Glue Myths: 1. End grain

hey this is patrick sullivan welcome to my shop everybody knows that gluing end grain is useless everything you read about glue points this out i don't want to single out wood magazine because every publication i've seen carries similar comments everybody agrees gluing end grain together is a rookie mistake this is just one of those universal truths that is so obvious that it can't be questioned is it true well stay tuned and see for yourself we're going to measure it very carefully the first task is to buy some beautiful clear straight grained lumber of various kinds and saw the boards in strips that are all three inches or 76 millimeters wide then they all go through my planer set at exactly three quarters of an inch or 19 millimeters next i cut them into perfect squares using a jig so that they're all identical in every dimension unlike some of my furniture joinery all the cuts are perfectly straight and perpendicular using a sharp 50-tooth carbide blade ultimately i'm going to test many glues but today we're just doing one it is titebond ii a glue made from polyvinyl acetate or pva titebond is the most popular wood glue in america but regardless of which pva glue you use the conclusions of this test should still be applicable there's a reason why so many woodworkers all over the world depend on pva glues most of the time and you probably already know those reasons they're strong cheap convenient fast easy to clean up they require no measuring or mixing and they sit patiently on your shelf for months without going bad now there are three ways to glue two blocks together side to side and decide and end to end the virtue of this particular setup is that regardless of the grain orientation the pair of blocks end up exactly the same size with the joint in exactly the same place this lets us compare the results directly and fairly without any calculations or adjustments i clamp up three to four pairs at a time in a parallel jaw clamp to keep the clamping forces centered i made some spacer blocks i've temporarily added blue tape to make it easier to identify them i wanted to keep the clamping pressure consistent so i built an electronic strain gauge specifically designed to measure clamping force it sends a signal to my computer so it can be easily read and saved glue came from a brand new bottle and i applied it to both surfaces at the same time i tried to be methodical about spreading the glue evenly and thoroughly someday in the future we might be able to test out just how important glue application is right now i just want to be sure that every surface is saturated evenly after gluing the blocks sit in the clamps for two hours and then are allowed to cure for two days that might be overkill but until i can perform reliable tests of the glue curate i want to be conservative when the joints are fully cured we will stress them in a hydraulic press that's specifically designed for this particular kind of test wood needs time to bend and adjust the loads and this setup lets us go at whatever speed the wood needs it also lets us see the breakage as it occurs the force applied to the joint is measured directly with an electronic load cell sitting under the hydraulic pump load cells are the definitive method used in scientific labs and the results record directly on a computer at 10 times per second so we can detect transient peak loads please note that this is not recording pressure like a hydraulic fluid gauge does the load cell reflects the total force on the wood piece okay enough about the test apparatus before we get into the testing i want to be sure that we're all on the same page with terminology wood derives its strength from cellulose fibers these fibers are laid down in layers of long strong cellulose tubes they're almost all running in the same general direction which is up and down the tree we'll call the direction of the cellulose tubes the grain direction tubes are held together with lignin lignin is an organic polymer much like glue or plastic it coats the tubes sticks them together and fills in most of the spaces if we were making fiberglass composites the lignin would be the resin species of wood produces slightly different versions of both the cellulose tubes and the lignin so we need to test a bunch of different species i've focused on popular woods that are readily available to me here in the united states wood glued side by side behaves in a predictable way invariably the wood splits between its long grain fibers before the glue joint breaks some of the fractures occurred very close to the glued joint but close inspection showed that the glue line itself remained intact the glue joints were stronger than the lignin in all of my samples what about wood glued in the side will the end grain weaken the joint surprisingly the opposite is true and the side joints are a little stronger than side to side the wood splits and of course the split is always in the piece with the side grain the increase in strength probably occurs because the end grain half of the sample stiffens it so there's less deformation for a given amount of force now the critical measurement joints glued end to end in every case the glue line actually fractured and the wood itself remained completely intact and now a big surprise the force required to break in grain joints was considerably higher than the force needed to break side grain joints instead of being vastly weaker as we've been taught end grain joints turn out to be about twice as strong as side grain joints on average out of curiosity just how strong is the long grain of the wood i cut samples from the same plain stock as my small squares i cut them twice as long so they have no joint at all it's fascinating to watch the wood fail along its long grain each species is a little different in general only the fibers on the top surface break and usually they don't all break at the same time the force required to break these samples was huge compared to the glued up pairs after measuring these forces we can now see what contributes the most to the strength of complex furniture joints the long cellulose fibers are so much stronger than either the ligament or the wood glue that the strongest joints are ones that maximize the amount of long grain that crosses the joint i'll do a video on this subject some later time rather than bombarding you with a lot of numbers let me try to show the results in a picture that might be more memorable the lines on this grid are 100 kilograms or 220 pounds apart for example the 2 is 200 kilograms and the 10 is 1000 kilograms or 1 metric ton the first bar is a summary of the results from fracture testing white pine the green box represents a range of values at which the lignan bonds fracture allowing the wood to split parallel to its long grain the gold box represents a range of values at which the glue itself fails without any breaks in the wood these numbers come from breaking end to end joints the red box is the force at which the long grain on a board with no joints at all broke it's obvious that the long grain strength is much larger than the glue strength here are the corresponding results for poplar and here are the results for cherry walnut maple and red oak now i don't want you to bog down in the numerical details unless you're a structural engineer they won't help you however there are several big picture conclusions that are critically important and that will help us to understand pva glue forever first the lignin strength varies from species to species but in general the lignin in most wood is roughly similar in strength second the glue strength also varies quite a bit from species to species but the important observation is that it is always stronger than the lignan strength there might be some species for which this is not true but i haven't found any such wood so far third the long grain is much much stronger than the glue in these tests it varied from three to eight times as strong as the glue how much stronger is the glue strength compared to the lignan strength well for pine it's two to three times as strong for poplar it's well over three times as strong for the rest of the woods it was roughly twice as strong wood glued side to side or endeside will always fail by splitting the lignan bonds which are much weaker than the glue bonds wood glued end to end will always fail by breaking the glue and since the glue strength is always stronger than the lignan strength end grain joints will always be stronger than side grain joints in general they'll be two to three times as strong despite all the bad press they've gotten over the years i realize that's not what the rest of the world believes but those are the results one of the things that bothered me when i tabulated these results is that the glue failure strength varied so much from species to species if the only thing that fails is the glue and all these samples are glued with the same glue applied under the same conditions with the same technique then shouldn't all the end grain samples fail at about the same value well it turns out pva glue does not form strong chemical bonds with the wood instead it penetrates into microscopic pores channels and defects and grips those irregularities tightly thus the strength of the bond is partly dependent on the internal microscopic structure of the wood fibers which varies from species to species take a look under my microscope if you break open end grain joints and look at them with your naked eye you really can't see much difference however if you look more closely you can just begin to see differences notice how the glue pulled away from one side of the joint on this maple sample the joint was very strong but when it failed it's separated almost completely from the wood this broken joint pine is a totally different story both sides of the joint have glue adherent in a very uniform pattern when we look at high magnification way more differences begin to appear compare this pine joint on the left with the maple joint on the right completely different or compare pine to walnut radically different again or to oak which has huge pores that often but not always grip the glue firmly cherry is different yet let's take one last look at glue adherence we're going to zoom in on this end grain walnut joint on some of the larger fragments of fractured glue you can see a perfect imprint of the mating surface i've highlighted several raised knobs of glue that correspond to pores that you can see on the opposite side of the joint for those viewers with a small screen i have enlarged a part of this image okay enough technology but before we quit how did this myth that end grain glue is fatally weak get traction i think we overlooked the fact that for the most part the in-grain joints we thought were so weak had a completely different geometry than the side grain joints that we thought were so strong it turns out that the geometry is critical take a look at these two glue joints same species of wood identical dimensions one is glued side grain to side grain and one is glued end grain to end grain but wait this is not what we mean by side grain joints no experienced woodworker would glue up wood like this when we picture side grain joints we picture gluing up the long edges together and when we picture n grain joints we picture gluing up the short edges together it turns out we were comparing apples to oranges when people think about end grain glue ups they're usually thinking about long narrow pieces glued at the small end this means long lever arms multiplying the forces on a tiny glue surface of course these badly designed glue joints sometimes fail but sometimes you're in for a surprise can you guess what will happen when i test these two long glue ups i easily broke the side grain piece using just a couple of fingers i could not break the end grain joint without bending it over my knee did you predict this outcome if not you failed the final exam maybe watch this video one more time okay so in conclusion ingrained joints are about twice as strong as side grain joints did this video blow up the myth of end grading glue for you i hope so is it possible for 95 of us veteran woodworkers to be mistaken apparently so and count me among the people who were taken in if you like this kind of information let me know tell your woodworker friends to check this video out for themselves i'm willing to do more of this research if enough people are interested personally i have lots of questions that are still unanswered and i am sure there are still more wood shop myths to be exploded as always thanks for watching

Channel: Patrick Sullivan

Views: 485,059

Keywords: Premiere_Elements_2021

Id: m7HxBa9WVis

Length: 16min 3sec (963 seconds)

Published: Fri Sep 03 2021