Proposed Specifications for LRFD Soil-Nailing Design and Construction (2011)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

D-1 D.1 Introduction This appendix presents a comparison of SNW designs based on the ASD and LRFD approaches. The comparison was between designs of identical cases and conditions of SNWs for both the ASD and LRFD approaches. Designs were performed using the computer programs SNAILZ (Caltrans, 2007) and GOLDNAIL (Golder, 1993), the two most commonly used computer programs for SNW design in the United States. Section D.2 provides a brief description of these two computer programs. Section D.3 provides an overview of the comparisons. Section D.4 contains results of a parametric study conducted to assess the design sensitivity to various factors. Section D.5 presents results of a comparison based on a design example presented in the FHWA Geotechnical Engineering Circular (GEC) 7 (Lazarte et al., 2003) of a SNW using the ASD method and designs for the same wall using the LRFD method. D.2 Computer Programs Used in Comparative Analyses D.2.1 SNAILZ Basic Features SNAILZ, developed by the California Department of Trans- portation (Caltrans, 2007), is an updated version of the pro- gram SNAIL (Caltrans, 1991) and is currently the most widely used program in the United States for the design of SNWs. The program is available through the public domain and can be downloaded free-of-charge from http://www.dot.ca.gov/ hq/esc/geotechlrequest.htm. Technical support is limited. SNAIL was originally in a Microsoft® DOS platform. SNAILZ runs within a Microsoft Windows® environment. SNAILZ is versatile as it allows the design engineer to consider various design scenarios and the most common elements that partic- ipate in the design of a SNW. The user can input nail bond and tensile resistances, as well as the facing resistance. Program Capabilities SNAILZ can model only two-dimensional wall geome- tries. It is based on the limit-equilibrium method and only achieves force equilibrium. Moment equilibrium is gener- ally not achieved in this program; therefore, results from SNAILZ are only approximate but are considered accept- able for design purposes. SNAILZ uses two-part planar wedges. It can model slip sur- faces with one wedge exiting the SNW toe and the other to the ground surface behind the modeled wall [Figure D-1(a)]. This is the most common scenario for SNWs. The program can also model approximately a slip surface extending behind and below the wall using a simplified passive earth pressure formulation for the section below the wall toe [Figure D-1(b)]. However, this solution approach is only approximate. Therefore, the slid- ing and basal heave limit states can be modeled only approxi- mately with this program. SNAILZ can model up to seven soil layers. Up to three points define the water table location, which for some ground- water conditions may not be sufficient. SNAILZ allows a max- imum of two uniform surcharge distributions behind the face of the wall. Therefore, the program may have limited capa- bilities to model complex stratigraphy and load conditions. For complex wall geometries, stratigraphy distributions, or load conditions, the design engineer may need to simplify actual conditions due to the program limitations. However, for most common conditions encountered in SNW design practice, this program produces acceptable results, even in relatively complex design situations. Input Parameters Parameters selected for input in SNAILZ include those related to reinforcement, loads, and soil. Reinforcement param- eters include nail head depth on the wall face, nail diameter, nail inclination, vertical and horizontal nail spacing, bar cross- sectional area, and nail tensile resistance. These parameters can A P P E N D I X D Comparison of ASD- and LRFD-Based Designs of Soil Nail Walls

D-2 be assigned either to individual nails or globally to all nails. Up to two uniform surcharge distributions can be input. Pseudo- static seismic loads can be considered in SNAILZ by entering horizontal and vertical seismic coefficients. Soil parameters include soil unit weight, soil cohesion, friction angle, and bond resistance. Soil nominal resistance is modeled in SNAILZ using the Mohr-Coulomb failure envelope model. Another input parameter that must be included is the facing resistance. Input data can be entered in the English or SI unit systems. Use of Computer Programs for LRFD Method Analyses SNAILZ is ASD based; therefore, SNAILZ strictly provides cal- culated global factors of safety, FSG, for overall stability. The program cannot be used to perform an analysis using LRFD methodologies unless simplifying assumptions are made and intermediate calculations are performed. The user may manually input reduced values of nail tensile, pullout, and facing resistances (i.e., nominal values multiplied by the corresponding resistance factors) before the program executes any computations. The user must use the “pre-factored” option available in SNAILZ for reduced values of nail tensile, pullout, and facing resist- ances. By selecting this option, only soil parameters (cohesion and tangent of friction angle) are affected by FSG, while the other resistances remain constant throughout the analysis. External loads (i.e., two uniform loads available in SNAILZ) can be entered pre-multiplied by a resistance factor. Earth loads cannot be entered pre-multiplied by a resistance factor. When the “pre-factored” option is selected and factored val- ues for resistance are entered, SNAILZ can provide equivalent results in ASD format or in a format resembling LRFD. How- ever, this is limited to the condition of load factor γ = 1.0. When SNAILZ is used to perform an LRFD-equivalent analysis for γ = 1.0, factored values of the nominal resistances must be entered. For this step, the nominal resistances of soil cohesion, cs, and the friction angle, ϕs, must be affected by multiplying manually these values by soil resistance factors. Note that in SNAILZ, ϕs, not the tangent of the angle (tan ϕs) is input. There- fore, an equivalent reduced friction angle (ϕs red) is computed as ϕs red = tan−1 [tan (ϕs) × φ] and entered. With these factored nominal resistances entered, the condition FSG = 1.0 in SNAILZ would represent a limit state for global stability. The ASD and “LRFD” modes would be equivalent in SNAILZ only for γ = 1.0. If load factors different than 1.0 were used in the “LRFD” format in SNAILZ, inconsistent results between the ASD and LRFD “modes” would be obtained. In addition, affecting soil loads with load factors different than 1.0 is not possible in SNAILZ. For example, an attempt to affect the soil unit weight by earth load factors (in general > 1.0) would also affect earth load effects on the resistance side and would ultimately produce inconsistent results between the ASD and the LRFD-equivalent analyses in SNAILZ. In summary, the only practical way to use SNAILZ with a LRFD format is to set all load factors equal to 1.0, which is con- sistent with a service limit state for overall stability as is currently adopted in the LRFD Bridge Design Specifications (AASHTO, 2007). For, load factors > 1.0, inconsistent results are obtained. D.2.2 GOLDNAIL Basic Features GOLDNAIL is a Windows-based proprietary program developed by Golder Associates (Golder, 1993). Although GOLDNAIL is not as commonly used as SNAILZ, the program offers more advanced analysis capabilities and options that allow considering a wider range of scenarios and material prop- erties than SNAILZ. The program is commercialized and some technical support can be obtained for a fee. Program Capabilities This program is two-dimensional and satisfies moment and force equilibriums. GOLDNAIL uses circular failure surfaces and analyzes SNWs as a series of slices instead of wedges. In GOLDNAIL, the sliding soil mass is divided into vertical slices, like is typically done in most slope-stability methods. The pro- gram iteratively modifies the normal stresses distribution at the base of the slices until force and moment equilibriums are obtained. The program constrains circular slip surfaces to pass through or above the SNW toe. Input data can be entered in the English or SI unit systems, or any other compatible unit system. Sliding and basal heave cannot be assessed using this program. GOLDNAIL may also be used to analyze unreinforced slopes and anchored walls. Figure D-1. Slip surfaces used in SNAIL: (a) two wedges through toe and (b) two wedges under toe.

D-3 GOLDNAIL allows analyzing SNWs using either an ASD- equivalent method or the LRFD method. For each of these methods, the program works in one of the three following calculation modes: (i) Design Mode; (ii) Factor of Safety Mode; and (iii) Nail Service Load Mode. In the Design Mode, the pro- gram is executed by modifying some of the factors controlling stability (e.g., nail length) until a target safety factor (ASD method) is calculated or the limit condition (LRFD method) is met. In the Factor of Safety Mode, a global factor of safety, FSG, is calculated using the ASD method or the limit condition is met (LRFD method) for a specified set of input parameters, including soil nail length. In the Nail Service Load Design Mode, the program provides the maximum in-service tensile forces in the soil nails that are used for the design of the nail bar diameter and facing characteristics resistances. Input Parameters Nail and soil parameters are similar to those entered in SNAILZ with a few exceptions. The program can model up to 13 soil layers, complex slopes and subsurface geometries, hori- zontal and vertical surcharge distributions, groundwater, and pseudo-static, horizontal seismic coefficients. The program only considers uniform spacing and inclination of the nails. Although this scenario is typical for most designs, this assump- tion may be too restrictive for some cases. Soil strength is mod- eled using a linear Mohr-Coulomb envelope with the option of using a bi-linear strength envelope. Therefore, if the bi-linear Mohr-Coulomb model option is used, additional sets of cohe- sion and friction values are needed. In addition, the program allows the input of both vertical and horizontal surcharge loads. Use of Program in the LRFD Method For the LRFD method, GOLDNAIL allows the user to input load and resistance factors directly into the program, and there is no need to pre-calculate manually factored resistances. The user can input load factors separately for soil weight, water weight, surcharge and seismic load. Reduction strength factors (i.e., equivalent to the inverse of safety factors) are also entered for other resistance components (i.e., facing or nail head resist- ance, nail tensile resistance, and bond or pullout resistance). When the ASD method is used in GOLDNAIL, safety factors are entered separately for cohesion and friction. D.3 Comparison of LRFD- and ASD-Based Designs Designs of SNWs using the LRFD and ASD approaches are compared in two manners in this section. First, a parametric analysis was performed in GOLDNAIL in the ASD and LRFD modes. The objective of this analysis was to assess differences of key design parameters (i.e., nail length, cross sectional area, facing resistances) using the ASD and LRFD modes in the same software to avoid potential inconsistencies. Several wall conditions were inspected and various factors that may influ- ence results were considered. Second, designs of a LRFD design example using GOLDNAIL and SNAILZ (with modi- fied input to emulate a LRFD mode) were compared to the ASD-based design made for the same design example pre- sented in Lazarte et al. (2003). The comparisons are presented in the following subsections. D.4 Parametric Study D.4.1 Description The influence of several factors that may affect the required nail length was evaluated using the LRFD and the ASD in a parametric study. To facilitate the comparisons, a uniform nail pattern and homogeneous soil profile were assumed. A wall of height H (Figure D-2) is reinforced with six rows of nails (inclination of 15 degrees) of uniform length, L. The param- eters analyzed included the wall height, soil friction angle, nail bond resistance, and surcharge. All results were compared against the results of a baseline case, whose parameters are indicated on Figure D-2 and Table D-1. In these analyses, the pullout limit state was assured by selecting an artificially high H Baseline Case H = 30 ft, ϕs = 35 deg, qu = 15 psi, Q = 0 Other Cases Variable Analyzed Wall height, H = 20, 30, and 40 ft Soil friction angle, ϕs = 28, 32, 35, and 38 deg Bond resistance, qu = 10, 15, 20, and 25 psi Surcharge, Q = 0, 250, and 1,000 psf For all cases: c = 0, γs = 120 pcf Figure D-2. Geometry of SNW in comparative analyses.

D-4 nail yield resistance and facing resistance. However, this is not a typical manner of analyzing SNWs. Because the calibrated pullout resistance factors are values that are close to 0.5 (a value that would have been derived through a calibration with safety factors), the results between the LRFD and ASD methods are expected to be similar. D.4.2 Results Results for the over 30 analyzed design cases are summarized in Table D-2. Results confirm what was expected: using the reliability-calibrated resistance factors of Chapter 3, the calcu- lated nails length that are required to satisfy design criteria are comparable using both the LRFD and ASD methods. For all cases considered, the calculated nail length is, on average, approximately 4 percent larger in the LRFD method. No single factor appears to have a significantly greater influence on the results. Slightly larger differences were obtained for large loads and for high nominal pullout or bond resistances. The largest difference obtained for nail length was approximately 8 percent. The soil nail loads calculated via the ASD or LRFD modes were similar, with differences on average less than 3 percent. As a result, it is expected that the differences in calculating the necessary nail cross sectional area and required facing resist- ance would be almost identical using either the ASD or LRFD methods for γ = 1.0. D.5 Example Design of a SNW D.5.1 Design Conditions In this design example (Figure D-3), the soil profile behind the proposed SNW and the project requirements are similar to those of the design example presented in Appendix D of FHWA GEC No. 7 (Lazarte et al., 2003). The objective of this exercise is to compare the results obtained from the two most common SNW software programs in the design of a wall with realistic conditions. The wall conditions are as follows. A 10-m (33-ft) high SNW is to be constructed as part of a roadway project. The road adjacent to the proposed wall is of low-to-medium Description Quantity Wall height (ft) 30 Wall batter (deg) 0 Number of soil nail levels 6 Nail diameter (in.) 1.128 Diameter of grouted hole (in.) 6 Nail inclination (deg) 15 Nail vertical spacing (ft) 5 Nail horizontal spacing (ft) 5 Soil unit weight (pcf) 120 Soil friction angle (deg) 35 Bond stress (psi) 15 Table D-1. Soil nail wall input parameters (baseline case). Required Length, L (ft) ASD LRFD LRFD to ASD Percent Difference (w/ respect to baseline case, %) Variable Compared Case Variable Value FSG = 1.5 φPO = 0.49 φPO = 0.47 φPO = 0.49 φPO = 0.47 – Baseline H = 30 ft, ϕs = 35° qu = 15 psi, Q = 0 23.43 24.14 24.48 3.03 4.48 1 H = 40 ft 31.24 32.18 32.63 3.01 4.45 Wall Height 2 H = 20 ft 15.62 16.16 16.31 3.46 4.42 1 ϕs = 28 ° 27.59 28.43 28.99 3.04 5.07 2 ϕs = 32 ° 25.22 25.99 26.51 3.05 5.11 Friction Angle 3 ϕs = 38 ° 21.64 22.29 22.74 3.00 5.08 1 qu = 10 psi 26.28 27.59 28.42 4.98 8.14 2 qu = 20 psi 18.93 19.39 19.78 2.43 4.49 Bond Resistance 3 qu = 25 psi 17.14 17.67 17.83 3.09 4.03 1 Q = 250 psf 36.09 37.18 38.69 3.02 7.20 Surcharge 2 Q = 500 psf 40.67 41.91 43.61 3.05 7.23 Note: φPO were calibrated for a reliability factor of 2.33 Table D-2. Comparison of required nail length using ASD and LRFD approaches.

D-5 traffic volume and is considered non-critical. A 7.3-m (24-ft) wide road will be constructed 3 m (9.8 ft) behind the wall. The wall is to be constructed in medium-dense silty sand with clay seams with the soil nails shown in Figure D-4. The parameters used for the SNW design are as follows: A. Wall Layout Wall height, H = 10 m (33 ft); Wall length Wall Length >> H; and Face batter, α = 0. B. Soil Nail Vertical and Horizontal Spacing, SH = SV = 1.5 m (5 ft). C. Soil Nail Inclination, i i = 20 degrees (for top row of nails to avoid utilities); and i = 15 degrees (for other nail rows). D. Soil Nail Length Distribution The soil nail length is variable as indicated by length ratios ri (see Figure D-4) E. Nail Yield Tensile Resistance, fy = 520 MPa (75 ksi) F. Soil Properties and Ground Conditions 1. Upper Silty Sand Deposit: ϕs = 33 degrees c′ = cs = 0 (conservative for long-term design conditions) γs = 18 kN/m3 (115 pcf) 2. Lower Silty Sand Deposit: ϕs = 39 degrees c′ = cs = 0 γs = 20 kN/m3 (125 pcf) 3. Groundwater: absent. Future Roa d Zone with future util it ie s SP T N Va lu e (b lows /3 00 mm) Corr ec ted an d No rm al iz ed SP T N Valu e, N 1 (blow s/30 0 mm ) Proposed Nail So il Wal l Medium De ns e Si lt y Sa nd with Cla y Se am s (SM) γ s = 18 kN /m 3 ϕ s = 33 de gr ee s De nse Fi ne to Coarse Silty Sa nd (S W) γ s = 20 kN/ m 3 ϕ s = 39 degree s Ve ry De ns e Fine to Coa rs e Si lty Sand (SM) Source: Lazarte et al. (2003) Figure D-3. Subsurface stratigraphy and design cross section.

D-6 G. Drill-Hole Diameter, DDH = 150 mm (6 in.) H. Bond Resistance: Upper Silty Sand: qu = 100 kPa (14.5 psi); and Lower Silty Sand: qu = 150 kPa (21.8 psi). I. Load Combination and Load Resistance Factors The combination of loads for the project conditions is adopted from AASHTO (2007) recommendations. The load combination considered is Service Limit I. The load combinations and load factors based on AASHTO (2007) recommendations are γ = 1.0. J. Facing Features See Table D-3. For a mesh 152 × 152 − MW19 × MW19 (6 × 6 − W2.9 × W2.9 mesh in English units) and using Table A.2 of Lazarte et al. (2003), the total reinforcement area per unit length at midspan is: At the nail, there are two No. 13 (No. 4) vertical and hori- zontal (waler) bars. Using Table A.3 of Lazarte et al. (2003), the total nominal area in each direction is: The total reinforcement area per unit length around the nails is: The reinforcement ratio at the nail head and at the mid- span, and the total ratio are calculated as: ρ ρ ρ n vn hn m vm TOT a b h a b h a b h a = × = × = × = 2 100 2 100 2 100 vn vm TOT a b h +( ) × = +( ) × × = 2 100 295 123 1 000 50 100ρ , 0 84. % a a A S a vn vm VW H vn = + = = × + = = 123 1 5 258 1 5 295 2 . . . mm m2 95 10 0 144× ( )− m m in. ft2 2. A AVW HW= = × = ( )2 129 258 0 4mm in.2 2. a avm hm= = = × ( )−123 1 23 10 0 0584mm m m m in ft2 2 2. . Elemen t Description Temporary Facing Permanent Facing Thickness ( h ) 100 mm (4 in.) 200 mm (8 in.) Facing Type Shotcrete CIP Concrete General Comp. Strength, f c 21 MPa (3,000 psi) 28 MPa (4,000 psi) Ty pe WWM Steel Bars Mesh Grade 420 (Grade 60) 420 (Grade 60) Reinforcement Denomination 152 × 152 MW 19 × MW 19 (6 × 6 - W2.9 × W2.9) No. 13 @ 300 mm (each way) [No. 4 @ 12 in. (each way)] Other Reinf. Ty pe Waler Bars 2 × 13 mm (2 × #8) – Ty pe 4 Headed-Studs 1 / 2 × 4 1 / 8 – Steel 250 MPa (Grade 420) – Length; L P = 225 mm (9 in.) – Bearing Plate Dimensions Thickness: t P = 25 mm (1 in.) – – Nominal Length: L s = 105 mm (4 in.) – Head Diameter: D H = 25.4 mm (1 in.) – Shaft Diameter: D S = 12.7 mm ( 1 / 2 in.) – Head Thickness: t H = 7.9 mm (0.3 in.) Headed-Studs Dimensions – Spacing: S S = 150 mm (6 in.) Table D-3. Facing features. δ = 20o δ = 15o r6,7 = 0.5 r4, 5 = 0.7 r1,2, 3 = 1 SV0 = 0.5 m SV = 1.5 m SVN = 0.5 m H = 10 m Nail 1 2 3 4 5 6 SM SW 7 L1 = length of upper nail row ri = Li/L1 = length ratio for nail "i" p = 12 kPa (250 psf) Source: Lazarte et al. (2003) Figure D-4. Non-uniform nail length pattern.

D-7 Resistance Factor Value Soil Shear Resistance φs 0.65 Nail Pullout Resistance, φPO 0.49 Nail Tendon Resistance, φT 0.56 Nail Head Resistance (flexure and punching shear), φFF (controls) 0.67 Nail Head Resistance (headed-stud in tension), φFH 0.50 Table D-4. Resistance factors for overall stability. D.5.2 Design Procedures Based on the recommendation of AASHTO (2007), the overall stability of the SNW is assessed using the load combi- nation for service limit state. The resistance factors in Table D-4 are used based on the recommendation of AASHTO (2007). For the pullout resistance factor, the values calibrated in Chapter 3 for sand are used. The overall stability of the SNW system is evaluated using SNW design software. The following are examples of design analysis results obtained using the programs SNAILZ and GOLDNAIL. GOLDNAIL 1) Define geometry of wall. Trial nail lengths are selected as follows: Nail Layers Trial Nail Length (ft) 1 through 3 30 4 and 5 21 6 and 7 15 Due to the limitations in the program, the nail inclination of nail layer #1 is selected to be the same as other layers (i.e., 15° instead of 20° as shown in Figure D-4). 2) Input the following parameters: To ensure that pullout failure controls over tensile or punching-shear failure, artificially large values of nail diam- eter and nail head resistance can be entered in GOLDNAIL. For consistency with the example in GEC No. 7, the follow- ing nail bar and head resistances are selected: • Threaded bar: No. 8, 25 mm diam., cross-sect. area = 510 mm2 (0.79 in.2) • Nail nomin. tensile resist. = 0.79 in.2 × 75 ksi = 59.3 kips • Nail head nominal resistance (for permanent facing) = 92 kip, from page D-28 (Lazarte et al., 2003) • Nail pullout nominal resistance (per linear ft): Upper silty sand: π × 6 in. × 1 ft × 12 in./ft × 14.5 psi = 3,280 lbs; and Lower silty sand: π × 6 in. × 1 ft × 12 in./ft × 21.8 psi = 4,931 lbs. • Soil Design Parameters: Upper Silty Sand Deposit: ϕs =33 degrees c′ =0 γs =115 pcf Lower Silty Sand Deposit: ϕs =39 degrees c′ =0 γs =125 pcf 3) Input the following load and resistance factors: In the safety factor screen, select LRFD mode. The load factors for water weight, soil weight surcharge, and seismic load are all selected to be 1.0. Input the resistance factors as shown in Table D-4. 4) Compute the necessary nail length and head resistance: In GOLDNAIL, run analysis using the design analysis mode. The required nail lengths to achieve a resistance-to- load ratio greater than 1.0 are calculated. Nail Layers Required Nail Length (ft) 1 through 3 32.7 4 and 5 22.9 6 and 7 16.3 The maximum force occurs in the lowermost nail at 32,070 kip (as obtained from the nail service mode in GOLDNAIL). Figure D-5 shows the calculated critical failure surface. SNAILZ In order to perform an analysis that resembles the LRFD for- mat in SNAILZ, resistances must be modified. Note that in this example, the service limit state is analyzed and all load factors are equal to 1.0. Below is a summary of the modified input parameters using a FS = 1.0 (i.e., an equivalent of the LRFD): Upper Silty Sand Deposit: ϕs = tan−1(0.65 tan33°) = 22.9° c′ = 0 × 0.65 = 0 γs = 115 × 1.0= 115 pcf qu = 14.5 psf (nominal value) BSF = 0.49 (Bond Stress Factor, equivalent to pullout resistance factor) q = 14.5 psf × 0.49 = 7.11 psf (factored value)

D-8 Lower Silty Sand Deposit: ϕs = tan−1(0.65 tan39°) = 27.8° c′ = 0 × 0.65 = 0 γs = 125 × 1.0 = 125 pcf qu = 21.8 psf (nominal value) BSF = 0.49 (Bond Stress Factor) q = 21.8 psf × 0.49 = 10.7 psf (factored value) Nail Head and Nail Tensile Resistances: Facing resistance = 92 (nominal) × 0.67 = 61.3 (kips); and Tensile resistance (force) = 59.3 (nominal) × 0.56 = 32.9 (ksi). Tensile resistance (stress) = 75 (nominal) × 0.56 = 41.7 (ksi). Nail lengths need to be computed in SNAILZ iteratively in different runs until a target factor of safety of 1.0 (i.e., a con- dition equivalent to the limit state) is achieved. Figure D-6 shows the critical failure surface calculated by SNAILZ. The required nail lengths as calculated with this procedure are listed below. Nail Layers Required Nail Length (ft) 1 through 3 34.1 4 and 5 23.9 6 and 7 17 The maximum calculated nail force is 32.7 kip (in the low- ermost nail). Note that these values are almost identical to those obtained using the ASD method according to GEC 7 (and using the pre- factored mode in SNAILZ). The comparison indicates that SNAILZ requires nails that are approximately 4 percent longer than those obtained using GOLDNAIL. The maximum nail forces in SNAILZ are approximately 2 percent larger than with GOLDNAIL. D.6 Discussion of Results Comparative analyses show that both the LRFD and ASD method provide comparable design values for soil nail walls under various conditions. Overall, the comparisons indicate that the required soil nail length calculated using the LRFD method and the proposed resistance factors are compara- ble with those obtained with the ASD method. For all cases considered, the length difference is on average approxi- mately 4 percent larger in the LRFD method. No factor appears to have greater influence than others do. Slightly larger differences were obtained for large loads and for high nominal pullout or bond resistances. The largest difference obtained in the comparative analysis was approximately 8 percent. In all cases, soil-nail loads calculated using either method are comparable, with a difference of less than about 3 percent. The analyses using the LRFD method with SNAILZ and GOLDNAIL show that the differences and nail loads are very small, 4 and 2 percent, respectively. Figure D-5. Critical failure surface calculated using GOLDNAIL.

D-9 D.7 Summary The comparative analyses confirm that the calculated quan- tities, including soil nail lengths and cross-sectional areas (as a function of the maximum soil nail force), as obtained using the LRFD and ASD methods are very similar. The reason for these similar trends, which were already apparent in Chapter 3, stem from the fact that the calibrated resistance factors for pullout are very similar to those that could have been obtained directly from a calibration using factors of safety. The differences were small between LRFD and ASD methods using the same pro- gram (i.e., GOLDNAIL) and between different programs using LRFD and ASD methods. Therefore, the calibration and com- parison demonstrate that the parameters currently used in practice should not be altered. Adopting the LRFD method and the calibrated resistance factors used herein would only result in a change of design format. However, the design would result in essentially the same quantities. A limitation of these compar- isons is that analyses have been performed for load factors equal to 1.0, per the current AASHTO LRFD practice of overall sta- bility. However, it is expected that slightly different results and design quantities would be obtained for conditions other than load factors = 1.0. References AASHTO (2007). LRFD Bridge Design Specifications, 4th Edition, Amer- ican Association of State Highway and Transportation Officials, Washington, D.C. Caltrans (1991). “A User’s Manual for the SNAIL Program, Version 2.02—Updated PC Version.” Division of New Technology, Material and Research, Office of Geotechnical Engineering, California Department of Transportation, Sacramento, California. Caltrans (2007). “A User’s Manual for the SNAILZ Program, Version 2.02—Updated PC Version.” Division of New Technology, Mate- rial and Research, Office of Geotechnical Engineering, California Department of Transportation, Sacramento, California. http:// www.dot.ca.gov/hq/esc/geotech Golder (1993). “GOLDNAIL Soil Nailing Design Program.” Golder Associates, Seattle, Washington. Lazarte, C. A., V. Elias, R. D. Espinoza, and P. J. Sabatini (2003). “Soil Nail Walls.” Geotechnical Engineering Circular No. 7, No. FHWA- IF-03-017, Federal Highway Administration, Washington, D.C. Figure D-6. Critical failure surface calculated using SNAILZ.