The article regarding clamping strategies in the current issue (v.194) makes a few statements that seem contrary to what I have learned through both reading and experience.
The assertion that a K-body clamp provides less pressure than a quick-grip clamp seems preposterous. The clamping pressure achieved by the various clamp designs has been covered in several magazines in recent years; don’t remember the details but I’m pretty certain the parallel k-body style clamps were MUCH stronger than a quick grip. Both Bessy and Jorgenson rate their parallel clamps at 1000 lbs.
Also, the suggested clamping pressures seem ridiculously high. I recall Titebond recommending roughly 100-200 psi depending on species and that the tightness of the joint was more important for a strong glue joint than the clamping pressure applied. Empirical evidence suggest that with well machined joints even minimal clamping pressure results in a glue joint stronger than the surrounding wood.
Check out example 2 on p. 39. NINE pipe clamps to glue a ~18″ long 12″ wide three board maple glue-up is outrageous and completely unnecessary in my experience.
Replies
"The assertion that a K-body clamp provides less pressure than a quick-grip clamp seems preposterous."
It does. There was an article recently somewhere (maybe Wood magazine?) reviewing parallel-jaw clamps, and they threw in a couple of quick clamps and pipe clamps for comparison. As I recall, the pipe clamps exerted the most force, as you would expect, and the quick clamps exerted much less force than the parallel-jaw clamps, by a factor of at least three or four.
Then again, the numbers in the article were "averages" among three people, and it could be that someone with small hands and skinny arms can do better with a quick clamp than with one where a handle needs to be turned, enough to skew the results.
Without more details, "average" values aren't very meaningful. The glue article had the same problem. The differences between the glues were small, and without seeing all of the measurements, there's no way to tell if the differences were statistically significant, or just sample-to-sample variation.
-Steve
P.S. Does the fact that I mentioned reading Wood magazine brand me in any way? I didn't buy it, honest.
Hi Zaphod,
Boy, I hope new woodworkers don't feel like they're going to need to start measuring clamping pressure.
After working wood for nearly 40 years the only time I had a glue line failure was because of too little clamping pressure. Unless you're using pneumatic or some other form of power clamping you would really have to go out of your way, using common shop clamps, to get too much pressure, in my experience.
Too many clamps is better than too few.
Paul
Speaking of too many clamps; has anyone ever used these clamping cauls? http://www.bowclamp.com/
It would seem like they would work but how well? Is the clamping pressure just as good as if you used multiple clamps?
mike
Hi Mike,
The basic idea is sound. It's success comes from the correct arch in relation to how springy the material is. Years ago I experimented with some shop made cauls similar to those but never used them much. They were always to long or too short. But they did work OK. I like looking at the joint when it's going together and deciding where it needs clamping.
I always use 3/4" pipe clamps for panel glue-ups. Those wimpy bar clamps shown in the video are best used for cabinet door glue-ups ( stile and rail ). At least that's how I use them.
Paul
"They were always to long or too short." That was the first thing that came to my mind while watching the video. Is too long really a problem? for instance, if the clamps were, say, 6" longer than your stock?? The clamp can slide along to any position, but I wonder about the physics of the pressure distribution.forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl ;-)
Hi forestgirl,
You're right. Too long wasn't really a problem, unless it was so long it got in the way somehow. I just found myself adding clamps in certain spots to really pull the joint together. I haven't used those commercial ones in the video. Maybe they work better. The fact that they're designed for those smaller clamps I don't think the pressure will be enough ( at least as much as I would like ). The idea does work, though. But I'm not sold on the use of just two clamps.
Paul
ps I wish I lived further north. Love your area
I read the article last night and I agree with you. The text states the the figures in the table are recommended glueline pressures but optimum pressure is about twice as much. So to glue maple I need to exert 2400 PSI on a flatsawn maple joint if I want the best outcome. That would mean I need 48 "I"beam clamps cinched to the max to glue a 36" panel of 3/4" thick maple!!! Guess I'd better go buy more clamps then.Does the logic extend to vacuum bag work too? It's widely used and is successful with just 14 PSI. Except for the sun, I don't think there is a place in the solar system with an atmosphere able to get close to 2400 psi. Glue open times on the sun are real short!
It's just about edge joining, not veneering.
QCI-
At the bottom of the Pacific Ocean in the Mariana Trench, there is 15751 psi. There are obvious drawbacks with using this method, though.Or, maybe you can use fewer clamps if you tighten them 40% higher like the editor from FHB?Now, off to observe the severe compression on radial parts of ash (LOL).
They're probably fine for attaching edging, where the edging material is more flexible than the caul, and strength isn't much of an issue. But I can't imagine that you could really generate sufficient pressure for a good edge-to-edge joint between two boards.
-Steve
Haven't seen the article, but clamping seems to be one of the elements of woodworking prone to freakish obsession. I think that when clamps are advertised as 1000 lbs., for example, people assume that pressures in that area are somehow useful. But I think it's just oneupsmanship in the advertising game. "Ours are better becasue they can generate 1050 lbs instead of just 1000!"
Would anyone think tires that can be inflated to 100 psi are necessary. or better than those rated at 95 psi???
As someone said, time is better spent on the mating surfaces. If ya gotta crank it to 500 lbs to close the joint something ain't right!
It's not about getting badly fitted mating surfaces, it's about getting the glue film down to the optimally thin thickness in part for strength, but also, and in my opinion more importantly, to eliminate any hint of visible glue line.
Thin thickness, or is it thick thinness, or just thinness? Steve, I have long admired your experience and wisdom in woodworking. I have heard that you shouldn't put so much clamping pressure on the glue joint that you force the glue out. Are you saying this isn't so. Please explain your thoughts.
I have heard that you shouldn't put so much clamping pressure on the glue joint that you force the glue out.
Yea, there's a whole legion of woodworkers that grew up hearding Norm say that on TV. Norm's a guy on TV; he's not a scientist or chemical engineer. And given the range of recommended pressures for PVA glues given by the glue manufacturers themselves and by gov't research labs, there's just no way someone in a home shop with some clamps could apply enough pressure to strave a joint, even if such a thing is possible (with industrial hydraulic presses, I dunno).
Note this applies to PVA glue. Other types; e.g., exopy, may depend on a thick glue line in order to work.
The FWW article is far from being the first to call for significantly higher glue clamping pressures for PVA glue than the "just enough to pull the joint together" idea. The glue manufacturers want lots of pressure, Bruce Hoadley reports similar figures in Understanding Wood, and the Forest Product Laboratory does too. And the author of the FWW article Roman Rabiej has published such research in peer reviewed scientific journals, though I haven't read it. Putting that all together suggests that the more is better, up to very high limits, is the right way to go. It would have been nice to have had a side bar on the serious research, but that wouldn't have been decipherable to more than a small fraction of the readers, therefore not likely to be a good use of editorial space for the magazine.
Barry does suggest one reason how widespread the idea is, though it undoubtedly pre-dates Norm. Perhaps it is something that comes from the hot hide glue era, but I don't know that. It may just come from an intuitive idea about glue that isn't true--lots of things around like that.
I think we may have missed something here - the "i", as in inch. Pressure per square inch.
It might be that the Quick Clamps have 1/5 the surface area of the K-Body . Place 20lbs on the Quick clamp "pad" and get around 20 lbs per sq. inch.
Put 20lbs on the K-body "Pad" and get 4 lbs per square inch. (sizes are rough guesses).
And I haven't read the article yet.
Edited 10/15/2007 12:32 pm ET by Womble
"Pressure per square inch."
Actually, force per square inch (equals pressure).
Anyway, the pressure at the clamp face is not relevant. The pressure at the glue joint is what counts. And that is obtained by dividing the force (not pressure) exerted at the clamp face by the area of the glue joint.
(There are also geometrical considerations, as shown by the "triangles of force" in the article.)
-Steve
How about ...... crank it down 'till the joint closes. :)
Paul
G'day all,
I have read the article and had serious misgivings as well. The pressure does seem high and one wonders if a kick back from a manufacturer is involved - hah hah. On the realistic side it is mentioned that the pipe and bar clamps do have the mechanical advantage of a lever. I also had a hard time with the pressure from the parallel clamps. I have a pressure force gauge at work so I tested one of my Besseys. At the end of the jaw all I could get was 410 pounds force and at the bottom close to 800 pounds force over repeated trials. Unfortunately, no Quick-Grips to play with.
It would seem that the Roman Rabiej really does have something to teach us about clamping. I'm tempted to do a few scientific tests of my own before raining on his parade anymore. Like he says "Most woodworkers have only the vaguest idea...."Just my couple of pennies worth...
I guess TO ALL.
We all try to rationalize this article that was an academic response that some staffer thought would be their star article! Neither(the staffer or the academic) knows the reality of wood working thus all the answers are wrong and we should just chalk it up to another bad FWW attempt to save money by not using premium folks ---but we still pay premium $$$ in subscriptions. It bothers me to be so negative but when I pays the money I expects the the best from the best. Paddy (is this so old world that it is passe?)
Edited 10/15/2007 10:38 pm ET by PADDYDAHAT
I'd say it was in fact a star article--it has done what a good article should do, generate a response. It's unfortunate that for some that response hasn't been thoughtful. But some horses just won't drink when led to water.
Without academic type research there is NO WAY to know whether joints couldn't be improved by using higher clamping pressures. You can know your joints haven't fallen apart over the years, but not whether they would have if more vigourously stressed. You can say it was "good enough", but there is no way to say prove those standards. I'll take science over "it's good enough for my father." Cars in 1955 were "good enough" but it's doubtful that any 2007 model isn't safer and more reliable.
I know people don't want to be confused by the facts, but the facts in the article are mostly old hat--just not so widely circulated. Bruce Hoadley, in 1980, tested clamping pressures obtainable from various models, and none of the straight handled models, such as the Bessey and its competitors, generated more than about 500 lbs. and most achieved less.
Glues don't work by having a film that grabs wood fibers on one side and wood fibers on the others and holding them together while the glue hardens. They don't work at this mechanical level but at a molecular level with a chemical attachment. It's research that determines that significant clamping pressure brings the parts into such proximity so that the structure of the molecular attractions can be formed as the glues solvent evaporates or as a chemical reaction takes place. Source: R. Bruce Hoadley, Understanding Wood, Taunton, 1980 p. 176-179.
This article struck me as a solution in search of a problem. There are plenty of challenges in the panel making process -- from stock prep to edge alignment at glue-up to flattening the panel after -- but joint failure is really not in the mix if you're past the absolute newbie stage.
And I'll keep all my "weak" K-bodies thanks very much. You all can have those quick grip Irwins - LOL.
Edited 10/16/2007 10:21 am ET by Samson
Steve and All,
I'm sure all woodworkers have there own habits when it comes to glue-ups and clamping. Is it really that important that clamp X has more force than clamp Y ? Use a system that works for your glue/ clamp/ hand strength combination. Just glue up some samples and break them. I prefer, once they're cured, to break the seam over the corner of a workbench or table rather than using a chisel as in the article.
To me , this isn't rocket science. It works or it doesn't. A quick test will tell you. Could it have been a stronger joint by varying the clamping pressure slightly? Maybe. But if the joint strength is that critical, I would mechanically reinforce it.
It's not a matter of being good enough, but rather Does it work for me?
Paul
And then there are all those woodworkers who don't use any glue, eliminates the whole problem altogether.
[...] I'm sure all woodworkers have there own habits when it comes to glue-ups and clamping. [...]
Chris Scholz
Atlanta, GA
Galoot-Tools
Hi Chris'
That's right. Take a lesson from Brian Boggs and let mother nature take over. Although I didn't realize there was a problem until I got sucked into this thread :)
Paul
It works or it doesn't? A quick test will tell? I don't believe either of those statements is true. Over its life joints in a panel face a variety of stresses as its environment changes--it dries and shrinks, it swells and expands. Something happens so that it's designed allowance for movement fails--such as finish "glues" the panel to its rails. Like all engineering, it's important to build in some allowance for the unusual, and for the effects of time. It's only by testing to failure that we have a clue as to how much margin exists. Otherwise you are just guessing. Science gives some answers--"we've always done it this way" stops progress.
Since there is very, very little cost in clamping more firmly, why is there resistance to stronger versus weaker? Why the Luddite attitude about improving technique? Too hard to admit any need for change?
By the way, just how do you mechanically reinforce an edge joint in a panel? Well glued joints will outperform, over the long term particularly, all the basic mechanical reinforcements such as dowels, biscuits, and particularly metallic fasteners.
Since there is very, very little cost in clamping more firmly, why is there resistance to stronger versus weaker? Why the Luddite attitude about improving technique? Too hard to admit any need for change?
I think what folks are reacting to is the notion in the article that home shops are not possibly clamping with enough pressure right now. While perhaps the joints achieved with such "inferior" clamping may not be "optimal" in terms of being the absolute strongest they could be, they are far stronger than they need to be for their applications. We often over-engineer in our woodworking. That's great, but it isn't always necessary.
Let's take a poll: assuming that wood movement has been accounted for in the design (e.g., a 4 foot wide solid table top has not been screwed to its apron), how many of us have ever seen edge joints fail?
By the way, just how do you mechanically reinforce an edge joint in a panel?
butterfly key? invisible cleat? batten?
Edited 10/16/2007 2:06 pm ET by Samson
I haven't seen many joints falling apart, but I sure have seen visible glue lines on pieces for sale by "custom" furniture makers at shows. (My wife usually nudges me before I say anything.) That may be a result of poor joint preparation and/or insufficient clamping pressures. But it does suggest that all is not well in the realm of edge jointing.
"That may be a result of poor joint preparation"
FWIW, in my experience, if I can see any "glue" in my glue lines (as opposed to just seeing an edge joint due to grain cues) it's because of mistakes or compromises (e.g., the board would be too thin if I planed all the twist out of it) I've made in prepping the edges or prepping the stock more generally.
Yes, I believe that it is true that it is just not possible to get an invisible joint without a well fitted joint.
However, I suspect that it is possible to see glue in the glue joint even with a well fitted wood if the "just enough pressure to see the first bit of squeeze out" techique were used. As I see it, the benefit of the FWW article is to lay that aside and to encourage joints to be firmly clamped. It's not about having to measure pressures or do calculations, the message is that you really would have to try hard to exert too much pressure on the glue line.
Following the message in the article doesn't add complications it just provides assurance that you really don't have to hold back when tightening clamps. And, if it's an excuse to buy a few more clamps, there isn't much wrong with that.
What!!??!! I can go buy more clamps!!!!I think you've hit the mark. Hear the message, don't argue all the details and get on with making beautiful things that won't fall apart - built of course with an appropriate amount of clamping force.Cheers All
Ya know when I read that article, I was left scratching my head as well. It flys in the face of what I had thought (bias), so I was not surprised. The bessy .vs. quickgrip was a shock to me. Then I started thinking back. I am a hobbyist, and I have not made tons of pieces, but one of the earliest piece I made is going on 22 years old, and it has not failed at any glue line (or any other place for that matter) and all I had back then was two cheapo pipe clamps and some Jorgy 3730 clamps with the little round pressure point.
My dad was just a carpenter, but always taught me to bring pressure to bear to close the joint good (squeezeout) and then a little more (it's a gut check feel for me). While far from scientific, I have never had an issue. I will say, that I do like the bessy's better than the jorgy's, but at times I've got all clamps in use and wish I had a dozen more no matter what brand they would be. Government's view of the economy could be summed up in a few short phrases: If it moves, tax it. If it keeps moving, regulate it. And if it stops moving, subsidize it.
How well do these findings jive with the 'how strong is your glue' article where different types of glue were tested? I think they also tested the fit of the M+T joints and I vaguely recall that the tightest where the strogest. M+Ts normally aren't clamped across the cheeks.
I think that the clamping preasure that most of us are applying in general work (not mortise and tenon) is adequate to make the joint stronger than the surrounding wood. What we have learned in v. 194 is that we can make the joint even that much MORE stronger than the surrounding wood by adding more clamps. Is this necessary? In most cases no, but still good to know.
Interesting thread... perhaps we could bring in some sharpening issues and really heat things up.My own view is that requiring clamping pressures in the geological range doesn't make a great deal of sense- we're converting two boards into one panel- not sandstone into marble.It is perhaps scientifically possible that a high pressure clamp will lead to a stronger or prettier joint, but in my experience, proper joint prep is a more important factor. A well prepared joint/edge that is clamped with modest pressure will look and hold up better than a poorly prepared one that is forced together under high pressure. Perhaps a well prepared joint/edge that is clamped at very high pressure will do even better, but I think the differences are marginal.Consider an antique hardwood door- such doors >100 years old are found all over New England and elsewhere. Many of these paneled doors- made from oak or mahogany, etc were constructed of matched panels and glued with hide glue. They were constantly open and closed, slammed and had to support their own very considerable weight hanging from 3 or 4 points (hinges). Yet few failed. I doubt if their makers had access to high pressure clamping techniques, let alone modern glues. Yet the doors endured. This suggests to me that there is a considerable margin of safety in a properly fitted and glued joint, even without extraordinary clamping/glue technology. I think there are many other examples in chairs and veneers made from the Federal period and before.While I can't discount the possibility that high clamping pressures and aliphatic glues will lead to an even stronger joint, I'm not sure the additional strength is really necessary.Glaucon
If you don't think too good, then don't think too much...
Hi Glaucon,
In a lot of areas " less is more". But when it comes to clamping wood " more is more", more or less. Except when your shop happens to be on the coast of Chad :>)
Paul
Edited 10/27/2007 2:05 pm ET by colebearanimals
<"But when it comes to clamping wood " more is more", more or less.">And this is based on what evidence or data?Why not hydraulic clamps or presses? Why stop there?People make these statements but I have not noticed an epidemic of properly prepped and glued joints failing. This is a solution in search of a problem... a marketing department's dream.Glaucon
If you don't think too good, then don't think too much...
Glaucon,
Lighten up. That was my attempt at being cute.I was merely agreeing with the article in that, in my opinion, you can use more clamping pressure with pva glues than most people think, based on my 35+ years of woodworking experience. I think there is plenty of leeway in clamping pressure on how it affects joint strength. But I like higher pressure because it really makes the joint disappear.
And yes, it is a solution in search of a problem. But then you would have understood my position if you had read my earlier posts.
Paul
You've left the subject behind when you bring up hide glue. Your example of hide glue doesn't support your case at all. The article was about clamping and PVA glue... he didn't say a word about hide glue.
A study of clamping and hide glue would be interesting, it would get rejected by anyone that didn't like the results, just like this one, but those of us willing to listen would know what is best.
Edward
<"Your example of hide glue doesn't support your case at all. The article was about clamping and PVA glue... he didn't say a word about hide glue.">My example was really an analogy. The point being that 200 years ago cabinet makers had to glue up panels and veneers with hide glue, and fairly basic clamping techniques. The glue chemistry was much less sophisticated, reproducible and reliable than the aliphatic resins that we use today. Oddly enough, many of these panels survive- even in applications where they received a good deal of abuse and are under tension most of the time.But now, despite having superior glues and plentiful clamps we are being told that we need clamping pressures formerly encountered only at the birth of stars to achieve acceptable results. To this I say booooo-gus. As I said before, this is a "solution" in search of a problem.Glaucon
If you don't think too good, then don't think too much...
"Oddly enough, many of these panels survive"
And many, many more have not. This reminds me of a remark that someone I know once made regarding the "fact" that the Romans had better architectural engineering skills than we do these days, since some of their works (aqueducts, the Pantheon, etc.) are still standing, whereas many modern structures have lasted only a few decades. Of course, they neglected the thousands upon thousands of Roman works that long ago crumbled away to dust.
Anecdotal evidence isn't.
-Steve
And where is the statistical mandate for using metamorphic clamping force?Perhaps my subscription to "The Proceedings of the National Academy of Clamping Technologists" lapsed, but I missed that passel of blinded, actively controlled, randomized, adjudicated, well-powered, independently funded, peer reviewed studies and the accompanying editorials supporting your position. Glaucon
If you don't think too good, then don't think too much...
Glaucon,
Hi. I'm the President of The National Academy of Clamping Technologists. If your subscription has lapsed we currently have a special offer for all you lackies! I am prepared to offer you a 50% discount on all back issues required to get you up to date.
Furthermore I can offer you a buxsome secretary who will come to your house and speed read all of them to you in your sleep, all at no extra charge!
Regards,
Bob @ Kidderville Acres - with toooooooooooooo much free time!
A Woodworkers mind should be the sharpest tool in the shop!
Edited 11/1/2007 3:18 pm ET by KiddervilleAcres
Hi President Bob,
I would like to become a Certified Clamping Technologist. Judging by the interest it could prove to be a very lucrative new field.
Funny stuff:)
Paul
My understanding is that, according to the Board of Directors of The Society we are all certified (or is it certifiable?). :>)
"And where is the statistical mandate for using metamorphic clamping force?"
Presumably in this paper:
RJ Rabiej, HD Behm (1992) The effect of clamping pressure and orthotropic wood structure on strength of glued boards, Wood and Fiber Science 24:260-273
Note that Prof. Rabiej is also the author of the FWW article. As I mentioned earlier in this thread, I tried to locate a copy of that paper, but none of the OhioLink libraries carries the journal, and an electronic version costs $25, which I wasn't willing to spend just to silence the naysayers.
Here is the abstract of that paper; draw your own conclusions:
Reference values for compression strength perpendicular to the grain were determined for radial and tangential sections of samples of sugar maple and ponderosa pine. Samples to be glued were matched according to specific gravity and orthotropic structure and bonded along the grain in tangential or radial sections. Magnitude of clamp pressure was controlled throughout a range of pressures commonly applied in industry, up to about 80% of the compression strength of the wood sample. Tests were conducted on the bonded samples to determine glueline shear strength and percent of wood failure at the bonded surfaces. Results were subjected to regression analysis to ascertain relationships. It was determined that clamping pressure had a different effect on both shear strength and percent of wood failure depending on species and orthotropic section. It is possible to maximize joint strength by applying proper clamping pressure. Results similar in direction but differing in magnitude were obtained with both PVAc and U-F adhesives. A generalized measure of clamping pressure was defined as the ratio of applied clamping pressure to the compression strength (CP/CS) of the wood section to be glued. Using this concept, the optimum clamping pressure for sugar maple was found to be 0.3 times compression strength using U-F glue and 0.5 times using PVAc glue. This approach to determining reliable clamping pressure data can lead to improved gluing practice and more precise testing procedures.
-Steve
<"And where is the statistical mandate for using metamorphic clamping force?"
Presumably in this paper:">Ummmm, no actually.1. No results appear in the abstract. Since you have not obtained the paper, and thus not read it or the results, seems odd that you would cite it in support of your argument. 2. The abstract states: "It was determined that clamping pressure had a different effect on both shear strength and percent of wood failure depending on species and orthotropic section. It is possible to maximize joint strength by applying proper clamping pressure."How much was this "different effect"? 10%? 20%? 5%? Is this a significant difference in real terms? My basement ceiling is supported by 6" steel I-beams on the bearing walls. Wouldn't 8" be stronger? How about 12"? But does it NEED to be much stronger- or is the tensile strength of steel in supporting a residential home so high that it is a distinction without a difference?3. How well did the laboratory conditions reflect real world practice? How was the joint prepared? How dry was the wood? In the end, very, very few panel joints, properly jointed and glued fail. Asking a WWr to apply high clamping pressure may simply focus the his/her efforts on a side issue- or cause the joint to be overstressed, mis-aligned or buckled. Using proper prep, sufficient glue and reasonable pressure makes it very unlikely that the joint will fail in the real world. This is because the adhesive safety margin of aliphatic glue is high- higher than the strength of the neighboring wood. I really don't care if a joint doesn't fail because it has 3 times the strength needed to bond, or 5 times. It still won't fail. I'd rather focus my attention on the truly important tasks in joinery than in trying to achieve some aeronautical level of glue performance.Glaucon
If you don't think too good, then don't think too much...
"No results appear in the abstract."
False. Look again. Hint: Try actually reading the abstract, rather than just copying and pasting pieces of it.
Since it appears that you don't understand the purpose of the abstract of a technical paper (and possibly haven't even read this one): It's supposed to summarize the topic of the paper and its conclusions, so that one can make a reasonably informed decision regarding whether the paper would be of interest or not. It's not intended to be a substitute for reading the paper itself. If you want to know more, read the paper; don't ask me to do your homework for you.
"Since you have not obtained the paper, and thus not read it or the results, seems odd that you would cite it in support of your argument."
You're tilting at windmills and grasping at straws. I have presented no arguments, just information. You asked a question; I answered it. (And yes, I feel that it is my obligation to respond to anti-intellectual blowhards.)
It's clear that you've already traveled a long way down the road to truthiness. I'm getting off at the next stop.
-Steve
Steve,You said:<"No results appear in the abstract."False. Look again. Hint: Try actually reading the abstract, rather than just copying and pasting pieces of it.Since it appears that you don't understand the purpose of the abstract of a technical paper (and possibly haven't even read this one): It's supposed to summarize the topic of the paper and its conclusions, so that one can make a reasonably informed decision regarding whether the paper would be of interest or not. It's not intended to be a substitute for reading the paper itself. If you want to know more, read the paper; don't ask me to do your homework for you.>Actually, I do know something about technical papers. I have been in academic medicine as a cardiologist for nearly 25 years. So I not only read papers... I write them... edit them... and review them.As for my original complaint, a lack of results in the abstract, I stand by my statement. The abstract does not contain any results; it contains a conclusion. To bring you up to speed about what "results" are, I have reprinted an abstract from a a medical journal below. Notice the section marked "METHODS and RESULTS". That's a clue for you. Note that in this section, the authors have included NUMBERS and PERCENTAGES and CONFIDENCE INTERVALS. We call those RESULTS. They are logical. Statistical. Sensible. The authors did not say, we gave patients with coronary artery disease an investigational medicine. It worked pretty good. We conclude it is nice.I have published similar papers in both medicine and physics, and results is results.I don't see any mention of quantification of strength or increases in glued joint effectiveness in the abstract you supplied. You say that you did not download the original paper, so you hadn't read that either. Thus all you have is a jibe that I can't read something that is not there. The authors may have made careful measurements and used well defined methods. But we can't know that.Moreover, the problem is that it is a technical paper... preclinical as it were. It is measuring (I guess) a laboratory induced standard, not a real world outcome. If you do a search on "clopidogrel" (see below, you will find hundreds of citations for laboratory and animal experiments using this drug. But they are not the reason the drug is used. The reason for that is the clinical outcome study in the citation. That's where the rubber meets the road.*******************************
Benefits and risks of the combination of clopidogrel and aspirin in patients undergoing surgical revascularization for non-ST-elevation acute coronary syndrome: the Clopidogrel in Unstable angina to prevent Recurrent ischemic Events (CURE) Trial.
Fox KA, Mehta SR, Peters R, Zhao F, Lakkis N, Gersh BJ, Yusuf S; Clopidogrel in Unstable angina to prevent Recurrent ischemic Events Trial.Royal Infirmary of Edinburgh, Edinburgh, UK. [email protected]BACKGROUND: Antiplatelet therapy and antithrombin therapy have been demonstrated to reduce the risk of cardiac events in patients presenting with acute coronary syndrome, yet all effective therapies also increase the risk of bleeding. METHODS AND RESULTS: In the Clopidogrel in Unstable angina to prevent Recurrent ischemic Events (CURE) trial, 12,562 patients were randomized to clopidogrel or placebo in addition to aspirin, and the primary outcome was cardiovascular (CV) death, myocardial infarction (MI), or stroke. The benefits were consistent among those undergoing percutaneous coronary intervention (PCI) [9.6% for clopidogrel, 13.2% for placebo; relative risk (RR), 0.72; 95% CI, 0.57 to 0.90], coronary artery bypass grafting (CABG) surgery (14.5% for clopidogrel 16.2% for placebo; RR, 0.89; 95% CI, 0.71 to 1.11), and medical therapy only (8.1% for clopidogrel, 10.0% for placebo; RR, 0.80; 95% CI, 0.69 to 0.92; test for interaction among strata, 0.53). For CABG during the initial hospitalization (530 for placebo, 485 for clopidogrel), the frequency of CV death, MI or stroke before CABG was 4.7% for placebo and 2.9% for clopidogrel (RR, 0.56; 95% CI, 0.29 to 1.08). For the entire study, there was a 1% excess of major bleeding but no significant excess of life-threatening bleeding. Among patients undergoing CABG, the rates of life-threatening bleeding were 5.6% for clopidogrel and 4.2% for placebo (RR, 1.30; 95% CI, 0.91 to 1.95; both nonsignificant). CONCLUSIONS: The benefits versus risks of early and long-term clopidogrel therapy (freedom from CV death, MI, stroke, or life-threatening bleeding) are similar in those undergoing revascularization (CABG or PCI) and in the study population as a whole. Overall, the benefits of starting clopidogrel on admission appear to outweigh the risks, even among those who proceed to CABG during the initial hospitalization.PMID: 15313956 [PubMed - indexed for MEDLINE]Glaucon
If you don't think too good, then don't think too much...
"I have been in academic medicine as a cardiologist for nearly 25 years."
The light bulb goes on! You're an M.D.! That explains everything....
-Steve
<"The light bulb goes on!"> Dimly, to be sure.<"You're an M.D.!">And you, no doubt, play one on TV.Glaucon
If you don't think too good, then don't think too much...
Justifying why an argument is going the way it is by stating that it is due to someone being a whatever professionally is a cop out and rude... If you want to continue your argument... some of us think it entertaining I would stay away from name calling as you totally show a loss of objectivity and your cool... is it worth it? I don't think so. This was a magazine article... I am an amateur wood worker... I learned a lot I did not previously know about from it. I know my panels have held up (all 60 something of them) but am interested in doing a better job. I was taught by a cabinet maker who said throw about 3 or 4 clamps on there and twist away on the handles till the gaps close if there are any little ones in there... Plainly the instruction was adequate for making panels but the panels made could have been better I am sure after reading the article. From now on I will spend a little more time and use a few more clamps... am I gonna get a pressure gauge out? No... Am I going to go buy 40 new pipe clamps... no... Am I going to keep watching you guys swing away on here to the point that all objectivity is lost and you both avoid knots (or one does) as his or her feelings are hurt. Probably... And for the record... I am a Real Estate Broker... who used to be a general contractor... and I have a 4 year degree in economics from a pretty good school and I grew up with a wood shop attached to the house...Drew
"Justifying why an argument is going the way it is by stating that it is due to someone being a whatever professionally is a cop out and rude."
It appears that you're missing the satirical tone of the ad hominem in my posts. This particular side discussion has nothing to do with clamping pressure; rather, it's about giving arrogant, browbeating bullies a taste of their own medicine. One begins by trying to be reasonable, but that doesn't always work. Sometimes, the military option is the only viable one.
-Steve
Steve-"...it's about giving arrogant, browbeating bullies a taste of their own medicine."Can you identify examples of arrogance, browbeating, or bullying? I would not characterize Glaucon's responses to you as anywhere near those aspects of behavior. When you are wrong, and someone points out that you are wrong, that's neither arrogance nor browbeating nor bullying.Don Brown
"Can you identify examples of arrogance, browbeating, or bullying?"
Sure. Look at 38062.63 or 38062.68.
"When you are wrong, and someone points out that you are wrong, that's neither arrogance nor browbeating nor bullying."
Are you referring to his assertion that the Rabiej abstract doesn't contain any results? If so, I'm sorry, but he's got you hoodwinked. The kind of stuff he talks about there is used only in the abstracts of certain kinds of medical papers, such as reports on clinical trials. I've never seen that kind of thing in scientific (as opposed to medical) abstracts. Here are some examples, from current issues of top-flight scientific journals:
From Applied Optics:
The zero-gravity surface figure of optics used in spaceborne astronomical instruments must be known to high accuracy, but earthbound metrology is typically corrupted by gravity sag. Generally, inference of the zero-gravity surface figure from a measurement made under normal gravity requires finite-element analysis (FEA), and for accurate results the mount forces must be well characterized. We describe how to infer the zero-gravity surface figure very precisely using the alternative classical technique of averaging pairs of measurements made with the direction of gravity reversed. We show that mount forces as well as gravity must be reversed between the two measurements and discuss how the St. Venant principle determines when a reversed mount force may be considered to be applied at the same place in the two orientations. Our approach requires no finite-element modeling and no detailed knowledge of mount forces other than the fact that they reverse and are applied at the same point in each orientation. If mount schemes are suitably chosen, zero-gravity optical surfaces may be inferred much more simply and more accurately than with FEA.
From Nature ("hedgehog" is the name of a gene, by the way):
Hedgehog (HH) morphogen is essential for metazoan development. The seven-transmembrane protein smoothened (SMO) transduces the HH signal across the plasma membrane, but how SMO is activated remains poorly understood. In Drosophila melanogaster, HH induces phosphorylation at multiple Ser/Thr residues in the SMO carboxy-terminal cytoplasmic tail, leading to its cell surface accumulation and activation. Here we provide evidence that phosphorylation activates SMO by inducing a conformational switch. This occurs by antagonizing multiple Arg clusters in the SMO cytoplasmic tail. The Arg clusters inhibit SMO by blocking its cell surface expression and keeping it in an inactive conformation that is maintained by intramolecular electrostatic interactions. HH-induced phosphorylation disrupts the interaction, and induces a conformational switch and dimerization of SMO cytoplasmic tails, which is essential for pathway activation. Increasing the number of mutations in the Arg clusters progressively activates SMO. Hence, by employing multiple Arg clusters as inhibitory elements counteracted by differential phosphorylation, SMO acts as a rheostat to translate graded HH signals into distinct responses.
From Physical Review Letters:
By means of analytical and numerical methods we analyze the phase diagram of polaritons in one-dimensional coupled cavities. We locate the phase boundary, discuss the behavior of the polariton compressibility and visibility fringes across the critical point, and find a nontrivial scaling of the phase boundary as a function of the number of atoms inside each cavity. We also predict the emergence of a polaritonic glassy phase when the number of atoms fluctuates from cavity to cavity.
Anyway, getting back to the topic at hand, I stand by my earlier statement; the abstract does present results, and useful ones at that. They are presented succinctly, but such is the nature of an abstract:
...Using this concept, the optimum clamping pressure for sugar maple was found to be 0.3 times compression strength using U-F glue and 0.5 times using PVAc glue...
-Steve
38062.86 in reply to 38062.85 "Can you identify examples of arrogance, browbeating, or bullying?"Sure. Look at 38062.63 or 38062.68."Steve-I have looked at those two posts, and there is nothing remotely like arrogance, browbeating or bullying in them.38062.86, addressed to someone other than you, asks for evidence for an assertion the poster made. Seems reasonable to me.38062.68 is sarcastic in tone, but not a,b, or b.You must have a very thin skin, Steve, to blow mere disagreement and insistence that conclusions be preceded by evidence as such a threat to your authority.Don
"I have looked at those two posts, and there is nothing remotely like arrogance, browbeating or bullying in them."
You are certainly free to interpret whatever you want however you want. Likewise, I am free to disagree with your interpretation.
"You must have a very thin skin, Steve, to blow mere disagreement and insistence that conclusions be preceded by evidence as such a threat to your authority."
Trust me, my skin is plenty thick. As for "authority," I never claimed any. Some people are reluctant to call a spade a spade. I'm not one of them.
-Steve
Steve-I asked if you had any evidence of arrogance, browbeating, and bullying. You referred me to two posts, neither of which provided any evidence in my reading. So please answer the question: What evidence in those messages (or any others) do you interpret as being arrogant, browbeating, or bullying? Enough of your vagueness.Don
"What evidence in those messages (or any others) do you interpret as being arrogant, browbeating, or bullying?"
Why not hydraulic clamps or presses? Why stop there?
Perhaps my subscription to "The Proceedings of the National Academy of Clamping Technologists" lapsed, but I missed that passel of blinded, actively controlled, randomized, adjudicated, well-powered, independently funded, peer reviewed studies and the accompanying editorials supporting your position.
But now, despite having superior glues and plentiful clamps we are being told that we need clamping pressures formerly encountered only at the birth of stars to achieve acceptable results. To this I say booooo-gus.
"Enough of your vagueness."
I'm not being vague (not intentionally, at least). I just find aplogists tiresome.
-Steve
Steve-My, you really are a sensitive sort if those are the kinds of remarks that make you feel browbeatingly bullied. I hope you find nothing but peace and quiet in your further exchanges on Knots.DonEdited for typo.Edited 11/2/2007 5:41 pm ET by DonaldCBrown
Edited 11/2/2007 5:41 pm ET by DonaldCBrown
"The authors did not say, we gave patients with coronary artery disease an investigational medicine. It worked pretty good. We conclude it is nice."
Dammit, you got me again! :-)
I too haven't read the article and can't answer your specific questions about the tests in the published article. They are of course the obvious questions to be asked. They are so obvious that there is a VERY high probability that they were answered to the satisfaction of the reviewers and editors of the professional journal which published the article. That is after all what peer-reviewed scientific journals are all about.
That doesn't mean that every thing published in peer-reviewed journals will be absolute truth. I've had fun writing papers poking holes in such articles. But, the only people who will have enough information to poke those holes are other experts in the specific field, either academic or industry.
To shift gears a bit, I would note that 200 years ago cabinetmakers when to considerable lengths to avoid glued up panels. There are some around, of course, but when ever possible single boards were used--for 36" tea table tops, or 24" deep desk sides, etc. We can't say this is entirely due to some fear that glued panels would fail, it's also easier, with hand tools, to dress a single board panel than to dress several boards to be glued. In any event, there isn't even much anecdotal evidence to the long term survival of panels glued with hot hide glue, which after all is considerably stronger than PVA glue.
Steve,I agree on several points. But note that this was a technical article carried out under laboratory conditions. Whether those findings have much relevance in the hurly-burly of the real world is an open question. We have cured cancer many times in lab mice, that doesn't mean those results are easily translated to patients. I still cannot tell from what has been proffered whether findings in clamping and gluing in a research environment will confer an advantage in actual WWring practice.As for the single panels of yore that you cite, I would say that if I had access to first growth hardwood lumber, rift or quartersawn, with growth rings spaced 1 mm apart in widths of 24" or 36"- I wouldn't be gluing up many panels either.Glaucon
If you don't think too good, then don't think too much...
"Oddly enough, many of these panels survive"
And many, many more have not.
Steve, I understand the statistical argument generally, but think you may be as guilty as Glaucon of assuming facts not in evidence.
I do not dispute that many pieces of old furniture that contained panels no longer exist. I suspect, however, that most no longer exist for reasons completely unrelated to the failure of the panel joints they contained (e.g., thrown away; lost in fire or flood; etc.). I further suspect that many edge joints in antiques panels that have shown signs of failure, would not have survived even with perfect clamping as the failure was caused by other things like (1) misuse (e.g., exposure to water -- ever seen a glued up cuttingboard left in a sink with water overnight?); (2) poor design choices in terms failing to allow for wood movement; or (3) poor craftsmanship in terms of edge preparation.
What evidence do you rely upon for asserting that "many many" panels from the past have failed for lack of proper clamping pressure?
I'd tell you my profession, but I'm afraid you'd dismiss my thoughts if I did. ;-)
"Steve, I understand the statistical argument generally, but think you may be as guilty as Glaucon of assuming facts not in evidence."
I think you missed my point...
"What evidence do you rely upon for asserting that "many many" panels from the past have failed for lack of proper clamping pressure?"
...right about there. I never said that panels of the past have failed from lack of clamping pressure. What I said was that the vast majority of the panels of the past have not survived. In other words, the inference is flawed: The fact that a small number of intact panels have survived to the present says nothing about what happened to panels of the past in general; no conclusions can be drawn regarding their viability from such a small sample.
The issue was the use of anectodal evidence, not the strength (or lack thereof) of hide-glued panels.
-Steve
The fact that a small number of intact panels have survived to the present says nothing about what happened to panels of the past in general.
On the contrary. It says quite a lot. Unless of course, you contend that the works that survived did so by some aspect that makes them different from other panels assembled 200 years ago.
And the number is not fairly characterized as "small." The continuing survival of thousands and thousands of antiques around the world says something.
Also, if chronic edge joint failure were really an issue with all but the "lucky few" antiques, it seems that we would know it in just the same way that we know that frame and panel is far superior to the chronically failing solid plank alternative.
Evidence exists in more forms than "scientific" and "anecdotal." Empiricism is indeed a powerful tool, but we need not wait for empirical proof before we accept things we know from wide experience. To wit, woodworkers' edge joints rarely fail in common use.
Pointing to the survival of thousands and thousands of antiques is interesting but evidence of very little, because among those antiques are very, very few examples of glued up panels, it's almost certainly a very small percentage, and without some "forensic conservation" it is impossible to say of how many of those few have survived only because the panels have been repair from previous failures. Finding edge joined panels are so rare in antiques, that finding a piece of furniture with them is presumptive that it is a reproduction or fake. It is a presumption can can be rebutted, of course with appropriate examination and provenance. There was some furniture made in the 18th. century with glued up panels. But, you will have to do some serious hunting if you want to find any in museums. (There is a Queen Anne birdcage tea table in the Metropolitan Museum of Art whose top was walnut glued from two pieces with 18" wide sapwood.)
I suppose you can call casual empiricism "evidence" but it is lousy evidence. So many of those kind of things that we "just know" turn out, with careful examination, to be false. Science offers the ONLY way to go past that sort of casual guesswork to more closely understand the true state of the world.
I suppose you can call casual empiricism "evidence" but it is lousy evidence. So many of those kind of things that we "just know" turn out, with careful examination, to be false. Science offers the ONLY way to go past that sort of casual guesswork to more closely understand the true state of the world.
What you denigrate as "casual empiricism," I would call valuable experience. Knowledge gained through experience (especially over generations) can be just as valid as the knowledge we gain from the scientific method and carefully controlled scientific experiments.
Who said anything about "just knowing" anything. Leave your strawmen out of this.
Same for your slight: "casual guesswork." You touch enough lit stoves, you learn they are hot - no guesswork at all. And you sure as hell don't need a well controlled experiment before you can reasonably decide whether to touch the next one.
The idea that formal science (scientific method) is the ONLY way to understand the true state of the world is rather narrow minded. I mean I love science and all, but I been around enough to know I've learned plenty of valuable truths in a variety of ways. I've also learned that there are plenty of things that are either literally impossible to test (the data cannot be obtained) or just far too expensive to test (the cost of obtaining the data is prohibitive). We rarely have perfect information. We rarely have statistical certainty. This does not mean that we are not often right in describing the various states of affairs.
But this is a tempest in a teapot. Edge jointing is just not that mysterious or problematic.
I agree that we rarely have perfect information, and often have to act in spite of not having such perfect evidence. I have no idea what statistical certainty even is--it's not a concept taught in statistics courses. I do know what statistically significant means, however. (At least to a fair degree, it is really a very complex topic.) And, what I believe is that there are a great many things that remain unknown. When we act on the basis of such incomplete evidence, whether tests are not feasible or whether we are ignorant of what tests others have done, is essentially guess work, whether it's called "experience" or not. How would we know that we were right in describing the various states of affairs, that what we know is truth and not something else, if we could not test the answers in some systematic way?
I spent much of my working life as a securities analyst, having to make recommendations based on insufficient knowledge and with short deadlines, armed only with some basic ways of looking at what little available data is available even with lots of digging. Some of the data are deliberately misleading, just to keep things interesting. Many millions are spent on such research, yet I would strongly assert that people who say they "know with certainty" that a given security will rise in price, or that they know what will happen to interest rates 6 months from now, or that they know the direction of the price of petroleum are charlatans. (As well as acting illegally if they work for broker/dealers.) I never confused having an opinion--the best I could come up with--with having knowledge. This just the same as understanding things from experience without a system to evaluate that experience. That's all science is, a way to turn experience into hypotheses and data that can then examined systematically and found to either be false or not false.
Mostly we just don't really know all that much. That's what makes life so interesting.
I appreciate your response, and agree with much of it. I have occassion to deal with economists performing econometric analyses using various regression models. What I meant by statistical certainty was -- loosely speaking -- results of something like a regression that give professional economists enough certainty to say they are satisfied that the empirical analysis has worked to provide a reliable answer (usually nothing much short of 99% from what I can tell).
You ask: How would we know that we were right in describing the various states of affairs, that what we know is truth and not something else, if we could not test the answers in some systematic way?
My point is simply that there are "systematic ways" other than use of the formal scientific method. For example, years of experience.
Here we are dealing with a binary proposition that is physically verifiable: the (millions?) of panels clamped with pressures far below those recommended in the article over the last several decades hold over time or they do not. Are you (or anyone else out there) aware of chronic issues of failing panels? I've never heard of such an epidemic, but I'm all ears. As such, I and others have remarked that this article is in some ways "a solution in search of a problem." More pressure might well produce a superior joint (and that's worth knowing), but such a superior joint is apparently rarely necessary (unless, as I've said, there is a problem of chronic panel failure of which I am unaware). Anyone? Anyone? .... Bueller? ;-)
I suspect that when a panel does fall apart there isn't a big rush to tell the world about the event. That alone makes them hard to uncover.
I've used and developed a bunch of those regression models--and if those econometricians were worth their salt they used terms like confidence, t or z scores and significance, and adjusted R-squared but never spoke of certainty, and likely not of terms like causality. I used quite a bit of these econometric techniques in the process of Piling it higher and Deeper in my economic studies.
So why do I distrust "experience"? Mostly because is so rarely systematic, with so little real follow up of results or tabulation of the meaning of what is observed.
Besides in this context I don't doubt a lot of substandard joints are strong enough to hold together for substantial periods. As far as I'm concerned mere survival isn't the real issue. Edge joints should do more than just survive. At various shows, I've seen enough visible glue lines on pieces made by supposedly professional custom furniture makers to question whether making good edge joints is such a routinely achieved task. A well made joint has NO GLUE LINE, only a place where some grain lines abruptly stop and start on what turns out to be the demarcation between two boards. To achieve that requires BOTH good joint preparation and good gluing practices. Full pressure gluing appears to be part of those good practices.
"And the number is not fairly characterized as 'small.' The continuing survival of thousands and thousands of antiques around the world says something."
"Small" as in "small percentage," not "small absolute number." The size of the sample population doesn't matter; what matters is whether it is representative of the whole population or not. And the fact that we're talking about glue joint strength (and thus the possibility of glue joint failure) is enough to raise suspicions about the representative quality of the surviving pieces, given that a glue joint failure is a reasonable reason for a piece not to survive.
Even in those pieces that have survived, I would wager that a significant fraction of them exhibit edge-to-edge glue joint failures--I've certainly seen it in museum pieces.
Empiricism is the starting point. It's what gets us to ask questions. But we really don't "know" anything until we test our empirical assumptions. (And even then, what we "know" is only an approximation to The Truth, until better observations and analyses occur.)
-Steve
Alright, you guys have worn me out. You two Steves are way smarter than me, I admit it. I'm glad to hear that you now know to use 50 quick grip clamps to glue up a super strong edge joint on a two foot panel. More power to ya.
I bought 20 of those closeout aluminum bar clamps from Lee Valley a month or so ago. At something like $6.50 each, they were quite the deal. And now I can put clamps every two inches along a 40" panel.
Eat your heart out. ;)
-Steve
Not smarter, just more stubborn. Interesting discussion just the same--thanks.
I don't own any quick clamps at least.
Cheer up- while others are spending 11 hours a day sharpening and clamping, you can be actually working wood.This bit reminds me of the "Anal Retentive Chef" routine on SNL (god bless Phil Hartmann)- who could never cook anything because he was always preparing to start.I'll wait until the epidemic of glue failed panels hits my neighborhood before before buying the 2,188 pipe clamps necessary to glue up my next end table project. In the meantime, I struggle along with my current clamps and cauls, and a little TB2.Glaucon
If you don't think too good, then don't think too much...
The argument regarding hide glue panels is really irrelevant towards the discussion. Despite both being glues, hide glue is not PVC glue.
I'm not a glue expert, but as far as I know hide glue is prized for it's reparability. Now that I think about it surviving panels could have survived because of that fact.
I'm sure if someone were to contact tightbond they would probably have the sort of clamping data given in the article.
<"But when it comes to clamping wood " more is more", more or less."><!----><!----><!---->
And this is based on what evidence or data?
<!----> <!---->
I think that it’s well established what the IDEAL clamping pressure is for a glue joint. It’s been covered in ‘Understanding Wood’, and various other books over the years. More is more, until you start to crush wood fibers (hence the more or less). Crushing wood fibers is unlikely in a wood shop.
<!----> <!---->
Oddly enough, many of these panels survive- even in applications where they received a good deal of abuse and are under tension most of the time.
<!----> <!---->
This is just anecdotal evidence, would you still feel the same if we had a ratio of the number of panels that failed? Say if for every panels the survives twenty panels failed, or thirty, or one hundred… Anecdotal evidence is always presented for more than its worth.
<!----> <!---->
This is a solution in search of a problem... a marketing department's dream.
<!----> <!---->
Marketing dream? Do you pay some sort of royalty for every turn on your clamps? Do you only own three clamps?
<!----> <!---->
I don’t get the resistance to the article. Yes we have superior glues, and no we’re not seeing a huge rash of joint failures (at least I’m not)… But all the article is about is getting the strongest glue joint, and the best way to do that is to ensure that you have enough clamps and that they’re tightened as best as you can.
<!----> <!---->For what its worth: I doubt we need the ideal glue strength for a majority of our furniture. Most will not undergo the stresses required to break the joint. I know my first panels are still together, and the were glued up with less than ideal prep on the edges and probably too few clamps… But then again why wouldn’t you want to get the best joint you can, especially when it cost you zero money and virtually no time. Just add an extra clamp or two, and tighten them the best you can.
"we are being told that we need clamping pressures formerly encountered only at the birth of stars to achieve acceptable results."
Wow! That made me laugh out loud. Thanks Glaucon!
I learned a lot from the article... one thing that I think will really help is the use of cauls in a panel clamp up... the 45 degree pressure cone bit was something I just have not considered...Drew
I have to weigh in on the side of "you gotta be kidding about that many clamps". Sounds kind of over-engineered to be. Engineering and engineered wood products certainly have their place, but I'm not going to be making heirloom furniture out of waferwood or OSB any time soon. Before I chuck my Bessey's, I need something more tangible than the joint is "more optimal". A fully cured PVA rubbed edge joint with a couple of nicely planed edges will break at least partly in the wood anyway. Besides aesthethics, maybe the thinner the glueline will have less creep? And if so how much less? That might be a reason to use so many clamps and PVA for bent laminations and skip the nasty clean-up of plastic resin glue.John O'Connell - JKO Handcrafted Woodworking
The more things change ...
We trained hard, but it seemed that every time we were beginning to form up into teams, we would be reorganized. I was to learn later in life that we tend to meet any new situation by reorganizing; and a wonderful method it can be for creating the illusion of progress while producing confusion, inefficiency, and demoralization.
Petronious Arbiter, 210 BC
Samson & Steve Schoene,
I believe that the glue provides a strong enough joint when edge-gluing. Biscuits or dowels merely provide alignment while cleats or battens only stabilize the top to prevent warping.Chris @ flairwoodworks
If anyone wants to toss their Besseys in favour of QuickGrips or other new clamps, I may reluctantly offer to find a good home for them.
No takers? Darn.Chris @ flairwoodworks
I believe that the glue provides a strong enough joint when edge-gluing.
Yeah, me too.
Biscuits or dowels merely provide alignment
I dunno. While I don't think that bisquits, dowels, or splines are necessary in most cases, I think they may well reinforce the joint somewhat. While an edge joint won't fail on the glue line, the adjacent wood might fail because the grain of both boards ends at the glue line - it does not continue across. Splines are especially strong because of the long grain to long grain surfaces in so many planes.
while cleats or battens only stabilize the top to prevent warping.
I agree that this their main aim in most cases. But if the aim is to strengthen a panel to better resist various loads and stresses, cleats and battens can help.
Also, you ignored butterfly keys. I dunno why you'd go to the trouble, but they certainly would provide reinforcement of an edge joint for the same reason they restrain checks.
Samson,
I ignored the butterfly keys simply because I have never used them and therefore have no actual experience.Chris @ flairwoodworks
>> Splines are especially strong because of the long grain to long grain surfaces in so many planes.Recognizing full well I am throwing gas on the fire, let me postulate that splines will almost always result in weakening a panel.First a couple of points. A well prepared and glued long grain to long grain joint is stronger than the wood itself. With glue alone,f a panel will fail at some point other than the glue line. When dowels or biscuits are used a panel will fail at some point other than the glue line. So, dowels and biscuits provide no additional strength to the glue line.Now to splines. When you use a spline, you first cut slots in the edge of both boards. Let's assume the slot is 1" deep on each edge. If you tried to use a 2" wide spline, applied glue and attempted to clamp the joint it would not close because the glue that accumulates at the bottom of the slot prevents the spline from fully seating. To deal with this the spline must be cut somewhat less wide than the combined depth of the slots--say 1 3/4". Now the glue has somewhere to go but the spline no longer bottoms in one or both slots. This open space in the bottom of the slot now has no wood (or strength) in it and it is now a weakened area in the wood. When pressure is applied to the panel, it will ALWAYS break along the bottom of one of slots. So, while the glue line joint itself will not fail, the strength of the overall panel has been compromised by this/these areas that contain less wood. So in attempting to "strengthen the glue line, the woodworker has actually weakened the panel. And, these areas run the full length of the panel.Some might argue that if the splines where accurately sized to allow them to bottom fully, then there would be no loss in strength. In theory, this is correct but from a practical point of view the excess glue still needs to go somewhere and glue itself has little strength. If you used epoxy glue which has some gap filling properties, then the area from the edge of the spline to the bottom of the slot would be as strong as solid wood.The point is that "strengthening the joint" is not what biscuits, dowels or splines do when edge to edge gluing. They strengthen the wood on either side of the joint therefore the breaking point will be beyond the area supported by the reinforcement. But, that is the same outcome if no support were added to a well made glue joint. In summary, using the standard woodworking adhesives in a properly prepared joint will give you a glue line that will not fail. The wood may fail along side of it but using other reinforcement just changes the location of the failure to some other point. Use biscuits, dowels, glue joint router bits and splines for alignment, not for strength in solid wood edge to edge joints.
Howie.........
A few things:
I've never used splines in my edge joints.
I will not be any sort of spline champion here (no bisquit or dowel champion either).
I agree they are mostly for alignment.
I think you make a decent point about the glue at the base issue potentially resulting in a void (having joined peices mate well in multiple planes is the perenial challenge in all joint making).
All that said, you may want to read Robert Lang's article in Issue 6 of "Woodworking:"
View Image
He states his opinion that splines are not necessary for strength, but do indeed add some.
I tend to agree. You seem to allow this as well: "if the splines where accurately sized to allow them to bottom fully, then there would be no loss in strength. In theory, this is correct ..." but quickly add that such fit cannot be achieved in the real world. Lang looks as though he got pretty close.
Strangely I was recently recycling a jointed panel from the nineteen-fifties, and just could not separate the joint lines, even though they were obvious from the grain directions.
As a last resort I used a bandsaw, and I do run metal cutting blades! The strips had been slot screwed, and nothing would have caused a joint failure. Good English quartersawn oak.
All,
One aspect of the whole PVA glue/clamping pressure/strength debate thus far has not been addressed: the manufacturer's expectations of the market they are selling to. Does anyone else think that maybe Franklin or Elmers understood what kind of clamps the typical home shop has in it, and will be used to clamp up the wood that their glue has been spread on? This product has not been designed and marketed in a vacuum. Would anyone care to bet that the mfrs didn't do a little product testing before they sent this stuff to market? Why re-invent the wheel here, folks? Ah, go ahead, it's fine with me.
Ray
The other sector not paying much attention to research is the clamp manufacturers. The new Besseys I bought earlier this year still included an instruction not to clamp too hard as that would lead to squeeze-out.
Jim
Jim,
I'm guessing that those same clamps also warn the user to wear eye protection. The point being that I would take the manufacture's word with a grain of salt.Chris @ flairwoodworks
>> Does anyone else think that maybe Franklin or Elmers understood what kind of clamps the typical home shop has in it, I'm willing to bet that their primary customer is the commercial shop, not the home workshop. PVA adhesives are the primary commercial furniture adhesive.Howie.........
Howie,
You are probably right.
Ray
Expert advice, please, am I using enough pressure?
You forgot the anvil, the horseshoes - and the horse!Frosty"I sometimes think we consider the good fortune of the early bird and overlook the bad fortune of the early worm." FDR - 1922
mufti,
Don't you have a few clamps you can use? I like the old fashioned cast iron C-clamps. They weigh more, and so, provide more pressure when piled on top of the work being glued.
Ray
Ray, I have various flat irons, the type they used to stick in the oven, odd scale weights and myself at sixteen stone to pile on top of glue ups. I just find it incredible that so much agonizing is spent onm everyday tasks!
mufti,
Yah, but it filled a few pages in the mag, dint it?
Ray
I'm a big fan of higher clamping pressures. You'll see that in my previous posts. That's the main reason I enjoyed that article.
You really should make yourself a few test pieces and break them. Last I checked I could easily see if the glue was sticking or not. Was it at optimum strength? Probably not. But I bet it was 99 % of it. I can tell you right now that I'm not going to start measuring my clamping pressure for that last 1%.
Has science improved woodworking? You bet. Better glues, finishes and a host of other products and analysis.
In the past 30 years I've made 1000's of raised panel doors, table tops. and other flat panels. Only once did I have a glue line failure that I know of. A small table I built for myself. The glue line came together so perfectly I didn't apply as much clamping pressure as I normally do.
I've found what works and what doesn't. So, I'll always do it this way because it works for me.
Paul
Paul,
Your test example is flawed. I used to always do this with the cutoffs from panels. All this shows is the weakest point in the wood, which may be a slight check or something similar... It does not show that the glue joint is stronger than the wood, only that it's stronger than the weakest point. By splitting with a chisel you're forcing it to break along the joint, and by examining how it breaks you can get your answer.
For what it's worth, I don't think clamping should be made any more complicated. The article covered some good clamping basics, but I'm not going to bring my calculator into the shop. For the most part I just tighten my clamps down as hard as I can and get on with my day.
Buster
Buster,
Hear hear! What I took from the article is that more pressure is good, and not a detriment. You won't find me out in my shop with pressure gauges. I'll just make sure to tighten the clamps as well as I'm able, maybe add a few more now that I've gotten the information from the article, and leave it at that.
Zolton* Some people say I have a problem because I drink hydraulic brake fluid. But I can stop any time I want.
Amen. Just about anything can be chased down the squirrel hole to the point of truely arcane information. I just don't think there's a point to reasearching the matter that far.
I agree. I did find the article interesting and I did get something from it. But I won't do much more than take care to tighten my clamps tighter and use more of them. In other words, I won't be measuring clamping pressures anytime soon...
Zolton* Some people say I have a problem because I drink hydraulic brake fluid. But I can stop any time I want.
I know people don't want to be confused by the facts, but the facts in the article are mostly old hat--just not so widely circulated. Bruce Hoadley, in 1980, tested clamping pressures obtainable from various models, and none of the straight handled models, such as the Bessey and its competitors, generated more than about 500 lbs. and most achieved less.
QFT. These clamping threads, both here and on other forums, are getting almost as contentious as dust collection threads. ;)
One thing these threads have illustrated to me is that there are two opposite extremes of personalities (with alot of folks somewhere inbetween).
On the one extreme you have what I'll label, for the purposes of discussion, the "scientists". These folks are always inquisitive, always skeptical, and rarely satisfied with an answer. They're the type of kid who took apart every toy handed to them, in order to figure out how they work. If they were lucky, they had a good science teacher early on that taught them how to apply this inquisitiveness through the Scientific Method, and hopefully they wound up with a career in the sciences or engineering.
These folks think the term "common sense" is, as Einstein once said, merely the accumulated prejudices built up over a lifetime, and isn't an excuse for not asking if there is something better.
At the other extreme there's what I'll label as the "ideologues" (maybe someone will suggest a better term). These guys know what they know; it works for them, so why bother going to the trouble of trying anything else? Their trademark phrase is "if it ain't broke, don't fix it". They are probably more comfortable and less stress-ridden in their worldview than the "scientists", and are willing to accept thing on "faith". "Common sense" is a good thing to people like this. 'not sure what careers are best for them, but people I know in law enforcement have said to me that they think that line of work tends to favor this sort of black and white philosophy.
So when the question "what is the best way to clamp a PVA joint" is asked, these two types will have two very different reactions. The "scientist" will react with "let's find out". They'll analyze the situation, conduct experiments, consult the peer-reiviewed literature, see if improvements can be made, publish the results and win the Nobel prize, etc. ;)
The "ideologue" would probably wonder why the question is being asked at all. They have a method that works, and may have worked for many, many years. It's worked for others they know also. So what's the point doing anything about it? At most, they may refer to an "authoritative source" for confirmation that what they're currently doing is OK. Norm's on TV, and he says just tighten them enough to get some squeeze- out. He's the world's most famous woodworker with millions of fans; that must count for something, right?
FWIW, I'm in the "scientist" camp, but I think the clamping issue needs to be put in context. The research that led to the optimum clamping procedure is just that - a search for the "optimum". This can be crucially important to a big industrial operation: e.g., how little glue can I get away with and still get a strong joint? Every 55 gallon drum of Titebond I don't have to buy goes to the bottom line. And with the razor-thin margins Wall-Mart's imposing on me, every penny counts. At the same time, I can't afford even a 1% return rate, either. Warranty issues due to failed joints, visible glue lines, or glue creep are unacceptable. So what is the absolutely optimized method of using PVA that'll give be the best result at the lowest possible cost? That's where this academic and scientific research is most helpful.
But how much of this is applicable to a hobbyist and small shop? Less than for the industrial user. To me, the key take-aways are:
-- With PVA, forget about worrying about clamping so tight you starve the joint. Ain't gonna happen. Crank those clamps down as hard as you want, guilt-free.
-- Gluing up narrow boards will require more clamps in order to get even pressure, regardless of whether a lower number of clamps would get an adequate total amount force on the joint. In other words, you're likely to need more clamps panel gluing 2" wide boards than you are 10" wide ones, if only due to the need for good pressure along the entire length of the joint.
-- Have problems with glue creep and/or visible glue lines on your panel glue-ups, even though you have well-jointed edges? You may want to analyze what magnitude of clamping pressures you're appying to the joint, and see if you can modify your technique to get closer to that of the article.
Neither the staffer etc. knows the reality of woodworking and everyone is wrong?
Why not step up and submit some ideas for correct articles Paddy. Be part of the solution you are looking for.
Which premium writers would you like to read?
The assertion that a K-body clamp provides less pressure than a quick-grip clamp seems preposterous
Apparently, Irwin has a new design for the quick-grip that lends itself to much higher pressures. Marc "the wood whisperer" recommends them.
There's been some tests posted on woodnet that shows the K-body's and Jets have trouble meeting that 1000 lb. rating. The handles are so small in diameter, people with average grip strength have trouble torquing them that tight. That's where the handle design on the pipe clamps as an advantage.
Say, does anyone know about those panel clamping systems? They attach to a wall with several pieces of apparatus that puts pressure on both the face of the boards to keep them in line, and then also on the edges of them to press them together. I cant remember the name, but I know Amazon sells, or used to sell them.
I know that they are more for production use, but I think the setup only had a clamp about every 16" or less. How much pressure can they generate?
Peter
www.jpswoodworking.com
I think any clamp other than a Quickgrip type provides more pressure. I can't begin to count how many times I had one of those things slip on me, even the bigger ones. For important clamping jobs, I don't use them, for quick hold downs, or to stop something from falling, then the quick clamps are useable.
Amen, i thought the article was downright weird. Whenever I have ramped up the clamping pressure I've had a heck of a time getting the panels to lay flat no matter how I place the clamps. I've also never had joint failures and I tend to use less pressure rather than more.
This forum post is now archived. Commenting has been disabled