If you’re doing these sorts of testing type things in the future, I would highly suggest doing at least 3 of each test, so in this case if you had done 3 of each type of joint with same wood and procedures. That can’t help bring to light the variability in the test/materials, which can be surprisingly large sometimes.
How about the following scenario: I'm building a wood frame for a deck. Would it be stronger to have the long stiles on the outside of the rails or vice versa?
@@markpowell6885 They are. The only end joint for strength in this plane for wood is the Japanese scarf joint (not the one he showed). Every one he tested should have been tested on the face plane, not the one he tested.
When the first board broke, I was scared by the sound of breaking because I didn't have psychological preparation. I hope to see more mortise and tenon tests. Thank you for your video.
I was wondering this , and also another setup where the pressure is applied on the wide flat side of wood (for example 10cm/4inch from the joint) and something's keeping the other side of joint sturdy. I think this is more similar to normal usage of wood.
Interesting! A few suggestions from others for a sequel. Most valuable: (1) biscuits make the butt joint weaker. Not significant on a long edge joint where there is lots of wood with biscuits spaced for alignment. (2) preglue! I didn’t see any mention of Dominos, but that would be purely academic (for most of us). I would like to see the tests repeated on the face with a caul to spread the pressure. Thanks for an interesting, informative video.
A lock miter joint that can be used at both 90 and 0 degrees would be an interesting thing to see tested. I would imagine it might come close to the finger joint, or better. I'll check and see if you try it some day.
It would be really nice to actually test the lap joint, scarf joint, and a piece of solid wood to see how they compare. Other ones I would like to see are spline, and tongue and groove.
I think your assumption on the scarf joint is incorrect. I think the glue would still fail first. And in lieu of the lap joint, you should test a 12:1 aircraft style scarf joint angled from the wide side of the board rather than the narrow side as you did before. That gets rid of the weak area you mentioned in the lap joint.
Interesting test Colin. I think most of it was testing the glue maybe. Joints with anything inside of them seemed to rob the endgrain of glue, and failed quicker than you may have thought. Plus I wonder if you had pressed down across the joint, and it's occupied reinforcements. IE: Across the Dowel, Biscuit, etc, if the joints wouldn't have been stronger? I agree finger joint made sense, and adding so much more glue surface, you got a stronger bonding.
Very Interesting, But I think that if you re-preform the same test with the same joints after the glue has set for a week or more, you might get different results.
I would prefer to see the strength of half-lap and scarf joints tested. A better test would be to secure one end of the boards with your strongest clamps, and hang weights on the end of the other board until the joint fails. Choosing the position of the fulcrum and length of the lever arms correctly should be enough to break even the stronger joints, so long as you have strong enough clamps to hold the other end.
Yes, those were the exact two joints I was most interested in seeing tested against each other so I was super-disappointed when he wasn't going to test them.
Matthias Wandel did this in a test, with a better measuring system than this Press. His numbers showed, that glue does NOT improve the strength of pocket hole joints. Some other woodworker also said the same. I think the woodwhisperer once also said, that glue does not improve the strength of pocket holes :=)
I just bought one. The place I bought it said it's awesome. I'm joining a brace between legs on a small table. I've cut one side.. maybe not the other. Should have went with dows.
Great video, thanks. Id also be interested to see a tensile test on joints. I have to join two pieces and Im thinking of using a dowel screw to keep them from pulling apart and keep the joint hidden. If you have a better solution please let me know? thanks again
Great info , Colin you must also do a Test without a joint on the same wood; so we know what is the breaking point of the wood without a joint and we can compare the strength, But once again thanks for sharing this info very helpfull.
I always enjoy these strength tests. Although I wonder if the dowel joint would have given different results of only two dowels were used. Thank you for the video!
Seems logical that the pre-glued sample is really strong: pre-gluing not only seals but penetrate the grains and thus serves as microscopic dowels. Now it depends on how deep the penetration was before the wood are clamped, that why some suggest thinning the glue first with some water.
The format of this test reminds me of a contest hosted by the mechanical drawing and woodshop classes in my high school. Build something 6" x 6" X 6" minimum out of balsa wood and glue that weighs less than 3 nickles to see which supports the most weight. I think the record was something like 350 pounds. They did this every year.
Again Colin comes thru with some great info, I don't come up to shortboards much in my projects but... I'm thinking of purchasing a finger bit or two on my next trip to the store
Surprised that the Biscuit joint was so easy to separate. Pre-gluing seems to be the way to go on Butt, or Half Lap joints. Educational and entertaining video. 🇬🇧
I believe I can help with some of the confusion as to why the biscuit and dowels weren't as strong as expected. Your test was set up such that bending would be the failure mode of each beam. The strength of a beam is a function not so much of how much material there is, but how that material is distributed. You can make a much stronger beam with less material just by arranging it efficiently. the material at the extreme top and bottom of the beam contributes the most to the beam's strength (side note: That's why I-beams are shaped the way they are. It's a very efficient way to get as much material as possible where it does the most good). So in this case, we see that the biscuit and dowels failed at lower loads than expected. That's because they didn't engage the beam fibers at the top and bottom of the beam the way the other three joints did. So even though the total beam was the same depth, at the point of load (right under the press), you had less *effective* depth to work with, and thus a much reduced strength. For the mathematically inclined, the reduction is proportional to the square of the depth, so it's a bigger difference than if it were proportional to the depth itself. Additionally, I think a big part of why the dowel and biscuit joint didn't crack the same way the others did is because they had those interior components. The dowels and the biscuits themselves didn't fail, so the beam was able to retain some strength just by bearing on those interior connectors and holding together with just friction. As sacha schnauzer pointed out, this has the desirable effect of being much safer failure modes than a full rupture as with the butt and finger joints.
I've become a bit of a fan for the biscuit joint in certain applications. I have done my own tests and have found quite different results. The right size biscuit properly fitted and glued into a properly glued joint certainly does add strength, not as much as a figure joint but way more than a straight butt joint. I must admit however that I would never use a large biscuit in a joint like the one in your demonstration.
Yes. One other consideration: The more complete test should compare edge to edge joining, and then the biscuits : wood percentage would perform completely different; also the end grain structure on biscuit slim profile might perform very differently then long grain.
So the go-to should be the lap joint if strength is the main objective. If making a scrap wood table would just glued joints be sufficient and forego the fancy joints because the sides are being glued?
I suspect the biscuit joint to perform much better if you applied the pressure to the face of the board rather than the edge. can you repeat the test but this time apply pressure to the face of each board. thanks of the videos. very informative
Been watching your stuff for a while and I like the content you provide. it is insightfull, sometimes funny, always informative - you don't need to do the 'amazed/shocked face for your thumbnails. It's not you and I bet you feel silly doing them. Just do what you do. Don't try and be something you're not.
I would like to see the strength of a dovetail if you have the time. I do mostly dovetails for friends and family and I have a hard time getting the dry fitted joints apart lol
It would be interesting to see what happens if you offset the joint so the joint is not directly under the pressure point. Would the non joined wood break before the joined part? Very interesting video! I love these!
Well learned something new but dont understand it how do u pre glue something also lol like welding the wood fails before the glue in welding the metal breakes b4 the weld does if things are done right lol
Whenever biscuit joints fail, the biscuit always seems to be fully intact, where as I would have expected it to kind of break apart? as the glue should be stonger than the wood right? Makes you wonder if the glue isn't bonding to the biscuits as best it could. I mean, I know a biscuit is no domino, or dowel, but it SHOULD do better in theory? it's just a loose tenon. It also is comparable to one of those thin finger joints you had, and is longer! .. so it should do better? whats going on?. Odd. I still think it's a bonding issue. One really interesting test would be to compare a standard biscuit vs one one you made yourself from solid wood. I reckon it would bond better and give better results.
My thoughts exactly. In theory the biscuit is a floating tenon that greatly increases the gluing surface and hence, in theory "should" perform better than a straight butt joint. I have noticed that some of my biscuits have a very smooth almost polished/waxy surface finish which may hinder the glue absorption? Other biscuits were dry with a very rough surface that immediately absorbed some of the glue and made for a very strong joint. When I first began using biscuits I glued some scraps up, let them dry for days and when I tested them the wood around the biscuit failed near the joint but not the joint itself. (Note: Just before I hit reply on my comment I had a memory that the wood may have been particle board. I know that particle board definitely fails before the biscuit as I have dismantled biscuit joined furniture and the material failed, not the joint.) It may be that biscuits are good but finicky with regards to moisture content, surface finish and the amount of glue used?
@@PaulScott_ Yes, something isn't right. It's almost like adding the biscuit weakens the joint to the point where it would probably be just as strong with the biscuit hole cut and then not actually having the biscuit (another interesting test).
In my opinion, biscuit joints aren't designed to add strength to a joint, they are for getting an exact alignment between two pieces that are being joined together. I know you stated that you made a particular effort to make sure it was all glued in the joint, but if that effort only concerned the biscuit and slot and not the end grain, then I can see why the first two joints bettered it. Great demonstration nonetheless.
This would be a better test for table tops. The type of failures may be different. In the tests about the tension is on the bottom of the board. Flat again the tension will be on the bottom and tension on the top. The wood joint will be put under different bending conditions. I would love to see the tests that you suggested.
Colin, I wonder if the glue was left to dry out for 24-72 hours would make finger joints even stronger. Also the whole glue technology asks for a deeper understanding. First how to help the penetration into and between wood fibers (like endgrain preglueing is one of its simpliest case). Then how thick the glue should be éhen applied, how much time to wait before joining the glued pieces, how strong hold them together, and let them apart, wood type, wood dryness, air humidity temp and glue drying time, glue hardening time, and so on. Finger joints are very very tight joints. I suspect some issue, too much glue squeeze out, less penetration. If you see the finger joints should be broken NOT AT THE JOINT SURFACE but anywhere else. I suspect almost zero penetration for that finger joints test, otherwise the joint surface would hold up and break anywhere else - knowing glue is harder than wood.
Hypothetical answer (i.e. just my personal speculation): Yes. Rationale: The logical extension of this idea is composite materials, and these are usually stronger in some way than their component materials. Plywood, for example.
Hello Colin! Very nice clip as usual! It would have been interesting to do the same test, but this time, using the other (width) side of the wood, and compare it with this test afterwards.
I was surprised the biscuit didn't do better. There was discussion below that biscuits have a waxy coating that reduces glue bond strength. This suggest some glues might work better than others with biscuits. As biscuits are used today I don't think strength of the joint is all that important. RUclips and woodworking literature is filled with testing that show that biscuit joints are weak. If the strength of the joint is an issue don't use biscuits. Pocket screws are similar. They are fast and good enough most places they are used. The biggest surprise for me was the dowel failures. I still am not sure what to make of that. I assumed you'd need to rip a board apart before a dowel would pull out. Not so I guess. The failure mechanism seems to be that the wood fibers fail where the dowel and the wood meet. Very surprising to me. But if that is the case then theoretically longer dowels would be stronger. I'm still betting on the dowels even if they lost this round to the finger joints. I would like to see a redo of this experiment with the dowel team getting a chance to make some changes.
I wonder how a jointless piece of wood would have performed as the control for the experiment? - That, a pre-glued finger joint and what about pocket holes perhaps? (Cheating I know, but still curious!)
this is the best place to learn new thing about wood and tricks to make things work better. i all way give this chanel high makes when the survey come up on how did you like the chanel
Would have been interesting to see you break a piece of wood. No joint, just a continuous piece of wood. Then we would have something to aspire to. Right now we are comparing "nothing" to something, but we don't know what the ultimate would be. No good idea how strong is strong enough. Surprising that some of the "improvements" weakened the joints. The biscuit is actually a cross grain connection in this case. Wonder what would happen if you brushed it's surface on a piece of medium sandpaper to get the "glaze" off the surface? Thanks for the test.
This was an interesting demonstration of SHEAR. In reality, any of the glued joints (beyond the first, 'glue and clamp') will sustain any practical application of force. An interesting follow-up, repeat the tests, except place a steel block between the press shaft and wood joint... say, three inches wide with 1.5" overlap each side of the joint. With the load distribution across a broader surface, you might find the results lean toward the biscuit or dowel joint. Why? Because of the mechanical advantage of either dowels or biscuits. This IS NOT a negative comment... I found it very interesting... and even more interesting if you carried it further with load distribution in lieu of a single point load! Thanks... I enjoyed the demonstration.
I had to re-watch that part of the video several times to understand what was going on, but in actual fact the joint was breaking apart and allowing the piston to move down slightly so the pressure didn’t go up, but at first it did look like the press had failed and not the joint.
More of a common need is to know how best to do “L” joints. And no one wants to do a detailed video about it, especially how to do it with a hardwood like Walnut. I tried Urea Formaldehyde Resin Glue, but applied too much clamping power and squeezed all the glue out of the joint, ruining the bond. I am going to Epoxy glue now, using only very light clamping. Colin, can you shed some light on this? Thanks good man! 💛😎
Just got this link- Slow-Mo made e laugh out loud. I'm a disabled builder & love pocket holes, so NO end grain to end grain for me & biscuits for alignment only. But i was curious about the other joints. My Daughter,♿Lisa🔧 started using dowels more after watching your other videos. She likes that dowels can be added for strength after a build. 👍She just got me to sign up for RUclips premium so i can watch your videos offline in my shop. She also reminded me to subscribe & ring the bell. I'm excited to be aboard when you hit your mark. Congrats in advance. your new friend♿Arch😎
HI, see that if consolidar this test, what I’m interpreting is the following, you are in fact seen glue failures in all the examples in my opinion. Consider that the rank os the result correlated to surface are in contact with the two pieces, the fact that non of the material failed, ie, failure correlated in the exact location of the flue area, neither the biscuit, the dowels, or the finges were deformed ( joint failure) but rated separated form the piece, one can concluidme this is really glue failure.
Of course, to make statistically valid comparisons, you would need to have multiple examples of each joint since the construction technique and wood properties could vary a bit between test pieces. But, that's still interesting information... None of those joints would I use in a load bearing situation unless there was a vertical member directly under the joint though. Another thing that sort of test rig would be useful for though would be testing the types of joints that we use on the legs of tables / workbenches -- assuming it can develop enough force to break the joints. By positioning the test piece at an angle, you could even test the resistance of a particular joint to "racking" (i.e. a side load causing the table to fold up and collapse).
That "pre-glue" result surprised me!
If you’re doing these sorts of testing type things in the future, I would highly suggest doing at least 3 of each test, so in this case if you had done 3 of each type of joint with same wood and procedures. That can’t help bring to light the variability in the test/materials, which can be surprisingly large sometimes.
3 to get the mean 😉
yes, and try hard to match cut and grain with all boards. i saw variation
Also, please add a non-joint variation for comparison. This illustrates how much weaker joining actually is.
Pre- Glue coming in #2! WOW
I know you addressed this, but particularly the scarf joint strength would be interesting to see
My first thought lol
Yes! Scarf vs half-lap is what we want!
Great demonstration, Colin! Thanks for sharing your findings with us.
How about the following scenario: I'm building a wood frame for a deck. Would it be stronger to have the long stiles on the outside of the rails or vice versa?
Very dramatic replays! I'd always had the idea that biscuits were intended more for alignment than increased strength. Thanks Colin.
I agree that biscuits are more for alignment than joint strength.
I'm gonna be using more dowels from now on.
@@markpowell6885 They are. The only end joint for strength in this plane for wood is the Japanese scarf joint (not the one he showed). Every one he tested should have been tested on the face plane, not the one he tested.
@@RobertErickson dowels and dominoes are best if not finger.
Brilliant! Always appreciate your videos Colin!
Yes, I’m here for the joint testing
When the first board broke, I was scared by the sound of breaking because I didn't have psychological preparation. I hope to see more mortise and tenon tests. Thank you for your video.
Very useful demonstration with important answers to a lot of questions.
How does this compare to a board that is whole?
I was wondering this , and also another setup where the pressure is applied on the wide flat side of wood (for example 10cm/4inch from the joint) and something's keeping the other side of joint sturdy. I think this is more similar to normal usage of wood.
Great video, Colin. I do feel you should re-test the biscuit joint.
Thanks for the great video Colin 👍 The testing gave some impressive results.
Interesting! A few suggestions from others for a sequel. Most valuable: (1) biscuits make the butt joint weaker. Not significant on a long edge joint where there is lots of wood with biscuits spaced for alignment. (2) preglue! I didn’t see any mention of Dominos, but that would be purely academic (for most of us). I would like to see the tests repeated on the face with a caul to spread the pressure. Thanks for an interesting, informative video.
Great video, Colin! Always a pleasure watching
You should have said “Always a pleasure, this time a pressure...”.
*AWESOME* test! Thank you very much Colin 😊👍
That’s very interesting. Thanks. I always enjoy your stuff and look forward to each episode
You need a much bigger sample to be able to draw any conclusion. That might answer your confusion with the biscuit join too.
Nice practicable vid. Answered my questions nicely. Thanks.
Thank you very much for the great information. I appreciate your effort in this wonderful educational video and wish you success and progress
Revealing demo! I will remember what you have shown. Thanks!
What about using for all glue ups, like cutting boards, table tops and the like?
A lock miter joint that can be used at both 90 and 0 degrees would be an interesting thing to see tested. I would imagine it might come close to the finger joint, or better. I'll check and see if you try it some day.
It would be really nice to actually test the lap joint, scarf joint, and a piece of solid wood to see how they compare. Other ones I would like to see are spline, and tongue and groove.
I think your assumption on the scarf joint is incorrect. I think the glue would still fail first. And in lieu of the lap joint, you should test a 12:1 aircraft style scarf joint angled from the wide side of the board rather than the narrow side as you did before. That gets rid of the weak area you mentioned in the lap joint.
Very interesting data from that testing... Thanks for doing the experiment 👊👍👍
Thanks for the test. I would also be interested in knowing the strength of each in the other direction.
Finally, a good comparison test, but now you have to do the same test, with the same joints, but on the horizontal plane.
Would be interesting to see dowel joint with glue on flat surfaces too, and not just in holes. Very good video Colin. What about pocket holes?
Interesting test Colin. I think most of it was testing the glue maybe. Joints with anything inside of them seemed to rob the endgrain of glue, and failed quicker than you may have thought. Plus I wonder if you had pressed down across the joint, and it's occupied reinforcements. IE: Across the Dowel, Biscuit, etc, if the joints wouldn't have been stronger? I agree finger joint made sense, and adding so much more glue surface, you got a stronger bonding.
Very Interesting, But I think that if you re-preform the same test with the same joints after the glue has set for a week or more, you might get different results.
Thanks Colin. Which glue did you use and how long was it left to cure/set?
Which glue did you use and how long was it left to cure/set?
I would prefer to see the strength of half-lap and scarf joints tested. A better test would be to secure one end of the boards with your strongest clamps, and hang weights on the end of the other board until the joint fails. Choosing the position of the fulcrum and length of the lever arms correctly should be enough to break even the stronger joints, so long as you have strong enough clamps to hold the other end.
Yes, those were the exact two joints I was most interested in seeing tested against each other so I was super-disappointed when he wasn't going to test them.
Awesome info! Thank you
Great help! Great content. Thanks!!
I wonder how much these numbers would change with a much larger sample, say 100 of each joint. Thanks Colin, Cheers!
Interesting video. I would not have thought that would be what the answers would be.
What about testing the same joints, but with the timber laying flat, rather than on narrow edge?
I would be curious to see pocket holes with and without glue.
Matthias Wandel did this in a test, with a better measuring system than this Press.
His numbers showed, that glue does NOT improve the strength of pocket hole joints. Some other woodworker also said the same. I think the woodwhisperer once also said, that glue does not improve the strength of pocket holes :=)
I've toyed with the idea of going Biscuit on a current project, but after seeing this, I've decided to stick with dowlels. Thanks for showing this!
I just bought one. The place I bought it said it's awesome. I'm joining a brace between legs on a small table. I've cut one side.. maybe not the other. Should have went with dows.
Great video, thanks. Id also be interested to see a tensile test on joints. I have to join two pieces and Im thinking of using a dowel screw to keep them from pulling apart and keep the joint hidden. If you have a better solution please let me know? thanks again
What was the psi of one solid piece?
Great info , Colin you must also do a Test without a joint on the same wood; so we know what is the breaking point of the wood without a joint and we can compare the strength, But once again thanks for sharing this info very helpfull.
Amazing video! Thanks 👍
I always enjoy these strength tests. Although I wonder if the dowel joint would have given different results of only two dowels were used. Thank you for the video!
I've never seen the pre-glue on the end grain technique before, that's pretty cool.
How do you do that, do you just wipe it on and wipe it off?
Seems logical that the pre-glued sample is really strong: pre-gluing not only seals but penetrate the grains and thus serves as microscopic dowels. Now it depends on how deep the penetration was before the wood are clamped, that why some suggest thinning the glue first with some water.
The format of this test reminds me of a contest hosted by the mechanical drawing and woodshop classes in my high school. Build something 6" x 6" X 6" minimum out of balsa wood and glue that weighs less than 3 nickles to see which supports the most weight. I think the record was something like 350 pounds. They did this every year.
Again Colin comes thru with some great info, I don't come up to shortboards much in my projects but... I'm thinking of purchasing a finger bit or two on my next trip to the store
I'd be interested to know why you skipped the half lap joint?
Surprised that the Biscuit joint was so easy to separate. Pre-gluing seems to be the way to go on Butt, or Half Lap joints. Educational and entertaining video. 🇬🇧
Hydraulic Press Channel meets woodworking....
I wonder if the biscuit would have done better if the pressure was perpendicular to the face of the wood.
I believe I can help with some of the confusion as to why the biscuit and dowels weren't as strong as expected.
Your test was set up such that bending would be the failure mode of each beam. The strength of a beam is a function not so much of how much material there is, but how that material is distributed. You can make a much stronger beam with less material just by arranging it efficiently. the material at the extreme top and bottom of the beam contributes the most to the beam's strength (side note: That's why I-beams are shaped the way they are. It's a very efficient way to get as much material as possible where it does the most good).
So in this case, we see that the biscuit and dowels failed at lower loads than expected. That's because they didn't engage the beam fibers at the top and bottom of the beam the way the other three joints did. So even though the total beam was the same depth, at the point of load (right under the press), you had less *effective* depth to work with, and thus a much reduced strength. For the mathematically inclined, the reduction is proportional to the square of the depth, so it's a bigger difference than if it were proportional to the depth itself.
Additionally, I think a big part of why the dowel and biscuit joint didn't crack the same way the others did is because they had those interior components. The dowels and the biscuits themselves didn't fail, so the beam was able to retain some strength just by bearing on those interior connectors and holding together with just friction. As
sacha schnauzer pointed out, this has the desirable effect of being much safer failure modes than a full rupture as with the butt and finger joints.
Would a longer dowels be stronger?
Great video!!
I've become a bit of a fan for the biscuit joint in certain applications. I have done my own tests and have found quite different results. The right size biscuit properly fitted and glued into a properly glued joint certainly does add strength, not as much as a figure joint but way more than a straight butt joint. I must admit however that I would never use a large biscuit in a joint like the one in your demonstration.
Yes. One other consideration: The more complete test should compare edge to edge joining, and then the biscuits : wood percentage would perform completely different; also the end grain structure on biscuit slim profile might perform very differently then long grain.
So the go-to should be the lap joint if strength is the main objective. If making a scrap wood table would just glued joints be sufficient and forego the fancy joints because the sides are being glued?
I suspect the biscuit joint to perform much better if you applied the pressure to the face of the board rather than the edge. can you repeat the test but this time apply pressure to the face of each board. thanks of the videos. very informative
Been watching your stuff for a while and I like the content you provide. it is insightfull, sometimes funny, always informative - you don't need to do the 'amazed/shocked face for your thumbnails. It's not you and I bet you feel silly doing them. Just do what you do. Don't try and be something you're not.
What about a pre glued, then regular finger joint using dowels?
So a finger joint with a dowel then no blow out
So, what effect would the direction the fingers are cut in (relative to the direction of force) have on the strength of the joint?
I would like to see the strength of a dovetail if you have the time. I do mostly dovetails for friends and family and I have a hard time getting the dry fitted joints apart lol
I just bought this bissket deal. Wondering now.
Thank you!
It would be interesting to see what happens if you offset the joint so the joint is not directly under the pressure point. Would the non joined wood break before the joined part?
Very interesting video! I love these!
I’d like to know what the board strength is by itself. This would tell the difference between a whole long board and a long board with a joint
Well learned something new but dont understand it how do u pre glue something also lol like welding the wood fails before the glue in welding the metal breakes b4 the weld does if things are done right lol
Ive got a 160T press at work and i always wondered if i press wood, would it become a super hard or too brittle
Whenever biscuit joints fail, the biscuit always seems to be fully intact, where as I would have expected it to kind of break apart? as the glue should be stonger than the wood right? Makes you wonder if the glue isn't bonding to the biscuits as best it could. I mean, I know a biscuit is no domino, or dowel, but it SHOULD do better in theory? it's just a loose tenon. It also is comparable to one of those thin finger joints you had, and is longer! .. so it should do better? whats going on?. Odd. I still think it's a bonding issue. One really interesting test would be to compare a standard biscuit vs one one you made yourself from solid wood. I reckon it would bond better and give better results.
My thoughts exactly. In theory the biscuit is a floating tenon that greatly increases the gluing surface and hence, in theory "should" perform better than a straight butt joint. I have noticed that some of my biscuits have a very smooth almost polished/waxy surface finish which may hinder the glue absorption? Other biscuits were dry with a very rough surface that immediately absorbed some of the glue and made for a very strong joint. When I first began using biscuits I glued some scraps up, let them dry for days and when I tested them the wood around the biscuit failed near the joint but not the joint itself. (Note: Just before I hit reply on my comment I had a memory that the wood may have been particle board. I know that particle board definitely fails before the biscuit as I have dismantled biscuit joined furniture and the material failed, not the joint.) It may be that biscuits are good but finicky with regards to moisture content, surface finish and the amount of glue used?
@@PaulScott_ Yes, something isn't right. It's almost like adding the biscuit weakens the joint to the point where it would probably be just as strong with the biscuit hole cut and then not actually having the biscuit (another interesting test).
Would like to have seen a test of solid wood. Interesting video, though. I enjoy all your videos.
In my opinion, biscuit joints aren't designed to add strength to a joint, they are for getting an exact alignment between two pieces that are being joined together.
I know you stated that you made a particular effort to make sure it was all glued in the joint, but if that effort only concerned the biscuit and slot and not the end grain, then I can see why the first two joints bettered it.
Great demonstration nonetheless.
If the pressure was applied to the flat surface instead of the edge, would the results be the same? It seems like the biscuit joint might do better.
This would be a better test for table tops. The type of failures may be different. In the tests about the tension is on the bottom of the board. Flat again the tension will be on the bottom and tension on the top. The wood joint will be put under different bending conditions. I would love to see the tests that you suggested.
What's the total length of the boards?
Colin, I wonder if the glue was left to dry out for 24-72 hours would make finger joints even stronger.
Also the whole glue technology asks for a deeper understanding. First how to help the penetration into and between wood fibers (like endgrain preglueing is one of its simpliest case). Then how thick the glue should be éhen applied, how much time to wait before joining the glued pieces, how strong hold them together, and let them apart, wood type, wood dryness, air humidity temp and glue drying time, glue hardening time, and so on.
Finger joints are very very tight joints. I suspect some issue, too much glue squeeze out, less penetration. If you see the finger joints should be broken NOT AT THE JOINT SURFACE but anywhere else. I suspect almost zero penetration for that finger joints test, otherwise the joint surface would hold up and break anywhere else - knowing glue is harder than wood.
Hypothetical question: if you cut a solid beam and put it back together with a lap joint, would it be stronger than if you had left it solid?
Hypothetical answer (i.e. just my personal speculation): Yes.
Rationale: The logical extension of this idea is composite materials, and these are usually stronger in some way than their component materials. Plywood, for example.
Hello Colin!
Very nice clip as usual!
It would have been interesting to do the same test, but this time, using the other (width) side of the wood, and compare it with this test afterwards.
I was surprised the biscuit didn't do better. There was discussion below that biscuits have a waxy coating that reduces glue bond strength. This suggest some glues might work better than others with biscuits. As biscuits are used today I don't think strength of the joint is all that important. RUclips and woodworking literature is filled with testing that show that biscuit joints are weak. If the strength of the joint is an issue don't use biscuits. Pocket screws are similar. They are fast and good enough most places they are used.
The biggest surprise for me was the dowel failures. I still am not sure what to make of that. I assumed you'd need to rip a board apart before a dowel would pull out. Not so I guess. The failure mechanism seems to be that the wood fibers fail where the dowel and the wood meet. Very surprising to me. But if that is the case then theoretically longer dowels would be stronger. I'm still betting on the dowels even if they lost this round to the finger joints. I would like to see a redo of this experiment with the dowel team getting a chance to make some changes.
I wonder how a jointless piece of wood would have performed as the control for the experiment? - That, a pre-glued finger joint and what about pocket holes perhaps? (Cheating I know, but still curious!)
this is the best place to learn new thing about wood and tricks to make things work better. i all way give this chanel high makes when the survey come up on how did you like the chanel
Interesting!! Thanks foto sharing...
Very interesting!!!!!!!!!!!!!!!!
Would have been interesting to see you break a piece of wood. No joint, just a continuous piece of wood. Then we would have something to aspire to. Right now we are comparing "nothing" to something, but we don't know what the ultimate would be. No good idea how strong is strong enough. Surprising that some of the "improvements" weakened the joints. The biscuit is actually a cross grain connection in this case. Wonder what would happen if you brushed it's surface on a piece of medium sandpaper to get the "glaze" off the surface? Thanks for the test.
I wonder how strong pocket screws are?
This was an interesting demonstration of SHEAR. In reality, any of the glued joints (beyond the first, 'glue and clamp') will sustain any practical application of force. An interesting follow-up, repeat the tests, except place a steel block between the press shaft and wood joint... say, three inches wide with 1.5" overlap each side of the joint. With the load distribution across a broader surface, you might find the results lean toward the biscuit or dowel joint. Why? Because of the mechanical advantage of either dowels or biscuits. This IS NOT a negative comment... I found it very interesting... and even more interesting if you carried it further with load distribution in lieu of a single point load! Thanks... I enjoyed the demonstration.
I don't understand. Neither the biscuit or the dowel joints failed in the video. But, you said they were the worst.
I agree. I don't understand why the gauge not going up is a failure of the joint. Seemed more like a failure of the press.
I had to re-watch that part of the video several times to understand what was going on, but in actual fact the joint was breaking apart and allowing the piston to move down slightly so the pressure didn’t go up, but at first it did look like the press had failed and not the joint.
More of a common need is to know how best to do “L” joints. And no one wants to do a detailed video about it, especially how to do it with a hardwood like Walnut. I tried Urea Formaldehyde Resin Glue, but applied too much clamping power and squeezed all the glue out of the joint, ruining the bond. I am going to Epoxy glue now, using only very light clamping. Colin, can you shed some light on this? Thanks good man! 💛😎
I was surprised by the biscuit joint. I would have expected it to do better.
Hi Colin, interesting test. Did you use white or yellow glue ? Thanks 🙂👍
It's crazy how much stronger simply pre-gluing makes a joint 🤯
What is the strength of a board not joined? Just a piece of wood?
Really interesting results, Colin!!! 😮
Great work!!!
Stay safe there with your family! 🖖😊
do People know what biscuit joints are for? They’ve gotten such a bad reputation because don’t get it.
Some of them really surprised me
Just got this link- Slow-Mo made e laugh out loud. I'm a disabled builder & love pocket holes, so NO end grain to end grain for me & biscuits for alignment only. But i was curious about the other joints. My Daughter,♿Lisa🔧 started using dowels more after watching your other videos. She likes that dowels can be added for strength after a build. 👍She just got me to sign up for RUclips premium so i can watch your videos offline in my shop. She also reminded me to subscribe & ring the bell. I'm excited to be aboard when you hit your mark. Congrats in advance. your new friend♿Arch😎
HI, see that if consolidar this test, what I’m interpreting is the following, you are in fact seen glue failures in all the examples in my opinion. Consider that the rank os the result correlated to surface are in contact with the two pieces, the fact that non of the material failed, ie, failure correlated in the exact location of the flue area, neither the biscuit, the dowels, or the finges were deformed ( joint failure) but rated separated form the piece, one can concluidme this is really glue failure.
Of course, to make statistically valid comparisons, you would need to have multiple examples of each joint since the construction technique and wood properties could vary a bit between test pieces. But, that's still interesting information...
None of those joints would I use in a load bearing situation unless there was a vertical member directly under the joint though.
Another thing that sort of test rig would be useful for though would be testing the types of joints that we use on the legs of tables / workbenches -- assuming it can develop enough force to break the joints. By positioning the test piece at an angle, you could even test the resistance of a particular joint to "racking" (i.e. a side load causing the table to fold up and collapse).
Very interesting,