A Withdrawal Reduction Factor for Structural Screws at Shallow Embedment Depths

To meet safety standards, it is advisable to implement a reduction factor of 0.8 when applying the reference withdrawal design values (RWDVs) for self-tapping screws with an effective thread penetration length (p_t) ranging from four to eight times their nominal diameter (D). This adjustment demonstrates prudence, given that the RWDVs are derived from test data focused on the most prevalent withdrawal scenarios in mass timber, where p_t is equal to or greater than 8D.

Approximately 3-minute read.

The reference withdrawal design values (RWDVs) prescribed in the Evaluation Service Reports 3178 and 3179 from the International Code Council Evaluation Service (ICC-ES) apply to ASSY^® self-tapping screws with an effective thread penetration length (p_t) of eight times or more their nominal diameter (D). For screws embedded at shallower depths of 4–8D, it is advisable to apply a reduction factor of 0.8 to the RWDVs. This modification ensures adherence to the established safety factor outlined in the ICC-ES Acceptance Criteria for Dowel-type Threaded Fasteners Used in Wood (AC233).

This recommendation is grounded in extensive comparisons of withdrawal resistance values at p_t = 4–16D. In mass timber construction, maintaining a p_t ≥ 8D is customary to ensure optimal withdrawal performance. Various issues, including micro-splitting, often arise at smaller p_t values, potentially compromising wood integrity and, consequently, withdrawal resistance. A direct result is a discontinuous correlation between p_t and capacity over a broad range of p_t values, prompting a prioritized approach.

To maximize applicability, RWDVs are derived from tests conducted under the most common withdrawal scenarios encountered in mass timber, specifically those involving p_t ≥ 8D. Incorporating data for p_t < 8D—for example, data for p_t = 4D, as shown in Figure 1—leads to an underestimation of capacity at p_t = 8D and 12D, as well as an overestimation at p_t = 16D. Notably, the deviation becomes increasingly pronounced as D increases, as evidenced by a comparison of Figure 1(a) and (b).
 

(a) D = 6 mm

(b) D = 8 mm

Figure 1. Relationship between withdrawal resistance and effective thread penetration length (expressed as a multiple of the nominal screw diameter D) for Douglas-fir glued laminated timber specimens (reference withdrawal design values not shown)

The discrepancies primarily stem from the disparity in the mechanism governing load response. At p_t = 8D and 12D, withdrawal resistance prevails, whereas at p_t = 4D and 16D, the screw behavior is predominantly influenced by splitting, due in part to test constraints with respect to specimen dimensions, and the interplay between tensile and withdrawal capacity, respectively. Given that situations involving p_t < 8D are infrequent and associated with intrinsic material complexities that introduce uncertainty, including such data into the determination of RWDVs would significantly diminish both estimation accuracy and efficiency.

In cases where economical and installation considerations warrant lower p_t values, a reduction factor can be applied to rectify potential overestimations resulting from the model established with higher p_t values.

Importantly, a p_t < 4D should be avoided in all cases, as achieving sufficient withdrawal performance is unlikely under this condition.

The consideration of p_t is crucial when employing the RWDVs for self-tapping screws. Situations where p_t falls in the range of 4–8D may necessitate the application of a reduction factor of 0.8 to satisfy safety requirements. For more information and design guidance, please contact our Technical Support Team. 🙂