how to affix bench seat and back rest (for wood movement)

[Bob Riefer]

I'm nearly finished with a bench for the dining table in our sunroom, and have some questions around affixing the seat and back rest. All parts are poplar.

The back rest is 1.5" thick x 6" wide x 7 feet long, and the seat is 1.5" thick x 12" wide x 7 feet long.

In both cases, a 1/4" deep dado in the back rest / seat will fit snugly over the back support / seat support.

My concern is that if I glue the dado connection, wood movement will cause problems over time. So I'm instead considering drilling oversized holes through the back support / seat support that would allow fasteners to move (and then I would plug over those holes).

Am I being too worried, or is this the right approach? (or is another approach preferred?)

Thanks!

[Jim Becker]

Some photos would be helpful for us to actually visualize what you are asking about... ;)

[Bob Riefer]

The two spots where you see the green tape are where the back rest and seat will be affixed.

If this were a table, this would be like the top sitting on the apron... which I would do with clips to allow movement. But clips in this case would be visible / not what I'm after.

If I glue the seat and back rest to these spots, I'm worried it will not allow for wood movement.

Does this make sense?

479586

[Andrew Hughes]

Don’t glue it.
Use a mechanical connection allow some space for the seat part to move and the back rest.
Very stylish bench Bob.
Good Luck

[Bob Riefer]

Thanks Andrew! I'm glad I confirmed my concerns.

I think I will drill stepped holes through the seat support and back support (aka the areas with green tape in the pic above) - the outer hole would accept a plug, and the inner would be oversized to allow the fastener some movement. The fastener would go through the hole and into the no-show side of the seat or back rest respectively.

[Dwayne Watt]

Is there a mortise for a stretcher not shown in this photo? Simply insetting the back and set into a dado is not going to provide much lateral strength.
I would mortise the seat and back rest into the supports and then add a stretcher on the lower part of the rear leg to get some lateral strength into this bench. You may have to get creative to raise the leg assembly to account for the offset of the bench seat otherwise the seat might get too low for comfort.

[Bob Riefer]

Hi Dwayne -
I'm either going 1 of 2 ways on this (would love opinions)

Diagonal brace from front leg to underside of the bench (on both left and right side) to prevent racking.

or

Stretcher from front leg to other front leg.


I thought the diagonal would look more inline with the design as I would cut a gentle arc to match the rest of the piece.


(any which way, I haven't cut those mortises yet)

[Lee Schierer]

The seasonal moisture changes for the back will be minimal Assuming a 5% moisture change in your house and that the screws are located 1" in from each edge of the back.
Seasonal Dimensional change (https://cdn.popularwoodworking.com/wp-content/uploads/2010/10/WoodMovement.pdf)= 5" x 5% x .00263 = .0657 (1/16") inches of change between the holes. For the 12" seat the over all dimension change would = 12 x 5 x .00263 = .158 inches or about 5/32" outside to outside.

[Bob Riefer]

Thank you Lee! I bookmarked that link for future. The wood movement being as minimal as this helps me understand how much "wiggle" is needed for each of the fasteners.

[Lee Schierer]

Thank you Lee! I bookmarked that link for future. The wood movement being as minimal as this helps me understand how much "wiggle" is needed for each of the fasteners.

You just need to figure out the max an min moisture levels in you house from season to season. I used 5% humidity variation, but your location may be different. The Philly area can be pretty humid in the summer.

[Dwayne Watt]

I would not put a stretcher at the front leg due to potential interference with the seated person's legs. The back leg is pulled in so far that a diagonal brace from it or the lower foot at the rear leg to the bench might look and function best. The brace could have a curve and angle to complement the rear leg. Thinking about it for a moment, the brace could be an arch from end to end of the bench. The arch could be fabricated by laminating (brick lay pattern) small-ish sections of 3/8" thick stock to approximate the general shape from which the arch could be cut. Mortise into the feet and into the bench then connect with floating tenons. Huge strength and complementary design.

[Anuj Prateek]

To attach back, I made a stopped dado. So the back rest kind of hangs on the support. Dado itself was made less snug.

I used epoxy (a blob towards top) and screws in oversized holes. Bench survived for a year at least outside and showed no signs of problems. Sold the house and left the bench back there so don't know the current state.

You can see some photos at https://sawmillcreek.org/showthread.php?277609-Outdoor-Bench&highlight=

Seat part is attached the same way.

[Tom Bender]

Dwayne I like the way you are thinking. How about running the arch from the back leg just below the seat support joint up to the back rest and back down on the other side. It could be 1" x 3" and made as a glued lamination for better grain display. Probably need a regular tenon if only 1 x 3. The challenge of chopping angled mortices would be a new learning experience, at least for me.

[Bob Riefer]

An update... My first version (pictured above) has turned out to be the prototype, and I'm onto version 2 today.

Long story short, I made the bench too tall by accident, and any efforts to shorten are only going to cheapen the look I'm after. I'm reminded that progressing at just about anything in life entails a willingness to fail and try again. So, I plan to re-use these for something else (tbd) and rebuild. Thankfully only using poplar so price impact is not too bad.

(btw... the seat and backrest can be re-used for version 2)

What I did learn is that with minor adjustment, the design can accommodate a simple stretcher under the front of the seat. Basically, if I make that front leg joint slightly bigger, I will be able to fit a stretcher there.

To make the stretcher blend with the design, while adding the desired racking resistance, the plan is to start wide at the point where joined to the legs with an arc across. When I blend the joints (power carving) the front leg will essentially look like it has two tree branches growing from it... one connecting to the back leg, and the other connecting across.

Rudimentary drawing:
479644

[Tom Bender]

Yeah, I've built a few chairs and the prototypes usually fail the sit test. One now serves as a plant stand.

[Bob Riefer]

:-) Glad I'm not the only one! The good news is that while the first time around (including prepping the rough lumber) took me about 5 hours, the second time only took me about 1.5 hours since I had it all figured out already.

This time I was able to plan the front leg joint to allow the stretcher connection noted above. I'll post pics soon

[Bob Riefer]

Here, I have it all mocked up and dry fit.

The goal with that front stretcher was to add a lot of strength laterally, so there are two 14 mm dominos making the connection on both sides - one from the stretcher into the seat rest, and another from the stretcher into the front leg.
479815

The other goal was for the stretcher to fit the design / look and essentially hide from the most-seen angle (the side view... because the front of the bench is hidden under the table it serves).

The final power carving to blend the joints is occurring today, but you can get the idea of the final look.

479816


Quite a beast to glue up!

479817