do you need to allow for wood movement on bench seat?

[Bob Riefer]

I built a bench for a friend... very simple... 1" thick x 12" wide x 6' long poplar plank is the sitting surface, mortise and tenon poplar base... All primed and painted (although in the pic here, it is unpainted still).

I currently have the top affixed to the base like you would with a table top (I use "z-shaped" hold downs that screw into underside of top and rest in a kerf groove in the apron) to allow the plank to move as needed.

Am I being overly nervous about wood movement across that small of a surface? That is... could I affix with pocket screws instead, or is the hold downs method more suitable?


Edit: perhaps the more useful question is this... is there a size "rule of thumb" (or similar general guidance) on when allowing movement is more critical?

Edit 2: Shoot, I meant to post this in "general woodworking".. too much multi-tasking for me today! sorry!


447713

[Paul F Franklin]

A quick calculation shows somewhere around 1/16 inch difference in movement between the top and the side rail over a typical humidity swing for yellow poplar. Since poplar is fairly soft you could probably get away with pocket screws, but since you already did the z fasteners, no reason not to leave them. If you want, you could screw the front two legs and let the top slide on the back two with humidity changes. Probably best to paint the top as a separate piece, otherwise the paint line between the top and base will likely crack over time.

[Andrew Seemann]

Somewhere around 6 inches with mechanical fasteners (nails screws), as there is usually a little bit of play with those, more if you counterbore one of the holes. Glue would be less, maybe around 4 inches. It depends on species and grain orientation, but those are probably as good of rules of thumb as any. Note that that is the distance between fasteners, not the overall width of the board.

Poplar is pretty stable, so screwing straight into the wood would probably be borderline on that that piece. I have several computer monitor shelves for the kids that are a pine 1 x 12 screwed into a 1 x 3 on edge and those haven't cracked or failed.

Elongating the holes that the screws go through works also, but Z clips or cleats are probably as good as anything. I wouldn't change what you already have to something else, unless you are worried about it being not strong enough to handle usage.

[Bob Riefer]

A quick calculation shows somewhere around 1/16 inch difference in movement between the top and the side rail over a typical humidity swing for yellow poplar. Since poplar is fairly soft you could probably get away with pocket screws, but since you already did the z fasteners, no reason not to leave them. If you want, you could screw the front two legs and let the top slide on the back two with humidity changes. Probably best to paint the top as a separate piece, otherwise the paint line between the top and base will likely crack over time.



What calculation are you running? I'd like to save that info for future reference if possible please.

[Bob Riefer]

Somewhere around 6 inches with mechanical fasteners (nails screws), as there is usually a little bit of play with those, more if you counterbore one of the holes. Glue would be less, maybe around 4 inches. It depends on species and grain orientation, but those are probably as good of rules of thumb as any. Note that that is the distance between fasteners, not the overall width of the board.

Poplar is pretty stable, so screwing straight into the wood would probably be borderline on that that piece. I have several computer monitor shelves for the kids that are a pine 1 x 12 screwed into a 1 x 3 on edge and those haven't cracked or failed.

Elongating the holes that the screws go through works also, but Z clips or cleats are probably as good as anything. I wouldn't change what you already have to something else, unless you are worried about it being not strong enough to handle usage.


Thanks both of you for chiming in so quickly. Sounds like the Z clips are good to go. Would it hurt to add one centered fastener at each end for good measure - this would still allow movement from that center point outwards in both directions right?

[Andrew Seemann]

Thanks both of you for chiming in so quickly. Sounds like the Z clips are good to go. Would it hurt to add one centered fastener at each end for good measure - this would still allow movement from that center point outwards in both directions right?

A center screw on the short aprons would probably be a good idea to keep the board centered, so you don't slide out of your Z clips. There is probably enough flex in the long aprons that you could put a screw in the center of them as well.

[Stan Calow]

I use the z-clips, and maybe a couple of figure eight fasteners (can move either way)in the centers. I can see that a bench like that would be more like to be picked up by the top and moved.

[Paul F Franklin]

What calculation are you running? I'd like to save that info for future reference if possible please.

I used one of the online wood movement calculators (there are many of them) to calculate the cross grain movement of 6" wide flat sawn yellow poplar when the moisture content varies from 7% to 15% (I just picked those values as representative of what you might seen around here in NE Ohio, PA shouldn't be too different). Some calculators let you enter relative humidity rather than moisture content. Anyway, that tells you how much the top might move across the width. Then looking at the base, you have any long grain movement in the top rail, which will be negligible (but you could use the calculator to figure it out), added to the cross grain movement of the two legs (which I treated as rift sawn). Subtract the Movement of the top from the movement of the top rail+legs assembly and you get a rough idea of the differential movement. It's all a bit of a game of averages, so any given piece/situation can vary, but it gets you in the ballpark.

[Bob Riefer]

I used one of the online wood movement calculators (there are many of them) to calculate the cross grain movement of 6" wide flat sawn yellow poplar when the moisture content varies from 7% to 15% (I just picked those values as representative of what you might seen around here in NE Ohio, PA shouldn't be too different). Some calculators let you enter relative humidity rather than moisture content. Anyway, that tells you how much the top might move across the width. Then looking at the base, you have any long grain movement in the top rail, which will be negligible (but you could use the calculator to figure it out), added to the cross grain movement of the two legs (which I treated as rift sawn). Subtract the Movement of the top from the movement of the top rail+legs assembly and you get a rough idea of the differential movement. It's all a bit of a game of averages, so any given piece/situation can vary, but it gets you in the ballpark.


Wow, thanks for this! Part of my notebook now :-)

[Tom Bender]

As Andrew says, crews in all four centers will work fine. It is a method I use without any other fasteners. Typically 2 screws in each center.

I'd be more concerned that the bench will tip over and maybe cause an injury. Maybe add a trestle foot on each end.