Can reinforcing a joint actually weaken it?
HTML-код
- Опубликовано: 9 мар 2022
- I have this theory that adding pins across a joint may actually cause it to fail sooner. I made a whole lot of test samples to test this out, but it didn't go as I had expected.
Хобби
As an aerospace engineer, I can tell you that bonded & rivetted joints are common. This 'belt & braces' idea is that if the glue fails, the rivets will keep it together. So the rivets do not make the joint stronger (how could they?), they are a secondary load path. Neither do they weaken the joint if it is designed correctly. So I think your experiment proved this quite nicely. Thanks.
Hey Richard...if you see this I would love for you to read and comment on my post should you have anything to add, questions, or perhaps a challenge of my perspective...What I can share here at this time is that adhesives are serving a different function (in most applications) than the rivets alone...They are working in concert but both are critical to the load paths of such critical joinery. See my other post here for my reasoning and thanks for the conversation should we have it...
The dovels also help with keeping things in place while the glue hardens and allows the structure to be used/loaded earlier than otherwise.
@@Gin-toki They can also weaken the wood as any "glued joint" should never be loaded until the glue is set if the purpose is a long-term function...As stated before...a joint is either glued...or it is pinned...doing both is seldom necessary and typically out of the context of traditional joinery...Understand that "pinned" in the application of them in "good practice" means a "drawn joint" and not just sticking a dowel in it...
Richard, do bonded joints have no deflection? The only way the rivets don't do anything is if the materials have absolutely no movement relative to one another.
@@MrGuitarbike Structurally a working diaphragm (e.g. floor, wall, roof, or related joinery in either furniture or a timber frame) has several elements that work in concert to withstand the shear, compression, bending, and related load paths...
If these are just tunneled, keyed, wedged, etc fixing modalities...working in concert with friction, gravity, and the geometry of the joint is what holds them together...without adhesive...just as we find in traditional furniture, which typically...was not...glued at all...!!!
When an adhesive is added (aka a 'bonded joint" as you put it) that is what is the primary element that holds it all together regardless of the load paths...deflection/deformation of the joint can and does take place under load, but the "rivets" in Richard's case is not overtly loaded as the adhesive is doing the brunt of the work (95% and probably greater) just as he suggested...
"The only way the rivets don't do anything is if the materials have absolutely no movement," is not really accurate...They are not the primary element holding the joinery (regardless of type) together...the adhesive is...and they can in, many examples, just like pegs can weaken a joint which is the point the video was making...It is a common misconception that "pegging joints" make them stronger, which is seldom the case. In timber frames, as just one example, they hold certain frame assemblies together during raising but after that do little for the overall strength of the frame...and continue just to keep certain joints tight...yet...even in this case, many forms of "pegged joints" over time are proven to weaken the frame and fail for many reasons...
Love the scientific testing vs gut feeling. I want the actual better thing, not what I think is better so this is great!
But you design for a safety factor so that neither failure mode is likely to occur in normal usage.
But as Matthias said, it isn't usually done for strength. Dowels, metal star dowels and wedges are often used in the industry to back up the glue joint. It's not so much about reinforcement but rather reassurance. Especially outdoors it is good to not only rely on glue.
Except real science experiments or run at room temperature and this is clearly not since he's in a heavy sweatshirt
@@osufwiffo It absolutely is at room temperature, just the temperature of that room specifically...
I about lost it when all those glue joints failed, having seen all your previous tests over the years. That shocked the hell out of me.
I guess if it's below 55F, I should just use pocket holes.
nah, if it's below 55F just go back inside and have a nice cuppa and wait for warmer weather.
@@sturmey1966 Uhh, I'd be waiting a long time in these parts!
Cool your jets on those pocket holes.
@@Don.Challenger I happen to really like them. I'm not afraid to admit it!
😂😂😂
Humidity needs to be looked at because I bet it's extremely dry in that workshop.
Even getting the wood damp before applying the glue so it has more time to penetrate into the wood etc
Wouldn't matter since they were all glued under the same conditions. The test results are just as valid.
@@d4rk0v3 I think what he wanted to say is that if the gluing was done under better conditions, the results would better show if dowels make the joint weaker or stronger, as opposed to now when it was mostly glue failure.
Yep, making the glue not a dry joint.
@@wobblysauce yes exactly!
Just curious if the overall results would be stronger.
I'm in search of whats best.
The science is sound.
Just curious if another variable would produce a better product.
@@d4rk0v3 Not really, if the glue is failing first then the joint isn't being tested and the results are meaningless to any properly glued joint.
What a serious lesson in READING THE BOTTLE!!! I never worried about temperature unless it was so cold that I could see my breath, but I will definitely be putting a thermometer in my shop now.
Thank you Mad Scientist of Digital displays and fibrous fractures!! You are living the dream!
-CY Castor
That first test really goes to show why it is that people still struggle to believe that glue is stronger than the wood. Bad glue ups are all too easy to do, plus of course glue can fail over time. Likewise adding the dowels, as mentioned in the conclusion. Depending on the task, some sort of back up for the joint, in case of glue failure, is going to be a crucial addition.
but still... it took over 60 kg to break the glued joints. And that was with some leverage. My conclusion: for furniture that does not need to be undestructible for 100 years it will suffice to just use glue. a small wooden brace in a corner or diagonaly will add much more strenght than a dowel or some fancy finger joints, i think. for a table or chairs that have a lot of movement in different directions i would take more consideration. For a book shelf ore some TV stand i just glue the butts together put a screw or dowel in (i don't have clamps long enough) and maybe brace some corners with small wooden triangles. never failed on me.
@@alias_not_needed Yeah, but fancy joints gets the babes.
I love your scientific approach to woodworking. I would've never even considered dowel pins contributing to a broken joint.
I would be particularly interested in seeing how construction adhesive performs. I use it for crude joinery sometimes and I'm always impressed by how effective it is.
This!!!
Heisz is that you?
It might not be as strong, but could probably flex more before it breaks resulting in higher toughness.
Have you noticed PVA glue is not like it used to be? I use liquid nails on bed repairs! If you reinforce properly ( every situation is different) it will last! As u found out from my son's bed repair. I couldn't believe how it broke after reinforcing and gluing? Then I busted him and his older brother wrestling and jumping on the bed! Lol. So I reinforced it to oblivion! Lol. If I have to do this again? I'm welding up a steel frame strong enough to drive my car and park on top!!! 😂
@@dragan3290 The bed is poorly built then. I can jump on one foot on one slat in the middle of the span on my bed that has a design flaw that allows all the slats to rest on a 2x2 that is screwed, nailed and glued. 180 lbs on 12 square inches very live load.
@
Gareth Baus Polyurethane construction adhesive typically outperforms wood glue, especially in loose joints, and also over time and under sustained live or dead load.
I believe the real advantage to dowels through tenons (besides aesthetics) was for draw boring the joint closed, rather than adding strength to the glue joint. I am also curious to see this test on hardwood and if it has the same results
RIVET a welded joint ;-))
Many thanks I learned something new - Also made me think of the blacksmiths shop, happy days 4th generation now ended, it was started in 1890 when everything was riveted or fire welded in the forge. I now work with wood and find it more difficult than metalwork.
Fantastic video! Made a claim, tested it, presented all data whether it supported the claim or not, learned a bunch about tangentially related material, presented possible errors in data, and presented further testing to be done. I wish I worked with engineers half as thorough as you
Your approach to woodworking has been so helpful for me this many years. Thank you!
I really enjoy when your experiments have unexpected outcomes. Always lots to learn!
When you built this testing rid I knew we were going to see some very cool tests and you have not disappointed!
This one was super interesting, please keep them coming!
The Wood by Wright channel did a very extensive glue test already and I believe you can download his spreadsheets. That info. could save you a lot of time deciding which glues you would want to test for your content.
I was thinking about that too, but then it would be interesting if Matthias could reproduce those results!
@@daniel_bohrer I was just thinking if James already figured out which ones are really trash than Matthias could focus his energies on better candidates.
@@daniel_bohrer reproducing results--a cornerstone of good science!
I'd imagine Project Farm has done something too
For your glue test, different humidity (besides temperature) would also be interesting!
Pretty interesting tests, Matthias! And pretty curious results as well! 😃
I'm definitely looking forward to see more of those!!!
Stay safe there with your family! 🖖😊
I am SO glad that the room temperature during the glue-up didn't seem to make a difference. My basement is a little colder than my glue recommends and I would hate to feel guilty every time I glued something. You have brought me much zen today, Mr. Wandel. Many thanks.
Really enjoying your home built “Instron” style test fixture. I do destructive testing for a living with an Instron and other types of load/displacement fixtures. Very impressed with the simplicity and accuracy of your fixture! Testing data always reveals some very interesting surprises!
pointing out the ease of pulling it and re-gluing with the absence of a dowel makes way more sense to me. neat test tho, curious to see the glue comparisons!
Excellent!! A new test series! Looking forward to it.
Interesting results, looking forward to the future tests.
I found the temperature variation in glue drying shocking! Would love to see a series of tests on glue strength curing at different temperatures, say every 5 degrees Celsius? For those of us who work in chilly environments this would be hugely useful knowledge!
Very interesting results! Didn't expect them to fail through the glue. Can't wait to see what else you test
Always feel happy to see you matthias wandel
This is great Matthias ! I love this testing you are giving me answers to questions I have had in my mind for years but never experimented like this to prove my theory's. I now have the answers I often sought after. thank you so much for taking the time to do all this and to be so precise. and for sharing. ECF
Really great video, informative, and a really surprising result. Looking forward to a host of follow on videos this one raises. Thanks!
Holy crap, I did NOT expect those results! Thanks a lot Matthias, super interesting!
I always enjoy your videos, thanks Matthias!
Quite surprising that the glue failed first, I always thought wood glue was extremely strong. I wonder if when you test the different types of glue there'd be a noticable trend testing each of them done at a wide range of different temperatures.
The glue failed because he glued it below the temperature requirement on the bottle. When he redid it at the end at 10c hotter, it was almost always the wood that broke instead.
This biggest lesson from this test shows me that temperature at time of gluing is very important.
Matthias's tests are back! YES!!! I love these videos!!
I love these testing/experiment videos!
Those are some excellent tests! So interesting!!
Neat, I like the dowels aesthetically so it's nice to know they don't weaken the joint
Pretty cool experiment it sounds like you need stronger and more consistent wood. Also better glue adhesion. Thank you for sharing
I always thought antique furniture used dowels with mortise & tenon joints because the glue that was used was known to fail over time. With the dowels it would keep the joint together. With the good glues we have now, it's really not necessary unless you like the look, or you are using it to draw bore the joint together without clamps.
i'm glad we got to see the experiment even if the result were inconclusive
This was very informative! I have really been enjoying this video series, keep it up!
I really enjoy these videos because the data doesn't lie and doesn't care about what you "think" is correct.
I've ALWAYS wondered about this. I love these videos!
oh man... you read my mind on testing glues. project farm did one on wood glues back in 2000 you could reference. But this discovery of the glue failures seems like the most exciting part of this video. Something to learn and discover together!!! maybe brand is one variable... maybe temperature of application is another... maybe clamping force is one... maybe curing time is another. super fun stuff!
Good lesson in glue up temperature. I have some glue joints that were glued up while it was too cold. I hope the rest of the joinery compensates. Will be more aware of this in the future.
Nice experiment Matthias! Thank you for sharing the video with us!💖👍😎JP
Thank you Matthias, I always though about dowels in these kind of joints but doubted the reinforcing effect. Now I a have a answer. Grüße aus Deutschland...
I need a break from all the super-techy stuff I watch usually... SO I end up watching the woodworking guy I like.... Aaaaaand he's using a freaking linux-controlled test-rig. I love it. :)
Look forward to the glue test! Temp, open time, and curing time are interesting variables, if not also degradation with age and gap filling.
Another great as well as informing YTV . Thanks !
Thanks for sharing these test it takes a lot of the guess work out. Great job awesome channel
Failure in a way that can be fixed is better than catastrophic failure, in my opinion. That argues against pins.
Drawbored joints are handy if you don’t have clamps, but otherwise not needed. A single pin could hold a joint together, but not need to be close to an edge to weaken the joint.
Well, a failure of the joint that can be fixed is often a catastrophic failure of the whole structure. Better cut new boards for the swings than search for the whole seat over the neighbour's fence, fighting off the guard dog ;)
A jig with wedges can work too, depending on goal with the joint.
you can remove old pins and fix that too...
but with pins you have crack 1 and then crack 2 ... guess how mutch you and your electrons can do in that extra time, before the 2.crack gives a catastropic failure.
Greatz from Germany
and have a nice Day
opo
Thanks for the testing! I always thought glue alone was not sufficient, so it's good to see it is actually stronger than I thought.
How about just one dowel in the center, so it has a maximum amount of wood around it? And it seems like you should try much bigger dowels. I think one 18mm diameter dowel in the center might be optimal for strength, but maybe a little strange looking.
stop it dowel guy
@@MrSubjetivo HAHA
I would only go for it, if you think it is a critical component, where you don't trust the glue. But ideally, you should do the glue up in warm environment to dry, clearly.
@@MrSubjetivo Matt V is an agent of Big Dowel.
Wouldn't a dowel in the center just weaken the joint because the piece could move around it? The dowel would act as a hinge in a sense... but it would be interesting to see.
Excellent testing setup, makes it pretty clear that they don't improve the joint strength. I'm glad you kept in the part about the cold glue, I would have never expected it to be so significant!
Personally I'd argue that the dowels holding the boards together if the glue fails as a negative thing. If it fails from a continued force (weight/gravity), the dowels are going to give way at the same time as the glue anyways. If the force isnt continued or reduced after the break (like your jack), your test shows it can simply hide the destroyed joint surprisingly well. At best it's going to give out anyways and potentiality cause even more damage, at worst it's an avoidable safety issue.
Agree with your findings. My own experience is that dowel etc don’t generally add strength and if too close to the edge of the material they actually make it weaker - cause you’ve removed wood and interrupted the grain. As you said, it’s harder to repair too. Dowels etc are useful for alignment so I do use them at times - but with great care
Very interesting! I would have never thought that the glue would fail so repeatable! Looking forward to the Glue-testing!
Same here
Tests similar to this one that I've seen had the load pulling down. Which leads to the question if that makes a difference. Also; many of the pinned joints are made where the holes do NOT line up and the pin, when forced into the holes, deforms AND pulls the joint together. Common in chairs and tables, where if the joints' glue fails you don't want total failure.
I think the real test here was glue temp being a factor. Great video as always.
Very useful info! It can help woodworkers like me!
Thank you!
I'm so glad to hear you say that you are going to experiment with different glues. I often see titebond being used and always wonder if it is as good as it is led to believe. it sure gets a lot of free advertising
Very interesting and thorough!
Beauty vid Mattias!
Matthias, Surprising glue failure results for the first “cold” setup joints. Good that is was much better a bond on the warmer setups. The dowels were also a bit more underwhelming than I thought they would be. Looking forward to the different glue tests.
Loved it.... from an old Quality Control Engineer.
In timber framing, dowels (pegs or trunnels) are used to hold mortise and tenon joints together (but wood glue is not used). It's also nice to be able to disassemble the joints. The pegs are also used to draw-bore joints together so they are tight. The point is that pegs are not used to solely strengthen a joint.
Great video! Good points at the end, too. I guess it depends on the use case. If it’s not that important whether the joint holds together, it would be more convenient to be able to just easily re-glue it.
I love all your informative videos
Fascinating. Thanks for sharing!
Great video can’t wait for the next one
How sooon?? Very soon! You are so analytical Matthias. As one engineer to another, I take my hat off to you every time sir. just wished we were neighbors.
Matthias is already ahead of other woodworkes again. Everybody builds more or less sophisticated equipment and furniture, but Matthias breaks stuff in a scientific way. :-) I like it. And according other comments, I'm not alone. The woodworking myth busters is definitely a promising direction as well as comparing various products/methods etc.
The thing the doweling bring to the party is slow down catastrophic failure. When the glue breaks, no real joint damage yet. Drill out pins and redo the glue up. No extra repair of damaged wood. Great thought about whether
they weaken the joint or not.
BTW, as stated by others here. nothing wrong with pocket hole joinery when used in the right situation. PH's can be stronger than doewls if put in the right places. Thing about it.
Awesome testing
It's always cool to see a new video from you, Matthias! The failed glue joints seemed to be very smooth, I wonder if it would help to "roughen" the glued parts so that the glue has a better chance to penetrate the wood. Or that the glue has more surface to hold onto.
Excellent experimental setup and very fair analysis. You could always do an ANOVA for group to group comparison! Boxplots are also pretty :)
Definitely agree with your point that I'd rather be able to reglue the joint than have my wood shatter using dowels
Interesting, I like your analysis of the joint with the dowel holding the piece together after the glue fails on the joint. Personally, I’d rather the joint held together with the dowels until I had time to repair the joint, it wouldn’t take much to go to the drill press and drill out a couple dowels.
I was shocked when the glue joints failed before the wood, even with the temperatures a little lower than the recommended amount. My first thought was maybe the glue was old but that didn’t seem to be the case. Id be interested in seeing what an old glue would break at compared to a glue that was fresh or still within its use before date.
I’m looking forward to seeing the results of the glue testing, I hope you use some of the different types of glue for the testing. There seems to be quite a few types for different applications for wood glues. Thanks!
Comment before watching: yes, absolutely
Upd: what a nice setup and convincing results! So it true to say that dowels are rather a safety measure against catastrophic failure
With that in mind, would be interesting to compare double-pinned setup with single-pinned (bigger dowel, crossectional area equivalent to two smaller) Theoretically might be stronger because of lower contact stress or weaker because of higher probability of grain defects around the dowel
And since everyone is commenting on the gluing technique, I'd recommend modifying the jig setup in order not to give such clean and even surfaces
Seriously looking forward to the glue tests!
This is really good!
Great work👌
Great vid as always
I'm definitely going to think twice before gluing in the cold now!
When you do your tests of glues it would be interesting to see how basic builders PVA (the stuff used for sealing plaster etc.) stands up compared to the more expensive types of actual wood glue.
Also seeing how cyanoacrilite (super glue) and hot melt glue (applied to preheated wood ofc) works would be nice. Preheating the surfaces does wonders for hot glue adhesion and strength.
I love your scientific testing. I look forward to the testing of glues! :)
Interesting testimony to glueing in the cold! Excited about the glue tests coming up!
How old was the glue? Had it been stored in a place that freezes?
Depending on the expected load angle on the joint, the dowels could be moved to more optimal locations. For example, the first dowel orientation, where the one dowel became a hinge, and the other caused a tear out of part of the tenon, move that dowel closer to the middle of the tenon. Probably keep it about the same distance from the edge on the bridle though.
Test joints with just dowels, and no glue? Just for comparison?
Thanks. All of the testing you have done is greatly appreciated.
I'd be interested to see how mistreating glues affect their performance too. You touched on the temperature here, and have done gaps and clamping force in the past...I wonder how it would work out gluing damp wood, or gluing dusty wood, or using previously-frozen glue, letting it cure in very-humid or very-dry space, etc.
That said, I am also interested in your "actual" projects too, so maybe we don't need to go all the way down the force testing jig.
Great video.
I know you're always thinking about optimizing for youtube. A suggestion: "before you tell me that i should test glues" - I think youtube parlance is "leave a note in the comments if you would like to see me test glues - if enough of you wan to see it, I'll do it". It does the thing you want (which is communicate you are thinking about it) but is also good for engagement.
In Sum, glue up in warm temps rather than cold.😉 Thank your for doing this.👍
One of the issues I've experienced with with joints cut with precise jigs is inadequate space for glue causing weak joints. I assign the root cause of "inadequate glue" as opposed to a "glue failure." As was identified earlier in the video, there was a lot of glue displaced on the cheeks of the tennon, where the best long grain adhesion occurs.
Fantastic content! 😁
I would like to see old titebond vs new fresh titebond. Excellent work by the way.
Most interesting!!! Great work! What about nails and screwes?
You are a good scientist! I had always heard that the glue is stronger than the wood but it seems perhaps not with the Elmers…I’m looking forward to the glue tests.
I love a good experiment!
Looking forward to glue testing!
This is great. Thanks
I think Project Farm did a pretty good glue test as well! Excited to see what your results look like
Excelente lo del experimento Matias, me parece bien los tarugos en forma diagonal para mejorar la resistencia a la tensión, también se recomienda aplicar el pegamento en ambas piezas a unir, entre más tiempo de secado mejor agarre tendrá, saludos
Thank you! Very interesting.