LumberJocks Woodworking Forum banner

Dowel joinery

2K views 21 replies 13 participants last post by  ed13 
#1 · (Edited by Moderator)
I'm a fan of mortise and tenon joints but I've used dowels for a couple small end tables. They seem to be pretty strong and are not that hard to setup but you have to be very accurate with them. I was browsing around for the strength of the joint and came across this:

https://www.canadianwoodworking.com/get-more/how-strong-dowel-joint

I'm curious on what other people's opinions are on dowel joints.
 
#2 ·
That''s a pretty cool article Matthias Wandel did a pretty cool analysis too on Youtube. Here is the link. Turns out doing things the old fashioned way works.

 
#3 ·
What a Rube Goldberg type apparatus Matthias came up with! Why not just use a hydraulic jack?

In most cases, strength tests like these are not of much value. What you are measuring is whether the joint is 10 or 20 times as strong as it needs to be or just 2 or 3 times as strong as necessary. I use a lot of pocket hole joints even thought I know they aren't as strong as dowel or M&T joints because they are good enough for what I need. Why waste time on anything more for something like shop cabinets?

I mostly use dowel joints where strength is important and I don't want the fasteners to show even though I know M&T is stronger. Once again, why waste time?

Festool claims that the Domino floating tenon machine makes stronger joints because they compare 1 Domino to 1 dowel. They even published a video about it. I thought it was ridiculous. I almost never use just 1 dowel in a joint. If you have a good jig and use as many as you can fit in a joint, it is a lot stronger than a Domino joint. If Festool wants to impress me, they need to perform a more realistic test.
 
#4 ·
Here's how it was explained to me: Imagine a classic M&T joining a rail and stile for a door. Look inside the mortise in the stile. The big surfaces that mate with the tenon cheeks are face grain and glue up strongly with the tenon cheeck face grain. The narrow sides of the mortise are end grain and, even though the corresponding face on the tenon is face grain, the glue-up isn't strong in tension because of the end grain.

Now consider a rail joined to a stile with a dowel. The dowel is face grain all the way around. Look inside the hole in the stile that will receive the dowel. The sides of the hole (width-wise) correspond to long fibers and can bond strongly. The top and bottom of the hole (along the length of the stile) are pure end grain, and will bond weakly. All the surface area of the hole between these two extremes are somewhere in between pure long grain and pure end grain, so they are intermediate in strength. So, even if you make equal surface area, dowel vs. traditional tenon, much of the area in the hole has an end grain component and can't be as strong as the classic tenon, which is a nearly perfect face-grain glue-up.
 
#5 ·
What a Rube Goldberg type apparatus Matthias came up with! Why not just use a hydraulic jack?

In most cases, strength tests like these are not of much value. What you are measuring is whether the joint is 10 or 20 times as strong as it needs to be or just 2 or 3 times as strong as necessary. I use a lot of pocket hole joints even thought I know they aren t as strong as dowel or M&T joints because they are good enough for what I need. Why waste time on anything more for something like shop cabinets?

I mostly use dowel joints where strength is important and I don t want the fasteners to show even though I know M&T is stronger. Once again, why waste time?

Festool claims that the Domino floating tenon machine makes stronger joints because they compare 1 Domino to 1 dowel. They even published a video about it. I thought it was ridiculous. I almost never use just 1 dowel in a joint. If you have a good jig and use as many as you can fit in a joint, it is a lot stronger than a Domino joint. If Festool wants to impress me, they need to perform a more realistic test.

- ArtMann
Yep- if I do dowel joints, I use 3 dowels per joint most of the time.
 
#6 ·
Here s how it was explained to me: Imagine a classic M&T joining a rail and stile for a door. Look inside the mortise in the stile. The big surfaces that mate with the tenon cheeks are face grain and glue up strongly with the tenon cheeck face grain. The narrow sides of the mortise are end grain and, even though the corresponding face on the tenon is face grain, the glue-up isn t strong in tension because of the end grain.

Now consider a rail joined to a stile with a dowel. The dowel is face grain all the way around. Look inside the hole in the stile that will receive the dowel. The sides of the hole (width-wise) correspond to long fibers and can bond strongly. The top and bottom of the hole (along the length of the stile) are pure end grain, and will bond weakly. All the surface area of the hole between these two extremes are somewhere in between pure long grain and pure end grain, so they are intermediate in strength. So, even if you make equal surface area, dowel vs. traditional tenon, much of the area in the hole has an end grain component and can t be as strong as the classic tenon, which is a nearly perfect face-grain glue-up.

- ed13
No doubt M&T is a really strong joint. Even better when pinned. It is nice to have such a variety of options for joinery. That is the great thing about woodworking. So many different ways of doing things.
 
#8 ·
I was told the festool domino is good because it takes the guesswork out of joinery. But yea, you should build your stuff with the stresses you expect to encounter. Alot of times I end up overscrewing, waterproofing, and finishing the underside of things when it's not needed. Don't be stupid like me.
 
#9 ·
Here s how it was explained to me: Imagine a classic M&T joining a rail and stile for a door. Look inside the mortise in the stile. The big surfaces that mate with the tenon cheeks are face grain and glue up strongly with the tenon cheeck face grain. The narrow sides of the mortise are end grain and, even though the corresponding face on the tenon is face grain, the glue-up isn t strong in tension because of the end grain.

Now consider a rail joined to a stile with a dowel. The dowel is face grain all the way around. Look inside the hole in the stile that will receive the dowel. The sides of the hole (width-wise) correspond to long fibers and can bond strongly. The top and bottom of the hole (along the length of the stile) are pure end grain, and will bond weakly. All the surface area of the hole between these two extremes are somewhere in between pure long grain and pure end grain, so they are intermediate in strength. So, even if you make equal surface area, dowel vs. traditional tenon, much of the area in the hole has an end grain component and can t be as strong as the classic tenon, which is a nearly perfect face-grain glue-up.

- ed13
Hmm, I see where you're coming from, but I think you're just repeating the same test as Festool that ArtMan mentioned in the post above yours. A nice thick tenon vs a single dowel.

If you size that tenon down to the size of a dowel, I think they would just about be equivalent in strength. If you look in the mortise/hole in the stile, you'd have end grain on top and bottom of the mortise/hole, but long grain on the sides and end. In the rail, you'd have long grain all the way around the mortise/hole and end grain only at the end of the hole which would be just as strong as a tenon of the same size.

And like ArtMan says in an earlier post, we only need to exceed the expected stress by a bit, 2 or 3 times vs 10 or 20 times. If you're in a production shop, dowels are much faster than m&t are a significant cost savings with only a little compromise on joint strength.

That said, if you enjoy making m&t, by all means continue to do so. Have fun and enjoy the craftsmanship.
 
#10 ·
Way back when i was experimenting with the strength of glue and glue joints, I found that the simple half lap joint is stronger than the M&T joints I made. The rail or stile piece would break before the glue joint. That was not true of M&T. I made cabinet doors that way back in the early 1980's and I haven't heard of a failure yet, although there could be. The problem with them is they are somewhat amateurish looking.
 
#11 ·
@ArtMann what made them look amateurish? If it's what I think it is, did you ever try a mitered half-lap? I'm guessing your half-lap divided something like 1/3 thickness against 2/3 thickness so that you could run a groove for the panel rather than being 1/2 thickness against half thickness?
 
#12 ·
Weird that dowels are coming back into fashion, memories are short. They fell out of favor because they don't hold over the long term but nowadays nobody keeps furniture long term so it probably doesn't matter. The new fluted dowels might hold up better, probably will.
 
#13 ·
Perhaps your dowel joints don't hold up well but I started using them in 1977 and I have yet to replace or repair a piece I built with them. I have heard that dowels don't hold up too but i have never seen any credible evidence of this. Can you please direct me to where I can investigate?
 
#14 ·
+1 on ArtMann. Why not just build the joint to the purpose needed. I seen people frown when I've told them some of my joints are doweled, and give a nod when I've told them that this joint was made with mortise & tenon. Some have been surprised when I've told them the joint was a locking miter joint. The article didn't mention anything about pinning a mortise & tenon joint.
 
#15 ·
I think it's interesting that the usual way to test joint strength is to glue up one joint, and bend it until it breaks. A chair or cabinet door would never have just one joint. It has 4 or 8 or more joints. Furniture failure is rarely from one peak force, but more often from repeated strains.

I don't know a better way to test joint strength, but I look at old furniture that has come apart and try to see what failed. A cross-pegged mortise and tenon could have the glue completely fail, and it would still be a sound joint. When a dowel fails, it lets the joint open up. In a way, loose tenons are just flat dowels, albeit with a little more surface area. For me, it's hard to beat the strength of an integral tenon.

Of course I do use biscuits and dowels occasionally, just not where a lot of strength is required.
 
#16 ·
Weird that dowels are coming back into fashion, memories are short. They fell out of favor because they don t hold over the long term but nowadays nobody keeps furniture long term so it probably doesn t matter. The new fluted dowels might hold up better, probably will.

- Woodknack
I think they got that reputation from a long time ago from being used by furniture factories who didn't use enough glue and didn't do a good job using dowels. Modern glues and multiple dowels per joint are really strong.
 
#18 ·
Every once and while someone give me a table or door for the wood. The last one was a oak table all doweled together. It came apart pretty easy. I even found areas where the glue was squirted right out of the bottle with a curved line. Those areas came apart very easily.
Maybe it was old glue
I don't use dowels from what I've seen they don't hold.
 
#19 ·
Every once and while someone gives me a table or door for the wood. The last one was a oak table all doweled together. It came apart pretty easy. I even found areas where the glue was squirted right out of the bottle with a curved line. Those areas came apart very easily.
Maybe it was old glue
I don't use dowels from what I've seen they don't hold.
 
#20 ·
Every once and while someone gives me a table or door for the wood. The last one was a oak table all doweled together. It came apart pretty easy. I even found areas where the glue was squirted right out of the bottle with a curved line. Those areas came apart very easily.
Maybe it was old glue
I don't use dowels from what I've seen they don't hold.

- Aj2
Just thinking if they didn't bother to spread the glue they probable didn't put any on the dowels either.
If they didn't the dowels served as a line up tool and nothing more.
 
#21 ·
When dowels are compared against M&T in a scientifically controlled situation, the M&T comes out better - - but not by a whole lot. What is it that changes over time to make the dowel joint less strong but does not affect the M&T? The evidence I have seen of failed dowel joints involves badly built furniture that wouldn't have survived better with any other kind of joint. Williamsburg, Virginia has many historic museum houses with doweled furniture that has survived unchanged since the 1600's. It is hard to refute that kind of evidence.
 
#22 ·
That's interesting (1600's). At that time, would they have made their own dowels from the exact same wood as the wood being joined rather than using manufactured dowels of some other wood? Were they using dry dowels in green (wet) holes? Would the seasoning method (air seasoned then vs. kiln now) matter? Did the old growth wood move differently vs. modern harvest? I don't know the answers to any of these.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top