Brickman Consulting

Wood Flooring Solutions.

What Causes Cupping in Wood Floors?

by Leave a Comment

Cross posted from Howard Brickman's article on Hardwood Floors Magazine: Inspector Blog.

The official topic of this blog will be an excruciating discussion of the term “cupping” for your consideration. But first…

My apologies for taking so long to get another submission ready, but for me writing is very hard work and extremely time consuming, and I marvel at those gifted individuals who can churn out wonderful written content on a regular schedule. I must admit that I’m not certain that I would be able to increase my output even if I were offered bushel baskets of legal tender to do it on a full-time basis. Time is finite and moves at an increasingly rapid pace. I remember watching the classroom clock as a youngster on Friday afternoons thinking that 4 o’clock would never arrive, and now Friday afternoons pass by faster than the pickets on a fence. But I digress…

If I were going to give an official definition for cupping it would be, “boards that are concave on the face.” There is a common misconception in the wood flooring bidness that all cupping is moisture-related and that pressure that develops due to swelling is the primary cause. Let’s explore some thought experiments.

Experiment 1: We place 10 S4S red oak boards ¾” x 4” x 72” edge-to-edge, which approximates a panel ¾” x 40” x 72”. Then we place pipe clamps at 3” intervals across the 40” dimension and tighten the clamps until a pressure of 200 pounds per square inch is reached. What do you think is going to happen to the shape of the individual boards?

Experiment 2: We place 10 pieces of red oak flooring ¾” x 4” x 72” edge-to-edge, which approximates a panel ¾” x 40” x 72”. Then we place pipe clamps at 3” intervals across the 40” dimension and tighten the clamps until a pressure of 200 pounds per square inch is reached. What do you think is going to happen to the shape of the individual boards?

Experiment 3: We nail 10 pieces of red oak flooring ¾” x 4” x 72” at a MC of 6-8% to a ¾”-thick plywood panel 48” x 72” at a MC of 6-8%. Then we place 1½” deck screws at 3” intervals into the first and the last boards so that they will be prevented from moving. We predrill the oak so that there will be no splitting. Then we place a bath towel on the face of the boards and saturate it with enough water to completely wet the towel but not have water puddling onto the surface of the flooring. Then we put a piece of 6-mil polyethylene over the towel to keep the water from evaporating. What do you think is going to happen to the shape of the individual boards?
 
Experiment 4: We nail 10 pieces of red oak flooring ¾” x 4” x 72” at a MC of 6-8% to a ¾”-thick plywood panel 48” x 72” at a MC of 14%-16%. Then we place 1½” deck screws at 3” intervals into the first and the last boards so that they will be prevented from moving. We predrill the oak so that there will be no splitting. Then we put a piece of 6-mil polyethylene covering the underside of the plywood to keep the water from evaporating. What do you think is going to happen to the shape of the individual boards?

Experiment 5: We nail 10 pieces of red oak flooring ¾” x 4” x 72” at a MC of 14-16% to a ¾”-thick plywood panel 48” x 72” at a MC of 6-8%. Then we place 1½” deck screws at 3” intervals into the first and the last boards so that they will be prevented from moving. We predrill the oak so that there will be no splitting. Then we put a piece of 6-mil polyethylene covering the underside of the plywood to keep the water from evaporating. What do you think is going to happen to the shape of the individual boards?

Experiment 6: We nail 10 pieces of red oak flooring ¾” x 4” x 72” at a MC of 14-16% to a ¾”-thick plywood panel 48” x 72” at a MC of 14%-16%. Then we place 1½” deck screws at 3” intervals into the first and the last boards so that they will be prevented from moving. We predrill the oak so that there will be no splitting. Then we put a piece of 6-mil polyethylene covering the underside of the plywood to keep the water from evaporating. What do you think is going to happen to the shape of the individual boards?

The great thing about these thought experiments is that we could actually do them. If you had a university or commercial testing company do these for you, it would cost many thousands of dollars. If someone wants to send me $12,479.00, I will do the experiments and send you a report on the results with cool pictures. In my next episode we will start to discuss the individual experiments. I want to thank Don Sgroi for the very thought provoking e-mail, which is the inspiration for what will I think be a very interesting series of blogs.

Calculating Shrink/Swell and Why It Matters

by Leave a Comment

Cross posted from Howard Brickman's article on Hardwood Floors Magazine: Inspector Blog.

Wood shrinks and swells when it loses and gains moisture content (MC). For example, if a piece of 2¼" wide plainsawn red oak flooring were to decrease in MC from 8% to 5%, the net change in MC would be 3%. Using standard values from the Wood Handbook Table 4-3 Shrinkage Values of Domestic Woods, the net change in dimension would be .021" (2.25" x .086 x .03 / .28 = .021"). Expressed as a fraction, .021" would be between 1/64" (.015625") and 1/32" (.03125").

How is this information useful? Let’s take a real-world scenario and show how a quantitative understanding of dimensional change helps us perform an analysis.

SCENARIO:
We look at a wood floor that we installed last year where the customer has called to complain about gaps between boards. As part of our normal procedure, we look at the surface of the floor to see if individual boards are flat, cupped (concave), or crowned (convex). In this case, the boards are still very flat. Then we determine the size, frequency and distribution of the gaps. We note the minimum and maximum gaps, then we eliminate the smallest and largest gaps to describe the range, which characterizes the majority of the gaps (80-90%). In statistics this is referred as “eliminating the outliers." Now we choose several locations where the gapping is the most severe and begin a series of accurate board-width measurements, along with MC of the individual boards. Our results are:

MC of all of the boards is less than 6%. We estimate the MC at 5% based on interior RH for the last three weeks using Wood Handbook Table 4-2 Moisture Content of Wood in Equilibrium With Stated Temperature and Relative Humidity. The widths of individual boards range from 1/64" to 1/32" less than the manufactured width of 2¼". The gaps are located between every board and range in size from 1/64" to 1/32".

Danger! FORMULA ALERT: IF YOU BECOME SHORT OF BREATH, BREAK OUT IN A COLD SWEAT, AND HAVE DILATED PUPILS WHEN YOU READ FORMULAS, please skip this section of the blog. For you brave souls, let’s proceed.

FIRST FORMULA (Change in Dimension)
Δ D (change in dimension) = Manufactured Width x St (Shrinkage factor from Wood Handbook) x Δ MC / .28

SECOND FORMULA (Change in Moisture Content)
Δ MC = [Δ D x .28] / [ Width x St ]

With these two formulas we can:

1 – Predict the amount that a board will swell or shrink (Δ D) and
2 – Estimate the magnitude of change in MC (Δ MC) based on the current width of the board.

IT’S SAFE TO START READING AGAIN. Danger over.

Now it’s time for some SHRINKAGE RULES:

Rule Numero Uno: If a board is less than its manufactured width, it has lost MC since it was manufactured.

Rule Numero Dos: If a board is exactly its manufactured width, it is at the same MC as at the time of manufacture.

Rule Numero Tres: If a board is greater than its manufactured width, it has gained MC since it was manufactured.

Applying Rule Numero Uno, we know that our boards that are smaller than the manufactured width have lost MC. Using the SECOND FORMULA for Δ MC, we can pretty accurately quantify the change in MC.

For boards that are 1/64" narrower than 2¼", the Δ MC is 2.26%:

Δ MC= [Δ D x .28 ] / [ Width x St ] Δ MC= [ 1/64" x .28 ] / [ 2.25" x .086 ] Δ MC= [ .015625" x .28 ] / [ 2.25" x .086 ] Δ MC= [ .004375 ] / [ .1935 ] Δ MC= .0226 = 2.26%

For boards that are 1/32" narrower than 2¼", the MC is 4.52%:

Δ MC= [Δ D x .28 ] / [ Width x St ] Δ MC= [ 1/32" x .28 ] / [ 2.25" x .086 ] Δ MC= [ .03125" x .28 ] / [ 2.25" x .086 ] Δ MC= [ .00875 ] / [ .1935 ] Δ MC= .0452 = 4.52%

If we add the Δ MC to our current 5%, the boards that are 1/64" narrow were originally at 7.26% (5% + 2.26% = 7.26%). The boards that are 1/32" narrow were originally at 9.42% (5% + 4.52% = 9.53%). This allows us to estimate MC at time of manufacture between 7.26% and 9.52%.

I find these quantitative methods to be useful tools when working through the analysis of a wood floor that has evidence of a change (or changes) in MC. In new construction there are frequently several MC changes, starting with the adsorption of excessive moisture from the subfloor, then the eventual drying during the following winter heating season.

Let’s explore how doing all this rigmarole calculating helps with analysis. Let’s change our scenario by a single factor: instead of gaps that range from 1/64" to 1/32", how about gaps that range from 1/32" to 3/64" with individual board shrinkage that ranges from 1/64" to 1/32"? We have already done the calculations on the board shrinkage, but that doesn’t account for the additional size of the gaps. SO… something besides seasonal low interior RH would have to be the cause of the increased size of the gaps. Maybe the flooring was left on the job to “acclimate" and picked up some excessive moisture before it was it was installed? Or… (to be continued)