in 1616 days


#77 posted 12-09-2012 07:11 PM

Sylvan you rely on the joint itself to provide the compression, because as soon as you provide compression to one side it makes the other side loose.

Most laminated beams that I have seen used in construction use nails to hold the boards together. My house has a beam that is 4 2×12’s joined together by nails and the ends of the boards all offset from each other.

View HorizontalMike's profile

HorizontalMike

6968 posts in 1665 days


#78 posted 12-10-2012 12:24 AM

All I know is that I live in a glass house, and that all is safe… Oh yeah, and my glue joints are perfect…

LOL! ;-)

-- HorizontalMike -- "Woodpeckers understand..."

View knockknock's profile

knockknock

252 posts in 924 days


#79 posted 12-10-2012 01:10 AM

“how do you apply clamping pressure in a tenon and mortise joint?”

If you are not draw-boring it, you clamp it so the shoulder joint closes.

View casual1carpenter's profile

casual1carpenter

353 posts in 1226 days


#80 posted 12-10-2012 04:16 PM

I asked this question of a Sr. Technical Specialist at “Franklin International” the producers of Titebond Products, I have included his response below.

“The gentleman is theoretically correct, that in the case of very thick pieces of wood, a lot of pressure is needed to bring wood together so there are no gaps between the boards.  I think the issue below is, he is not giving the dimension that is needed for estimating clamping pressure.  Since the face of the board is 4 inches thick and 96 inches long, he is missing the critical gluing dimension of how wide is the board.  The width of the board is the dimension that needs the clamping force to bring the boards together.  A board that is only 1 or 2 inches wide will need a lot less force than one that is 4 inches wide.  That is why you see a lot of benches made from small width boards.
 
But let’s get away from the theoretical clamping forces and let me explain why we clamp wood when gluing.  Most wood glues are approximately 50% water.  This means that when they dry, they will shrink in size.  Clamping is needed to keep the boards in contact with the wood glue as it shrinks, dries and builds strength.  Since we typically glue one side of a board, that board is gaining moisture and swelling.  This will cause the wood to want to warp.  The clamping is needed to keep the wood from warping apart during this process.  The theoretical pressure is a gauge of just how strong a thick board will want to move.
 
If the boards are flat sawn, they may want to move quite a bit and extra clamping pressure (as close to theoretical as can be gotten within reason) may be needed.  Quarter sawn boards tend not to warp as badly and less clamping pressure may be needed.  The key is to keep the boards together during the drying process, no more and no less.  I cannot give you the exact numbers as this will vary with species and type of cut and how the grain runs in the board.”

View Nicholas Hall's profile

Nicholas Hall

348 posts in 857 days


#81 posted 12-10-2012 08:22 PM

I got really curious over the weekend and decided I’d pony up the money to download a pdf of the original Wood and Fiber Science journal article by Roman Rabiej: The Effect of Clamping Pressure and Orthotropic Wood Structure on Strength of Glued Bonds. Interestingly, Rabiej was a research scientist a Titebond for many years before entering academia, so he has a fairly credible perspective. The best thing about the paper is that it plots the curve of joint strength as a function of clamping pressure.

As Steve Peterson Pointed out earlier in this thread, the important thing is not the optimal number, but the actual plotted curve between clamping PSI and the actual strength of the glue bond. Put differently, the “optimal” clamping PSI is much less important than the adequate PSI for practical purposes. The purpose of Rabiej’s paper was to plot the curve between clamping psi and joint strength for maple lumber and ponderosa pine lumber using PVA glue, taking into consideration whether the wood grain was flatsawn (tangential) or quartersawn (radial). He started first with just a “rub joint” at 0 psi, and then testing at regular intervals all the way to 3000psi. The results were fairly conclusive, particularly for workbenches. Here is a graph of the glue strength curve as a function of clamping PSI for Maple (radial=quartersawn glue face and tangential=flatsawn glue face):

Summary:

1. The theoretically optimal clamping pressure is the point at which the strength of the bond exceeds the strength of the wood. This clamping pressure corresponds with the table from Fine Woodworking magazine multiplied by 2 (ie 2400 psi for flatsawn Maple glue joint).

2. It is possible to increase the strength of glue bond 50% by using the “optimal pressure” recommended in the Fine Woodworking article, rather than a 0 PSI rub joint in the case of Maple.

3. In the case of maple, a rub joint on a flatsawn joint face creates a joint strength of 25MPa (3600psi) shear strength. A joint at the optimal clamping pressure of 2400psi yielded a joint strength of 36MPa (5200psi). Translated to the 384 sq in surface area of the workbench top glue joint, this would mean that 0 psi will yield a joint strength of 1,382,400 lbs, whereas optimal clamping pressure will yield a glue joint with a strength of 2,000,000 lbs.

Conclusion:

Workbenches are indeed woefully underclamped during glue-up if the bench will subsequently be used for projects that incorporate decorative dynamite, and occasional detonations occur. In any other imaginable case, workbench tops are sufficiently clamped during glue-ups if more than a half a dozen clamps pipe clamps are used; where the primary purpose of these 6 clamps is to ensure that excess glue is squeezed out of the joint, and that nothing shifts during glue-up.

-- 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 patcollins's profile

patcollins

1004 posts in 1616 days


#82 posted 12-10-2012 11:18 PM

Just what I thought. And they charge people to tell them something that they didn’t need to know in the first place…...

View Steve Peterson's profile

Steve Peterson

263 posts in 1833 days


#83 posted 12-11-2012 05:47 PM

Nicholas, That is exactly the graph that is needed to solve this problem. You can get the absolute maximum glue strength at 2400psi. And you can still get about 66% as much glue strength at 0psi as long as there is complete wood to wood contact.

For an average woodworker, the proper clamp pressure is whatever is needed to bring the wood together. Additional pressure will only result in a slight increase in glue strength. A low pressure glue joint is still going to be stronger than the wood.

Now I can rest knowing that my workbench is not going to fall apart.

-- Steve

View HorizontalMike's profile

HorizontalMike

6968 posts in 1665 days


#84 posted 12-11-2012 06:30 PM

Steve,
You might even get to rest ON your workbench, as well… ;-)

-- HorizontalMike -- "Woodpeckers understand..."

View Sylvain's profile

Sylvain

590 posts in 1250 days


#85 posted 12-11-2012 08:03 PM

My question was with a tong in cheek.
It was just to show that the theory is not applicable for such a joint.

Although the gluing is effective for those joints.

-- Sylvain, Brussels, Belgium, Europe - The more I learn, the more there is to learn

View Nicholas Hall's profile

Nicholas Hall

348 posts in 857 days


#86 posted 12-12-2012 01:53 AM

Steve: Yeah, the chart is pretty helpful. I looked everywhere on the web for this information but I wasn’t able to find it anywhere. It was worth the money to buy the article just to get this chart. It makes me feel a lot better about using my 10 clamps for my bench glueup!

-- 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 bondogaposis's profile

bondogaposis

2753 posts in 1102 days


#87 posted 12-12-2012 02:15 AM

When I build my bench. it’s going to be a split top with each section being 12” wide. That’s about 6 boards per section. Would it be a terrible idea to recess lag bolts every 12” or so on the 4 inner boards before clamping?

You’ll have to keep them in mind when you drill for things like holes for holdfasts. It would be easy to forget that they are there.

-- Bondo Gaposis

View vipond33's profile

vipond33

1405 posts in 1248 days


#88 posted 12-12-2012 03:31 AM

From casual1carpenter 80. “The key is to keep the boards together during the drying process”

One thing not mentioned much about getting great joints, dead tight and strong – is to give them some serious drying time.

At work we take clamps off quickly and then handle gingerly, we use heated cauls, anything to keep moving. With critical stuff though, it dries under clamps overnight and there’s no point anyway in trying to scrape or sand the joint earlier on as it is moisture raised, and will shrink back to show a hollow on the finished surface. The strength also rises dramatically while you sleep.
I read somewhere that when irreplaceable violins etc. are repaired and re-glued they use shaped sand bags and highly specialized light pressure clamps, and leave them on for 2 weeks.

As we are serious amateurs here for the most part or just doing this as an extension of our daily work – we got lots of time. Nailing a good glue-line is real satisfaction, and there’s been a ton of good info offered here.
gene

-- gene@toronto.ontario.canada : dovetail free since '53, critiques always welcome.

Have your say...

You must be signed in to reply.

« prev 1 2
88 replies


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