Joint Strength- Biscuits vs. Dowels vs. Loose Tenons?

[Gator95]

Is there a general consensus on the strength of a 90* butt joint reinforced with these three common means? I've read conflicting reports googling it.

When using high quality modern glues, are each of the 3 'good enough' for reinforcing butt joints in furnature frames and tables construction?

For purposes of discussion, assume fluted dowels like these: http://www.rockler.com/product.cfm?p...&filter=dowels
with 2-3 per joint.

Assume 2 #20 biscuits per joint.

For Loose tenons, assume a 3/8" x 1.5" long x 1" deep mortice on each side of joint and a hardwood loose tenon of same species as wood being joined.

[Russianwolf]

You'll need a lot more info. How thick is the stock being joined? What is the material?

Some woods will be more likely to break with the loose tennon as you've removed too much of the wood to insert it, giving the dowel or biscuit joint an advantage. Other woods won't have this problem as they are inherently stronger.

Also, you mention furniture frames and tables. Well, for face frames any and all will do fine since they usually don't carry much of a load. For a table or chair that will carry a load it can matter.

So to answer, you need a specific example to get to a specific answer. All three have situations in which they will work fine, and situations in which they won't do well.

[pelligrini]

I never really read all the articles I've seen on the strengths very closely. I do seem to remember that biscuits don't really add much strength to a joint.

[Bill in Buena Park]

I believe biscuits don't provide good "flex strength" - but I use 'em for 90dg butt joints, and have great results - but then again, I use biscuits like spline - I place one every 3-4 lin in of the joint. Have never had a joint "pull apart".

[JimD]

In the tests Wood magazine did, the mortise and tenon was the strongest of the joints you name and the biscuits the weakest. They used a single #10, however and two dowels, I think they were 3/8. More surface area would increase the strength of the biscuit joint. They were joining pretty narrow pieces where a #20 was too wide. A loose mortise and tenon is not weaker or stronger than a more traditional mortise and tenon as long as it fits well and is appropriately glued.

I've had good luck with interior doors made with as many biscuits as I could fit in. I made one with dowels and it did not hold together as well. Maybe it was me. My most recent doors used mortise and tenon and are holding together very well.

For little lightly stressed stuff like face frames for cabinets, I tend to use pocket screws. For load bearing situations I try and use mortise and tenon (usually traditional but I also do loose - I have a small hollow chisel mortiser). If the pieces are wide and the load is not real high, I will use biscuits. For the joints you name where there is only wood, you would have a decent idea of the strength difference if you look at the cross sectional area difference for shear loads and the surface area difference for pull apart loads. This is probably a bit kind to dowels because you do not have ideal long grain to long grain for as much of the area but it would give you an ideal of the relative strength.

Jim

[cgallery]

I think biscuits have gotten a bad wrap.

First, those joint torture tests isolate a single joint and pull it or shear it apart. Valid, certainly, if we're talking about the construction of a chair, where the forces can be substantial.

But in terms of casework, no concern whatsoever. Sure, the other methods are probably stronger than a typical biscuit. But once I put two sides, top, bottom, and back together (typical five-sided box), the strength of each joint becomes depended on the unit.

With a back on, there is no racking. Remove the back and rack a unit with biscuits, and you're now racking four joints. Still very strong.

And most of the joint torture tests I've seen indicate a failure in the biscuit itself. The biscuit simply breaks. There are, however, plywood biscuits ("normal" biscuits are made of beech, I believe). The plywood biscuits, I'm told and I've read, are much stronger. Never used them, though, because I've been happy with standard biscuits.

My Makita biscuit joiner is fast and precise. It absolutely helps in aligning in the money plane.

I'm not saying biscuits are the solutioni to every wood joint. But they are great some of the time. A lot of the time, for me.

[drumpriest]

I just posted this a couple of days ago.

http://www.youtube.com/watch?v=DhLfb7m9Fug

Mortise and Tenon is pretty much the strongest joint. This is assuming an integral tenon, which is not the same as loose tenon. Some test have shown that loose tenon has roughly 95% of the strength of integral tenon. This is a broad percantage, of course, as type of wood matters here. For instance, a wood that is very straight grain, with long fibers will be stronger with integral tenon in comparison to something with curly figure, or wavy figure. The latter will compare much like loose tenon.

The real question is what is "good enough", for instance, an non reinforced miter failted at 139 lbs of pressure, with reinforcement you get into 500+ lbs pressure, and some M&T pull apart as high as 4700 lbs. So when making a picture frame, are you going to apply more than 139 lbs of force to a single corner? Probably not, adding a biscuit can't hurt, but even if it doesn't help but for alignment, who cares??

[cabinetman]

Since the question is non specific other than a 90 deg butt joint in furniture frames and tables, I'm adding a M&T.

IMO, an M&T is the strongest of joints.

A loose tenon would be second on my list. I don't consider them as strong as a M&T, as there is a glue joint on both ends instead of one end.

I vote dowels third on the list. In actuality, a dowel can be considered a loose tenon.

And last on my list are biscuits. They may help in general alignment, but IMO, do not insure alignment. When used in connection with cauls, clamps, and wedges if needed, biscuits may provide the joint from pulling straight apart. They have relatively little torsional strength. I would rather use splines instead of biscuits.
.

[JimD]

I hessitate a little to come back to this because I think I'm splitting hairs somewhat in this response but properly made joints do not fail at the glue line. They fail in the wood. For that reason, I think loose mortise and tenon are as strong as traditional - if and only if they fit well. I leave a little space on the ends of the loose mortises sometimes to give me a little wiggle room on the alignment during assembly and if you do that, you loose a little strength.

I also agree with cggallery on biscuits. They are better for many applications than the Wood magazine test would indicate. The issue with their test as far as being a general guide is the narrow width of the boards and resultant use of a single #10 biscuit. There was little glueing area and little cross section to the biscuit so the joint was not real strong - but still noticably stronger than the joint without reinforcement. The slot made by the biscuit slot cutter is a noticably good fit to a biscuit which then swells with glue contact so you get a really good tight joint - but even a #20 biscuit is not a lot of surface area or cross sectional area. Dowels give you more cross section but I question whether at least I can get as good a fit in a dowel hole as I get with a biscuit slot cutter. In other words, I think at least for me and probably typically, the fit of the joint is better with a biscuit than with a dowel. That is only one factor. I also use splines sometimes but the fit is not as tight when I do a spline as I get routinely with a biscuit. So I trade more surface area for a looser fit.

The type joint I am least likely to use, if you can't tell already is a dowel joint. My success with them has been limited and I think there are better options. Some of you appear to feel about the same way about biscuits. I'm happy we have choices.

Jim