Nh lie Tice ae ie ee a | PROCEDURE FOR RAILWAY TRACK GRANULAR LAYER THICKNESS DETERMINATION REPORT NO. R-898 i>) Research and Test DepartmentSussmann PROCEDURE FOR RAILWAY TRACK GRANULAR LAYER THICKNESS DETERMINATION REPORT NO. R-898 By Dingging Li Theodore R. Sussmann, Jr. and Ernest T. Selig Association of American Railroads Transportation Technology Center Pueblo, Colorado October 1996DISCLAIMER This report is disseminated by the Association of American Railroads (AAR) for informational purposes only and is given to, and is accepted by, the recipient at the recipient's sole risk. The AAR makes no representation or warranties, either expressed or implied, with respect to this report or its contents. The AAR assumes no liability to anyone for special, collateral, exemplary, indirect, incidental, consequential, or any other kind of damages resulting from the use or application of this report or its contents. Any attempt to apply the information contained in this report is made at the recipient's own tisk. Copyright© 1996 by ASSOCIATION OF AMERICAN RAILROADS, All rights reserved. No part of this publication may be copied or distributed, transmitted, transcribed, stored in a retrieval ‘system, or translated in any language, in any form or by any means, electronic, mechanical, magnetic, manual or otherwise, or disclosed to third parties without the express written permission of the Association of American Railroads,1. Report No. 2. Report Dato 3.Period Covered R898 October 1996 1994 to 1995 4.Tite and Subtitio. Procedure for Railway Track Granular Layer Thickness Determination 5. Authors D.Li, T.R.SussmannJr.* and E. T. Selig* 6. Performing Organization Name and Address ‘T.Type of Report Association of American Railroads Research ‘Transportation Technology Center P.O. Box 11130 Pueblo, CO 81001 ‘8. Contractor Grant No. ‘Sponsoring Agency Name and Addross 10. No. of Pages. 2 Association of American Railroads ‘Transportation Technology Center P.O. Box 11130 11, Number of References Pueblo, CO 81001 il 42, Supplementary Notes “*Department of Civil Engineering, University of Massachusetts, Amherst, MA 13. Abstract ‘The design methodology for railway track granular layer thickness has been developed to prevent subgrade failures under} repeated dynamic traffic loads. This report discusses the design procedures and gives the design charts for various ballast and subgrade soil conditions. The methodology is believed to provide the best approach currently available and design results have been consistent with field results. 17. Subject Torms 48. Avalability Statement Granular layer thickness design Association of American Railroads: Subgrade failures Publication Order Processing Repeated dynamic traffic loads COG, Sth Floor 50 F Street, NW Washington, D.C. 20001EXECUTIVE SUMMARY Under the Association of American Railroads’ (AAR) Heavy Axle Load (HAL) Implementation program, design methodology and design charts have been developed to determine the granular layer thickness required to prevent railway subgrade failures under repeated traffic loads. This methodology, along with the AAR’s newly developed cone penetrometer, is intended to reduce the maintenance costs associated with the operation of HAL traffic. The design methodology is believed to provide the best approach currently available and the results are consistent with the limited available field data. However, further verification of the design methods by field data and experiments is desirable. The main design parameters considered in the design methods include: Dynamic wheel loads, - Tonnage by million gross tons (MGT), + Moduli of granular materials and subgrade soil, and + Subgrade soil type and soil compressive strength. Proper granular layer thickness is required to distribute and reduce the traffic load induced stresses to the subgrade. With the growing trends of railroads towards heavy axle loads and higher train speeds, higher dynamic wheel loads will be exerted on the track structure. Without sufficient granular layer thickness, the subgrade will experience higher repeated stresses, which in turn may lead to subgrade failures. On the other hand, redundant ballast thickness will lead to higher costs. Granular layer thickness is defined as the combined thickness of ballast and subballast between the subgrade surface and the tie bottom. The subgrade failures to be prevented or remedied are subgrade progressive shear failure (subgradesqueezing) and excessive plastic deformation (ballast pocket). These two types of subgrade failures are the most common soft subgrade failures under repeated heavy axle loads. Other types of subgrade problems such as subgrade mud pumping are not covered by the design methods. To obtain satisfactory track performance, adequate track drainage must also be provided with the proper granular layer thickness.TABLE OF CONTENTS 1.0 Introduction... 00... eee eee ean eters cee sel 2.0 Background . . 3.0 Design Traffic 4.0 Properties of Materials 4.1 Granular Layer . 4.2 Subgrade 5.0 Design Criteria . 5.1 Criterion 1. 5.2 Criterion 2 .. 6.0 Design Procedures 6.1 Method 1 . : 6.2 Method 2 ....... 7.0 Design Examples... 7.4 Method 4 7.2 Method 2. 8.0 Comparisons with Field Results . 8.1 Low Track Modulus Test Track ...... Fs 8.2 Burlington Northern Hot Mix Asphait Test Site 8.3 AMTRAK Sites Locvveteeeeeees 9.0 Summary and Recommended Future Studies ................-5 . 28 ‘Acknowledgments 30 References 31 APPENDIX A: Charts for Design Method 1 APPENDIX B: Charts for Design Method 2TABLE OF EXHIBITS Exhibit 1. Characteristics of Typical Cars and Locomotives in North America ....... 6 Exhibit 2. Suggested Values of Soil Parameters a, b, and m for Various Soil Types .............. 0202202005 mee eeee eens 7 Exhibit 3. Depth Measurement Definitions . . eres hae Re 8 see 8 Exhibit 4. Material Properties of Granular Material and Subgrade Soil ............ 9 Exhibit 5. Subgrade Progressive Shear Failure - 10 Exhibit 6. Excessive Subgrade Plastic Deformation ....... eet ent Exhibit 7. Design Flow Chart for Granular Layer Thickness - Design Method 1..... 14 Exhibit 8. Design Flow Chart for Granular Layer Thickness - Design Method 2..... 16 Exhibit 9. Example Granular Layer Thickness Design ............... seeeeeee WT Exhibit 10. Design Results for TLTM and LTM Track Granular Layer Thickness .... 20 Exhibit 11. Comparison of Profile Degradations ..... 3 Rss Kama aR a Hee peeenen 22} Exhibit 12. Design Results of Track Granular Layer Thickness For BN HMA Site ... 24 Exhibit 13. Track Roughness Developments ..... 2.2.2.2... 0.0. ecceeeeee eee 25 Exhibit 14. Traffic parameters for AMTRAK Sites . 27 Exhibit 15. Evaluation of Track Granular layer Thickness for AMTRAK Sites ...... 284.0 INTRODUCTION Proper granular layer thickness is required to reduce the traffic load induced stresses to the subgrade. With growing trends of railroads towards heavy axle loads and higher train speeds, higher dynamic wheel loads will be exerted on the track structure. Without sufficient granular layer thickness, the subgrade will experience higher repeated stresses, which in turn may lead to excessive subgrade deformation and subgrade failures. On the other hand, redundant ballast thickness will lead to high costs. Under the Association of American Railroads’ (AAR) Heavy Axle Load (HAL) Implementation program, design methods and design charts have been developed to determine the granular layer thickness required to prevent or remedy railroad subgrade failures under repeated traffic loads. Granular layer thickness is defined as the combined thickness of ballast and subballast layers between the subgrade surface and the tie bottom. The subgrade failures to be prevented or remedied are progressive shear failure (subgrade squeezing) and excessive plastic deformation (ballast pocket). These two types of subgrade problems are the most common soft subgrade failures under heavy axle loads.? The design procedures and design charts given in this report are applied to conventional, ballasted track structure. The main design parameters considered in the design methods include dynamic wheel loads, tonnage by million gross tons, moduli of granular materials and subgrade soil, and subgrade soil type and compressive strength. The design methodology presented in this report is believed to provide the best approach currently available. Results are consistent with the limited field data, illustrated later in this report. However, further verification and modification by field practice are highly desirable.2.0 BACKGROUND Operations of heavy axle loads (HAL) will result in high stresses in track and subgrade. The high stresses in the subgrade, if exceeding subgrade soil strength, will in turn lead to rapid subgrade deformation or subgrade failures. As discussed in a separate AAR report,’ some tracks built on soft to medium subgrades have already experienced problems such as progressive shear failure (subgrade squeezing) and excessive plastic deformation (ballast pocket). The number of problem track locations with high subgrade deformation rates would inevitably increase due to ever increasing axle loads and train speeds, if no measures are taken to reduce subgrade stresses or to improve subgrade strength. ‘There are several approaches to reducing subgrade stresses, including reducing tie spacing, increasing rail weight, and using a hot mix asphalt (HMA) layer above the subballast. However, increasing granular layer thickness is likely to be the most effective and economical approach.’ Presently, the recommended ballast thickness for main line tracks of most North American railroads varies from 6 to 12 inches and from 4 to 12 inches for the subballast thickness. However, based upon economic considerations, a uniform granular layer thickness for many miles of tracks may never accommodate train induced dynamic stresses for locations with soft to medium subgrade supports. On the other hand, a uniform granular layer thickness for many miles of track based upon the worst subgrade conditions may not be cost effective. To design a minimum granular layer thickness for preventing excessive subgrade deformation and failures, a number of design methods with various limitations have been provided in the literature. For example, of several equations recommended by the manual of the American Railway Engineering Association (AREA) the following classical Talbot Equation is used more frequently: a 16.8P,where +H = granular layer thickness (in.), P, = allowable subgrade pressure (psi), with an allowable subgrade pressure of 20 psi recommended by AREA, and P,, = vertical stress applied on the ballast surface (psi), with a maximum ballast stress of 65 psi or 85 psi (often assumed for wood tie or concrete tie tracks). The above empirical equation was developed based on field tests conducted during the 1910s and 1920s. Its application, however, represents an over-simplified situation for tracks under heavier axle loads and higher train speeds. Subgrade soil conditions, effects of repeated dynamic loads, and ballast layer quality are not reflected at all in this equation. Because subgrades actually vary widely in strength, using a universal allowable subgrade pressure of 20 psi will lead to difficulty in soft soil conditions and will be too conservative in strong soil conditions. In the 1970's, British Railways developed a “threshold stress” design method for selecting granular layer thickness.’ This method was recognized worldwide as the most rational. It’s objective is to limit the stress on the subgrade soils to less than a “threshold” stress in order to protect against subgrade failure by excessive plastic deformation. The threshold stress is determined from repeated load tests in which the cumulative strain of the soil is noted as a function of the number of loading cycles applied. At applied stress levels above the threshold stress level, the rate of cumulative plastic deformation of the soil will be extremely rapid. At applied stress levels below the threshold stress level, the rate of the cumulative deformation is small. The threshold stress (or threshold strength) for cohesive subgrade soils is usually lower than soil static strength. In practice, the threshold strength is often assumed to be equal to 50 percent of the soil static strength. Based on the subgrade soil “threshold stress” concept, design charts have been developed by British Railways for selecting granular layer thickness for various subgrade soil conditions and axle loads.In considering the effect of repeated stresses on subgrade soil strength, the “threshold stress” design method was the most rational. The design charts, however, were developed using a track model which neglected the effect of a much higher stiffness of the top granular layer. The track model assumed a homogeneous half space for all ballast, subballast and subgrade layers. As a result, the design often gives a larger required granular layer thickness because of the higher calculated stresses in the subgrade. Another important limitation with the “threshold stress” method is that the design method uses only the single maximum axle load as the design load, without considering the effect of cumulative tonnage. In other words, if the maximum axle loads are the same for both tracks, a track with an annual tonnage of 10 MGT (million gross tons) would be designed with the same granular layer thickness as a track with an annual tonnage of 100 MGT. The design methods and design charts presented in this report were developed based upon two primary foundations: a validated track model GEOTRACK® for stress calculation in the track arid subgrade due to dynamic wheel loads, and extensive test results of subgrade soils under repeated loadings. GEOTRACK is an elastic, multi-layer and three-dimensional model for analyzing stresses and deformation in the rails, ties, ballast, subballast and subgrade layers under multi-axle loads. GEOTRACK was used. in the development of design charts for the stress analysis in the subgrade under multi- axle loads. Since a track subgrade is always subjected to repeated stress applications, the findings regarding subgrade failures and deformation based on extensive test results ‘were used to characterize the influence of soil type, soil physical state (such as moisture content), and repeated stresses on subgrade soil strength. In this report, however, only final design procedures and design charts are given. The underlying theory and extensive subgrade soil test results used to characterize subgrade soil strengths can be found in References (3) and (4), The design methods and design charts presented in this report can be used to design new tracks as well as to remediate existing tracks with subgrade deformation problems. However, when the subgrade soil is too soft, there is a limitation of using large granular 4layer thickness. As shown by design charts listed in the Appendices for clayey subgrade soils, when subgrade soil strength becomes very low, a slight decrease in soil strength may require a very large increase in granular layer thickness. When a subgrade is too soft, the more effective and economical method for long term remedy is to improve subgrade soil strength although increasing granular layer thickness may provide a short- term remedy (unfortunately it has been the practice often adopted in track maintenance). Otherwise, a subgrade problem known as “ballast pocket” will develop as more ballast is repeatedly added to accommodate a too soft subgrade support. Because a ballast pocket traps water, the situation will often perpetuate itself. Methods for improving subgrade soil strengths can be found in References (1) and (2), among others. As always, a proper track drainage system should be included with an adequate granular layer thickness design. Poor drainage contributes to a very soft subgrade condition under which even large granular layer thickness will not provide a ong term remedy. It may also result in higher subgrade stresses, which will in turn worsen subgrade conditions’ This report will not address track drainage system design. More information can be found in References (1) and (2). 3.0 DESIGN TRAFFIC The design methodology emphasizes the influence of repeated traffic loads on subgrade performance. Thus, the design traffic parameters are: + static wheel loads, + train speed, and traffic as represented by MGT (million gross tons). These parameters are converted into two design variables: the design dynamic wheel load, P,, and the total number of repeated load applications, N, for the design period. The design period is defined as the time for the track to carry the desired amount of traffic without having the subgrade strength influence the track performancebeyond a degree specified. This specified degree should be chosen based on maintenance costs and traffic speed restriction considerations. A dynamic wheel load, Pz, corresponding to a static wheel load, Py, is calculated by the AREA recommended equation as follows:* Py = (1+ 258Yyp, 2 where V= train speed (mph), and D = wheel diameter (inches). Characteristics of typical freight cars and locomotives in North America are given in Exhibit 1. Exhibit 1. Characteristics of Typical Cars and Locomotives in North America Gross Rail Load Wheel Nominal Static | Axles Axle (Car Name) Diameter Wheel Load per Spacing (in) (ibs) Vehicle |__(in.) 315,000 Ib (125-ton) 38 39,000 4 72 286,000 Ib (110-ton) 36 36,000 4 70 263,000 Ib (100-ton) 36 33,000 4 70 225,000 Ib (70-ton) 33, 28,000 4 68 6-axle Locomotive 40 31,000 6 134 4-axle Locomotive 40 32,000 4 112 The number of repeated applications of any wheel load,P,, during the design period is determined by the following equation (assuming 4 axle passes under two adjacent bogies from two adjacent cars as one loading cycle for the subgrade):“w, ® where T, = total traffic tonnage for the wheel load, P,, for the design period in the same unit as Py. To represent the influence of all levels of wheel load on subgrade performance, the following equation is used to convert N, cycles of wheel load, Py, to N? cycles of the design wheel load,P,:° py NP = Nil @ Py where m, b = parameters dependent upon soil type, as given in Exhibit 2. Exhibit 2. Suggested Values of Soil Parameters a, b, and m for Various Soil Types Soil Type at b m CH (fat clay) 12 0.18 2.4 CL (lean clay) 14 0.16 2.0 MH (elastic silt) 0.84 0.13 2.0 ML (silt) 0.64 0.10 17 * “a” values are needed in the later sections. Thus, the total number of load applications, N, for the design load, Py, is calculated by: N= NP +. +NP+ (5)4.0 PROPERTIES OF MATERIALS Exhibit 3 shows the simplified granular layer overlying the subgrade. The subgrade is simplified as one layer overlying a rigid layer. The granular layer thickness, H, includes both ballast and subballast layers. The material property considered for the granular layer is resilient modulus, E,,. The material properties considered for the subgrade layer include resilient modulus, E,, soil compressive strength, 0,, and soil type. 4.1 GRANULAR LAYER For simplicity, the ballast and subballast layers are combined as one layer, which is referred to as the granular layer. Analysis by Li’ showed that the material property of the granular layer, which influences stress level in the subgrade most, is the resilient modulus of this layer. A stiffer granular layer will result in a lower stress level in the subgrade caused by repeated traffic loading. Granular layer (B) Deformable subgrade layer (©, E,, soil type) Rigid layer Exhibit 3. Depth Measurement Definitions Many factors affect the resilient modulus of the granular layer, including the ballast type, degree of fouling, stress state, and compaction level of the granular layer. In Exhibit 4, some typical values of resilient modulus for the granular layer are given corresponding to the quality of the ballast and subballast layers." 8Exhibit 4, Material Properties of Granular Material and Subgrade Soil Waterial Condition | lon Modulus Compaseslis Granular Material Good 80,000 - Intermediate 40,000 = Poor 20,000 * ‘Subgrade_ ‘Stiff 10,000 - 20,000 ‘30-50 ‘Medium ‘4,000 - 10,000 15-30 Soft 1,000 - 4,000. 5-15 4.2 SUBGRADE The subgrade is simplified as one deformable layer overlying a rigid layer, as illustrated in Exhibit 3. The material properties considered for the deformable subgrade layer are resilient modulus, soil compressive strength, and soil type. In general, other conditions being the same, a stiffer subgrade will result in a higher stress level inthe subgrade. However, this higher level of stress will be compensated for by the corresponding higher strength of the subgrade soil. For a stiffer subgrade, the increase in soil strength is generally greater than the increase in subgrade stress level. Exhibit 4, above, gives the ranges of resilient modulus and soil compressive strength for various subgrade soil conditions.’ 5.0 DESIGN CRITERIA Two criteria are used to design the granular layer thickness. One criterion is intended to prevent subgrade progressive shear failure and the other criterion is intended to prevent excessive subgrade plastic deformation. A detailed explanation of subgrade progressive shear failure and excessive subgrade plastic deformation is given in a separate AAR report? Both criteria should be evaluated to determine the one that gives the largest granular layer thickness in each case.In this report, the procedure based on the criterion for preventing subgrade progressive shear failure is Design Method 1, and the procedure based on the criterion for preventing excessive subgrade plastic deformation is Design Method 2. 5.1 CRITERION 4 Exhibit 5 illustrates the development of subgrade progressive shear failure (subgrade squeezing). For a soft subgrade, subgrade progressive shear failure begins at the subgrade surface and gradually develops to a greater extent under repeated loading. The causes and influence on track performance of this type of subgrade failure are described in Reference (2). Exhibit 5. Subgrade Progressive Shear Failure The design criterion for preventing subgrade progressive shear failure is intended to limit the total cumulative plastic strain at the subgrade surface to an allowable level for the design period. The criterion is expressed by the following equation: & < Boe © where €, = total cumulative plastic strain at the subgrade surface for the design period, and allowable plastic strain at the subgrade surface for the design period. 105.2 CRITERION 2 Exhibit 6 illustrates subgrade excessive plastic deformation leading to ballast pockets. Detailed discussion of the causes and influence on track performance of this type of subgrade failure is given in Reference (2). To prevent subgrade excessive plastic deformation, the design criterion is to limit the total cumulative plastic deformation of the deformable subgrade layer to an allowable level for the design period. This criterion is expressed by the following equation: PsP, O) where p = total cumulative plastic deformation of the subgrade layer for the design period, and p, = allowable plastic deformation of the subgrade layer for the design period. (2) Longiudinal Seetion Exhibit 6, Excessive Subgrade Plastic Deformation "6.0 DESIGN PROCEDURES This section describes the design procedures for the two different design criteria. The charts developed for Design Methods 1 and 2 are given in Appendices A and B, respectively. 6.1. METHOD 4 Exhibit 7 shows the flow chart for determining granular layer thickness based on the criterion of limiting cumulative plastic strain at the subgrade surface. The procedure consists of the following three steps: (1) Prepare the information required for the design, including: Traffic condition - Dynamic wheel load must be determined using Equation 2, and the number of load repetitions for the design period must be determined using Equation 3. If there are several major groups of wheel load magnitudes, the corresponding dynamic wheel loads and the numbers of load repetitions should be determined separately. Then Equations 4 and 5 can be used to determine the total equivalent number of load repetitions, N, for the design dynamic wheel load, P, Design criterion - The magnitude of the allowable cumulative plastic strain at the subgrade surface, €,,, must be determined for a certain number of load repetitions (i., for the design period), Subgrade characteristics - This requires the determination of subgrade soil type, soil static compressive strength, 0,, and soil resilient modulus, E,. Granular material - This requires the determination of resilient modulus, E, for the granular material. (2) Determine the allowable deviator stress at the subgrade surface as follows: Select the portion of Chart A1 in Appendix A corresponding to the subgrade soil type from Step 1. orLocate the point on that chart corresponding to the values of €,, and N from Step 1. Obtain the value of B for this point. Multiply B by the soil strength, 0,, from Step 1 to obtain the allowable deviator stress, Oy. (8) Select the required granular layer thickness to prevent subgrade progressive shear failure as follows: Calculate the strain influence factor, J, by yA tp @) where gq = allowable deviator stress at the subgrade surface determined from Step 2, P,, = design dynamic wheel load determined from Step 1, and A dimensionless. The area factor A is 1000 square inches when the area factor used to make the strain influence factor English unit is used, or 0.645 square meter when the metric unit is used. Select Chart A2, A3, or Ad in Appendix A for which the E, corresponds most closely to the value of granular layer resilient modulus selected in Step 1. Locate the point on that chart corresponding to /, and the subgrade resilient modulus from Step 1. Obtain the value of H/L for that point. Multiply H/L by the length factor L to get the required granular layer thickness, H. The length factor is used to make the design charts dimensionless. L is equal to 6 inches for the English unit or 0.152 meter for the metric unit. BChart Al in Appendix A can be skipped if the allowable deviator stress at the subgrade surface is directly selected by some means without using the allowable cumulative plastic strain at the subgrade surface for the design period. In this case, proceed to Step 3. Step 2 ‘Granular Li ‘Thickness, H by Chart AT Allowable cuales Traffic Conditions Exhibit 7. Design Flow Chart for Granular Layer Thickneee - Design Method 1 “46.2 METHOD 2 Exhibit 8 shows the flow chart for the procedure based on the criterion which limits total plastic deformation of the subgrade layer. Again, the granular layer thickness design consists of the following three steps: (1) Prepare all the information required for the design: In addition to the information required in Design Method 1, Design Method 2 requires knowledge of the thickness of the deformable subgrade layer, T. The allowable cumulative plastic strain at the subgrade surface for Design Method 1 is replaced by the allowable total plastic deformation of the subgrade layer for Design Method 2. (2) Calculate the deformation influence factor, /,, by the following equation: a{ P+)" yo 0,A where —_p, = allowable total subgrade plastic deformation for the design period, tal equivalent number of load repetitions during the design period, = design dynamic wheel load, soil compressive strength, a,m, b = material parameters (see Exhibit 2), and x100 0) A,L = length factor and area factor, as defined previously. (8) Select the required granular layer thickness to prevent excessive subgrade plastic deformation as follows: Select the design chart in Appendix B which best corresponds to the proper soil type, subgrade resilient modulus and granular layer resilient modulus. 15Calculate T/L. Locate the point on that chart corresponding to J, and T/L. Obtain the value of H/L for that point. Multiply H/L by the length factor, L, to get granular layer thickness, H. Granular Material Resilient Modus, I =a Deformation inuenee Facto, tp Exhibit 8. Design Flow Chart for Granular Layer ‘Thickness - Design Method 27.0 DESIGN EXAMPLES The design procedures ‘or Design Methods 1 and 2 are illustrated in this section using an example of a test track built on a soft subgrade under repeated heavy axle loadings, i.e,, the example of the AAR’s Low Track Modulus (LTM) test track. Exhibit 9 gives the information needed for the granular layer thickness design. Exhibit 9. Example Granular Layer Thickness Design Design Parameters Values Remarks Traffic conditions | Static wheel load P, = 39 kips FAST HAL train Train speed V = 40 mph consist Tonnage = 60 MGT (ie., design period) Wheel diameter D = 38 in ‘Subgrade soil Soil fype: CH clay (a= 1.2,m=24,andb= | Mississippi characteristics 0.18 from Table 2) Buckshot clay Compressive strength a, = 13 psi Resilient modulus E, = 2000 psi ‘Subgrade thickness T = 5 ft Granular material _| Resilient modulus E, = 40,000 psi Intermediate Design criteria Allowable cumulative plastic strain Based on = 2 percent preliminary LTM Or, allowable total plastic deformation: track performance 9,51 inch results 7.1 METHOD 1 Step 1: The first step is to obtain the information needed for the design, i.e,, traffic condition, subgrade soil characteristics, granular material characteristics, and design criterion (See Exhibit 9). The dynamic wheel load is determined by Equation 2 fo be: Py = 39 (1 + 0.33x40/38) = 53 (kips); and the equivalent number of load repetitions is determined by Equation 3 to be: N = 60x1,000,000x2,000/8/39,000 = 380,000. v7Step 2: For the allowable cumulative plastic strain, ¢,, = 2 percent, and N = 380,000, the allowable B (0,,/0,) is found to be 0.52, from design chart, Al. Thus, the allowable 0.52x 0, = 0.52x13 = 6.8 (psi). deviator stress, 04 Again, if the allowable deviator stress at the subgrade surface can be determined directly, skip Steps 1 and 2 and proceed to Step 3. Step 3: The strain influence factor is determined by Equation 8 to be: |, = 6.8x1000/53000=0.128. From design Chart A3, the required H/L is found to be 4.3 for E, = 40,000 psi, E, = 2000 psi, and |, = 0.128. Therefore the required granular layer thickness, H = 4.3x6 = 26 inches. 7.2 METHOD 2 Step: Again, using the information given in Exhibit 9, the dynamic wheel load is determined by Equation 2 to be: P, = 39 (1 + 0.33x40/38) = 53 (kips); and the equivalent number of load repetitions is determined by Equation 3 to be: N = 60x1000000x2000/8/39000 = 380,000. Step 2: By Equation 9, the deformation influence factor is determined to be: |, = 100x(1/6) /{1.2(53,000/13/1000)2.4x380,00040.18) = 0.047. ‘Step 3: From design Chart B17 for E, = 40,000 psi, E, = 2000 psi, T/L=5x12/6 = 10, and |, = 0.047, the required H/L is found to be 4.6. Therefore, the required granular layer thickness, H = 4.6x6 = 28 inches. For this example, the required granular layer thickness is 28 inches, ie., the larger of the two granular layer thicknesses determined using the two design procedures. 88.0 COMPARISONS WITH FIELD RESULTS The design methodology presented in this report has been applied in several field cases. This section will give comparisons between design results and field test results, 8.1 LOW TRACK MODULUS TEST TRACK A low track modulus (LTM) test track 600 feet long was constructed on a soft soil subgrade in 1991 at the FAST (Facility for Accelerated Service Testing) test track. Prior to the LTM track construction, a trial low track modulus (TLTM) test track of 100 feet was built and tested to evaluate the feasibility of constructing and testing a longer LTM track, The major purpose of the LTM test track was to study the effect of a soft subgrade on track performance under repeated heavy axle loading. Since the natural subgrade soil at FAST is a silty sand and does not represent a soft subgrade soil, a trench of 12 feet in width was excavated in the natural subgrade to a depth of 5 feet from the ground surface, and backfilled with the Mississippi buckshot clay (PI = 40-45, LL = 60-70). The clay was compacted with a specified moisture content and dry density to achieve a subgrade of low stiffness. Although the specified soil moisture content for both the TLIM and LTM subgrades was 30 percent, the actual average moisture content was 29.0 percent for the TLTM track subgrade and 32.7 percent for the LTM track subgrade. The corresponding soil static strength for the LTM subgrade was approximately 13 psi, and 24 psi for the TLTM track subgrade. The corresponding resilient modulus of the clay was approximately 2,000 to 3,000 psi for the LTM track subgrade, and 6,000 to 8,000 psi for the TLTM subgrade. Exhibit 10 summarizes the required information for the design and the corresponding design results using Design Method 2. For more information regarding the LTM track and subgrade conditions, readers can refer to References (3) and (7). 19Exhibit 10. Design Results for TLTM and LTM Track Granular Layer Thicknesses Design Parameters Values Traffic conditions. 39 kips \V=40 mph Design traffic = 60 MGT Wheel diameter D = 98 inches Granular material E, = 40,000 psi Design erterion P, = Linch Subgrade characteristics | Subgrade thickness T = 5 feet Subgrade soil type: fat clay LIM ™™ 3 psi o, = 24 psi 2000 psi E, = 6000 psi Design results The design methodology presented in this report was not available when the TLTM and LIM tracks were initially built. However, field measurements of track performance and subgrade conditions were conducted throughout the LTM test, which provided an excellent case to evaluate the design methodology presented in this report. Furthermore, in Phase 2 tests of the LTM test track, which were conducted in 1993, the design results were applied directly in the reconstruction of the LTM granular layer after the LTM subgrade failed in Phase 1. The original adopted thickness of the granular layer for both the LTM track and LTM Phase 1 track was 18 inches (12-inch ballast layer plus 6-inch subballast layer) based on an assumption of a soil moisture content of 30 percent and a minimum compaction level of 90 percent standard maximum dry density. The TLTM track, with an 18-inch granular layer thickness, was able to sustain repeated heavy axle loading for 20approximately 53 MGT without generating excessive plastic deformation in the subgrade. However, the LTM track, with the same granular layer design thickness, had much greater difficulty maintaining the required track surface geometry. The subgrade for the LTM track experienced rapid progressive shear failure (squeezing). As a result, the test track required frequent tamping and surfacing. At approximately 62.3 MGT, traffic was stopped to allow for rebuilding the test track. Based on the design results in Exhibit 10 and the originally adopted thickness for the TLTM and LTM Phase 1 tracks, a brief comparison between the required granular thicknesses indicates that the adopted thickness of 18 inches for the TLTM track was larger than the required thickness (12 inches). The adopted thickness of 18 inches for the LTM Phase 1 track was, however, much smaller than the required thickness (28 inches). Thus, the TLTM track should have been to be able to maintain the required track geometry, while the LTM Phase 1 track should not. Obviously, the field test results were consistent with the design results. ‘After the failure of the LTM subgrade, the test section was rebuilt based on the design results using the design methodology presented in this report. The granular layer thickness was increased to 27 inches (12-inch ballast and 15-inch subballast). Traffic then resumed and measurements and observations showed little track geometry degradation until 9.3 MGT when the heavy rainfall during train operations caused sudden deterioration of the track geometry. The causes for this track failure were discussed in an AAR Technical Digest’ A thick subballast layer (15 inches) with relatively low permeability was saturated for an extended period due to poor track drainage. Under repeated dynamic axle loads, the clay subgrade was subjected to higher stress intensity due to the reduced loading spreading capacity of the subballast layer. The liquefaction of the saturated subballast layer under dynamic loadings was also considered the cause leading to the strength reduction of the subballast layer. 24Exhibit 11 shows a comparison of profile degradation with traffic for the three tests. In this figure, the calculated profile roughnesses, based on track geometry measurements, are plotted versus traffic up to 14 MGT (see Reference (7) for roughness calculations). Although the TLTM track had 53 MGT of tonnage and the LTM track had 62.3 MGT of traffic, a comparison is not made beyond 14 MGT in Exhibit 11. This is because the TLTM never required tamping and surfacing prior to 53 MGT. LTM Phase 1 track did, however, require frequent tamping and surfacing following 12.4 MGT. As the results in Exhibit 11 show, the profile degradation with traffic is higher for the LTM Phase 1 track than for both the LTM Phase 2 and TLTM tracks. 0.12 0.10 0.08 Profile Roughness (in.) ° 8 0.04 -e- 1™ 0.02 —™- LTN phase 4 —4— LTM phase 2 0.00 ~ o 2 4 6 8 0 12 14 Cumulative Traffic (MGT) Exhibit 11. Comparison of Profile Degradations 228.2 BURLINGTON NORTHERN HOT MIX ASPHALT TEST SITE A section of Burlington Northern (BN) Railroad track approximately one mile long near ‘Alps, New Mexico, experienced continuous development of differential track settlement and mud pumping problems. Frequent track maintenance was required to maintain a proper track geometry under a traffic of approximately 60 MGT annually. BN decided that a major upgrade of the track structure had to take place to permit normal train traffic and to reduce maintenance costs. Field and laboratory investigations performed in 1994 indicated that the major causes for the track problems we the subgrade soil attrition with mud pumping and/or the subgrade soil progressive shear failure.” Remedies recommended include: an adequate granular layer thickness to reduce stresses transmitted to the subgrade, a layer of subballast for preventing mud pumping, proper track drainage, and the altemative hot mix asphalt layer (HMA) to the larger granular thickness?" For compariscn, the entire section of the track was divided into three zones: two zones with two different HMA layer thicknesses and one zone with the conventional granular layer (ie, ballast plus subballast). See References’ (8-10) for more details on investigations and remedy recommendations. The subgrade soils were generally clays of high plasticity with low strength and stiffness when saturated or nearly saturated with water. Water was found to have access to the subgrade, thus leading to subgrade soil softening. The subgrade soil strength also varied from location to location. The subgrade soil strength for the zone where the conventional granular layer thickness was planned ranged from 10 to 15 psi. For the granular thickness design, Exhibit 12 gives design information and design results using Design Method 1. 23Exhibit 12. Design Results of Track Granular Layer Thickness for BN HMA Site Design Parameters Values Traffic conditions P, = 88 kips V=35 mph ‘Annual traffic = 60 MGT Wheel diameter D = 36 inches Granular material E, = 40,000 psi Design criterion 1% for a total traffic of § years ‘Subgrade characteristics | Soil type: fat clay 9, = 10 psi E, = 2000 psi Design results H = 23 inches The final adopted granular layer thickness was 21 inches (16-inch ballast layer and 5-inch subballast layer) for the control zone with the conventional granular layer. Exhibit 13 shows the measured rail profile roughness performance for the three alternative track structures for traffic up to 100 MGT. As this figure shows, the zone with the conventional granular layer thickness showed a higher increase in roughness. However, after 1.5 years of traffic and about 90 MGT, the accumulated roughness appears to be only moderate. If present rates of roughness accumulation continue, tamping will not be required for another two years (at about 200 MGT after construction). 240.04 Regular granular layer > a 4 in. HMA Sg 2937 | a gin. HMA 3 3 8 £ 5 3 e 2 g e é 0.00 ™ ttt TT 0 10 20 30 40 50 60 70 80 90 Cumulative Traffic (MGT) Exhibit 13. Track Roughness Developments 8.3 AMTRAK SITES Two track sites on the AMTRAK Northeast Corridor between Philadelphia and Baltimore required frequent track maintenance. One site is in Edgewood, Maryland. The tracks at this site have experienced recurring development of differential settlement and require production tamping at least twice a year and additional spot tamping over approximately 6 miles. Remediation measures such as lime slurry injection and geotextile application had been taken since 1984, but were not successful. The other site is located in Aberdeen, Maryland. At this site, one track approximately 200 feet long in a multi- track system had a mud pumping problem. Geometry deterioration was not a primary concer. A section of track roughly 100 feet long was undercut and the fouled ballast 25was replaced in 1991. However, the pumping problem was extending and progressing from where it occurred before. To determine the major causes of the track foundation problems and the material characteristics for the problem remedy design, field investigations and laboratory tests were conducted in 1994. The findings were discussed in a separate report.” For the Edgewood site, the major cause for the differential track settlement was the non-uniform soft subgrade support, resulting from variation in subgrade soil strength aggravated by water trapped along the length of the track. Subgrade progressive shear failure caused significant squeezing and heave of the subgrade soils across the track. Deep ballast pockets were also observed at several locations. Poor track drainage due to fouled ballast also contributed to the problem since rainfall water could not drain out of the track easily, thus causing softening of the subgrade soils. For the Aberdeen site, ballast breakdown and abrasion appeared to be the major reason for the ballast fouling. Mud pumping occurred when the hydraulic actions caused by cyclic traffic loading and the free water in the ballast ejected the mud slurry onto the ballast surface. The subgrade is strong and no progressive shear failure or ballast pockets were observed. Also, there was not apparently any infiltration of the subgrade soils into the ballast. Remedy designs included evaluation of the granular layer thicknesses for both the Edgewood and Aberdeen sites. Results using the design methodology presented in this report and the comparison with field observations are described below. The traffic is a mixture of passenger trains with speeds up to 120 mph and freight trains with the heaviest wheel weight of 35 kips. Exhibit 14 gives the traffic parameters used for designs for these two sites. Note that Equation 2 is not needed to calculate dynamic wheel loads since both static wheel loads and dynamic wheel loads were available. 26Exhibit 14, Traffic Parameters for AMTRAK Sites Trattic type | Annual Static Loading Traffic Wheel load _| Percentage | Wheel load _| Percentage Freight 37 MGT 40 kips 50 35 kips 40 15 kips 50 10 kips 60 Passenger | 15 MGT 35 kips 20 15 kips 100 20 kips 80 The subgrade soils at the Edgewood site were generally lean clay with unconfined compressive strength in the range of 7 to 12 psi. The subgrade soils at the Aberdeen site were either sandy lean clay or silty sand with gravel. The unconfined compressive strength varied from 14 to 42 psi. Exhibit 15 gives the subgrade soil properties required for design for both sites. Other required information for design and the corresponding design results are also given in Exhibit 15. Based on the design results, the required granular layer thickness for the Edgewood site should be more than 40 inches, while the required granular layer thickness for the Aberdeen site is only 15 inches. The difference in required granular layer thickness between these two sites is a result of the difference in subgrade soil strength. However, the existing granular layer thickness for the Edgewood site was found to be only 20 to 31 inches using field cross trench measurements and cone penetration tests (CPT). Obviously, the existing granular layer thickness was not sufficient to reduce traffic load induced stresses in the subgrade below the subgrade soil strength. Consequently, significant subgrade progressive shear failure and ballast pockets occurred, causing significant differential track settlement. arOn the other hand, the existing granular layer thickness for the Aberdeen site ranged from 26 to 30 inches, which is larger than the required granular layer thickness, based on design (15 inches). As discussed previously, there were no subgrade failures found at the Aberdeen site. The traffic induced stresses transmitted to the subgrade are lower than the subgrade soil strength. Exhibit 15. Evaluation of Track Granular Layer Thickness for AMTRAK Sites Design and Comparison Edgewood Site Aberdeen Site Parameters Subgrade Lean clay (CL) Sandy lean clay (CL) characteristics +12 psi 0, = 1410 42 psi 000 psi 000 psi Granular materials E, = 4000 psi E,=4000 psi Design criteria 8% for 10 years, or = 3% for 10 years, or px = 1 inch for 10 years Inch for 10 years Required thickness 29 inches by Method 1, 9 inches by Method 1, by design 42 inches by Method 2 15 inches by Method 2 Existing thickness 20 t0 31 inches based on _| 26 to 30 inches based on field field crossing trenching and | cross trenching results CPT results ‘Actual subgrade conditions | Excessive subgrade No subgrade failures squeezing and heaving, ballast pockets 9.0 SUMMARY AND RECOMMENDED FUTURE STUDIES The development of the design methodology and design charts presented in this report was based upon two main foundations: (1) a validated track model GEOTRACK for stress analysis in track and subgrade due to multi-axle loads and (2) extensive test results of various subgrade soils under repeated stress applications. Compared with 28existing design methods in the literature for track granular layer thickness, the design methods presented in this report consider not only the effects of the maximum wheel load, but also the effects of cumulative tonnage on track performance. The major track and subgrade variables represented include subgrade soil type, soil strength and stiffness, and granular layer stiffness, each of which significantly affects the selection of the granular layer thickness required for preventing subgrade failures. The subgrade failures to be prevented are the two most common types for soft subgrades, ie., subgrade progressive shear failure (squeezing) and excessive plastic deformation (ballast pocket). Although the limited number of available field results considered have been consistent with the results derived using the new design methodology, further verification and modification are needed. Recommended future studies include: (1) Selection of design period: This will address what length of design period should be used for various classes of railroad tracks when determining the granular layer thickness. Considerations will be given to linking the design period to a typical ballast production tamping. (2) Selection of allowable plastic deformation (total settlement): For Design Method 2, the magnitude of allowable cumulative subgrade settlement for a design period needs to be studied corresponding to different classes of railroad track. Consideration will be given to linking the allowable subgrade settlement to the allowable track geometry deviation for various classes of track. (3) Selection of allowable cumulative plastic strain at the subgrade surface: The allowable deviator stress at the subgrade surface based on Design Method 1 may be an alternative to using allowable cumulative plastic strain at the subgrade surface. However, strain is more fundamental, is independent of soil type, and is more directly linked to geometry change. This method needs to determine appropriate allowable strain for various classes of tracks. (4) Comparison of design results with field results: Although preliminary results by the design methods were consistent with the limited available field results, 29further comparisons are necessary to verify and improve the design methodology developed. Such comparisons are needed to help determine allowable strain and deformation in Methods 1 and 2. (©) Design software development: Presently, selecting a minimum granular layer thickness requires following the design procedures and using the design charts given in this report. A future development to the design procedures and charts will be a computer program which will generate the required design results and recommendations based on the input design parameters. 10.0 ACKNOWLEDGEMENTS The authors wish to thank Dr. Steven Chrismer, Senior Engineer, AAR, for his encouragements throughout the writing of this report. Support from Semih Kalay, Chief Technical Officer, AAR, is key to the completion of this report and is greatly appreciated. The authors also wish to thank Drs. Jim Lundgren, Assistant Vice President, Railroad Technology, AAR, and Al Reinschmidt, Senior Assistant Vice President, Technical Services, AAR, for their valuable advice in finalizing this report. 30REFERENCES (1) Selig, E. T. and J.W. Waters (1994) Track Geotechnology and Substructure Management, Thomas Telford Ltd., London, England. (2) Li, D. and E. T. Selig (1995), Evaluation and Remediation of Potential Railway Subgrade Problems under Repeated Heavy Axle Loads, R-884, Association of American Railroads, July. (3) Li, D. (1994), Railway Track Granular Layer Thickness Design Based on Subgrade Performance under Repeated Loading, Ph.D. Dissertation, Department of Civil Engineering, University of Massachusetts, Amherst. (4) Li, D. and E. T. Selig (1996), Cumulative Plastic Deformation for Fine-Grained Subgrade Soils, to be published in December issue in Journal of Geotechnical Engineering, ASCE. (5) Read, D. and D. Li (1995), Subballast Considerations for Heavy Axle Load Trackage, Technical Digest 95-028, Association of American Railroads. (6) American Railway Engineering Association (1995), Manual for Railway Engineering. (7) Ebersohn, W. (1995), Substructure Influence on Track Maintenance Requirements, Ph.D. Dissertation, Department of Civil Engineering, University of Massachusetts, Amherst. (8) Li, D. and E. T. Selig (1994), Investigation of BN Track Subgrade Problems Near Alps, NM, Consulting Report for AAR and BN Railroad, Amherst, Massachusetts, July. (9) Li, D. and E. T. Selig (1994), Remediation of BN Track Subgrade Problems Near Alps, NM, Consulting Report for AAR and BN Railroad, Amherst, Massachusetts, December. (10) Chrismer, S. M, V. Terrill, and D. Read (1996), Hot Mix Asphalt Underlayment Test on Burlington Northern Railroad, R-892, Association of American Railroads, July. (11) Li, D. And E. T. Selig (1994), Investigation of AMTRAK Track Foundation Problems in Edgewood and Aberdeen, MD, Consulting report to AAR and AMTRAK, December. ”APPENDIX A Charts for Design Method 1«, (%) N=10 10 16 peers , , 12 eer ' ; + Soil Type: CH (fat clay) Soil Type: CL (lean clay) 10k 4 2 sf a 8 6b J 4p © 4 o ‘a 2b q 0 1 L 1 o L ! 1 00 02 04 06 08 00 02 04 06 08 B= (oo, B=(6fo,) 6 T i T 4 roy T T 5 |. Soll Type: MH (elastic silt Soil Type: ML (silt) 4 3 2 1 0 i t 1 00° «02 «(04+ «(0608 00 02 04 06 08 B= (Jo) B= (Jo) Chart AlEy = 80,000 psi = 550 MPa Chart A2WL 'b: 4,000 psi = 28 MPa : 8,000 psi = 55 MPa 4: 16,000 psi = 110 MPa| E, = 40000 psi = 280 MPa 1 L 1 L ‘Chart A3a: 2,000 psi= 14 MPa b: 4,000 psi= 28 MPa c: 8,000 psi = 55 MPa d: 16,000 psi = 110 MPa E, = 20,000 psi= 140 MPa 4APPENDIX B. Charts for Design Method 20 0.05 0.10 0.15 0.20 0.25 T T Chart BL Ey, = 80,000 psi Soil Type: CH 3 4 Chart B2 E 's E,, = 80,000 psi = 550 MPa 9 Soil Type: CL 10 tHL gE, Soil Type: ML Chart B3 Soil Type: MH Chart B4§.00__ 0.05 0.10 0.15 0.20 0.25 T T 1 r T 15 2b 3b 4b Chart BS S5F 6b 7H 8h E, = 4,000 psi= 28 MPa 4 5 E,, = 80,000 psi =550 MPa Soil Type: CH 7 Chart B6 Soil Type: CLHL 0 O01 02 03 04 0.5 06 0.7 08 09 10 Soil Type: ML fcstotreodiatisasl 000.02 0.04 0.060.080.1002. Soil Type: MH pede td Chart B7 Chart BS2009.05 0.10 ots 0.20 0.25 0.30 Chart B9 | Chart B10 Soil Type: CLil inda dt ao RN BRaEeF §-00_ 0.02 0.04 0.06 0.08 0.10 0.12 0.14 Toor T T T T T a b ce defgi 2 4 TL 3 2 4 = E b=4 o c=8 | 4 d=12 = 55 e=16| 4 £=26 6E 2-36] 4 Soil Type: MH Chart Bil Chart B12f.00_ 0.05 0.10 0.15 0.20 0.25 0.30 0.35 T T ToT T T 1 ab ec de fg 2b q 3h 4 TL 4p a=2 Be b ¢ | 4 Chart B13 d=12 er e=16| 4 £=26 7h 2-36) 4 BE E, = 16,000 psi=110MPa 80,000 psi = 550 MPa. oF Soil Type: CH | 1 1 L 1 1 02 0.4 06 0.8 10 1 ab c de fel 2B 4 3b 4 4E a=2 4 b=4 aa e=8 | 4 Chart BI4 d=12 e=16 6E teas] 4 TE ge 4 8h 6,000 psi= 110MPa | 50 MPa 9 10BL mmo Roce win iad promo nn Beas Ey = 16,000 psi= 110 MPa E,, = 80,000 Soil Type: MH Chart B15 Chart B16§.00__ 0.04 0.08 0.12 0.16 0.20 0.24 0.28 T T T T T T 7 a ob c de fg Chart B17 Chart B18 wooo RN me OTe BRas 1 E, = 2,000 psi=14 MPa E,, = 40,000 psi=280 MPa Soil Type: CL0.0 0.2 0.4 0.6 0.8 1.0 0 T T T 1 2 3 4 sf Chart B19 6k 75 81 E,, = 40,000 psi = D Soil Type: ML 10 9.00, 0.02 0.04 0.06 0.08 0.10 0.12 0.14 r + : Toor T 1 a ob c de fg] ™m 3 2 4] 4 8 {>| | Chart B20 16 26 36] |AIL WL a 0.05 0.10 0.15 0.20 T T T c de fg E,, = 40,000 ps Soil Type: CH E, = 4,000 psi=28 MPa E,, = 40,000 psi Soil Type: CL 0.25 0.30 T Chart B21 Chart B221.0 12 TL 4 Chart B23 Cee wa mee RRaBTEN me aoce Chart B242000.05 0.10 0.15 0.20 0.25 0.30 0.35 TT T ab c de fg Chart B25 Chart B26 Soil Type: CL L L 1 L1 1 E,= 8,000 psi EF 40,000 p: Soil Type: MH 1 Chart B27 Chart B28Chart B29 a b e doe f g 1p 2b 4 3h 4 4b q sos Chart B30 6E 4 TE 4 oo Ey = 16,000 psi of E,, = 40,000 psi = 280 MPa } Soil Type: CLHIL en Ss 16,000 psi = 110 MPa EF 40,000 psi = 280 MPa Soil Type: MH mRomenoes wir ine ween oan RSaS Chart B31 Chart B32HL piano nN BRaS mmenece Vung E,, = 20,000 psi = 140 MPa Soil Type: CH 4 E, = 2,000 psi = 14 MPa E,, = 20,000 psi = 140 MPa Soil Type: CL Chart B33 Chart B34H/L 12 .00 0,02 0.04 0.06 E, = 2,000 psi=14 MPa E,, = 20,000 psi = 140 MPa Soil Type: ML L 0.08 0.10 0.12 0.14 0.16 Tot ab c defg 8 5 wm menoce Men Chart B35 Chart B36one O01 0.2 0.3, 0.4 7 ab oc de fe j 2 4 3b 4 4b a=2 q b=4 SE c=8 d=12 e=16 &F £=26 q =36 7 ce 4 8 E, = 4,000 psi=28 MPa 4 E,, = 20,000 psi = 140 MPa Soil Type: CH Ey = 4,000 psi E,, = 20,000 psi Soil Type: CL Chart B37 | Chart B38HL Soil Type: MH 1 \ ab © de E Soil Type: ML y 1 : 1 1 1 00 0.05 0.10 0.15 0.20 T ; r a bc defg bE TL 4 a=2 b=4 4 c=8 d=12 e=16 | £=26 E 2-36, 4 Chart B39 Chart B40Chart B41 Bone AN momonecs via Exaots) Chart B42 Soil Type: CL L L 1 LHL HL E, = 8,000 psi E,, = 20,000 psi Soil Type: ML 55 MPa 140 MPa 0.04 0.08 meno ee iow ne Soil Type: MH. Chart B43 Chart B44E, = 16,000 p E,, = 20,000 psi Soil Type: CH L 1 10 ae T Ex = 16,000 ps E, = 20,000 psi Soil Type: CL Chart B45 Chart B46Chart B47 Eg = 16,000 psi= 110 MPa E,, = 20,000 psi = 140 MPal Soil Type: ML ge 00s oto ts 0.20 0.25 T r T ib abcde fg j 2k 4 3h 4 4b 4 gs Chart B48 6 4 7b j BE 4 > 0,000 psi = 140 MPa Soil Type: MH 10 L 1 1