NH - How would this work? Assuming someone puts a floor down shortly after raising... those wedges will need to be cut flush, assuming that the joint is open enough to insert one.

As the frame dries it will still shrink away from the wedges... no?

We are having a similar discussion on the Forestry forum timber framing section.

Here is a link to that discussion: Link to discussion-click here http://www.forestryforum.com/board/index.php/topic,41639.0.html

Here is what I posted about the subject:

The problem with using dovetail joints for floor joists are that the dovetail is going to shrink.

If you're looking down at the dovetail from above the shrinkage is going to happen from left to right. This makes the dovetail smaller than the dovetail pocket cut into the side of a sill, for example. If the dovetail tenon is smaller then the dovetail pocket then the joist can pull out of the pocket.



In the above drawing the red dimension is what is going to shrink. To offset the expected shrinkage hardwood wedges are driven in on the sides shown here at the blue lines.

I hope that helps.

Jim

Thanks Jim.
I don't use them (never have) - but know folks have used them from the Benson Book effect.

I'm curious to help out Kevin (the OP), and curious as to how NH's wedges work... which I don't see working for long unless the flooring stays off and you come back 6 months or a year later and tap tap them in.

Kevin's frame design doesn't rely on the wedged dovetail to resist pull out (his bent / wall design does this) - so he'll be OK, assuming he designed the dovetail tenon stout enough to take the floor load. But - there should be no assumption that notching out that material and then replacing it with the tenon will give you any compressive advantage across the top of the timber.

I'm a fan of one of the early golden rules of joinery - 'Thou shalt remove only as much wood as needed...'

(followed by 'Thou shalt not covet thy neighbor's tools...'

Mike:
Personally I don't like this type of joint and I don't usually use it.
As mentioned in the other site thread, they rely on the eastern white pine to compress when they drive in the wedges and then when it drys out to expand some and not let the joist pull out of the timber it's connected to.
So further pounding in of the wedges is not done even if it is needed.

Holding of the frame together by other timbers is best.

My approach to joinery would be different, but if 7 x 7 is the required beam to carry the load, I would suggest that you oversize the tie so a 7 x 7 remains uncut in the cross section under the joist.

Please pardon my inexperienced opinion, but first, it sounds like, given that his floor to ceiling height goes from 2 to 5 feet, there isn't much room to maneuver or to get as much weight up on this storage floor as Kevin suggests, unless he is storing cast iron engine blocks and heads up there. Is Kevin over building this floor? Granted, you may want this sized joists for looks/proportion from underneath. Now, you cannot see the dovetails once finished, and given the frame structure, it doesn't seem that he would need dovetails to hold force along the joists length, the frame does that if I understand the short spans described. You only need the joist to hold the vertical weight. A simple drop in joist, no pinning, should work fine, and save a lot of unnecessary work of the dovetail joint(which should be housed if doing it). What do you think Mike and Jim and all? Is Kevin over complicating this?

My understanding is that the loft is 10'X30', but we are NOT talking 30'free span, but broken up into 10'X10' floor sections, supported by ties and posts every 10 feet across the span. Which would imply to me that drop in joists would work fine.

To really simplify it just skip the pocket and run long 20' timber on top of the 7x7.

As for the wedges going into the gap and depressing the wood to have it shrink and still gap after, I don't know, I have never taken one apart to see what it looks like after 15 years.

A curiosity I see when handling logs just before they go onto the saw mill is the grapple bucket putting dimples in the log, after the log is sawn and set for a period of time the dimple protrudes out beyond the sawn surface of the timber or board. This is similar to the hammer track left after missing a nail, the elephant tracks, if you wet the spot with spit it will smooth out in time.

Could the same thing happen with these wedges driven hard into the dovetail joints?

Tim

When I first did the math for timber sizes I was looking at code requirements for first floor loading given that the loft will be packed pretty much full of household belongings. I wasn't (am not) sure how much all that will weigh, so I went conservative and plugged in 50 psf (live plus dead loads). The reality, however, is that there is only a total of 1000 cubic feet up there, spread across 3 ten-foot spans. The number I came up with (from moving companies) for "typical" household stuff is 7 lbs/cubic foot. So, if I stuff the loft at that weight for every available inch, I'd have 7000 lbs spread across four 7x7 ties and fifteen 4x6 floor joists. The ties in the middle would have a maximum of 2333 lbs on them - i.e. less than half the weight of the 50 psf number I started with. Given that, the 7x7 ties seem to be okay, but I wasn't sure how much I'd be weakening them with the joist pockets.

There are no permits or inspections at all required here on my land so I'm not looking to satisfy the building inspectors. I just want to be sure that it's gonna last for the farm.

As for the dovetails, given that I'm still a rookie timber framer, I felt that this equipment shed would serve as a good practice frame in preparation for the new main barn that I'm planning on building a couple years down the road. I also thought that, given the lack of room in the 7x7 beam to created pegged lap joints, the dovetail would be an alternative for securing the tenon in a more limited space. I have the time to cut them, but it sounds as if there is some sentiment among framers they are not a particularly useful joint. Am I reading that right?

I'm about as fond of dovetails as I am of Hemlock, never use either.

Tim's on it there with the overriding joists. Never understood why folks get hung up on having everything come together at the same height, in houses even, but especially outbuildings. I use continuous joists whenever possible, though cogged into the ties an inch or three so they are both self-situating and so they do tie the bents together and help make the building a cohesive unit.

In this case there is a three-fold advantage, with the joists, up go the connecting girts, avoiding a three-way joint and the weakening of the posts removing that much section at one point represents. Up also goes the moment, (IMO a twofoot kneewall is pushing it even on a ten foot building) it would be a balance act with a relief cut to a barefaced tenon on the CG's, but if you make the girts planer with the raised joists and tie all the way across with ten foot decking, you will spread the moment over more square inches of joinery/timber and shift the bending moment up, and effectively shorten the kneewall. It would at the same time provide ready and full access to the dovetail tenons wedge

Will, I agree with the dovetails and hemlock, and I cringe when someone wants more that 2' on the knee wall.

I am not sure I follow the last two sentences of the last paragraph, almost but not fully.

I have been raising the tie beam on the first or last bent and tusk tenoning the long joists into it with either no very shallow daps to locate the joist. Trying to reduce joinery, time, wear on the body, money and so on.

The other part is I am leaving out the CG, less wood removed at the critical point where the post and tie intersect. I use longer braces as well, bracing to the top plate. To support the flooring I run the boards to the wall and fasten a support under after the boarding is done, there is not much load in there anyway.

It would be interesting the hear from folks that are using wedged dovetailed joist or purlins.

Tim