Sign up

Bending fatigue life of metal-plate-connected joints in furniture-grade pine plywood.
Subject:
Jointer (Woodworking machine) (Testing)
Fatigue testing machines (Testing)
Materials (Fatigue)
Materials (Testing)
Authors:
Zhang, Jilei
Yu, Youshan
Quin, Franklin
Pub Date:
11/01/2006
Publication:
Name: Forest Products Journal Publisher: Forest Products Society Audience: Trade Format: Magazine/Journal Subject: Business; Forest products industry Copyright: COPYRIGHT 2006 Forest Products Society ISSN: 0015-7473
Issue:
Date: Nov-Dec, 2006 Source Volume: 56 Source Issue: 11-12
Topic:
Event Code: 330 Product information
Product:
SIC Code: 2511 Wood household furniture
Geographic:
Geographic Scope: United States Geographic Code: 1USA United States

Accession Number:
157098559
Full Text:
Abstract

This study investigated fatigue performances of T-shaped, end-to-side, metal-plate-connected (MPC) joints in furniture-grade pine plywood. Tested joints were subjected to one-sided cyclic stepped bending loads. The purpose of the study was to obtain joint static to fatigue moment capacity ratios. Performance test results showed that a MPC plywood joint would fail within 25,000 cycles when a stepped load level reached 46 percent of the static moment capacity of the tested joint. Joints failed mainly due to tooth fatigue shear at the roots. The static to fatigue moment capacity ratio for tested joints averaged 2.5 with a coefficient of variation of 11 percent and a range of 2.2 to 3.1.

**********

Metal plates are chosen to connect highly stressed joints, such as front stump-front rail and bottom side rail-back post joints in upholstered furniture frame constructions (e.g., three-seat sofas, two-seat loveseats, and recliners) because metal-plate-connected (MPC) joints yield high moment resistance capacity. Limited information is available concerning static and fatigue moment capacities of MPC joints in wood composites for furniture frame construction, especially fatigue performance data.

One-sided constant fatigue testing of MPC furniture joints in solid wood (Eckelman 1980) indicated that the fatigue life of joints loaded to 3/8 of their ultimate bending moments amounted to an average of 22,000 cycles, while at one-fourth of ultimate moment the joint life is about 100,000 cycles. As the load level was reduced from one-fourth to one-sixth of ultimate, the life of the joint increased to a point where it may be regarded as infinite with respect to the life of the furniture.

The fatigue life of T-shaped. end-to-side, two-pin dowel joints was investigated (Zhang et al. 2003) by subjecting them to one-side constant and stepped cyclic bending loads. The ratios of static moment capacities of tested joints to their corresponding passed stepped moment levels averaged 2.2 with a coefficient of variation (COV) of 13 percent, and a range of 2.1 to 2.6. Major failure modes observed were dowels sheared parallel to the grain and dowels withdrawn from joint members.

Fatigue properties of wood butt joints with MPC in timber joining systems were evaluated under reversed and non-reversed cyclic loading (Hayashi et al. 1980). It reported that the ratios of fatigue strength as a percentage of static tensile strength were 75.4, 55.9, 38.2, and 24.7 percent for [10.sup.4], [10.sup.5], [10.sup.6], and [10.sup.7] cycles of non-reversed loading, respectively, while they were 52.2, 40.1, 29.8, and 20.5 percent for reversed loading conditions.

The influence of wood density on the mechanical behavior of M PC joints under cyclic loading was analyzed by Moura et al. (1995). Fatigue tests were made in a non-reversed load schedule (varying tension) following a sinusoidal function. During the fatigue tests, two different types of failure were observed. Type I--all nails were pulled out and wood fiber around the nails was deeply damaged. This rupture occurred on samples that performed a low number of cycles (< 2,000). Type II some teeth were sheared and the rest pulled out. It seemed that some teeth failed under fatigue first, increasing the load on the remaining ones sufficiently to pull them out before fatigue could set in. This failure mode was observed for the samples that sustained at least 2,500 cycles. It was concluded that the endurance limit can be relatively easy to estimate and lies at around 40 percent of ultimate static tensile strength corrected by individual densities. Tooth shear first occurred by 2,100 cycles. Beyond that cycle threshold, shearing of teeth was the mare parameter determining joint life.

Strength design of upholstered furniture frames in order to satisfy durability performance test standards, such as the General Service Administration (GSA) test regimen FNAE-80 to 214 A (GSA 1998), requires information about their joint fatigue performance data, especially performance characteristics under cyclic stepped loads (variable amplitude loading). This is due to the fact that the performance test regimen is based on a zero-to-maximum (one-sided or non-reversed) cyclic stepped fatigue load method rather than a static or constant amplitude cycling load method (Eckelman 1988a, 1988b; Eckelman and Zhang 1995).

Therefore, in order to design MPC furniture joints making use of the stepped load effects and satisfying different stepped load levels of furniture performance tests, it is necessary to study relationships between the static moment resistance capacity of the joints and their fatigue resistance capacity. Previous studies (Zhang et al. 2003) show that fatigue life of dowel joints can be estimated by moment-number of cyclic to failure curves, which are related to the static moment resistant capacity of the joints. Also, the ratio of the joint's static moment capacity to its fatigue capacity can be used to establish the relation between fatigue and static moment resistance for furniture joints. By testing joints with different configurations, this ratio will represent a wide range of moment resistance capacities.

This paper reports on a bending fatigue life study of MPC joints in plywood that was the continuation of the static moment capacity study reported by Zhang et al. (2005). The objective of this study was to determine the ratio of the static moment capacities of MPC furniture joints to their fatigue moment capacities, and to describe the failure modes of joints being subjected to stepped fatigue loads.

Materials and methods

The configuration of the T-shaped, end-to-side, MPC joint specimens for this study is shown in Figure 1. Each specimen consisted of two principal structural members, a post and a rail, joined together with a pair of metal plates symmetrically attached to both faces of the members, i.e., an equal number of teeth were pressed into both the rail and post. Plates were placed so that weak sections resist tensile loads as shown in Figure 1.

[FIGURE 1 OMITTED]

One type of furniture grade, 3/4-inch-thick seven-ply southern yellow pine plywood was included in this study. Also, one type of tooth configuration and steel material were chosen for metal plates in this study. Metal plate specifications (supplied by a manufacturer) and plywood materials properties are given in Table 1. The plywood and metal plates were the same as those used in the previous static moment capacity study (Zhang et al. 2005). The plywood properties were obtained in accordance with the procedures described in ASTM D 4761 (ASTM 2001a) and ASTM D 3501 (ASTM 2001b).

The columns I through 5 of Table 2 summarize dimensions of joint members and metal-plate connectors of each tested specimen group and also the average static moment capacity of each group. The average moment capacity of each joint group was estimated based on the results of a previously tested, same joint configuration group from a static moment capacity study (Zhang et al. 2005). These joint groups were selected with the intention of including different moment capacity levels from 3,000 to 10,000 lb-in, and also different joint member and plate sizes within the same moment capacity group. Joint specimens connected with 3-inch-wide metal plates represented the joint group that failed due to tooth withdrawal when they were tested with static loads, and specimens with 4.5-inch-wide metal plates represented the group that failed with the metal plate yielding in tension (Zhang et al. 2005).

Joint fatigue tests were conducted with a specially designed air cylinder and pipe rack system shown in Figure 2. Joint specimens were tested using two loading schedules, referencing the Arm-Outward test and Backrest Frame test loading schedules. An assumption was made that the metal-plate connectors were used to connect stump-to-front rail joints and back post-to-bottom side rail joints for a two-seat loveseat frame without a middle upright added onto the back frame system.

[FIGURE 2 OMITTED]

In the case of the Backrest Frame test, two loads were applied to the frame top back rail. The magnitude of each load followed the backrest flame test schedule as shown in the first column of Table 3. The loading assumption was that those two loads were applied to the top back rail which was simply supported at two ends, where they were connected to two back post top ends. Therefore, two reaction forces acting at the top back rail ends could be viewed as two end loads acting on the two back posts. The magnitude of each of these two loads was equal to half of the total load, which is equal to the load given in the first column of Table 3. The first fatigue load applied to the 30-inch-long rail joint specimen was 75 pounds with a moment arm of 26 inches, which yielded a moment of 1,950 lb-in. Therefore, the joint specimens were subjected to the stepped loads given in the third column of Table 3, i.e., an initial moment of 1,950 lb-in, was applied to the specimen at a rate of 20 cycles per minute for 25,000 cycles. After the prescribed number of cycles was completed, the moment was increased to the next loading level of 2,600 lb-in for another 25,000 cycles. This loading process was continued until the joint specimen failed to resist the load.

For the 16-inch-long rail joint specimens, the Arm-Outward test load schedule was applied, which is given in the first column of Table 4. Due to the fact that a single load was applied to the arm, the magnitude of its load was directly applied to joint test. The third column of Table 4 shows the stepped load schedule applied to the joints, i.e., the moment arm was set to 12 inches, assuming the moment was resisted by the stump-to-front rail joint.

In all tests, a one-sided fatigue load was applied to the joint by an air cylinder at a rate of 20 cycles per minute. A programmable logic controller and counter system recorded the number of cycles completed. Limit switches stopped the test when a joint suffered disabling damage. For each of two joint groups, specimens were tested within their referenced furniture testing load schedule. Loading ranges were 75 to 150 lb, and 50 to 200 lb, for Backrest Frame and Arm-Outward tests, respectively. If the joints passed all testing levels in the furniture schedule loading ranges, then tests continued on the extended load schedule as given in Tables 3 and 4 until the joints failed. Failure modes and loads, and cumulative cycle numbers, were recorded.

Results and discussion

Mode of failure

Three types of failure modes occurred in the tests: tooth shear off at the roots, tooth withdrawal from plywood, and post member compression. Tooth shear-off failure occurred mostly in the bottom half of the plate, presumably on the tension side, as shown in Figure 3. Tooth failure was also seen at the last one or two columns on the top half of the plate, presumably the compression side. No obvious wood compression failure was observed around the teeth which remained in the wood caused by tooth shear off. Tooth withdrawal failure occurred at the top half of the plate, accompanied by ply material torn off and attached to the teeth. It might be assumed that teeth failed first due to fatigue, which increased the load on the remaining teeth sufficiently to pull them out before fatigue could set in. No metal plate yield failure on the tension side was observed. The tension side failure of fatigue tests was different from the mode of the joint static test, which was tooth withdrawal from the tension side and metal-plate yielding on the tension side (Zhang et al. 2005).

[FIGURE 3 OMITTED]

Static to fatigue moment capacity ratio

Averaged values of passed, failed fatigue moments, and cumulative cycles to failure of tested joints with their COV are summarized in the "Fatigue" section of Table 2. The ratios of the static moment capacity of each tested joint set to its corresponding passed and failed fatigue moment are calculated and summarized in "Ratio" section of Table 2.

The ratio of "static moment" to "failed moment capacity" for 10 tested joint groups averaged 2.2, with a COV of 12 percent and a range of 1.9 to 2.8. The average ratio indicated that MPC joints in pine plywood failed within 25,000 cycles under a load level of 46 percent of their static moment capacities after they were subjected to a series of cyclic stepped loads. Under stepped loads with a maximum magnitude of 46 percent of static moment capacity, the fatigue moment resistance of MPC joints may be governed by metal tooth configurations and plate material fatigue properties, rather than plate tooth lateral withdrawal resistance capacity.

The ratio of 10 "static moments" to "passed moment capacities" averaged 2.5, with a COV of 11 percent and a range of 2.2 to 3.1. These results were compared with the results from a bending fatigue life study (Zhang et al. 2003) of two--pin dowel joints, which had an averaged static to fatigue moment capacity ratio of 2.2, a COV of 13 percent, and a range of 2.1 to 2.6. These results suggest that the average ratio of static to fatigue moment capacity for design of upholstered furniture frame joints considering fatigue effect can be set to various values for different types of fastening systems, such as 2.5 for the metal-plate connectors tested in this study and 2.2 for wood dowels.

The fatigue moment resistances of other types of commonly used furniture joints, such as end-to-side, in-plane gusset plate joints, end-to-face, and notched type joints, need to be investigated under the cyclic stepped load conditions. Fatigue moment resistance of gusset plate joints can be governed by the fatigue properties of wood, glue, and staple resisting lateral shear forces. For end-to-face, notched type joints, moment resistances to fatigue loads are governed by the properties of wood, glue, and staples resisting direct withdrawal forces. Therefore, ratios of static moment resistance capacities to fatigue moment capacities for these two types of joints, of which moment capacities depend on staple holding power and glue bonding strength, could be compared to dowel and MPC joints. Information obtained from these studies of joints subjected to cyclic stepped loads could supply furniture engineers with practical data to use in the rational design of upholstered furniture frames to meet desired durability requirements.

Conclusions

Fatigue performances of T-shaped, end-to-side, MPC joints in furniture grade pine plywood were investigated by subjecting them to one-sided cyclic stepped bending moments. Experimental results showed that MPC joints in pine plywood would fail within 25,000 cycles under the load level of 46 percent of their static moment capacities after they were subjected to a series of cyclic stepped loads. The average static to fatigue moment ratio for the design of MPC joints in an upholstered furniture frame would be set to 2.5, i.e., it might be advisable to design the MPC joints so that they will not be loaded to more than 40 percent of their static moment capacity.

Failure modes of these joints indicated that the resistance capacity to fatigue moments was governed by metal tooth resistance to lateral fatigue load, when these joints were subjected to one-sided cyclic stepped loads with a maximum moment under 46 percent of the static moment of tested joints. Joints mainly failed by tooth shear off at the roots followed by tooth withdrawal from plywood. These failure modes were different from those in static bending tests, where joints failed mainly due to tooth withdrawal or metal-plate yielding.

The average ratio of static moment capacity to fatigue moment for MPC joints tested in this study is different from wood dowel-type joints. Therefore, fatigue performances of other commonly seen types, such as gusset plate and notching based joints, need to be investigated in order to obtain appropriate ratios of the static moment resistance of the joints to their fatigue moment capacities. Subsequently, those values would be compared with ratio values of MPC and dowel joints.

Literature cited

American Society for Testing and Materials (ASTM). 2001a. Standard test methods for mechanical properties of lumber and wood-base structural material. ASTM D 4761-96. ASTM, West Conshohocken, PA.

-- 2001b. Standard test methods for wood-based structural panels in compression. ASTM D 3501-94. ASTM, West Conshohocken, PA.

Eckelman, C.A. 1980. The bending strength of furniture joints constructed with metal tooth connector plates. Inter. J. of Furniture Res. 2(1): 12-14 and 2(2):40-42.

-- 1988a. Performance testing of furniture. Part I. Underlying concepts. Forest Prod. J. 38(3):44-48.

-- 1988b. Performance testing of furniture. Part II. A multipurpose universal structural performance test method. Forest Prod. J. 38(4):13-18.

-- and J. Zhang. 1995. Uses of the General Serv. Administration performance test method for upholstered furniture in the engineering of upholstered furniture flames. Holz als Roh- und Werkstoff 53: 261-267.

General Service Administration (GSA). 1998. Upholstered furniture test method. FNAE-80-214A. Furniture Commodity Center, Federal Supply Service, Washington, DC.

Hayashi, T., H. Sasaki, and M. Masuda. 1980. Fatigue properties of wood butt joints with metal plate connectors. Forest Prod. J. 30(2):49-54.

Moura, J.D.D.M., C. Bastian, G. Duchanois, J.M. Leban, and P. Triboulot. 1995. The influence of wood density on metal-plate connector mechanical behavior under cyclic loading. Forest Prod. J. 45(11/12): 74-82.

Zhang, J., G. Li, and T. Sellers, Jr. 2003. Bending fatigue life of two-pin dowel joints in furniture grade pine plywood. Forest Prod. J. 53(9): 33-39.

--, Y. Yu, and F. Quin. 2005. Moment capacity of metal-plate-connected joints in furniture grade pine plywood. Forest Prod. J. 55(5): 45-51.

Jilei Zhang * Youshan Yu Franklin Quin

The authors are, respectively, Associate Professor, Forest Products Lab., Mississippi State Univ., Mississippi State, MS (jzhang@cfr.msstate.edu): Associate Professor, Architecture and Building Design Inst., Univ. of Sci. and Technology, Nanjing, China: and Research Associate, Forest Products Lab., Mississippi State Univ., Mississippi State, MS (fquin@cfr.msstate.edu). This paper was received for publication in September 2005. Article No. 10114.

* Forest Products Society Member.

[c] Forest Products Society 2006. Forest Prod. J. 56(11/12):62-66.
Table 1.--Metal plate specifications and plywood properties. (a)

Property                                           Specification

Metal plate
  Thickness (gage number)                          0.036 in. (20)
  Tooth density                                 10 teeth/[in..sup.2]
  Slot width                                          1/8 in.
  Slot length                                         1/2 in.
  Tooth length                                        5/16 in.
  Teeth configuration                                 in-line
  Tension effectiveness ratio                           0.73
  Yield tensile strength                        40 x [10.sup.3] psi
  Ultimate tensile strength                     52 x [10.sup.3] psi
Plywood
  Ultimate compressive strength perpendicular
    to face ply grain                           2.9 x [10.sup.3] psi
  Ultimate compressive strength parallel to     5.1 x [10.sup.3] psi
    face ply grain
  Modulus of rupture edge-wise                  6.7 x [10.sup.3] psi
  Modulus of elasticity edge-wise               1.0 x [10.sup.6] psi
  Moisture content (%)                                   6

(a) 1 in. = 25.4 min; 1 psi = 6.895 kPa.

Table 2.--Summary of dimensions, replications, and static moment
values of tested specimens and their corresponding fatigue
performance results. (a)

            Joint dimensions

     Rail         Plate-plate

Length   Width   Width   Length   Replication

                 (in.)

Subjected to 12-inch moment arm schedule

16        4.5     1.6      3           5
          4.5     2.4      3           5
          4.5     3.2      3           5
           6      3.2      3           5
           6      2.4     4.5          5
          7.5     2.4      3           5
          4.5     3.2     4.5          5

Subjected to 26-inch moment arm schedule

30        7.5     3.2      3           5
           6      3.2     4.5          5
          7.5     2.4     4.5          5

Static                            Fatigue

Static       Passed       Failed        Cumulative
moment       moment       moment    cycles to failure

            (lb-in.)

Subjected to 12-inch moment arm schedule

3,163    1,320 (12) (b)   1,620     75,000 + 18,868 (17)
4,531      1,680 (10)     1,980    125,000 + 2,795 (10)
5,381      2,040 (7)      2,340    150,000 + 8,866 (6)
7,135      3,000 (7)      3,300    225,000 + 8,529 (9)
7,526      3,180 (18)     3,480    225,000 + 23,668 (18)
7,654      2,460 (5)      2,760    175,000 + 23,549 (7)
7,824      3,600 (0)      3,900    275,000 + 11,164 (3)

Subjected to 26-inch moment arm schedule

8,760      3,770 (8)      4,420    100,000 + 923 (9)
9,828      3,770 (14)     4,420    100,000 + 5,455 (23)
9,912      4,550 (17)     5,200    125,000 + 11,057 (17)

 Ratio

Static/   Static/
passed    failed

Subjected to 12-inch moment arm schedule

 2.40      1.95
 2.70      2.29
 2.64      2.30
 2.38      2.16
 2.37      2.16
 3.11      2.77
 2.17      2.01

Subjected to 26-inch moment arm schedule

 2.32      1.98
 2.61      2.22
 2.18      1.91

(a) 1 in. = 25.4 mm; 1 lb-in. = 0. 113 N x m.

(a) Values in parentheses are coefficients of variations

Table 3.--Cyclic stepped load schedule for testing 30-inch-long
rail joint group with fatigue load applied using 26-inch moment
arm by referencing GSA Backrest Frame testing schedule. (a)

Backrest frame
    test               Joint test

       No. of                     Cumulative
Load   loads    Load   Moments      cycles

(lb)            (lb)   (lb-in.)

75       2       75     1,950       25,000
100      2      100     2,600       50,000
125      2      125     3,250       75,000
150      2      150     3,900      100,000

Extended test

175             175     4,550      125,000
200             200     5,200      150,000
225             225     5,850      175,000

(a) 1 lb.= 4.448 N; 1 lb-in. = 0.113 N x m.

Table 4.--Cyclic stepped load schedule for testing 16-inch-long
rail joint group with the fatigue load applied using 12-inch
moment arm by referencing to GSA Arm-Outward testing schedule. (a)

Arm-outward test       Join test

       No. of                     Cumulative
Load   loads    Load   Moments      cycles
(lb)            (lb)   (lb-in.)

50     1         50      600        25,000
75     1         75      900        50,000
100    1        100     1,200       75,000
125    1        125     1,500      100,000
150    1        150     1,800      125,000
175    1        175     2,100      150,000
200    1        200     2,400      175,000

Extended test

225             225     2,700      200,000
250             250     3,000      225,000
275             275     3,300      250,000
300             300     3,600      275,000
325             325     3,900      300,000
350             350     4,200      325,000

(a) 1 lb = 4.448 N; lb-in. = 0.113 N x m.
Gale Copyright:
Copyright 2006 Gale, Cengage Learning. All rights reserved.