Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

Support WoodCentral when buying things you need

How Much Is Not Enough?

Posts

How Much Is Not Enough?

#1

How Much Is Not Enough?

Mark Mandell - Gone Round In New Jersey

Assume frame and floating panel construction in EMC Walnut with the panel 3/4" thick and 26" wide. The wood will, naturally, move in and out of the stiles with seasonal MC changes.

Question: How much panel material would you insert into the stiles? 1/4", 3/8", 1/2"? More? Rails and stiles will be glued M&T's and the center of the panels, top and bottom, will be pegged.

TIA

Re: How Much Is Not Enough?

#2

Re: How Much Is Not Enough?

Ken Oakley-Sunny St. Cloud, FL

When she asks "Is it in?"

Re: How Much Is Not Enough?

#3

JL

Re: How Much Is Not Enough?

JL

What does the Wood Movement calculator tell you?

I doubt as if a 12" piece (figuring on two halves) is going to move over 1/4" even with wide varying humidity and a "reasonable" temperature fluctuation.

I did have a pinned 15" panel vary something like 5/16" overall. I also had a 36" oak ramp increase something like 5/8" overall. With the oak, I used 3/8" dadoes - the only reason it expanded that much was because they put it in a storage facility - and it was a humid summer.

Re: How Much Is Not Enough?

#4

Re: How Much Is Not Enough?

John Veerkamp

If the wood is dry and humidity low, then 1/4 is plenty, if the reverse is true I'd go 3/8. Of course your mileage may vary depending on the wood species.

Re: How Much Is Not Enough?

#5

Re: How Much Is Not Enough?

Gary Smyth

I have a desk top made of cherry (about 2002) 5/4 x 29"w x 41"l, with a breadboard edge fastened solidly only in the center. In winter (when it was assembled) the edge was made flush with the field. Currently, I'm estimating that the field is 1/32 SHY of the edge of the breadboard width edge on either side. In the humid summer (PA), the field may expand 1/8" on either side of the crossgrained breadboard edge. Walnut may have a different expansion rate, but perhaps this will give you an idea.

Re: How Much Is Not Enough?

#6

Moses Yoder

Re: How Much Is Not Enough?

Moses Yoder

The panel will probably move about 1/8" on each side. Since you are building it in the spring it is most likely to shrink in the winter. I would put a 1/2" groove in the frame, use 1/4" rubber space balls, and insert the panel 5/16" into the frame. If you do without space balls, you could put a 3/8" groove in the frame and insert the panel 1/4".

Re: How Much Is Not Enough?

#7

Ellis Walentine

Perennial question, and a bit complicated

Ellis Walentine

This is the classic woodworker question, how to plan for wood movement. It actually depends on a lot of variables: the species of wood, the cut of the wood (quartered, flatsawn, rift), the width of the piece, the current moisture content, and the expected maximum and minimum MC down the line, plus whatever safety factor you want to add.

To calculate wood movement within the typical seasonal MC range, you need to know the shrinkage coefficient for that species and cut. Radial and tangential coefficients are different, with radial usually being significantly smaller than tangential. The difference between them is generally proportional to the amount of warp you might expect in that board.

I get the shrinkage coefficients in the Wood Handbook, originally published by the USDA Forest Products Laboratory and now in the public domain. Lee Valley sells it under the name Wood Handbook, Wood as an Engineering Material. According to the Wood Handbook, the shrinkage coefficient for quartersawn walnut (CR) within the 6% to 14% MC range is .00190. Multiply that by the width of your piece and the expected difference in MC to get the amount of movement. Here's the equation:

ΔD = D1 (CR (MF - M1))

If your piece is 26" wide and currently at 10% MC, and you think it could, in the driest heating season, get down to 6%,

D1 = 26

MF = 6

M1 = 10

the equation is

ΔD = 26 (.00190 ( 6 - 10 ))

ΔD = 26 (.00190 ( -4 ))

ΔD = -.1976"

If your current MC is 14%, double this figure.

The upshot is that if this is actually radially sawn and if it is a relatively typical piece of walnut, you're not going to see more than about 3/8" of total expansion/contraction across this MC range; and since the range is likely to be even smaller, the movement is likely to be even less. Here's where the safety factor comes in. Why cut it so close that a slight difference in one variable will ruin your whole calculation? Plan for the worst, which would be a total of .3952" (or more, if you think the piece will ever live in drier or wetter climates, or if the grain isn't strictly radial).

The actual dimensioning of parts -- panel widths, tongue & groove dimensions -- relies on the current MC of the piece. If it is at the high end (14%, and I hope not), you can expect most or all of the movement to be shrinkage, and much less to be expansion. In this case, you don't need to leave much (if any) headroom at the bottom of your side grooves for expansion, but you do want to allow for contraction of up to .3952" overall, or more depending on your safety factor, call it 1/2". That means that you need at least 1/4" tongues on the sides of your panel to keep them from pulling out of the grooves if your 14% wood dries to 6% at some point. So, to design for this worst-case scenario, you'd want tongues even wider, say 3/8", and grooves to match. And, you'd want the panel fairly close to bottomed out in the grooves, since it is not going to expand.

The opposite is true for wood at the low end of the MC range. If your wood is at 6% now (freshly KD stock), it can only expand from here unless you move to Phoenix or the Mojave Desert. So, you need to allow plenty of headroom in your grooves for expansion.

Obviously, you don't want to see huge gaps between frame and panel, so it's best to work with wood that is around the middle of the expected MC range, so the gaps can tighten up in the summer but not look overly large in the winter.

Another phenomenon of wood movement is that annual swings are highest in year 1 and decline significantly after that. So, if you make it to year two and the panel hasn't fallen out, you probably don't have any further worries. About the panel at least. :)

Sorry for the nosebleed. It's actually pretty easy to predict this stuff from past experience with different woods and widths, but it doesn't hurt to have a grasp of all the dimensions of the problem and the calculations involved.

You can also work from the "Green to 6%" coefficients. Hoadley explains all that in Understanding Wood.

Ellis

Re: How Much Is Not Enough?

#8

An Article For An Answer!

Mark Mandell - Gone Round In New Jersey

Thank you, Boss! This should go in the Articles Section for easy reference.

To complete the facts: 20 of the 26" width is quarter sawn. The stock (walnut) is air-dried to 11% and has been stacked in my basement (dehumidified) shop for more than 12 months. Piece is going to Brooklyn.

I'll use 3/8" tongues.

Thanks again.

MM

Re: How Much Is Not Enough?

#9

Re: Wood Handbook *LINK*

Bob Entwisle in NJ

Here is a link to all 509 pages of the 2010 edition.

Bob in NJ


http://www.fpl.fs.fed.us/documnts/fplgtr/fpl_gtr190.pdf

Re: How Much Is Not Enough?

#10

Only a guide...the "fine print"

Bill Tindall

The tables list average values calculated from a selection of samples. These data are useful in predicting that beech is a less stable wood than walnut, but they are much less useful in predicting what will happen to any particular board. The range of shrinkage within a species can be quite large. As an example Period furniture makers are finding that the mahogany now being harvested is much less stable than mahogany in the past.

Re: How Much Is Not Enough?

#11

Re: Wood Handbook

Steve Schoene

By the way, US Govt documents such as the Handbook are public domain as soon as they are published. It doesn't assert copyright.

If the gov't doesn't want something published it just stamps it SECRET or the like.