There are several steps in “checking” the MC of incoming lumber. First and foremost is determining what is acceptable for your operation. Readers of this column know that I like 6.8 percent MC average for most hardwoods most of the time. Further, most readers know that I would like to see all the pieces of lumber within + or - 1.0 percent MC. (Note that I use decimals here; we need to be accurate about MC levels and not just something like “6 to 8 percent MC.”)
 
So once you have determined the correct MC for your products and for your ultimate customers, the next step is to measure the MC of either every piece in the load or of a random sample of pieces from the load. Measuring every piece in a load with an in-line meter is awesome, and is effective for eliminating wet pieces that will cause manufacturing and product defects (plus if you have your own kilns, it can develop moisture profiles and let you know how the kilns are working and what and where maintenance is needed). But such a meter is very expensive. In a small operation, we could have a person measure the MC of every piece, but that is not practical for most operations.
 
 

KNOWLEDGE CENTER

Find more Wood Doctor at FDMCdigital.com

Gene Wengert, aka The Wood Doctor, troubleshoots wood related problems, and explores lumber and veneer qualities and performance, species by species, in Wood Explorer, inside FDMC's Knowledge Center.


The moisture is measured with a portable, hand-held electronic moisture meter. The pin type leaves holes and is slow, so for this operation, in most cases a pinless meter is the best choice; it is six times faster and leaves no holes. However, if you will not unstack the lumber to measure the MC, then a pin meter is the only choice for a tight stack. That is, the pinless cannot get to the lumber inside a tight stack; but, the pin meter can be used on edge boards. Also, we do not want the moisture of the top layer or bottom layer, but want the moisture of the pieces inside the pack.
 
The pinless meter must be calibrated for the species and thickness of the lumber being measured. Make sure that you measure one piece at a time with air underneath the piece being measured. Read the meter instructions. When using the pinless hand meter for any pieces that seem too wet or too dry, the pieces can be checked with the pin meter to confirm the MC. Outside factors, like wood density variation, can make the pinless give a misleading reading sometimes.
 
So, how many readings should we take in order to get a good estimate of the MC in the entire load? The answer is 30 random pieces of lumber measured in one place away from the edges and ends. With 30 pieces, we will have an excellent estimate of the average MC of the whole load. (Similar to the elections, a small sample of 900 people out of millions of voters gives us an excellent estimate of who will win.) Many moisture meters have a memory and also have a button that you can push to give you the average of the readings you have saved. Some can even download to a computer using a cable or going wireless. In addition, these built-in computers (or you could manually enter the data into an inexpensive calculator) can calculate the spread of the data, using a number called the standard deviation. So, even though we may have 10,000 pieces in a kiln load, from 30 random pieces we can determine quite accurately the average and the spread.
 
The standard deviation (SD) is a neat number. If you take the average MC and then subtract the SD from the average and then add the SD to the average, you will have an interval that includes 2/3 of all the pieces in the load. (Example: If the average is 7.2 and the SD = 0.3, then two-thirds of the pieces in toehold will be between 6.9 percent MC to 7.5 percent MC.) This is an estimate, a very good estimate with moistures from 30 random pieces. If you want to know the complete picture precisely than you have to measure the MC of all 10,000 pieces. Even better than this first estimate, if we subtract two SD from the average and add two SD, we have an interval that includes about 95 percent of all the values in the load, even though we measured only 30 pieces. Finally, if we have three SD subtracted and added, we have essentially 100 percent of the MC values.
 
You have to admit that this is really cool! Especially when you factor in that probably three-quarters of all manufacturing problems are because of incorrect MCs. After instituting a program like this, I had one client that reduced the callbacks for wintertime splitting and open joint problems from 300 to just three. Another client reduced manufacturing defects to under 2 percent. What is really neat is that this program does not require a large capital investment. All that is required are two medium-priced meters ($300 each or more), a little employee time (which will save a lot of employee time later in manufacturing) and care in collecting and processing the data.
 
(Technical Info: This technique is called Statistical Process Control (SPC), Six Sigma (SD is a sigma and three SD on each side of the average or six sigma is 100 percent of the range) and several other names. Oftentimes, local community colleges have practical courses in this subject. The technique allies to many different operations in manufacturing and not just moisture measurement.)
 
Having said all this, I must add that even if the incoming MC is perfect, it is still necessary to store the lumber at the correct humidity to prevent MC changes. When this cannot be done, wrapping the lumber in plastic or using a tight shipping container trailer will accomplish the same thing. The lumber will not change MC for years.
 
Also, if you do find some lumber that is a bit too high in MC, use a pin meter with insulated needs and drive the needles to the center thickness of the lumber. This core value will be the MC when the lumber left the dry kiln in 99 percent of the cases. That is, the core MC does not change very quickly.

 

Have something to say? Share your thoughts with us in the comments below.