Expansion Lesson - Woodworking Talk

Gravel7 · 08-07-2009, 05:09 PM

I am having a really hard time fully understanding the best way to join pieces of wood and allow for expansion. Can anyone point me in the direction of a good web article or maybe a book? Or if someone wants to give some example here, then GREAT!

Here are some examples of where the simplest joint makes me wonder:

Attaching a single 12" wide board (say pine) to another board...just a butt joint screwed in or maybe in a dado with screws.

Or

What if i was using two 12" boards, glued up, for a dresser top; attaching this to the dresser? Ive read about elongating the holes. Isn't that kind of weak? And wouldn't that mean DO NOT GLUE? And how would elongating the holes help if I'm going to screw it down nice and tight anyway?

And for tenons; Ive read, "somewhat snug fit all around with just a little extra clearance in the back for excess glue". Also read to leave a 1/16 on the width...so which is it?

Seems like a simple thing, but I can't seem to wrap my head around it

.

Thanks for any explanations or references!!

phinds · 08-08-2009, 07:05 AM

Think of wood as a compacted bundle of strings. If you get it wet, it isn't going to stretch out very much in length (maybe a couple of percent) but it's going to get a lot fatter (could be as much as 10 percent). If you butt the end of one bundle against the edge of another bundle, you're matching 2% change against 10% change --- that's the fundamental problem.

Solutions I'll leave to others since I'm not good at joinery.

Paul

Gravel7 · 08-10-2009, 08:58 AM

reply Paul. I can't seem to educate myself on this. I have purchased to books, one specifically on joinery, and neither really touched this.

Tony B · 09-26-2009, 10:30 AM

Quote:

Originally Posted by Gravel7

I am having a really hard time fully understanding the best way to join pieces of wood and allow for expansion. Can anyone point me in the direction of a good web article or maybe a book? Or if someone wants to give some example here, then GREAT!

Here are some examples of where the simplest joint makes me wonder:

Attaching a single 12" wide board (say pine) to another board...just a butt joint screwed in or maybe in a dado with screws.
If you are going to join the 12" boards at 90* to eachother, that is, the 12" wide bpard to the other12" part, then both 12" boards will expand and contract equally and no allowance for movement is required.
TonyB

Or

What if i was using two 12" boards, glued up, for a dresser top; attaching this to the dresser? Ive read about elongating the holes. Isn't that kind of weak? And wouldn't that mean DO NOT GLUE? And how would elongating the holes help if I'm going to screw it down nice and tight anyway?
Assuming you glued up the 2 12" boards side by side (long Grain to long grain) in effect you now have on 24" wide board attaching to a 24" dresser side made of same kind of wood. Again, no allowance would be required. If on the other hand attaching to the dresser top with the grain running at 90* you would elongate the hole in the glue block attached to the side and not elongate the hole in the top. In the latter case, you would not glue the top down. The top would be screwed snuggly but not forced in place. The top then would be allowed to move slightly from side to side. Generally, this type of design is exceptionally rare and I doubt if you would see it anywhere but home made.
Tony B

And for tenons; Ive read, "somewhat snug fit all around with just a little extra clearance in the back for excess glue". Also read to leave a 1/16 on the width...so which is it?
Leave a small gap in the back or you wont be able to push the tenon in all of the way due to trapped air and excess glue being trapped in the rear of the tenon. Leaving a 1/16" gap on the width is not only unnecessary but is also poor joinery. The expansion/contraction of a small tenon is negligable in the open air, when sealed in a relative climate controlled situation of a mortise, it is immeasurabe for all intents and purposes.
Tony B

Seems like a simple thing, but I can't seem to wrap my head around it

.

Thanks for any explanations or references!!

When attaching tops such as table tops to aprons, the grain is running 90* and one suggestion is shown in my attached sketch.
Tony B

rrich · 09-26-2009, 04:12 PM

IIRC, starting from ideal moisture content, 6%, you have to account for both shrinkage and expansion due to changes in climate.

The change, up to about 9% or down to about 3%, is about 1/8" for most wood that would be used in furniture over 12 inches. So when thinking of a table top, 36" wide, the total anticipated wood movement is about 3/8". The 3/8" seems like a lot but it can be halved. Simply securely attach the center of the table top to the frame. Then each side of the table top will only move about 3/16".

There are many ways to allow for this movement. Probably the easiest and cheapest is to attach the table top to cleats using #8 screws and a washer. Use a 1/4" drill bit to make the hole through the cleat and rock the drill while drilling and elongate the hole in the direction of anticipated wood movement.

Next cheapest would be to use either "Z" clips or figure "8" metal attachment devices.

Finally there are router bits designed to cut slots that a screw will ride in. You just route a small groove in the cleat and install the holding screw in the center of the groove.

With any of these methods, put a bit of paraffin wax on the two surfaces that are expected to move relative to each other. This will reduce any strain.

Above all

APPLY FINISH EQUALLY TO BOTH SIDES OF THE TABLE TOP!

This will make the table top less susceptible to warping or cupping.

John in Tennessee · 09-26-2009, 04:17 PM

Got a idea in my feeble mind..

NKYDarrell · 09-26-2009, 08:58 PM

hope this helps.

Tony B · 09-26-2009, 09:26 PM

Made error, see next post

Tony B · 09-26-2009, 09:44 PM

Quote:

Originally Posted by NKYDarrell

hope this helps.

Unless I am missing something, is the above picture suggesting that the table on the left is secured in just 2 points? And, the other 2 tables are supported in one straight line? There is also no explanation of how to attach the other ends. Also, the snug attachment should be made on the ends of the tables and not along the direction of expansion. Reason being is that as the table expands/contracts the attachment points slide in grooves sideways and not dig in head on. If you also add attachment points along the direction of movement as I oftentimes do, these points should be relatively loose and their only function would be to prevent warping and cupping.

NKYDarrell · 09-26-2009, 09:46 PM

well, first off wood expands across the grain which is in my mind, the primary illustration shown. as far as securing the top, I would think that that means a solid connection where the red dots are and for the rest of the table top, using clips or some sort of connector that allows for movement.

Tony, I hope this image helps you relate to your sketch posted above.

Tony B · 09-26-2009, 10:23 PM

Yes, Thank You.
My freehand drawing skills are not exactly up to par. LOL

That leaves us with one more question......does Gravel7 understand any of this?

Gravel7 · 09-27-2009, 08:47 AM

Wow, thanks for all of the replies. I thought this post was gone forever. When I first posted it, almost 2 months ago, noone really replied and now all of the sudden all of my questions were answered.
I do understand and really appreciate it!

Thanks Tony B for your involved response that brought this post back to life!!

Mike

rrich · 09-27-2009, 04:28 PM

Quote:

Originally Posted by NKYDarrell

hope this helps.

Beating the dead horse a bit more...

The purpose of the solid attachment of the table top is to both control the direction of expansion and to reduce that apparent effect of expansion. The red dots are the points of solid attachment. (Could even be glued.)

The table at the left, the apparent expansion is cut in half (not really) because all the expansion is out from the middle. So half of the total expansion is on each side of the middle.

In the middle table, all of the expansion is from the back and toward the curved front. This expansion will be less noticeable due to the curve.

The right table has the expansion moving away from the front of the table.

In all three examples, the woodworker is controlling the direction and effects of expansion.

Regarding the power of wood expansion has been known for centuries. In early granite quarries, holes were drilled in the granite. Then dowels the exact size of the holes were forced into the holes and saturated with water. As the wood expanded the granite cracked and a block was removed from the quarry. During the winter months, dowels were not used and water alone was used. As the water froze the ice performed a similar function.