Log in to view the full article
A more educated complainer will admit they know that wood varies in dimensions with moisture and will have acclimated the wood to the local job site, and even verified that the subfloor and wood flooring are within the industry standards (the strip is within 4% and plank within 2% of the subfloor moisture level).
A newly educated expert may quote the new NWFA standard:
• for flooring with an actual width less than or equal to 3 1⁄4”, the standard is + or - 0.005”.
• for flooring greater than 3 1⁄4”, the standard is + or - 0.008”.
The maximum width variation allowed, then, is 0.010” (ten-thousandths or one-hundreth inch) for strip and 0.016” (sixteen-thousandths or one sixty-fourth inch) for planks?
I challenge readers: Open a box of uninstalled solid plank (3 1⁄4” or wider) and measure the boards. Widths will vary by 0.020” to 0.030” or more (this applies to any brand, any species, any moisture content, any time)!
How can a board milled within a few thousandths of an inch by the manufacturer have variations much larger even before installation? Because it’s wood!
How to Inspect a Width-Variation Complaint
Here are some tips:
• Toss the tape measures and get the calipers and calculators. What “eyeballs” as 1⁄16” or 1⁄8” will measure as much less. A caliper ($15-30) can accurately measure to 1⁄1000” and reads the same in everyone’s hands, stopping the arguments (trying to calculate in fractions like 5⁄32”, 7⁄64”, etc., will frustrate all but high school math teachers!).
• Let the complainer pick and measure the worst uninstalled boards, not the gaps in the floors.
• Many factors can cause gaps: width variations (yes), but also uneven starting lines (especially walls), bowed boards, and uneven nailing. If you have to measure gaps, use mechanic’s feeler gauges, not eyeballs.
• Never measure wood without measuring moisture, for the wood changes exactly with the moisture change. This exact number is called the dimensional change coefficient; labeled CR for radial (quartersawn) and CT for tangential (flatsawn). It measures how much 1” of wood will change for 1% change in moisture content (between 6% and 14%). It is listed in the current Forest Products Laboratory Wood Handbook Table 13-5 for 100 species or more (PDFs of the Handbook chapters are available for free online).
• Remember that the new NWFA Standard allows a manufacturing MC range of 6-10%.
To think about dimensional change, take two quartersawn red oak boards—one at 6% MC, one at 10%. Mill them into identical 3 1⁄4” boards. If the 10% board equalizes to 6%, it shrinks by 4 x 3.25” x .00158 (% moisture change x width x CR), or 0.021”.
Take two flatsawn red oak boards—one at 6% MC, one at 10%. Mill them into identical 3 1⁄4” boards. If the 10% board equalizes to 6% it shrinks by 4 x 3.25 x .00369 (% change x width x CT), or .048” (between 1⁄32” and 1⁄16”).
This width variation will stay the same if both boards go to 4%, 8%, 10%, 12% or any MC; it is the moisture difference at time of manufacture that creates the width variations.
Some boards will vary in moisture from end to end; a board milled at 6% MC at one end and 10% at the other will vary in width from end to end.
Some boards will vary in grain from end to end; a board that is radial at one end and tangential at the other can vary in width if the moisture changes.
How can a manufacturer minimize width variation complaints?
1) Control moisture to a narrower range than the maximum 6-10%. A 4% range in red oak will allow almost a 1.5% maximum width variation (4 x .00369); the old NOFMA/NWFA published range of 6-9% allows about a 1% variation. Narrowing the moisture range to 2% means boards vary +/- by one third of one percent or less; but still +/- 0.008” on a 2 1⁄4” strip, 0.018” on a 5” plank!
2) Use radial grain (quartersawn) lumber; it varies by less than half the number of tangential (0.00158/0.00369).
3) Use more stable species. Beech (CT 0.00431) varies more than red oak (0.00369); cherry (0.00248) varies less.
4) Use eased/beveled edges to minimize visibility of gaps/width variations.
5) Be honest with consumers’ expectations! “Prefinished flooring will vary in width by up to 1⁄16” or 1% coming out of the box; this is not a defect!” Wide planks will show larger gaps.
How can a retailer/installer fix these width variations?
They can’t! But ...
1) Educate the consumer. Prefinished plank cannot be installed perfectly tight. Unfinished flooring has the same variations and gaps, but they are usually filled during finishing.
2) Start straight! When starting installations use the centerline starting method described in the NWFA Installation Guidelines.
3) For fussy consumers and wide planks, use tools that tighten the boards together, and remember that putty/filler is your friend. Don’t try to sort planks for width unless you’re compulsive and paid for it!
