Anchor Bolt Steel Strength in Annular Stand-Off Base Plate Connectionsdocs.trb.org/prp/14-3892.pdf · 1 Anchor Bolt Steel Strength in Annular Stand-Off Base Plate Connections 2 3

McBride, Cook, Prevatt, Potter 1

Anchor Bolt Steel Strength in Annular Stand-Off Base Plate Connections 1 2 By : 3 Kenton E. McBride1, Ronald A. Cook, Ph.D.2, David O. Prevatt3, William Potter4 4 5

6 1. Corresponding author: Doctoral Student, University of Florida, Department of Civil & Coastal Engineering, P.O. 7

116580, Gainesville, FL 32611 (Tel: 352-392-9537 Ext. 1503, E-mail: [email protected]) 8 9 2. Professor, University of Florida, Department of Civil & Coastal Engineering, P.O. 116580, Gainesville, FL 32611 10

(Tel: 352-392-9537 Ext. 1507, Fax: 352-392-3394, E-mail: [email protected]) 11 12 3. Associate Professor, University of Florida, Department of Civil & Coastal Engineering, P.O. 116580, Gainesville, 13

FL 32611 (Tel: 352-392-9537 Ext. 1495, Fax: 352-392-3394, E-mail: [email protected]) 14 15 4. Senior Structures Design Engineer, Florida Department of Transportation, Tallahassee, FL 32399 (Tel: 850-921-16

7100, E-mail: [email protected]) 17 18 Submitted : 15 November 2013 19 Word count : 4909 words, 10 Tables/Figures 20

TRB 2014 Annual Meeting Paper revised from original submittal.


ABSTRACT 1

Anchor bolts in annular stand-off base plates connecting cantilever sign and signal structures to concrete foundations 2 may experience high shear from base plate torsion and direct shear forces. A three-phase experimental study 3 evaluated the steel shear strength of anchor bolts in grouted and ungrouted annular stand-off base plate connections 4 to concrete. Phase 1 included flush-mounted and ungrouted stand-off base plates with 5/8 in. and 1 in. diameter bolts 5 loaded in direct shear. Phase 2 employed a novel torsion test approach on ten circular groups of six 5/8 in. diameter 6 bolts installed in flush-mounted, ungrouted stand-off, and grouted stand-off base plates along with three ungrouted 7 groups of three 1 in. diameter bolts. Phase 3 comprised one ungrouted and three grouted full-scale annular base plate 8 tests of six 1.25 in. diameter bolts under predominantly torsion loading. Results from this study suggest that 9 AASHTO’s provisions allowing bolt bending stresses to be ignored for anchor bolts with exposed length less than 10 one bolt diameter overestimates ultimate strength. The beam model suggested by AASHTO accounting for anchor 11 bolt bending fits experimental data well and is thus recommended for all ungrouted annular stand-off base plates. 12 ACI’s 0.8 reduction factor for steel shear strength of anchor bolts in grouted connections was found to be 13 conservative and is also recommended. 14



INTRODUCTION 1

Highway sign and signal structures are typically connected to concrete foundations via an annular base plate with 2 anchor bolts extending into the foundation. The majority of such connections include nuts below the base plate that 3 are used to level the structure on-site. The space between the top of the concrete surface and the bottom of the base 4 plate may or may not be filled with a grout pad. Figure 1 shows an ungrouted annular stand-off base plate with 5 leveling nuts. For the purposes of this paper, “exposed length” is defined as the distance from the concrete surface to 6 the bottom of the leveling nut and “stand-off distance” is defined as the distance from the concrete surface to the 7 bottom of the base plate. 8

9 FIGURE 1 Ungrouted annular stand-off base plate connection. 10

Loads on sign and signal structures are predominantly induced by wind or seismic events and self-weight. 11 In cantilever sign structures, a resultant lateral load produces global overturning moment, torsion, and direct shear at 12 the base of the pole, which are transferred through the annular base plate to anchor bolts through combinations of 13 axial and shear forces. Axial forces in individual bolts within the group are produced by equally shared axial base 14 plate loads and unequally distributed loads from base plate overturning moment based on anchor bolt position 15 relative to the neutral axis of global bending. Shear force on individual anchor bolts is the resultant vector from 16 equally distributed direct shear and resolved global torsion. An elliptical relationship between combined tension and 17 shear stresses as shown in Equations 1 and 2 has been found appropriate in predicting individual bolt capacity in 18 bolted steel-to-steel connections by Chesson et al. (1) and in flush-mounted base plate connections to concrete by 19 Cook and Klingner (2). AASHTO (3), AISC (4), and ACI (5) include provisions for combined tension and shear 20 using variations of the elliptical equation represented in terms of stress in Equations 1 and 2. 21

�𝑓𝑓𝑣𝑣𝐹𝐹𝑣𝑣�2

+ �𝑓𝑓𝑡𝑡𝐹𝐹𝑡𝑡�2

≤ 1.0 (1)

�𝑓𝑓𝑣𝑣𝐹𝐹𝑣𝑣�2

+ �𝑓𝑓𝑐𝑐𝐹𝐹𝑐𝑐�2

≤ 1.0 (2)

where 𝑓𝑓𝑣𝑣 is the shear stress acting on the bolt, 𝑓𝑓𝑡𝑡 and 𝑓𝑓𝑐𝑐 are the tensile and compressive stresses acting on the bolt, 22 and 𝐹𝐹𝑣𝑣 ,𝐹𝐹𝑡𝑡, and ,𝐹𝐹𝑐𝑐 are the allowable shear, tensile, and compressive stresses. 𝐹𝐹𝑣𝑣 is taken as 0.6𝐹𝐹𝑡𝑡 in AASHTO’s 23 LRFD Specifications (3) and AASHTO LTS-6 (6) following recommendations by Kulak et al. (7). 24

In stand-off base plate connections, shear forces act over the exposed length of the anchor bolts, producing 25 local bending stresses in the anchor bolts at the top and bottom of the exposed length in addition to the shear stresses 26 acting over the bolt area, resulting in reduced stiffness and strength. Based on recommendations in NCHRP 412 (8), 27 AASHTO (6) Section 5.17.4.3 allows anchor bolt strength in ungrouted stand-off base plate connections to be taken 28 as the strength of anchor bolts in analogous flush-mounted base plate connections for exposed lengths less than one 29 nominal anchor bolt diameter, beyond which tension and compression stresses from bolt bending must be added to 30 the axial stress component of Equations 1 and 2. AASHTO (3) C6.13.2.12 states that strength should be multiplied 31 by a 0.8 factor as appropriate to account for “oversize holes and other factors.” Where grout pads are present, ACI 32 318-11 (4) Section D.6.1.3 specifies that the flush-mounted steel shear strength of anchor bolts be multiplied by a 33 0.8 reduction factor. AASHTO (3) 5.17.3.3 states that grout pads shall not be designed to contribute to strength, 34 resulting in an equivalent design to an ungrouted base plate of the same stand-off distance. 35



The ultimate strength of anchors in stand-off base plate connections has been studied in various forms. Lin 1 et al. (9) conducted steel-to-steel tests under shear loading, restraining vertical displacement in a double-test 2 assembly. Adihardjo and Soltis (10) and Nakashima (11) studied grouted single-bolt connections under various 3 combinations of applied shear and tensile loads. Fichtner (12), Gresnigt et al. (13) and Gomez et al. (14) tested 4 multiple-bolt connections in grouted rectangular base plates under predominantly shear loading (12) and combined 5 shear and tension loading (13 and 14). Grosser (15) studied shear behavior of anchor bolts in a flush-mounted base 6 plate varying concrete strength and steel properties, the effects of which may be amplified in stand-off base plates. 7 None of this previous work isolated the direct shear behavior and strength observed in anchor bolts in circular or 8 annular stand-off base plates. 9

This paper presents results from a three-phase Florida Department of Transportation research study (16) of 10 the shear strength of anchor bolts in ungrouted and grouted stand-off base plates. Ultimate strengths predicted using 11 existing AASHTO (6) and ACI (4) methodology are evaluated against test results. Modifications to AASHTO (6) 12 provisions for estimating anchor bolt strength in annular stand-off base plates are proposed. 13

EXPERIMENTAL SETUP 14

Table 1 provides the matrix of reported tests. Major test variables were connection type (i.e. ungrouted or grouted) 15 and base plate stand-off distance. Minor test variables were test method, hole size, anchor bolt diameter, top nut 16 pretensioning, and the presence of a Fiber-Reinforced Polymer (FRP) retrofit around a grouted base plate. In Table 17 1, 𝑑𝑑𝑏𝑏 is the nominal anchor bolt diameter, 𝑛𝑛 is the number of anchor bolts in a test, 𝑙𝑙𝐵𝐵𝐵𝐵 is the base plate stand-off 18 distance, and 𝑙𝑙𝐿𝐿𝐿𝐿 is the exposed length of the bolt. 19

All tests used fully threaded cast-in-place (CIP) ASTM F1554 Grade 55 anchor bolts headed with ASTM 20 A563 heavy hex nuts embedded in FDOT Class IV Drilled Shaft concrete with a specified 28-day compressive 21 strength of 4000 psi. Sufficient edge distance and embedment depth were achieved to preclude concrete failure. 22 Where present, block reinforcement was placed away from regions of concrete spall development. In flush-mounted 23 tests the base plate rested on the concrete surface with an intermediate 1/32 in. Teflon layer to minimize friction. All 24 ungrouted and grouted stand-off base plate tests included a nut and washer above and below the base plate. Bolt 25 tension and additional displacement readings are provided in (16). 26



TABLE 1 Overall Test Matrix 1

Phase Seta Reps 𝑑𝑑𝑏𝑏 (in.)

Hole Size (in.) 𝑛𝑛 Connection

Typeb Top Nut

Tightnessc 𝑙𝑙𝐵𝐵𝐵𝐵𝑑𝑑𝑏𝑏

𝑙𝑙𝐿𝐿𝐿𝐿𝑑𝑑𝑏𝑏

1 DS1 8 0.625 0.63 1 FM FT 0 na 1 DS2 6 0.625 0.63 1 U FT 1.2d 0

1 DS3 2 0.625 0.63 1 U FT 1.6 0.4 1 DS4 10 0.625 0.63 1 U FT 2 0.8 1 DS5 4 0.625 0.63 1 U FT 3 1.8 1 DS6 5 0.625 0.63 1 U FT 4 2.8 1 DS7 4 0.625 0.63 2 FM FT 0 na 1 DS8 5 0.625 0.63 2 U FT 2 0.75 1 DS9 4 1 1.01 1 FM FT 0 na 2 T1 2 0.625 0.81 6 FM FT 0 na 2 T2 2 0.625 0.81 6 U TOTN 2 0.8 2 T3 1 0.625 0.81 6 U TOTN 4 2.8 2 T4 1 0.625 0.81 6 U FT 2 0.8 2 T5 2 0.625 0.81 6 G TOTN 2 0.8 2 T6 1 0.625 0.81 6 G TOTN 4 2.8 2 T7 1 0.625 0.81 6 GF TOTN 4 2.8 2 T8 1 1 1.25 3 U FT 0 na 2 T9 1 1 1.25 3 U TOTN 2 0.75 2 T10 1 1 1.25 3 U TOTN 4 2.75 3 FS1 1 1.25 1.75 6 U TOTN 2.3 1 3 FS2 1 1.25 1.75 6 G TOTN 2.3 1 3 FS3 1 1.25 1.75 6 G TOTN 4.3 3 3 FS4 1 1.25 1.75 6 G TOTN 4.3 3

na - Not Applicable 2 a DS - Direct Shear, T-Torsion, FS - Full-Scale 3 b FM - Flush-Mounted, U - Ungrouted, G - Grouted, GF - Grouted with FRP Retrofit 4 c FT - Finger-Tightened, TOTN - Turn-of-the-Nut 5 d Note: Lowest possible stand-off with leveling nut and washer 6

Phase 1: Direct Shear Tests 7

Steel shear strength of anchor bolts were tested in the direct shear setup shown in Figure 2. Bolts were installed in 8 tight holes (1/16 in. oversize) in a 1.18 in. thick rectangular base plate. Load was applied following ASTM E488 9 protocol through a 7/8 in. diameter ASTM A193 Grade B7 threaded rod. In tests with base plate stand-off, the 10 loading rod was attached to the plate through a hinged clevis connection, while in flush-mounted tests the loading 11 rod was threaded directly into the plate. Nuts were installed in a “finger-tight” condition at approximately 20 in.-lb. 12

A majority of direct shear tests were conducted using the single-bolt setup shown in Figure 2(a), which 13 included a roller to restrain base plate rotation during load application. To test the effect of the roller, the double-bolt 14 setup shown in Figure 2(b) was employed in two test series. At high displacements, the roller in single-bolt tests 15 restrained vertical displacements, producing slightly higher stiffnesses than in equivalent double-bolt tests. 16 However, ultimate strength was not significantly affected and the use of several stand-off distances in these tests 17 helped identify reduction trends in the ultimate strength. 18



(a)

(b)

FIGURE 2 Phase 1 experimental setup for (a) one-anchor tests and (b) two-anchor tests (not to scale). 1

Phase 2: Scaled Torsion Testing 2

A test method was developed to reproduce the shear loading on anchor bolts in ungrouted and grouted annular base 3 plate connections subjected to torsional loading. By applying torsion to bolt groups, shear loading on individual 4 bolts within the groups was maintained throughout the loading process, even as the base plate moved downward due 5 to bolt geometry at high displacements. Ten groups of six 5/8 in. diameter anchor bolts were tested to determine the 6 effects of stand-off distance, ungrouted or grouted condition, anchor bolt pretensioning, and a Fiber-Reinforced 7 Polymer (FRP) wrap around a grout pad. Three groups of 1 in. diameter anchor bolts were tested to investigate size 8 effect. 9

Tests containing six 5/8 in. diameter anchor bolts were designed to represent scaled versions of field 10 annular base plate connections for sign and signal structures. The base plate assembly comprised a 12.5 in. diameter, 11 0.625 in. thick steel base plate, a 12 in. HSS 7.500 x 0.500 pipe stub, and a 2 x 14 x 1 in. thick rectangular “loading 12 wing.” Six 0.81 in. diameter and three 1.25 in. diameter oversize holes were cut in the circular base plate following 13 AASHTO (3) specifications for “shear holes” for 5/8 in. and 1 in. diameter anchor bolts, respectively. The minimum 14 recommended base plate thickness by NCHRP 412 (6) of one nominal anchor bolt diameter was chosen. To 15 maintain this for 1 in. diameter anchor bolt tests, 0.375 in. plates were welded to the base plate at the 1.25 in. hole 16 locations. After its component parts were welded together, the base plate assembly was case-hardened. 17

The Phase 2 test setup is shown in Figure 3. Anchor bolt groups were installed in 46.5 x 46.5 x 12 in. deep 18 reinforced concrete blocks. Six-bolt groups were installed four to a block and the larger-diameter three-bolt groups 19 were installed one per block. Rolling reaction frames included two C10 x 15.3 channel sections separated by a 3 in. 20 gap for load application and restraint. The frames supported independent hydraulic actuators connected to the base 21 plate assembly “loading wing” by a 1 in. diameter ASTM A193 B7 threaded rod and a 1 in. diameter clevis pin 22 connection. Each frame reacted against the opposite side of the concrete block via a 1.5 in. diameter ASTM A193 23 B7 tension rod through a PVC duct as shown in Figure 3. No external support was used. 24

Specimen and component preparation varied by type of test. Top nuts were pretensioned using the turn-of-25 the-nut method in nine of ten stand-off base plate tests in a star pattern. Test T4 investigated the effect of removing 26 pretension by finger-tightening bolts to approximately 20 lb.-in. Flush-mounted tests were not pretensioned in order 27 to minimize friction contributions to strength. In grouted tests, nonshrink high-strength grout pads were installed 28 flush with the perimeter of the base plate after bolt pretensioning and concrete surface preparation, which entailed 29 removal of debris/laitance and a “saturated surface dry” condition at the time of grout installation. Grout was mixed 30 to an ASTM C939 flow rate of 20-25 seconds and installed per manufacturer’s instructions. Clear plastic formwork 31 was removed at 24 hours, after which the grout surface was painted with a product-specific curing compound and 32 wrapped with wet paper towels. Grouted tests were conducted at seven days after grout placement. 33



Torsion was applied through a force couple supplied by independently operated 120 kip capacity hydraulic 1 actuators. Applied load to the actuators was measured by moment-compensating through-hole load cells. A step-2 and-hold approach was used in the linear range of load-displacement behavior, after which hydraulic flow to the 3 actuators was supplied monotonically until anchor bolt rupture. Load cell data were viewed in real time by test 4 operators to ensure equal load between actuators throughout the test. Displacement readings were taken by Linear 5 Variable Differential Transducers (LVDTs) oriented parallel to the applied load. 6

7 FIGURE 3 Phase 2 experimental setup. 8

Phase 3: Full-scale Torsion-Dominant Testing 9

Four tests were conducted on groups of six 1.25 in. diameter anchor bolts under predominantly torsion loading. As a 10 baseline, test FS1 was left ungrouted with 𝑙𝑙𝐿𝐿𝐿𝐿 = 1𝑑𝑑𝑏𝑏, the maximum exposed length permitted by AASHTO (3) 11 without reduction in shear load capacity from bolt bending. Test FS2 was identical to FS1 with the addition of a 12 grout pad. An “extreme” exposed length value of 3𝑑𝑑𝑏𝑏 was chosen in grouted tests FS3 and FS4. Top nuts were 13 pretensioned using the turn-of-the-nut method and grout was treated as in the Phase 2 experimental description. 14

Each test specimen was composed of a 6 ft. tall x 10 ft. wide x 3 ft. deep reinforced concrete block 15 containing the anchor bolt group in the center. The block was tied to a structural floor using steel beams and 1.5 in. 16 diameter threaded rods as shown in Figure 4. The loading assembly included a 24 in. diameter, 1 in. thick circular 17 base plate with 1.75 in. diameter holes, the oversize value permitted by current FDOT standards. For test FS4, a 15.5 18 in. diameter circle was cut out of the center of the base plate to investigate annular vs. circular grouted base plate 19 performance. The base plate was welded to HSS 16.000 x 0.500 steel pipe assembly containing a 23 in. stub welded 20 to a 120 in. loading arm through a 45 degree cut from each member and a 0.5 in. thick rectangular plate. 21

The location of load application produced a 108 in. torsion arm and 17.625 in. moment arm relative to the 22 centroid of the bolt group and the top of the anchor bolt exposed length, respectively. Displacement-controlled load 23 was applied at a constant hydraulic flow rate through a pinned connection to an actuator supported by the strong 24 floor of the laboratory at a rate corresponding to approximately 150 pounds per second in the linear portion of each 25



test. Load was measured with a 50-kip load cell threaded into the actuator. Displacement was measured using two 1 string potentiometers connected at mid-thickness of the base plate. 2

3 FIGURE 4 Phase 3 experimental setup. 4

RESULTS 5

A summary of results from all phases in the study is provided in Table 2. The ultimate shear load reported for each 6 test, 𝑉𝑉𝑢𝑢,𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡, is expressed as the maximum applied shear load per bolt and ultimate displacement, 𝛿𝛿𝑢𝑢, corresponds to 7 this load. Phase 3 ultimate load is expressed as the maximum resultant shear from direct shear and torsion load at 8 failure. Phase 2 and Phase 3 𝑉𝑉𝑢𝑢,𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡 values reflect adjusted values based on displaced geometry and self-weight of 9 components. 𝛿𝛿𝑢𝑢 values reported for Phase 2 and Phase 3 reflect the distance traveled along the circumference of the 10 bolt group circle. 𝑇𝑇𝑢𝑢 and 𝐹𝐹𝑢𝑢,𝑡𝑡 represent the measured ultimate tension strength and stress corresponding to the 11 appropriate anchor bolt batch for a given test. 12

Concrete strengths were measured according to ASTM C39 and grout strengths were measured according 13 to ASTM C939. The 28-day concrete strengths for Phase 1 and Phase 2 testing were 4990 psi and 6360 psi, 14 respectively. Concrete strength was measured on the day of every Phase 3 test; in order, FS2-FS4 strengths were 15 6590 psi, 7160 psi, 7140 psi, and 8460 psi. Grout strengths for Phase 2 tests T5, T6-B, and T7 were 5630 psi, 7480 16 psi, and 6230 psi. Phase 3FS2-FS4 grout strengths were 9100 psi, 9010 psi, and 8170 psi, respectively. The ultimate 17 tension strength of anchor bolt threaded rod material was determined by testing specimens from every unique batch 18 according to ASTM F606. 19



TABLE 2 Summary of experimental results 1

Phase Set Reps 𝑙𝑙𝐵𝐵𝐵𝐵𝑑𝑑𝑏𝑏


𝑇𝑇𝑢𝑢 (kip)

𝑉𝑉𝑢𝑢,𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡

(kip) COV 𝑉𝑉𝑢𝑢,𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡

𝑇𝑇𝑢𝑢 𝛿𝛿𝑢𝑢

(in.) 𝛿𝛿𝑢𝑢𝑑𝑑𝑏𝑏

1 DS1 8 0 na 21.1 12.3 3% 0.59 0.18 0.29 1 DS2 6 1.2 0 21.1 11.3 1% 0.54 0.56 0.89 1 DS3 2 1.6 0.4 21.1 9.8 na 0.46 0.77 1.23 1 DS4 10 2 0.8 21.1 7.9 6% 0.37 0.72 1.15 1 DS5 4 3 1.8 21.1 5.6 6% 0.27 0.64 1.03 1 DS6 5 4 2.8 21.1 4.4 10% 0.21 0.74 1.18 1 DS7 4 0 Na 21.1 11.4 5% 0.54 0.17 0.27 1 DS8 5 2 0.75 21.1 6.9 3% 0.33 0.64 1.02 1 DS9 4 0 Na 55.8 33.5 1% 0.60 0.31 0.31 2 T1 2 0 Na 20.7 11.1 na 0.54 0.30 0.30 2 T2 2 2 0.8 20.7 6.5 na 0.31 0.56 0.56 2 T3 1 4 2.8 20.7 4.6 na 0.22 0.73 0.73 2 T4 1 2 0.8 20.7 7.4 na 0.36 0.60 0.60 2 T5 1 2 0.8 20.7 10.9 na 0.53 0.56 0.56 2 T6 2a 4 2.8 20.7 13.1 na 0.63 1.03 1.03 2 T7 1 4 2.8 20.7 14.8 na 0.71 0.44 0.44 2 T8 1 0 Na 55.8 31.7 na 0.57 0.39 0.39 2 T9 1 2 0.75 55.8 18.0 na 0.32 0.88 0.88 2 T10 1 4 2.75 55.8 10.5 na 0.19 1.57 1.57 3 FS1 1 2.3 1 84.4 27.1 na 0.32 1.89 1.52 3 FS2 1 2.3 1 84.4 59.1 na 0.70 2.12 1.70 3 FS3 1 4.3 3 84.4 46.9 na 0.56 3.57 2.86 3 FS4 1 4.3 3 84.4 52.8 na 0.63 2.53 2.02

na - Not Applicable 2 a T6-A was not run to failure; all T6 results reflect T6-B 3

4 Figure 5 provides load-displacement behavior of representative individual tests from each phase. Bolt 5

displacement values, 𝛿𝛿𝑏𝑏, are normalized by respective nominal bolt diameters and the shear load per bolt, 𝑉𝑉𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡 , is 6 normalized by respective 𝑇𝑇𝑢𝑢 values. One test, T6-A, was not run to failure. Its load-displacement behavior, however, 7 is still displayed in Figure 5(c). 8



1

(a)

(b)

(c)

(d)

FIGURE 5 Load-displacement behavior for (a) selected direct shear tests, (b) ungrouted torsion tests, (c) 2 grouted torsion tests, and (d) full-scale tests. 3

DISCUSSION OF RESULTS 4

General Load-Displacement Behavior for Major Test Variables 5

Flush-mounted tests demonstrated linear-elastic behavior until approximately 0.35𝑇𝑇𝑢𝑢, beyond which stiffness 6 degraded until ultimate load as concrete spalling and inelastic steel deformation developed. Aside from T1-B (see 7 hole oversize discussion), ultimate load values were commensurate with the code-specified value of 0.6𝑇𝑇𝑢𝑢. 8

Ungrouted stand-off base plate tests showed linear-elastic behavior to approximately 0.15𝑇𝑇𝑢𝑢 and 0.08𝑇𝑇𝑢𝑢 for 9 2D and 4D tests, respectively, followed by a ductile inelastic phase as the bolts deformed laterally over their 10 exposed lengths. Ultimate displacements of the 2D and 4D tests were on the order of one bolt diameter for both with 11 the exception of the 1 in. diameter 4D torsion test (T10) and the full-scale 2.3D test (see bolt size discussion). 12

Grouted stand-off base plate tests exhibited markedly different behavior than their flush-mounted 13 counterparts. Initial behavior in all cases was very stiff until grout cracking, after which behavior diverged between 14 tests. The two tests run at lesser stand-off distances, T5 and FS2, transitioned smoothly into an approximately linear 15 inelastic range, degrading only moderately until failure. Tests run at higher stand-off distances, T6-A, T6-B, FS3, 16 and FS4, were split into two distinct patterns of behavior. Shortly after grout cracking, T6-B and FS4 entered a 17 phase of stiffness approximately parallel to the ungrouted 4D stand-off test, T3. After significant deformation, these 18 tests transferred into a final phase of increased stiffness leading to ultimate load. T6-A and FS3, however, contained 19 additional segments of elevated stiffness and associated strength gain immediately following grout cracking, 20 reaching local maxima then losing load until displacements roughly correspondent to the T6-B and FS4 transition 21 into a final phase of elevated stiffness. The post-cracking behavior observed in T6-A and FS3 is believed to have 22



been caused by shear interlock between cracked sections in the grout pad that, when overcome, caused the tests to 1 retrograde to the baseline behavior observed in T6-B and FS4 equivalents. At high lateral displacements, downward 2 displacement of the base plate attributable to anchor bolt bending geometry “clamped” the grout pad, which in turn 3 offered vertical restraint, generating significant friction forces until anchor bolt rupture below the leveling nut. The 4 ultimate load of all grouted tests ranged from approximately 0.55𝑇𝑇𝑢𝑢 to 0.7𝑇𝑇𝑢𝑢, all above the ACI (4) allowable 5 strength of 0.8 times the bolt shear strength of 0.6𝑇𝑇𝑢𝑢resulting in 0.48𝑇𝑇𝑢𝑢, suggesting that this value is conservative. 6

Effect of Minor Experimental Variables on Load-Displacement Behavior 7

While interactions between minor variables cannot be quantified without further investigation, inferences of their 8 impact on testing can be gleaned from observation of failure load and load-displacement behavior. Potential 9 contributions to the study’s major test variables are identified for the following minor study variables. 10

• Test Method: Stiffnesses in the inelastic phase of single-bolt 2D and 4D stand-off direct shear tests 11 are observably higher than in the torsion tests. Double-bolt direct shear behavior, however, closely 12 resembled that seen in torsion tests. Because the magnitudes of ultimate load were commensurate 13 between Phase 1 and Phase 2 equivalents, it was determined that Phase 1 ultimate load values are 14 applicable to the shear strength of field-installed annular base plate connections and valid for 15 comparison. 16

• Hole Oversize: Effects of hole oversize were observed only in flush-mounted torsion tests T1-A and 17 T1-B. While the two tests were nearly identical through the initial slip and linear phases of behavior, 18 the stiffness of T1-B degraded at a lower displacement and failed at lower ultimate load (0.51𝑇𝑇𝑢𝑢) than 19 T1-A (0.56𝑇𝑇𝑢𝑢), in good agreement with the AASHTO (3) commentary recommendation for a 0.8 factor 20 for oversize holes. The effects of slip in oversize holes diminished with the larger displacements 21 observed in stand-off tests. 22

• Bolt Diameter: Aside from flush-mounted 5/8 in. torsion tests (see hole oversize discussion), ultimate 23 loads normalized by respective 𝑇𝑇𝑢𝑢 values were comparable between equivalent ungrouted stand-off 24 tests with 5/8 in. and 1 in. diameter bolts. Thus, it has been determined that the results are 25 representative of the full range of anchor bolt diameters in ungrouted annular base plates. Grouted tests 26 with 5/8 in. and 1.25 in. diameter bolts exhibited similar load-displacement behavior at comparable 27 stand-off distances. 28

• Bolt Pretensioning: The effect of bolt pretensioning can be inferred directly from ungrouted 2D stand-29 off torsion tests T2 (pretensioned) and T4 (finger-tightened). Base plate slip occurred immediately in 30 T4, while no slip is apparent throughout the T2 curve. T4 linear elastic stiffness was lower, likely 31 owing to additional bolt bending within the thickness of the base plate. Surprisingly, however, the 32 inelastic phase was stiffer in T4 than T2 and the ultimate load was higher (0.36𝑇𝑇𝑢𝑢vs. 0.31𝑇𝑇𝑢𝑢). It is 33 hypothesized that the same intra-plate bending that caused lower stiffness in the linear region allowed 34 greater tensile pickup from bolt geometry. Thus, the consequences of absence/loss of pretension in 35 ungrouted stand-off base plate connections may not be critical beyond serviceability considerations. 36

• FRP Grout Pad Retrofit. Initial behavior of the grouted 4D stand-off torsion test with an FRP retrofit, 37 T7, was similar to other grouted tests. After apparent grout cracking at 0.3𝑇𝑇𝑢𝑢, however, radial 38 displacement of grout material was restrained by the FRP wrap and the test entered a phase of stiffness 39 equivalent to flush-mounted tests until approximately 0.6𝑇𝑇𝑢𝑢. A final phase of stiffness commenced 40 until explosive failure. The smooth sheared failure surface of the anchor bolts at the top surface of the 41 grout pad was identical to that observed in flush-mounted tests. This result shows promise for 42 retrofitting anchor bolts in existing ungrouted and grouted stand-off base plates to flush-mounted 43 strength or higher. 44

Other variables including base plate thickness and straight-line (Phase 1) vs. circumferential (Phases 2 and 45 3) in-test displacement, were determined to have had minimal impact on the results. 46

Evaluation of Existing Design Specifications for Anchor Bolt Strength in Annular Stand-off Base Plates 47

Equations within this section are expressed for an individual bolt positioned for worst-case values of tension and 48 resolved shear within a group in a stand-off base plate exposed to shear (𝑉𝑉𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔), overturning moment (𝑀𝑀𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔), 49 torsion (𝑇𝑇𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔), and axial (𝑁𝑁𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔) forces. 50



AASHTO (6) states that anchor bolt exposed lengths may be accounted for using a model of a doubly 1 moment-restrained beam free to displace laterally, placing resulting axial stresses from bolt bending into the axial 2 stress component as shown in Equation 3 for tension. This method is applied to both ungrouted and grouted base 3 plates with 𝑙𝑙𝐿𝐿𝐿𝐿 greater than one bolt diameter. 4

5

�𝑓𝑓𝑡𝑡,1 + 𝑓𝑓𝑡𝑡,2

𝐹𝐹𝑡𝑡�2

+ �𝑓𝑓𝑣𝑣𝐹𝐹𝑣𝑣�2

≤ 1 (3)

6 𝑓𝑓𝑡𝑡,1 is defined as the individual bolt tensile stress from global equilibrium, conservatively assumed to be 7

oriented perpendicular to the axis of overturning moment in the following expression: 8 9

𝑓𝑓𝑡𝑡,1 =𝑀𝑀𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔

𝑆𝑆𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔+

𝑁𝑁𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔(𝑛𝑛)(𝐴𝐴𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡)

(4)

𝑆𝑆𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔 =𝑛𝑛 ∗ 𝑟𝑟𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔

2 (5)

10 𝑓𝑓𝑡𝑡,2 is defined as the individual bolt tensile stress produced by bolt bending over the exposed length: 11 12

𝑓𝑓𝑡𝑡,2 =𝑚𝑚𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡

𝑆𝑆𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡

(6)

𝑚𝑚𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡 =𝑉𝑉𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡𝑙𝑙𝐿𝐿𝐿𝐿

2 (7)

𝑉𝑉𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡 =𝑉𝑉𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔𝑛𝑛

+𝑇𝑇𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔

(𝑟𝑟𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔)(𝑛𝑛) (8)

where 𝑛𝑛, 𝑑𝑑𝐿𝐿𝐿𝐿, and 𝑉𝑉𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡 are defined earlier, 𝐴𝐴𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡 is the net tensile area of the threaded portion of the bolt, 𝑓𝑓𝑣𝑣 = 13 𝑉𝑉𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡/𝐴𝐴𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡, 𝑆𝑆𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔 is the section modulus of a bolt positioned for maximum global moment-induced tension, 𝑚𝑚𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡 14 is the bolt-level moment produced by 𝑉𝑉𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡 acting over 𝑙𝑙𝐿𝐿𝐿𝐿, 𝑆𝑆𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡 is the section modulus of the bolt, and 𝑟𝑟𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑔𝑔 is the 15 radius of the bolt group. For comparison to test data 𝐹𝐹𝑡𝑡 is taken as the measured ultimate strength 𝐹𝐹𝑢𝑢,𝑡𝑡, 𝐹𝐹𝑣𝑣 is taken as 16 0.6𝐹𝐹𝑢𝑢,𝑡𝑡 in accordance with AASHTO (3) and AASHTO (6), and 𝑆𝑆𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡 is replaced by 𝑍𝑍𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡, the plastic section 17 modulus of the bolt. 18

For grouted base plates, the ACI 318 (4) 0.8 shear strength reduction factor results in 𝐹𝐹𝑣𝑣,𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑡𝑡𝑔𝑔𝑑𝑑 =19 (0.8)(0.6)𝐹𝐹𝑡𝑡 and 20

�𝑓𝑓𝑡𝑡,1

𝐹𝐹𝑡𝑡�2

+ �𝑓𝑓𝑣𝑣

𝐹𝐹𝑣𝑣,𝑔𝑔𝑔𝑔𝑏𝑏𝑢𝑢𝑡𝑡𝑔𝑔𝑑𝑑�2

≤ 1 (9)

Table 3 provides mean values for each experimental dataset against AASHTO (6) predicted ultimate values 21 and proposed modifications. Values using the interaction equation were found by solving Equations 3 and 9 for 𝑉𝑉𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡 22 using Equations 4 through 8 and in-text definitions. Proposed modifications to AASHTO (6) used in Table 3 are as 23 follows: 24

• The AASHTO (6) beam bending methodology given in Equation 3 is applied to all ungrouted 25 stand-off tests. 26

• The AASHTO (3) 0.8 shear resistance reduction factor for oversize holes is applied to 𝐹𝐹𝑣𝑣 for 27 ungrouted stand-off tests. 28

• Grouted tests are treated as flush-mounted (i.e., no bolt bending considerations) with the ACI (4) 29 0.8 shear strength reduction factor and no additional reduction factor for oversize holes. 30

31 32



TABLE 3 Comparisons between design methods 1

Phase Set 𝑙𝑙𝐵𝐵𝐵𝐵𝑑𝑑𝑏𝑏


𝐹𝐹𝑣𝑣𝐹𝐹𝑇𝑇

(proposed)

𝑉𝑉𝑢𝑢,𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡

(kip)

AASHTO (Code-Specified)

AASHTO (Proposed Modifications)

𝑉𝑉𝑢𝑢(6)

(kip) 𝑉𝑉𝑢𝑢,𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡

𝑉𝑉𝑢𝑢(6)

𝑉𝑉𝑢𝑢,𝑔𝑔𝑔𝑔𝑏𝑏𝑔𝑔𝑏𝑏𝑝𝑝𝑔𝑔𝑑𝑑

(kip) 𝑉𝑉𝑢𝑢,𝑏𝑏𝑏𝑏𝑙𝑙𝑡𝑡

𝑉𝑉𝑢𝑢,𝑔𝑔𝑔𝑔𝑏𝑏𝑔𝑔𝑏𝑏𝑝𝑝𝑔𝑔𝑑𝑑

1 DS1 0 na 0.6 12.3 12.7 0.97 12.7 0.97 1 DS2 1.25 0 0.6 11.3 12.7 0.89 12.7 0.89 1 DS3 1.6 0.4 0.6 9.8 12.7 0.77 10.6 0.93 1 DS4 2 0.8 0.6 7.9 12.7 0.62 7.7 1.03 1 DS5 3 1.8 0.6 5.6 12.7 0.44 4.0 1.38 1 DS6 4 2.8 0.6 4.4 2.6 0.35 2.7 1.65 1 DS7 0 na 0.6 11.4 12.7 0.90 12.7 0.90 1 DS8 2 0.75 0.6 6.9 12.7 0.55 7.7 0.90 1 DS9 0 na 0.6 33.5 33.5 1.00 33.5 1.00 2 T1 0 na (0.8a)(0.6) 11.1 12.4 0.89 10.0 1.11 2 T2 2 0.8 (0.8a)(0.6) 6.5 12.4 0.52 6.8 0.95 2 T3 4 2.8 (0.8a)(0.6) 4.6 2.6 1.75 2.6 1.77 2 T4 2 0.8 (0.8a)(0.6) 7.4 12.4 0.59 6.8 1.08 2 T5 2 0.8 (0.8b)(0.6) 10.9 12.4 0.88 10.0 1.10 2 T6 4 2.8 (0.8b)(0.6) 13.1 2.6 4.95 10.0 1.31 2 T7 4 2.8 (0.8b)(0.6) 14.8 2.6 5.61 10.0 1.49 2 T8 0 na (0.8a)(0.6) 31.7 33.5 0.95 26.8 1.18 2 T9 2 0.75 (0.8a)(0.6) 18.0 33.5 0.54 18.7 0.97 2 T10 4 2.75 (0.8a)(0.6) 10.5 7.3 1.45 7.2 1.47 3 FS1 2.3 1 (0.8a)(0.6) 27.1 26.9 1.01 25.0 1.08 3 FS2 2.3 1 (0.8b)(0.6) 59.1 26.9 2.19 40.5 1.46 3 FS3 4.3 3 (0.8b)(0.6) 46.9 10.4 4.52 40.5 1.16 3 FS4 4.3 3 (0.8b)(0.6) 52.8 10.4 5.08 40.5 1.30

na - Not Applicable 2 a Reduction factor for oversize holes per (3) 3 b Reduction factor for grouted base plates per (4) 4

5 In Figure 6 (a), ungrouted results are shown with AASHTO (6) code-predicted ultimate strength and two 6

predictive curves over all values of 𝑙𝑙𝐿𝐿𝐿𝐿 using Equation 3 with 𝐹𝐹𝑣𝑣 = 0.6𝐹𝐹𝑡𝑡 (Tight Holes) and 𝐹𝐹𝑣𝑣 = 0.48𝐹𝐹𝑡𝑡 (Oversize 7 Holes), respectively. Figure 6 (b) normalizes ungrouted results by the AASHTO (6) code-predicted strength and (c) 8 normalizes ungrouted test results by the applicable proposed modifications for each test. Figure 7 shows similarly 9 normalized grouted test results against the AASHTO code-predicted ultimate strength and the ACI (4) predicted 10 ultimate strength. In Figure 6 (a) and Figure 7, separate x-axes are given for 𝑙𝑙𝐵𝐵𝐵𝐵 and 𝑙𝑙𝐿𝐿𝐿𝐿. However, because the 11 difference between 𝑙𝑙𝐵𝐵𝐵𝐵 and 𝑙𝑙𝐿𝐿𝐿𝐿 is defined by the combined thickness of the leveling nut and leveling washer, 12 exposed lengths of DS9 and T8-10 and stand-off distances for full-scale results are not properly reflected by Figure 13 6. Refer to Table 1 for these values. 14



1 (a) 2

(b)

(c)

FIGURE 6 Ungrouted experimental results normalized by (a) direct tension strength, (b) AASHTO (6) 3 ultimate values (c) proposed modifications to AASHTO (6). 4



1

2 FIGURE 7 Grouted experimental results normalized by direct tension strength. 3

CONCLUSIONS AND RECOMMENDATIONS 4

The AASHTO (6) allowance for ignoring bolt bending for exposed lengths less than one bolt diameter is 5 unconservative in predicting ultimate anchor bolt strength in ungrouted stand-off base plates. Recommended 6 modifications to AASHTO (6) apply its beam bending model to anchor bolts in all ungrouted annular stand-off 7 connections and incorporate the 0.8 shear strength reduction factor for hole oversize from AASHTO (3). Future 8 work will include development of an expanded model building upon this approach using an adjusted effective 9 exposed length value and secondary geometry of the anchor bolts. 10

Grout pads significantly increase the steel shear capacity of anchor bolts in stand-off base plates, albeit at 11 high levels of connection deformation. The ACI (4) anchor bolt shear strength reduction factor of 0.8 for grouted 12 base plates was found to be conservative against test results. Existing AASHTO (6) specifications stating that a 13 grout pad shall not be factored in the strength of the connection, then, may be overly conservative. The application 14 of Equation 9 for the steel strength of anchor bolts in grouted annular base plate is recommended for properly 15 installed high-strength nonshrink grout pads with a thickness of up to four nominal anchor bolt diameters. For 16 existing ungrouted annular base plates with inadequate design strength using Equation 3, grout pad retrofits 17 designed using Equation 9 are recommended. 18



ACKNOWLEDGEMENTS 1

The authors are grateful to the Florida Department of Transportation for funding this study under contract BDK75 2 TWO 977-49. 3

REFERENCES 4

1. Chesson, E., N. L. Faustino, and W.H. Munse. High-Strength Bolts Subjected to Tension and Shear. ASCE 5 Journal of the Structural Division, Vol. 91, No. ST 5, 1965, pp. 155-180. 6

2. Cook, R. A. and R. E. Klingner. Ductile Multiple-Anchor Steel-to-Concrete Connections. ASCE Journal of 7 Structural Engineering, Vol. 118, No. 6, 1992, pp. 1645-1665. 8

3. AASHTO LRFD Bridge Design Specifications. AASHTO, Washington, D.C., 2013. 9

4. American Institute of Steel Construction. Steel Construction Manual, 14th Ed. AISC, Chicago, IL, 2010. 10

5. American Concrete Institute. Building Code Requirements for Structural Concrete. ACI 318-11. Farmington 11 Hills, Mich., 2011. 12

6. Standard Specifications for Structural Supports for Highway Signs, Luminaires, and Traffic Signals, 6th ed. 13 AASHTO, Washington D.C., 2013. 14

7. Kulak, G. L., J. W. Fisher, J. H. Struik. Guide to Design Criteria for Bolted and Riveted Joints. John Wiley & 15 Sons, Inc., New York, 1987. 16

8. Kaczinski, M. R., R. J. Dexter, J. P. Van Dien. NCHRP Report 412: Fatigue-Resistant Design of 17 Cantilevered Signal, Sign and Light Supports. TRB, National Research Council, 1998. 18

9. Lin, Z., D. Petersen, J. Zhao, Y. Tian. Simulation and Design of Exposed Anchor Bolts in Shear, Geneva, 19 CH. International Journal of Theoretical and Applied Multiscale Mechanics, Vol. 2, No. 2, 2011, pp. 111-20 129. 21

10. Adihardjo, R. and L. Soltis. Combined Shear and Tension on Grouted Base Details. AISC Engineering 22 Journal, Vol. 16, No. 1, 1979, pp. 23-26. 23

11. Nakashima, S. Mechanical Characteristics of Exposed Portions of Anchor Bolts in Steel Column Bases Under 24 Combined Tension and Shear. Journal of Constructional Steel Research, Vol. 46, No. 1, 1998, pp. 262-263. 25

12. Fichtner, P. R. Investigations on the Behavior of Group Fixtures Taking into Account Anchor Plate Thickness 26 and a Layer of Mortar (in German). Ph.D. Dissertation, University of Stuttgart, Stuttgart, DE, 2011. 27

13. Gresnigt, N., A. Romeijn, F. Wald, and M. Steenhuis. Column Bases in Shear and Normal Force, Delft, NL. 28 HERON, Vol. 53, No. 1, 2008, pp. 87-108. 29

14. Gomez, I. R., A. M. Kanvinde, and G. G. Deierlein. Experimental Investigation of Shear Transfer in Exposed 30 Column Base Connections. AISC Engineering Journal, Vol. 48, No. 4, 2011, pp. 245-264. 31

15. Grosser, P. R. Load-Bearing Behavior of Anchorages Subjected to Shear and Torsion Loading in Uncracked 32 Concrete. Ph.D. Dissertation, University of Stuttgart, Stuttgart, DE, 2012. 33

16. Cook, R.A., K. E. McBride, D. O. Prevatt. Florida Department of Transportation Report BDK75 977-49: 34 Steel Shear Strength of Anchors with Stand-Off Base Plates, 2013. 35

36


Documents

Anchor Bolt Steel Strength in Annular Stand-Off Base Plate Connectionsdocs.trb.org/prp/14-3892.pdf · 1 Anchor Bolt Steel Strength in Annular Stand-Off Base Plate Connections 2 3