It's the word key that throws me here. I'm thinking key as in key stone, absorbing pressure from two sides co pressure, com-pression. It could be that I'm not familiar enough with your timber framing jargon in this case.

As for splitting, I don't think it's an issue of the peg, round or not. It is another matter altogether, could be in the area of accuracy and appropriateness of the joinery, something like that. Take the simple example of hammering a nail into a board. If I hammer a relatively small nail close to the end of a far thicker board of say spruce it's likely to split that board. At the same time, if I clamp a thin strip of lattice, of the same wood, in a vice, (I could just as well pre-drill for the nail. Or don't forget blunting the tip, that one works wonders, or even lubricating the nail, sometimes I use that one too), I can hammer a big nail just as close to the end take it from the vice and there will be no weakening of the wood relative to the nail. If I split the wood then the split is just as likely to go around the nail/hole. In other words it's the wood structure influencing the split not the nail or peg in our case.

I think there is a notion of the mechanics of the joinery and the structure and properties of the wood going on here that may be intuitive but is incorrect.

Another thing is the matter of which looks better, round-like or square-ish. And to add to that, in my examples up there the openings for the pegs are off-set, so some drawn effect taking place on those. I was surprised that there was no splitting with them.

Greetings,

Don Wagstaff

Don, from the TF Glossary

KEY. Small element, usually wedge-shaped, used to lock
a joint or, if a shear key, to prevent sliding of one member
over another.
KEYED BEAM. Two or more beams laminated together
with crossgrain keys let in between the beams to prevent
slippage during bending.
SHEAR BLOCK. Wood block dapped (q.v.) partially to
adjoining parallel laminae in a built-up chord, designed to
resist shear between the two members or to transfer load
around a discontinuity such as a scarf, and properly oriented
parallel to the grain, so that shear block end grain bears
upon chord end grain.
SHEAR KEY. Wood block oriented perpendicular to
(across) the grain. Easier to assemble, and can be tightened
if wedge-shaped, but not so resistant to compression
as a shear block (q.v.).

In my usage a key has a flat surface oriented square to the force, except as a sloping shear block. The example of the rotated key is just a simple visual example of how sloping surfaces can impart force, a peg has a sloping surface that varies from parallel to square to parallel from the direction of force.

So, how does riving force matter? In a quiet frame not so much, the pegs function as place holders however in a tension generating frame with open joinery. such as a frame using free tenons, I do have concerns over the chance that the tenon will split. So in that case, square pegs or keys seen like a better choice.

In Swiss architecture, it is very common for sills to be joined together with a keyed double tenon. The tenons pass far out the other side, and are secured in place with a strong key, creating a joint which can move no where.

The preferred practice in this situation is that the key be square and straight, not tapered. This requires the joint to be very carefully formed as you cannot rely on a wedged key to draw it together. But the thought is that it is stronger this way. Also, a straight key does not work itself out over time.

And then, on framed buildings, pegging is often totally absent.

For you reading pleasure, the dissertation "Investigation of Through-Tenon Keys on the Tensile Strength of Mortise and Tenon Joints" by Lance David Shields

http://scholar.lib.vt.edu/theses/available/etd-07112011-100606/unrestricted/Shields_LD_T_2011.pdf

Jim

I only breezed about the pdf, and saw no mention of round keys. I would be very interested in info on round keys. Were they ever used in the past and if not why not?

Split ring connectors would be one instance of a widely used round shear keys. NDS will have useful specs regarding their use.

Pegs are roundish, too. Same effects. In the link Jim provided I noticed nice clear stock for their test, for predictable constancy. Where are the test with knots at strategic locations? Say, the key placed right where a knot is, or have a knot just below the key in the shear plane. Any test with this, how about at any of the guild events where they have busted joints? Results? Is it worth it to take the time to lay out keys on through tenons with knots in mind?

Round keys have also been used between tie and top plates, at the edge of the tie, aiding in withdrawal of the tie from the top plate as thrust is applied to the wall from the rafters. Most likely bored and installed after the structure was assembled.
Seen in HATJ A Graphic Guide, pages 12 and 13, fig. 15,16 &17. They are not labeled as keys but pegs.

Hello,

They've crunched the numbers, but I think they've cooked the books. I'm a bit with Tim on this one, not being such a reductionist particularly in the matter of wood.

Greetings,

Don Wagstaff