Truss Lumber: Checking the Moisture Content

The lack of attention to the moisture content (MC) of lumber used in truss manufacturing can cause or contribute to in-service performance issues. Mr. Glenn Traylor’s fine article in the September issue of The Advertiser, “Does Your Lumber Have a Shelf Life?”, was greatly appreciated, as Glenn once again provided practical educational contributions. In this article, Glenn offered 18 items that CMs should consider for the management of truss lumber and the second recommendation caught my attention:

“2. Lumber should be inspected for moisture content.”

For me, the suggestion begs the question—how should lumber MC be inspected?

I’ve often been told by various construction-related parties that they “check the moisture” by lifting it. While this practice works for lumber freshly cut from a live tree, it is not appropriate for truss lumber. Moisture content (MC) in percent is defined by the following equation: MC = 100 (Weight of water in the wood/Weight of oven-dry wood). So, with this definition in mind, we can consider how to evaluate MC and how it affects lumber in use.

Green Lumber

At the time lumber is sawn from logs, the MC is typically 150% up to 200%. For the more common MC of 150%, the weight of water in the wood is 1.5 times its dry weight. Hence, freshly cut firewood will indeed feel heavy when lifted. To a lesser extent, commonly available pressure treated (PT) lumber will feel heavier when it is lifted as the treating process adds a water chemical solution to the lumber.

MC of Truss Lumber Background

Dry lumber, by definition, is lumber having a MC of less than or equal to 19% (MC≤19%). Tabulated design values for most truss lumber in the U.S. are based on dry lumber. Lumber having a MC > 19% is partially seasoned and requires a downward adjustment of tabulated design values based on factors in the National Design Specification for Wood Construction (NDS) Supplement.

The impact of lumber MC between 19–30% is substantial in terms of design stresses, shrinkage, and creep deflection in-service. At 30%, the lumber is assumed to be “saturated”, thus higher moisture contents only make the lumber heavier and contribute to additional creep deflection in-service. MC values between 30–200% do not impact shrinkage potential.

Practical Implications of Non-Compliant Truss Lumber

Case 1—MC non-compliant at time of truss fabrication

When the truss plates are installed in lumber with a MC > 19%, plate “griping values” must be reduced by 20% per the ANSI/TPI 1–2014 Standard [See Section 6.4.6.3 Moisture Content>19% at the Time of Fabrication and Table 6.4-4 Wet Service (C_M)]. Without the MC adjustment, a truss joint connection could be overloaded by up to 25%.

Case 2—MC non-compliant when trusses are installed and in-service

Creep deflection is the primary structural issue. Creep deflection is affected by two variables—sustained load level and MC when load is applied. According to ANSI/TPI 1–2014, the “creep factor” for deflection due to sustained loading is “3.0 for trusses using green lumber or for wet service conditions.”

While it is not possible to list all possible ways additional creep deflection due to higher MC can impact in-service performance, low-slope roof trusses (flat roofs) and floors trusses come to mind. For parallel chord roof trusses, creep deflection can exacerbate “ponding” potential. This subject has been studied in depth by Scott Coffman, P.E. and reported in STRUCTURE Magazine in a two-part article, ¼ in 12 Design Slope and Water Drainage and Low-Slope Roof and Deck Design Considerations.

For 4x2 floor trusses, one concern is additional creep deflection due to greater sustained load levels, such as hard surfaces, kitchen islands, and heavy appliances. See the article by Scott Coffman, P.E. in this issue for an in-depth discussion of design for in-service floor deflections.

Another truss lumber MC concern is the potential for excessive settlement of the exterior walls in multi-family construction up to five stories. Assuming the 4x2 floor trusses are not top chord bearing, 14 pieces of flatwise 2x_ framing would be involved in a calculation for potential settlement of the wall below the 5^th floor ceiling. To evaluate the impact of elevated truss chord and plate MC on potential wall settlement, a rule-of-thumb for shrinkage perpendicular-to-grain found in lumber grade rule books is: 4% Change in MC = 1% Change in size.

Using the five-story example cited above and assuming the wall plates and floor truss chords have an elevated MC of 8%, additional wall settlement in-service would be calculated as follows: (%Change in MC)/4 x 0.01 x (Lumber thickness involved); (8% MC increase)/4 x 0.01 x (14 x 1.5 ins.) = 0.42 ins.

Excessive wall framing settlement can cause a wide variety of compatibility issues with the wall sheathing and cladding assembly.

See the WoodWorks^TM publication, Accommodating Shrinkage in Multi-Story Wood-Frame Structures, for an in-depth discussion of some of the issues.

Measuring Truss Lumber MC

The challenge for the CM (and contractors as well) is to detect lumber having a MC between 19 and 30% as lumber may not be properly dried when delivered or become hydrated after receipt.

• As will be demonstrated, the use of a hand-held moisture meter (Figure 1) is the only way to obtain a meaningful measurement of MC in a truss plant environment.

Calculating the Weight of a 2x4x16 Southern Pine

The key to understanding that “lifting” lumber to gauge MC will not provide accurate information is to first know the natural variation of the density of dry wood. For example, we measured the specific gravity SG (or density) of a large sample of 2x4 southern pine (SP) and it ranges from about 0.40 to 0.70. Of course, most of the samples were around 0.55, which is the published value for SP. From these data, the weight of pieces of lumber with assumed MC can be quickly calculated.

The volume in cubic feet (V, ft³) of the example 2x4x16 is calculated as: (1.5”/12 ft.) x (3.5”/12 ft.) x (16 ft.) = 0.58 ft³. The approximate weight of a piece of truss lumber with volume V, and assumed SG and MC, is given by the equation: Weight= V (SGx62.4) x (1+MC/100), where 62.4 is the weight of a cubic foot of water in lbs./ft³, the basis of the SG measure.

The following table gives the weight of the example 2x4x16 based on assuming a lumber MC of 15, 19, 25 and 30% for three cases, SG 0.4, 0.55, and 0.70.

• Using the apparent “weight” of the 2x4x16 as felt by the “lifting” method, what conclusions are likely to be reached regarding the MC of the lumber?

Table 1. Calculated weight in pounds of pieces of lumber (2x4x16) having combinations of MC and SG.

For example, the calculated weight data in the table shows that a piece of lumber with SG equal to 0.70 and 19% MC weighs 30.1 pounds. Note that the four weights for the MC compliant cases highlighted in green exceed four of the weight values for non-compliant lumber highlighted in red, indicating non-compliant lumber would not be detected by the “lifting method” for monitoring lumber MC. Moreover, assuming the accuracy of the lifting method is 10% for the dense lumber (0.70 SG), none of the non-complaint lumber would be rejected using 30 lbs. as a benchmark for acceptable MC.

In conclusion, the natural variation of lumber SG masks the difference in the weight of a piece of lumber with a MC value in the range of 19 to 30%.

Recommendations for Checking MC of Truss Lumber

In addition to the recommendations given by Glenn Traylor in his article last month, I suggest the following at the time of delivery:

1. Using a hand-held moisture meter, bunks of truss lumber (including FRT and KDAT lumber) at the time of delivery should be tested for MC and recorded by size and mill number located on the grade stamp.
2. Additional attention should be given to FRT lumber because it is typically processed by another party after it is manufactured and dried to the grade-marked MC specification.
3. Only meters with a “calibration device” should be considered. The meter accuracy should be checked on a scheduled basis.
4. Based on my experience with hand-held moisture meters checked against the scientific method per ASTM D4444, I believe the accuracy of a calibrated hand-held meter for 2x_ Southern Pine (comprised of Shortleaf, Slash, Loblolly, and Longleaf pines) lumber is about plus/minus 3%.
5. MC readings can be corrected for species when used for different species. The species adjustment factors for hardwoods and softwoods should be included with the meter calibration device obtained from the manufacturer.
6. Surface water or visible water from a recent rain shower can affect MC readings, thus the lumber surface must be allowed to dry before obtaining an accurate reading.
7. When elevated MC is observed upon delivery, the mill (lumber supplier) should be contacted and made aware of the issue. This practice should encourage better MC quality control before the lumber reaches your truss plant.

Following these steps will help in any assessment of MC before parallel chord roof trusses and floor trusses are in-service and subjected to sustained loads, typically “uniform dead loads” plus substantial concentrated loads.

Learning More About Lumber

The most authoritative document on truss lumber is the American Softwood Lumber Standard PS 20-15 and it can be used to answer most questions about lumber as a product, manufacturing requirements, and related commerce. For example, in the context of this article, Section 6.2.1 Seasoning provisions defines the meaning of non-compliant lumber with respect to moisture content:

6.2.1 General—The grading rules shall include provisions regulating lumber seasoning and moisture content. … The restrictions on the moisture content of seasoned lumber shall apply at the time of shipment, at the time of dressing (if dressed lumber is involved), and at the time of any reinspection (if moisture content is involved in the reinspection), as provided in the applicable grading rules. Any piece exceeding the allowable moisture content of the applicable grade provisions shall be considered out of compliance. (Underline for emphasis.)

While not referenced in PS 20-15, the ASTM standards used to calculate design values for lumber rely on an assumed average MC of KD19 lumber to be 15%. CMs can use the ASTM 15% MC-number to get a feel for what the average MC of well-dried lumber should be. In other words, samples of KD19, MC19, and S-DRY lumber should have an average MC of about 15%, and a maximum value of 19%.

Frank Woeste, P.E., is Professor Emeritus, Virginia Tech and a wood construction consultant. Frank, along with his colleagues, have developed and presented continuing education programs for more than 30 years. Mark your calendar for our next course, May 21-22, 2019. Visit http://www.cpe.vt.edu/sdwnds/index.html for details.