Humidity, Temperature, Wood Moisture Content, and Wood Movement

Basic technical information and an extended discussion of the way wood expands and contracts in response to changes in temperature and humidity. November 13, 2009

From Gene Wengert, forum technical advisor:
From time to time, there are postings that involve the movement of wood. Here is some information that may be helpful. Wood does not shrink or swell in use except if its MC changes. Its MC changes when the RH changes. (Of course, if the wood is at the wrong MC when first put into use, it will adjust to achieve equilibrium with its environment and therefore may shrink or swell initially quite a bit.) Temperature alone does not cause any significant size change in wood. Heating does causes moisture changes to occur faster.

The basic relationship between MC and RH is given below, with a third column for the equilibrium moisture content (EMC) which is a property of the air.
0% RH = 0% MC = 0% EMC
30% RH = 6% MC = 6% EMC
50% RH = 9% MC = 9% EMC
65% RH = 12% MC = 12% EMC
80% RH = 16% MC = 16% EMC
99% RH = 28% MC (approx.) = 28% EMC (approx.)

Note that most heated homes and offices will run 6% EMC in the wintertime and even a bit lower. In the summertime, 9% EMC is common. Outside in most of North America, the outside is 12% EMC, summer and winter. In coastal locations, 16% EMC outside is common.

When air is heated, its RH drops unless moisture is added (that is, the air is humidified). For example, if it is 30 degrees F outside and 100% RH = 28% EMC; snowing or foggy perhaps) and that outside air is brought into a home or office and heated, the following will be seen:
Heated to 40 F will result in 68% RH and 13% EMC.

Heated to 50 F will result in 47% RH and 9% EMC.
Heated to 60 F will result in 34% RH and 7% EMC.
Heated to 70 F will result in 24% RH and 5% EMC.

Note in the above that a home with plants, cooking, bathroom showers and so on will add moisture, so these values might be a bit low, but not much. Of course, using a humidifier will increase these values; offices are not humidified too often, although such higher humidity would help keep static off of paper used in high speed copy machines, etc.

Because purchased wood is often around 9% MC (actually too high, but that is reality for some folks; I suggest 7.5% maximum MC) and a shop can run around 5% EMC, some folks are tempted to increase the shop air to about 50% RH. Indeed this will eliminate problems in manufacturing, but it actually only postpones the shrinkage until the customer gets the wood in his/her dry home or office. Do not over-humidify a shop.

Checking and recording on paper the MC of wood you are using and the RH of your shop using fairly inexpensive instruments is a prudent thing to do. It provides documentation to prove that you did the right thing regarding MC so that subsequent MC problems are not your fault - assuming you would not use wood that had incorrect MC readings and that you would not keep a shop at the wrong RH or EMC. (Sometimes this approach is called CYA.)

Once a piece of wood or a wood product is put into a fairly tight container, including wrapped in plastic or into a closed truck or trailer van (that is, liquid water cannot get in and there is little RH exchange with the outside), the MC will not change in transit or storage, regardless of the temperature. For 100 pounds of wood to change by 2% MC will require the addition or loss of one quart of water...that is a lot of water to be brought into a container by merely exchanging a small amount of air with the outside. It would involve bringing in more than 1000 cubic feet of humid air and this is just for one 100 pound piece of cabinetry, furniture or flooring. I hope this helps. Does anyone have any questions?

Forum Responses
(Architectural Woodworking Forum)
From contributor J:
Wood does expand and contract with temperature. Doors I built and installed in south Florida move quite a bit with temperatures at over 100 degrees in the summer to sometimes in the 30's in the winter. This is with keeping a MC (depending on species) of 6% year-round.

I learned the hard way (as we all do) to increase my astragal/active reveal to 3/16" on outswing doors. This may raise some eyebrows and it has with a couple of builders. But even if I rely on a perfect install, doors seem to do funny things in the heat regardless of MC or RH.

In a chart I found called "Rates of Thermal Movement for Common Materials" (coefficient of expansion) published by AAMA states: "Range of movement based on a100 degree change using an 8 foot section of material". Range of movement for wood: 1/64" to 1/32". In most situations those numbers are negligible, but I believe it's not to be overlooked. For what it's worth, spend the extra money for a good meter.

From Gene Wengert, forum technical advisor:
Regarding moisture meters, a meter costing around $200 will be adequate. Higher cost meters (over $1000) have memories, easily changed species parameters, temperature compensation, and so on. For measuring RH, Radio Shack has one for about $25 that is adequate. It takes a 5% RH change for a 1% MC change, so RH measurement does not need to be very accurate. For thermal expansion and contraction, there is some confusion. The values you quote are for 0% MC wood. In the Wood Handbook (the bible for wood information), the expansion across the grain is .00002 per degree F for bone dry wood.

However, we do not deal with dry wood. The problem is that when wood is heated its response to RH changes; that is, the EMC vs. RH curves change; wood that is at 28% MC at 100% RH at room temperature will be only around 22% MC at 200 F. So with heating at constant RH, the MC will change and this MC change causes wood to change size almost immediately. Of course, as I stated, the heating of air will lower the RH and this effect will be larger than the EMC effect, but these moisture changes take time to occur.

So, it is virtually impossible to heat wood, such as a door, without changing its MC. For wood with moisture in it, here is what the Wood Handbook says: "Wood that contains moisture reacts differently to varying temperature than does dry wood. When moist wood is heated, it tends to expand because of normal thermal expansion and to shrink because of loss in moisture content. Unless the wood is very dry initially (perhaps 3% or 4% moisture content or less), shrinkage caused by moisture loss on heating will be greater than thermal expansion, so the net dimensional change on heating will be negative. Wood at intermediate moisture levels (about 8% to 20%) will expand when first heated, then gradually shrink to a volume smaller than the initial volume as the wood gradually loses water while in the heated condition."

From Gene Wengert, forum technical advisor:
Regarding a 36" wide piece of wood, such as in a door, the change in width (across the grain) will be about 1% for a 4% MC change (depending on species). This is 0.36" for a 4% MC change or 0.09" for a 1% MC change, or about 3/32" for a 1% MC change, or about 1/32 for a 0.3% MC change, or about 1/64" for a 0.16% MC change. I would question whether the 1/64" to 1/32" change you note is due to a small moisture change, our moisture meters cannot measure moisture that accurately, especially when there is a large temperature change. Also, the pin type meter cannot measure under 6.5% MC with good accuracy. In other words, it is impossible to keep a door at 6.00% MC from 32 F to 100 F. The MC could easily range to 6.3% MC and it would not be able to be measured. (Incidentally, 6% MC requires 30% RH which seems awfully low for southern Florida.) Bottom line is that heat (or cooling) alone is not an issue in wood behavior; rather it is the moisture.

From contributor Z:
Gene, thank you for this summary. You mention that pinned meters are not that good below 6.5%MC. I have an issue with a thick drum shell product I make in hard maple in Washington state, then ship to southern California for CNC shaping. I equilibrate the shells to 7% (35%RH), and bag them for shipping. I've suggested that they equilibrate them in southern CA before they machine them, as the machine shop is not humidity controlled. They experience more than a little bit of chip out from the CNC router. I had them ship some back to me so that I could check MC. It was 4.8%. But, now after your comment on low accuracy at low MCs, I'm thinking it might have been even lower. Is there a non-destructive method to check these low MCís? I'd use the microwave method if I had something to test, but then I'm thinking that method would require a more accurate scale than I have for those very low MCs. I believe you've mentioned in past forum posts that dry wood chips more, so I'm thinking of doing the machining operation in WA instead of CA to machine them at 7% instead of below 5%. This isn't currently practical due to my customer location, available machinery and machining expertise, but I'm looking at it.

I can get around the low accuracy of pinned meters in one way, just by controlling the humidity. Mine is at 35%. I visited the southern CA machine shop last month and was stunned to see a RH reading of below 5% on an instrument in an adjacent building. If that is true, could the MC in my product be 1% after a few weeks of equilibration?

This leads to a question on finishing at these low MCís. My customer does their finishing work in southern CA as well, without humidity control. Might I anticipate problems here as well? They are used to working with laminates, but not thick solid hardwoods so maybe some issues will come up. The most damaging issue I can think of here is that if they finish the products at 5%RH, then ship around the world where any RH could be encountered, later swelling might pop joints or hardware.

From Gene Wengert, forum technical advisor:
The reason that pin meters do not work at lower than about 6.5% MC is that the resistance is extremely high in dry wood, so high that it is next to impossible to measure - billions of ohmís. Static electricity will dominate. The resistance of the electronics will be larger than the wood, etc. Also, a bit of surface moisture or contamination of the probes can lead to a reading that is not correct. The inability to read under 6.5% MC with inaccuracy does not reflect poorly on the meter manufacturer; it is just the high resistance of wood and its variability due to surface moisture, etc. as mentioned.

The pinless meter does read to lower MCís well. Specifically to your case, as long as the wood is in a bag, the MC is constant. Your 4.8% MC reading is not at all accurate. The MC could have been 6.1% MC, 4.8% MC or somewhere in between or even drier. We just do not know or have any idea. One idea for you is to weigh the wood piece when you put it into a bag. This will establish the weight at 7.0% MC. Divide the weight by 1.070 and you will have the oven-dry weight estimate. Your customer can then check the MC by having the piece weighed again before they begin processing (grocery store, meat market etc.).

It is nearly impossible to estimate the time involved for the piece to change MC when exposed to a drier EMC. It would be very unusual to see an environment of 5% RH, although it does happen (I have seen it in Denver, CO, but it is transient.) If they are really that dry and are shipping to the rest of the world, they need to take a small area and make a plastic enclosed room and then put a humidifier in that room set for 35% RH. They would then store all the work that they have in progress (that is, after they take it out of the shipping bag from you) in this room, except for the hours they are working on it. Once finished, it can stay out for a longer time, but even then, it would be best to put it back into this room if it is not wrapped in a plastic bag. They could also, whenever long storage is anticipated and they do not have room in this RH controlled room, wrap the piece in a plastic bag, which will prevent any moisture change from that point onward until the piece is unwrapped. This procedure is easy, inexpensive, effective, but does require a bit of dedication to these guidelines. Does all this make sense to you?

Hard maple is known to have a lot of chip out as the grain is never perfectly straight. So, in small spots we can be planing or machining against the grain and that gives us tear-out. This defect is more common with dull steel or with carbide tools as well. I suspect that you will have chip-out as well, unless you have very sharp tools, good feed rates, etc.

From contributor Z:

Thanks Gene. The low humidity reading is suspect. The meter was wall mounted in a section of their plant located with heated presses. They do a lot of veneer work in there, but they have for decades and seem to live with it. I've suggested a humidity controlled room for storage, but unless defects show up, I don't think they'll do it. A lot of the reason is that the machine room is in a drier part of the state, the finishing room is not humidity controlled and the product will change MC anyway after finishing, in retail showrooms, and in use. If it can't survive these normal environmental changes, it won't be a good product. I'll make that storage recommendation again though, for the CYA benefit. I'll ask them if they have a good lab scale to give me weights back. I weigh them before bagging and packing them anyway, more CYA in case defects show up after long periods of sub-optimal storage and I can perhaps point to storage conditions. Though nothing has developed in a year, knock on solid wood.

These parts are curved. Is there a pinless meter that works on curved parts? For hard maple, are you recommending steel over carbide? There certainly is non-uniform grain in a lot of the chip out areas. I find that there may also be denser or harder areas that are not obvious defects, but may cause chip out. They may also be areas of grain slope change that just isn't obvious from observing the surface.

From Gene Wengert, forum technical advisor:
The grain dips in HM are very subtle. Yes, HSS is preferred, but the frequent sharpening has made some tolerate defects. Is the curve always the same? If so, I believe that Wagner can make a pinless meter for you. It is mainly fast moisture changes that result in defects. Once a good finish is on the product, the moisture changes will tend to be slow enough that they will not be as frequent (we hope). We would see most showrooms, etc. with 305 RH to 50% RH. So, if they have exceptionally dry areas, that is abnormal and would justify a humidity controlled room.

From contributor J:
I have some related issues I'm curious about. First, what general advice might you give about how control shop/lumber storage RH in the summertime? I'm in the Great Lakes area, where summers tend to be very hot and humid but with an occasional hot and dry spell. Dehumidifiers aren't a great option because they add heat to an already uncomfortable room. Air conditioning can improve comfort and lower humidity, but they are typically controlled by a thermostat rather than a humidistat, so they're a rather blunt instrument for this purpose. And of course, both are expensive to run so I'd like to use such tools as efficiently as possible.

Secondly, can you suggest a method or rule of thumb for estimating how quickly a piece of lumber might respond to changes in RH? I'm especially curious about how the thickness of a piece of lumber might relate to the speed of dimensional changes. I understand that there are a lot of variables here, but it seems obvious that a piece of veneer will respond very quickly while a butcherblock countertop might shrink or expand much more slowly. I guess I'm wondering whether large pieces of lumber might change dimensions slowly enough that they never come to equilibrium during a given season, i.e. that the fluctuations in RH between dry winters and damp summers might happen so quickly that the wood never moves as much as simple calculations might suggest.

From Gene Wengert, forum technical advisor:
First, the outside RH, even in the Lake States and also in the South, averages around 65% RH summertime and wintertime (12% EMC). Our bodies say differently, but a check of the weather will confirm my statements for most of the USA. Of course, just before sunrise, there might be 100% RH, but as the air heats, the RH drops, giving us the 65% RH average. Certainly, there can be a few days when it is more humid, but then there will be a few days when it is less.

Since in a shop we do not cool off during the early morning and reach 100% RH, the average in a shop with the windows open will be a bit drier, maybe 9 to 11% EMC. If we need it to be drier (in the old days without central heating, being drier was not an issue) then we can add a dehumidifier or add heat. Adding heat in the summertime is not a good idea, so then we can dehumidify. Our first attempt may be to run the DH at 45 to 50% RH when the dust system and doors are closed, including all day on the weekends. (Of course, the DH unit must be large enough to actually achieve the desired conditions within a few hours.) We also could air condition (money). Finally, we could build a small plastic lined room or area where we keep all the wood when it is not being worked on. Once a piece is finished, then we also need to put it in the room or wrap it well in plastic to prevent MC changes.

Let's come back to the heating idea. I do have one customer in Houston that bought an old 18-wheeler trailer van. He painted it a dark color and then stores wood in the van. A small fan assures that the hot (and therefore dry) air is circulated throughout the trailer. The fan is run only when the sun is shining.

From contributor Z:
I set up an insulated chamber, 4'x4'x8' with a humidistat from a greenhouse supplier to equilibrate parts to 7%MC. You can set the desired RH (to say 35%), then plug in a heater or dehumidifer to drive the RH down when the humidity rises. If you need to raise humidity, there is a second plug in so that could also plug in a humidifier. You can plug a fan in too if that achieves the desired results. The humidistat powers one outlet or the other, depending on whether you are trying raise or lower RH. I recall that the humidistat was not cheap - somewhere between $100 and $200 I think. Gene, what do you think of this approach?

From Gene Wengert, forum technical advisor:
Itís a perfect approach. Incidentally, you should never need a humidifier, as if you do nothing (unless there is a lot of solar heating), the unit will equalize (as it cools off) with the outside at about 11% or 12% EMC.

From contributor S:
To contributor J: we put an A/C unit in our shop that is controlled by a humidistat an a thermostat. In the summer we set the humidistat to 50% and leave the thermostat control off. On odd days in the spring or fall when it was humid and a little on the cool side, we might have to turn the heat on for ten minutes. On a humid 95 degree day our shop might be 80 degrees but 50% humidity.

A/C and dehumidifiers are the same thing only an A/C unit puts the heat outside so you might as well get the added production of cool employees. The shop has 20 foot ceilings so we hung the unit at one end and used a 3" diameter sock running the length of the shop to distribute the cool dry air. Two big pleated filters on the intake let us use the unit as an air cleaner too. We hardly notice the difference in the electric bill from summer to winter.

From contributor J:
Gene, is there a simple formula one might apply to determine how much the room temperature might need to be raised to achieve a particular drop in RH, given current temperature and RH? For example, given that my shop is currently at 58F and 48% RH, how can I calculate where my thermostat needs to be set to bring the RH down to 35%? Contributor S - thanks for those thoughts. I somehow missed your post earlier.

From Gene Wengert, forum technical advisor:

To contributor J: the technique is mentioned in Drying Lumber Hardwood (Fig 2.5, p. 19) which is here at WOODWEB if you do not have a copy. As a rough rule of thumb, 10 degrees F of heat will lower the EMC by 2% EMC or lower the RH by about 10% RH.