TOP_01_1_010A_Vehicle_Test_Course_Severity.pdf

REPORT DOCUMENTATION PAGE

Form Approved

OMB No. 0704-0188

The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Department of Defense, Washington Headquarters Services, Directorate for information on Operations and Reports (0704- 0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number.

PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.

1. REPORT DATE (DD-MM-YYYY)

12-12-2017

2. REPORT TYPE

Final

3. DATES COVERED (From - To)

4. TITLE AND SUBTITLE

Test Operations Procedure (TOP) 01-1-010A Vehicle Test Course Severity (Surface Roughness)

5a. CONTRACT NUMBER

5b. GRANT NUMBER

5c. PROGRAM ELEMENT NUMBER

6. AUTHORS 5d. PROJECT NUMBER

5e. TASK NUMBER

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)

U.S. Army Yuma Proving Ground Yuma Test Center 301 C Street Yuma, AZ 85365-9498

Automotive Directorate (CSTE-DTC-AT-AD) U.S. Army Aberdeen Test Center 400 Colleran Road Aberdeen Proving Ground, MD 21005-5059

8. PERFORMING ORGANIZATION

REPORT NUMBER

TOP 01-1-010A

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)

Policy and Standardization Division (CSTE-TM) U.S. Army Test and Evaluation Command 2202 Aberdeen Boulevard Aberdeen Proving Ground, MD 21005-5001

10. SPONSOR/MONITOR’S

ACRONYM(S)

11. SPONSOR/MONITOR’S REPORT

NUMBER(S)

Same as item 8

12. DISTRIBUTION/AVAILABILITY STATEMENT

Distribution Statement A. Approved for public release; distribution is unlimited.

13. SUPPLEMENTARY NOTES

Defense Technical Information Center (DTIC), AD No.:

This TOP supersedes TOP 01-1-010 CN1, Vehicle Test Course Severity (Surface Roughness), dated 4 January 2012.

Marginal notations are not used in this revision to identify changes, with respect to the previous issue, due to the extent of the changes.

14. ABSTRACT

Provides a method of analyzing vehicle test course surface roughness. Describes use of profilometers and the conversion of profilometer data to power spectral density (wave number spectra) curves. Includes wave number spectra and other metrics of vehicle endurance test courses at ATEC Test Centers.

15. SUBJECT TERMS

Automotive test course Roughness of roads Wave number spectrum Test course surface roughness Profilometer Vehicle test course

16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF

ABSTRACT

SAR

18. NUMBER

OF

PAGES

19a. NAME OF RESPONSIBLE PERSON

a. REPORT B. ABSTRACT C. THIS PAGE Unclassified Unclassified Unclassified 19b. TELEPHONE NUMBER (include area code)

Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std. Z39-18

(This page is intentionally blank.)

U.S. ARMY TEST AND EVALUATION COMMAND

TEST OPERATIONS PROCEDURE

*Test Operations Procedure 01-1-010A 12 December 2017 DTIC AD No.

VEHICLE TEST COURSE SEVERITY (SURFACE ROUGHNESS)

Page Paragraph 1. SCOPE

2. FACILITIES

3. MONITORING TEST COURSE SURFACE

ROUGHNESS

4. TEST COURSE SEVERITY AND GRADE FROM THE

SURFACE PROFILE

4.1 Test Course Root Mean Square (RMS)

4.2 Other Surface Profile Statistics

4.3 International Roughness Index (IRI)

4.4 Longitudinal Grades

4.5 Statistical Assessments of Elevation Profiles and Terrain Severity Parameters

5. WAVE NUMBER SPECTRUM ANALYSIS OF GROUND

VEHICLE TEST COURSES

5.1 Aberdeen Test Center (ATC) Courses

5.2 Yuma Test Center (YTC) Courses

5.3 Tropic Regions Test Center (TRTC) - Horoko Test Courses 45

6. SOIL CLASSIFICATION

7. ROAD SURFACE CLASSIFICATION

APPENDIX A. ATC LASER PROFILER ..................................................... A-1

B. YTC LASER PROFILOMETER .......................................... B-1

C. TEST COURSE INSPECTION VEHICLE .......................... C-1

D. ATC CART PROFILOMETER ............................................ D-1

E. ABBREVIATIONS ............................................................... E-1 F. REFERENCES ...................................................................... F-1

G. APPROVAL AUTHORITY .................................................. G-1

* This TOP supersedes TOP 01-1-010 CN1, Vehicle Test Course Severity (Surface Roughness), dated 4 January 2012.

Approved for public release; distribution unlimited.

TOP 01-1-010A

12 December 2017

1. SCOPE.

a. Vehicle test course severity is a product of many attributes used to define a test course.

Among these attributes are surface roughness, grade, side slope, mud/dust, soil strength, radius of curvature, etc. This Test Operations Procedure (TOP) describes methods for determining vehicle test course surface roughness and grades by analysis of profilometer data as well as methods used for course maintenance. The calculated roughness metrics include normal displacement root mean square (RMS); wave number spectrum (WNS); international roughness index (IRI); and grade. This TOP includes the calculated metrics of vehicle endurance test courses at the U.S. Army Aberdeen Test Center (ATC), the Yuma Test Center (YTC), and the Tropic Regions Test Center (TRTC). These values are used in conjunction with the course descriptive information in TOP 01-1-011B1** and the vehicle group terrain type recommendations in TOP 02-2-506A2 when developing a mission profile for endurance test planning.

b. Consistent course roughness is essential to properly analyze and compare reliability and durability data obtained years apart. Even when maintained, test course surface roughness will change depending on weather conditions and traffic density. Proper maintenance is needed in order to minimize the amount of variation that will occur.

2. FACILITIES.

Specific descriptions of the U.S. Army Test and Evaluation Command (ATEC) Test Center facilities are provided in Section 5, in the following paragraphs:

a. ATC courses, paragraph 5.1.

b. YTC courses, paragraph 5.2.

c. TRTC courses, paragraph 5.3.

3. MONITORING TEST COURSE SURFACE ROUGHNESS.

Test course roughness is monitored using several techniques that have been developed to track course roughness and minimize the variability of that roughness over time and to trigger timely and appropriate maintenance actions.

a. Hazardous course conditions are reported immediately by test vehicle operators or course inspection personnel upon detection. Hazardous conditions can include large potholes, standing water, snow and ice, wash-outs, or large rocks. Depending on the severity of the hazard, the course may be closed until repairs are made. Hazardous conditions are generally addressed within 24 hours.

** Superscript numbers correspond to Appendix F, References.

b. Courses are inspected at least monthly (except for those not regularly used). The courses are reviewed for hazards and excessive wear. At ATC, this also includes the use of a High-Mobility Multipurpose Wheeled Vehicle (HMMWV) operated at a pre-established speed for each course; the vehicle is instrumented to record suspension displacement and acceleration, which is compared to established roughness bands (see Appendix C). YTC also has an instrumented HMMWV, though it is no longer used regularly for course roughness assessment except if the YTC Profiler is disabled.

c. Courses are profiled periodically at ATC and YTC (see Appendices A, B, and D for a description of test center profilometers and processing techniques). Most courses are profiled monthly, with the exception of some of the longer courses at YTC which are profiled quarterly.

(Note that profiling intervals are objectives that are dependent on equipment functionality, scheduling conflicts and weather permissibility.) The profiles are then analyzed for roughness, including using the wave number spectrum technique. Increases in roughness may signify the need to change course maintenance patterns. For example, secondary gravel roads at YTC go through a cycle of washboard build-up then grading. The washboard is graded when the wave number spectrum shows the amplitude at the washboard wave number (0.3 to 0.5 cycles per foot) to be approaching 5.0 x 10-3 ft2/(cycles/foot).

4. TEST COURSE SEVERITY AND GRADE FROM THE SURFACE PROFILE.

a. The surface profiles of the test center endurance courses are measured on a periodic basis using profilometers. Off road profiling equipment is not very common, with each system being rather complicated and having characteristics which influence the final results. For decades ATC used a cart based profiler, as described in Appendix D. This system has been replaced by the rear mounted laser profiler system described in Appendix A. YTC uses a front mounted laser profiler system described in Appendix B.

b. The profilometer field measurements are processed to generate test course profiles, which present the test course elevations as a function of distance traveled along the course. Data are presented at discrete increments along the length, with a sample every 7.6 centimeters (cm) (3 inches (in.)) being very typical. These uniformly sampled spatial histories are analogous to signals in the time history domain. Many of the standard analysis techniques that are applied in the time domain are applicable in the spatial domain. Test course profile data are used to inform course maintenance decisions, are provided to modelers, are used to characterize the test courses for endurance test planning, and are retained for historical comparative purposes.

4.1 Test Course Root Mean Square (RMS).

a. The normal displacement RMS is a frequently referenced metric used to characterize terrain for military systems. The RMS is used in the North Atlantic Treaty Organization (NATO) Reference Mobility Model, ride quality specifications, many Operational Mode Summary/Mission Profile (OMS/MP) reports, and endurance test planning. Because an RMS value can be determined through a variety of processing routines, it is important to describe how RMS is calculated in this document. RMS values should not necessarily be compared if they are calculated from different methods.

b. Essentially the RMS is derived by first calculating the test course WNS. (WNS plots are presented throughout Section 5 of this document.) The WNS is the spatial domain analog to a Power Spectra Density (PSD) from the time/frequency domain. The WNS is typically calculated using Welch’s method (with standard parameters of 256 feet (ft) blocksize, with 50% overlap, using a Hann window.) The x-axis of the WNS, or wave number, which has units of cycles per foot (or meter), is the inverse of the test course wavelength and is analogous to frequency in the time domain. The Y-axis of the wave number spectrum has units of vertical displacement squared divided by the analysis bandwidth (cycles per foot or cycles per meter).

This is usually expressed as ft2/cycle/ft. The test course RMS roughness is the square root of the area under the wave number spectrum between two wave numbers used as integration limits.

The RMS value is, therefore, dependent upon both the WNS spectral shape and the integration limits.

c. While the definition of the integration bandwidths can be application specific, it has been generally accepted for military endurance testing that the RMS determination will be made between wavelengths of approximately to 0.15 to 20 meters (m) (0.5 to 64 ft). These integration bounds are partly arbitrary but are intended to cover the wavelengths which could influence a vehicle suspension at speeds typically used for endurance tests.

d. The ground excitation frequency can be estimated by multiplying the WNS wavenumber by the test item speed. It may be appropriate to use a longer wavelength integration bound if a test item is expected to traverse a course at a higher speed. Shorter wavelength content will excite subsystems with higher frequency inputs; longer wavelength content will excite subsystems with lower frequency inputs.

e. It should be noted that the course RMS process has switched post-processing filters recently in order to improve the capture of long-wavelength (near 64-ft) energy in the WNS.

This change has resulted in an increase in reported RMS values for many courses since the implementation of the filter (circa 2015). Some of the reported RMS values have increased on the order of 10-50% due to this effect.

4.2 Other Surface Profile Statistics.

a. While RMS is a useful metric, in that it correlates to surface roughness and is well-established within the Army, it is not adequate to fully describe a test course as it relates to the surface profile’s effects on the test item. There are many reasons for this.

(1) As described above, the RMS value is dependent upon the WNS curve shape, but the WNS curve shape is not dependent upon the RMS value. This means that two different courses with significantly different WNS content can generate the same RMS values.

(2) The RMS calculations overemphasize the contributions of long-wavelength content (up to 64 feet) and underemphasize the shorter wavelengths. The content in the shorter wavelengths may have a greater influence on specific subsystems, as discussed in the section on

IRI.

(3) The RMS calculation process “averages out” transient features; this means transient course content such as bumps and potholes, which can have a serious influence on a test item, will have little influence on an RMS value.

(4) Course grades, side slopes, and curvature create effects distinct from course roughness, and are therefore not described by RMS. Whereas roughness has an effect on suspension and fatigue characteristics, grades will have an influence on powertrains and braking.

Similarly side slopes and curvature primarily (and often interdependently) affect steering, side loading, and lateral stability.

(5) RMS does not convey how a test item would respond to a course at a given speed.

This is an impossible task for any single profile based metric given the variety of test items which will respond in different ways to the same inputs.

b. For the reasons listed above more metrics than a simple RMS value should be conveyed when describing a test course.

4.3 International Roughness Index (IRI).

a. As mentioned, RMS is a useful metric in that it correlates to course roughness and its use is well-established within the Army. In addition to RMS, there are well-established methods in the civilian world that approach the concept of course roughness using different analytic methods, one of which is the IRI.

b. The IRI was developed in the 1980’s to provide a common standard of roughness analysis for roads and highways throughout the world, and has been in wide use throughout the world since that time. The IRI algorithm (as described in World Bank Technical Paper Number

453) processes longitudinal-elevation road profiles by essentially simulating the response of a heavily damped quarter-car model traveling on the road at a fixed speed. The standard simulation speed is 80 kilometers per hour (km/h) (49.7 miles per hour (mph)). The IRI algorithm produces a filtered response of the suspension displacement (or rate) between the quarter-car sprung and unsprung masses, as well as the “Average IRI” (AVEIRI) metric. The AVEIRI characterizes how much the simulated suspension displaces over a given distance.

Specifically it is the average of the absolute value of the filter response. A road which is “rougher” will result in more suspension displacement per mile. AVEIRI values for military off-road endurance courses can be much higher than those typically analyzed in civilian studies that generally concentrate on primary road surfaces.

c. Compared with RMS, the IRI is superior at characterizing how much the terrain would cause a plausible vehicle suspension response. The IRI acts as a frequency domain transfer function, filtering out the long wavelength content which very often drives RMS values. One of the weakness of the IRI comes from its development in the civilian world, where civilian roads, vehicles, and speeds may differ significantly from those in the military. Furthermore the AVEIRI broadly characterizes suspension displacement response, which is certainly an important focus when maintaining civilian roads and highways, but ignores other ways a vehicle may respond.

d. IRI should be viewed side-by-side with RMS as a metric that characterizes roughness, but with more focus on the shorter wavelengths of a course. An additional benefit of IRI is that it is an established metric, which enables a comparison of test center courses with other roads which have been measured with the IRI. Finally, AVEIRI can be useful to differentiate two courses that may have comparable RMS values but different AVEIRI values.

e. The varied terrains traversed by military vehicles relative to civilian vehicles results in a high variance in speeds. IRI, being a velocity-dependent parameter, is therefore presented at three speeds in this report: the standard IRI speed of 80 km/hr (49.7 mph), which is most appropriate for high-speed primary and secondary roads; a speed of 40 km/hr (24.9 mph), more appropriate for rougher secondary and trails terrain; and a speed of 25 km/hr (15.5 mph), which would be generally appropriate for cross-country terrain. Using the terminology presented in World Bank Technical Paper Number 45, the different IRI parameters would be designated Reference Average Rectified Slope (RARS) 80, RARS 40, and RARS 25.

4.4 Longitudinal Grades.

The longitudinal grades along the course length can be calculated and statistically characterized.

Longitudinal grade is an important metric in that many vehicle specifications indicate a maximum grade that can be traversed. The longitudinal grade profiles are useful for assessing powertrain, longitudinal chassis loading, and braking. Grade is simply the change in height over a given distance change (“rise over run”). For the analysis in this document, a distance change of 64 ft was used. A distance of 64 ft was considered an appropriate length for typical off-road speeds wherein a significant change in momentum might occur, thus requiring accelerator or brake input to maintain speed. The longitudinal grade values of a profile were calculated along the length of the course by simply taking the difference in elevation between the sample point 32 feet before and 32 feet after each location.

4.5 Statistical Assessments of Elevation Profiles and Terrain Severity Parameters.

Superior characterization of the terrain can be achieved with a more in-depth statistical analysis of elevation profiles. This may include calculating statistics of the elevation profiles themselves, or presenting statistics on test course characteristics as they change along the length of the course.

a. Calculating the standard deviation, skewness, kurtosis, and percentile ranges is a simple procedure. Generally the elevation profile is high pass filtered prior to statistical analysis.

The skewness and kurtosis values are scaled by the standard deviation value. A test course with a purely Gaussian profile will have a skewness value of 0 and a kurtosis value of 3. The skewness of an elevation profile indicates the degree to which the elevation profile deviates more positively or negatively from the mean. For example, Perryman No. 3 Course (at ATC) has a skewness of approximately 1.15 while Perryman No. 2 Course has a skewness of approximately

0.09. The kurtosis value characterizes the extremal deviations from the mean, relative to the standard deviation. For example, Perryman A Course and Perryman No. 1 Course are comparable in many respects, however their kurtosis values differ (respectively 5.39 and 12.39).

This is indicative of the nature of their deviations (Perryman No. 1 Course has sharper potholes.)

b. The test course surface profile parameters introduced (RMS, IRI, Grades), can be better understood when keeping in mind their variations along the course length. This is most notably the case in the longer YTC endurance courses. Several statistics help provide a better overall picture of each course. For example, maximum RMS and maximum Grades may help determine the limitations of a course as it relates to the vehicle’s specification, but the means of these parameters may provide a better indication of the average effect on the vehicles. For each YTC test course, the following statistics were calculated for each roughness parameter based on 1024-ft section lengths: minimum, mean, and maximum. For grades, the following statistics were calculated: root-mean-square average (standard deviation) and maximum absolute value.

5. WAVE NUMBER SPECTRUM ANALYSIS OF GROUND VEHICLE TEST COURSES.

The various endurance test courses at ATC and YTC are regularly measured for their surface roughness using profilometer systems. The courses measured and the resulting wave number spectra are described in this Section by each test center. These curves are plotted in terms of power per unit spatial frequency versus spatial frequency. Hence, the units will be ft2/cycle per ft = ft3 for the ordinate and cycles per ft = ft-1 for the abscissa. These spectra are general characterizations of the surface roughness of each test course and should be considered typical for that course. Activities that require actual surface roughness at the time of a test (e.g., model validation, ride quality) should conduct a terrain measurement at the time of the test. The wave number spectra for each course are presented in the following paragraphs. Sufficient data exists such that percentile spectra (10th and 90th) were computed in addition to the average spectrum (note that the average and percentile bands do not reflect any one profile event, but rather these statistics for each frequency bin). Based on the existing data sets and maintenance practices, the test course spectral content can reasonably be expected to be within those spectral bounds 80% of the time. Excursions above or below the percentile spectra should not be significant except on courses where significant washboarding is allowed to develop before grading. As can be expected, longer wavelength content is less sensitive to wear, weather, and maintenance. As such the shorter wavelength content shows a greater range between percentile spectra.

a. The ways in which the spectral content influences a vehicle depends on the speed at which the test item is traversing the test course. For example when traveling at 25 mph (36.67 feet per second (ft/sec)), terrain content at the 30 ft wavelength will excite a vehicle with a frequency of 1.2 Hertz (Hz), while content at the 3.67 ft wavelength will excite the vehicle with a frequency of 10 Hz. Changing the speed will proportionally change the frequency those wavelengths excite. If the item under test is understood well enough to know how frequency content influences its various subsystems, and the test course operation speeds are known, then insights can be gained as to how the test course spectral content will influence the test item. In general, low frequency content will affect universal joints, transmissions, and other drive train components. Higher frequency content will affect wheel bearings, shock absorbers, and other suspension and structural components. (The low- and high-frequency effects should also be evident when observing how the IRI relates to the RMS value.)

b. Terrain or test courses with noticeable periodicities will have WNS curves with distinct peaks or spikes, which are most noticeable when plotting the vertical axis on the linear scale. Periodicities with wavelengths less than 6-ft are generally considered washboarding, while periodicities at wavelengths greater than 6-ft are referred to as undulations (or similar terms). Extremely small periodicities (less than 1-foot wavelengths) are most often soft-soil track pad indentations and generally do not affect vehicle vibrations.

c. Generally the WNS content is broad spectrum in nature. Occasionally focus is paid to the general slope of the WNS, as presented in the log-log domain. Two courses may have the same slope but will have different RMS values if their curves are offset from each other. WNS curves generally slope downward from left to right, with the imperfect fits to averaged WNS curves having slope values typically ranging from -2.5 to -1. A slope of -2 would indicate the terrain has, on average, an amplitude-to-wavelength ratio that is a constant; steeper slopes (less than -2) indicate more emphasis on long-wavelength content; and shallower slopes indicating an emphasis on short-wavelength content. The downward slope makes sense when considering the WNS content at a wavelength is proportional to the profile height variance at that wavelength. A height change of 2 in. every 2-ft is significant, while a height change of 2 in. every 20-ft is less significant.

5.1 Aberdeen Test Center (ATC) Courses.

5.1.1 Summary.

a. The nature of the terrain at ATC, and its associated severity, is greatly influenced by the climate and local geography. Munson Test Area (MTA), Perryman Test Area (PTA), and The Automotive Engineering Test Facility (ATEF) are all situated in relatively low lying areas located near the Chesapeake Bay. Churchville Test Area (CTA) courses are located more inland, where naturally occurring hills enabled the creation of courses with extended grades. Natural creeks and streams run in close proximity to many of the PTA courses. Precipitation (rain, sleet, or snow) will accumulate and can remain on the ground for days after it falls. The degree of moisture content in the ground (from dusty to oversaturate) will influence the wear patterns that test operations will produce. Refer to TOP 01-1-011B Vehicle Test Facilities at Aberdeen Test Center and Yuma Test Center for more information.

b. A more detailed description of ATC courses is given in the following sections. Tables characterizing ATC course RMS, IRI, statistics, and Grade values are respectively presented in Tables 1 through 4. The RMS values presented in Table 1 are derived from the WNS curves, as discussed in paragraph 4.1. The Standard Deviation of RMS characterizes the degree to which the course measurements may vary over time. The elevation profiles which lead to Tables 1 and 3 were high pass filtered (forward and backward) through a 4th order high pass Butterworth filter with a cut off frequency of 1 cycle/128 ft. For some of the test courses, the higher speeds used in the IRI analysis are unrealistically high. These higher speed IRI values are still presented, though noted, in case they provide any informational value. The maximum Grade values presented in Table 4 indicate the single largest Grade a vehicle will climb (or descend) while traversing the course, while the Standard Deviation and 80% Range of the Grade values characterizes how much of the test course the test item spends “at grade”.

TABLE 1. SUMMARY OF ATC COURSE RMS VALUES

COURSE NAME AVERAGE

RMS, in.

STANDARD DEVIATION

OF RMS, in.

3-Mile Straight-away 0.09 0.01 Belgian Block 0.69 0.03 Improved Gravel Road 0.25 0.07 Perryman A 0.41 0.07 Perryman No. 1 0.38 0.04 Perryman No. 2 0.81 0.08 Perryman No. 3 2.98 0.07 Perryman No. 5 2.96 0.07 Churchville C 0.22 0.04 Churchville B 1.61 0.08 ATEF Paved 0.05 0.01 ATEF Gravel 0.17 0.03 Churchville B Smooth 0.39 0.04 Churchville B Rough 2.55 0.15

TABLE 2. SUMMARY OF ATC COURSE IRI VALUES

COURSE NAME

AVERAGE IRI AT SPEED, in./mile 50 mph 25 mph 15 mph

3-Mile Straight-away 169 255 299 Belgian Block 1788 1961 2144 Improved Gravel Road 382 494 559 Perryman A 611 761 850 Perryman No. 1 579 702 770 Perryman No. 2 1118 1257 1350 Perryman No. 3 2521 1892 1588 Perryman No. 5 6397 6000 6171 Churchville C 339 527 648 Churchville B 1130 1064 1080 ATEF Paved 76 122 158 ATEF Gravel 265 522 736 Churchville B Smooth 556 700 815 Churchville B Rough 2195 1721 1539

TABLE 3. SUMMARY OF ATC COURSE STATISTICS

COURSE NAME STANDARD

DEVIATION, in. SKEWNESS KURTOSIS 80%

RANGE, in.

3-Mile Straight-away 0.13 -0.47 9.77 0.30 Belgian Block 0.76 -0.06 3.52 1.91 Improved Gravel Road 0.42 -0.19 4.71 1.01 Perryman A 0.61 -0.31 5.39 1.43 Perryman No. 1 0.60 -0.30 12.93 1.41 Perryman No. 2 1.25 0.09 5.84 2.94 Perryman No. 3 5.65 1.15 4.93 12.42 Perryman No. 5 3.12 0.22 2.57 8.27 Churchville C 0.38 0.18 18.41 0.68 Churchville B 3.09 2.08 13.50 4.58 ATEF Paved 0.08 -0.21 5.80 0.18 ATEF Gravel 0.21 0.58 7.27 0.49 Churchville B Smooth 1.07 0.77 14.27 1.84 Churchville B Rough 5.03 1.36 5.51 11.95

TABLE 4. SUMMARY OF ATC GRADES

COURSE NAME

GRADE (%)

Max Std. Dev.

3-Mile Straight-away 1.1 0.4 Belgian Block 3.9 1.6 Improved Gravel Road 3.3 1.2 Perryman A 1.9 0.5 Perryman No. 1 3.5 0.8 Perryman No. 2 5.9 1.6 Perryman No. 3 8.0 2.3 Perryman No. 5 2.2 0.7 Churchville C 11.2 5.0 Churchville B 29.3 9.9 ATEF Paved 0.7 0.4 ATEF Gravel 1.0 0.4 Churchville B Smooth 29.3 9.7 Churchville B Rough 23.2 10.1

5.1.2 3-Mile Straight-away.

This is a continuous asphalt course located in the PTA, consisting of a 5 km (3 miles (mi)) straightway with banked turnaround loops at each end for tests requiring uninterrupted operations such as cooling tests, operation at high speed, etc. The course has a speed limit of 80 km/hr (50 mph) on the straight section (which may be exceeded by written waiver) and 40 km/hr (25 mph) in the turnaround loops. While this course is frequently used for endurance and performance testing, the course is not frequently profiled due to the expectation that the profile of a paved course does not change very quickly. A representative WNS of this course is presented in Figure 1.

Figure 1. Wave Number Spectrum of Perryman Paved Course.

5.1.3 Munson Improved Gravel Road.

This road in the MTA is a loop of about 3 km (2 mi) with left and right curves, and some distinct straight sections, with a noticeable crown. The surface is compacted “Bank Run” gravel, maintained by grading. Frequently portions of this course are used to form a contiguous loop with other courses in the MTA, such as when a vehicle runs the Gravel – Belgian Block loop.

The typical operational speed limit is 56 km/hr (35 mph). The nominal RMS of the course is

0.25 in., though that is likely to vary greatly throughout a year given exposure to inclement weather and high mileage wear exposure from a variety of vehicles. The WNS of Munson Improved Gravel is presented in Figure 2.

Figure 2. Munson Gravel Course spectrum.

5.1.4 Perryman A Course.

This is a generally level course used for tests of various types of vehicles, with frequent useage by tracked vehicles, consisting of a 3.9-km (2.4-mi) loop. This road has sharp and sweeping turns typical of unimproved country roads. A distinct straight section runs parallel to the 3-Mile Straight-away. The course is maintained by grading and filling with native soil. Under wet conditions, severe mud is present; when dry, the course is extremely dusty. The typical operational speed limit is 56 km/hr (35 mph). The WNS of this course is presented in Figure 3.

Figure 3. Perryman A Course spectrum.

5.1.5 Perryman No. 1 Course.

This is a generally level course of moderate irregularity with a substantial roadbed primarily of quarry spall and bank run gravel. The 8.4-km (5.2-mi) loop contains both sharp and sweeping curves, with a distinct straight section parallel to the 3-Mile Straight-away. The surface ranges from smooth to rough; roughness is due to potholes, washboard, and rutting. Potholes and other sharp depressions are limited in depth by filling in with crushed stone. The typical operational speed limit is 56 km/hr (35 mph). The WNS of this course is presented in Figure 4.

Figure 4. Perryman No. 1 Course spectrum.

5.1.6 Perryman No. 2 Course.

This course is laid out in a loop of moderately irregular terrain. Surfaces range from smooth to rough, with sweeping turns resulting in a 2.9-km (1.8-mi) loop. Perryman No. 2 is not as level as Perryman A or Perryman No. 1, but has relatively mild grades when compared to Churchville Test Area courses or many YTC courses. The course is made up of loose native soil. A mixture of soil and stone is used to fill in holes during maintenance. Wear and weather can lead to some stone exposure. Under wet conditions, severe mud is present; when dry, the course is extremely dusty. The typical operational speed limit is 40 km/hr (25 mph). The WNS of this course is presented in Figure 5.

Figure 5. Perryman No. 2 Course spectrum.

5.1.7 Perryman No. 3 Course.

This is a rough course of native soil similar to the soil of Perryman No. 2 Course. Mud ranges from light (with free water) to cohesive. Under wet conditions, severe mud is present; when dry, the course is extremely dusty. The courses consists of approximately 74 humps with an average peak-to-peak spacing of 61 m (200 ft), with spacing’s ranging from approximately 37 m (120 ft) to 90 m (295 ft) apart. The humps have a typical height of about 1 m (3 ft), with possible heights ranging from 0.5 m (1.5 ft) to 1.4 m (4.5 ft). The roughness between the humps is generally comparable to that of Perryman No. 2. However, a notable characteristic is that the Average IRI value for Perryman No. 3 increases as speed increases (unlike most other courses, such as Perryman No. 2). The typical operational speed limit is 32 km/hr (20 mph). The WNS of Perryman No. 3 Course is presented in Figure 6.

Figure 6. Perryman No. 3 Course spectrum.

5.1.8 Perryman No. 4 Course.

This is a tract of native soil making up extremely rough terrain including marshy areas with swamp-type vegetation. The profile is a series of main repetitive humps spaced in a pattern. The swampy and deformable nature of the soil means maintenance must be a concerted effort. The extreme marshy character and vegetation make profile measurements impossible. Therefore a wave number spectrum is not available.

5.1.9 Perryman No. 5 Course.

This is a relatively level and straight course, 300 m (1000 ft) in length, intended to be extremely rough which causes large suspension displacements at slow speeds. Constructed from gravel and large stone, Perryman No. 5 was originally built to be a mobility course, though later reconstruction iterations allowed the course to be used for endurance testing. The course is intentionally built to induce roll inputs by having the “left” side be out of phase relative to the “right” side. Maintenance of this course requires concerted effort because of the large amplitude changes within short distances. The course is not used for endurance testing as frequently as the other test courses, therefore it is not profiled as frequently. Operational speeds are typically set through jury rides prior to testing to take into account course conditions at the time of the test.

Typically speeds are 25 km/hr (15 mph) or less. Relative to other courses, Perryman No. 5 contains more spectral content in the 1.5 m (5 ft) to 9 m (30 ft) wavelength range. The WNS presented in Figure 7 should only be used to give a sense of the nominal roughness.

Figure 7. Perryman No. 5 Course spectrum.

5.1.10 Churchville B Course.

This course consists of broad and tight turns making a 5.8-km (3.6-mi) loop. The course has a wear surface of “Bank Run” gravel that is similar to Munson Improved Gravel Road. Some of the turns occur at the base of hills, causing additional stresses on the vehicles. This course contains the most severe grades at ATC, with the largest grade of 29 percent. Churchville B Course also has the most frequent grade changes along course length, with the transitions between grades (cresting, leveling off, etc.) being more abrupt relative to other courses. Stop signs are located at various locations around the course which causes more severe loading on brake systems and on power trains as the vehicle must regain momentum to climb the hills.

a. The course contains an approximately 1 mile long “rough” section, which consists of approximately 37 “hump” features. Because many of the humps are built onto the grades, the “peak” of a hump is not necessarily higher than any part of the course further up the hill.

Instead, the elevation profile of the humps on downward grades feature a brief segment of leveling off followed by a brief segment of steeper grade. The humps have a nominal length of 12 m (40 ft) to 15 m (50 ft). The hump features become more apparent when a profile is processed with a high pass filter to remove the elevation of the hill, with hump heights ranging from approximately 0.4 m (1.2 ft) to 0.8 m (2.6 ft). The rough section also has the characteristic of the average IRI increasing as speed increases, caused by the humps inducing more suspension movement as the vehicle moves faster. The remainder of the Churchville B (the “smooth” section) has average IRI responses comparable to Perryman No. 1.

b. The typical operational speed limit is 56 km/hr (35 mph), with an expectation that reducing speeds in the “rough” section may be necessary. The WNS of just the “rough” section of Churchville B Course is presented in Figure 8, while the WNS of the entire course is presented in Figure 9.

Figure 8. Churchville B Course, “rough” section spectrum.

Figure 9. Churchville B Course, entire course spectrum.

5.1.11 Churchville C Course.

Churchville C Course is a 2.4-km (1-1/2-mi) secondary road test course with long consistent grades up to 10 percent and turnarounds at each end. The wear surface consists of “crusher run” gravel and is generally free of significant ruts or potholes. WNS of Churchville C Course is comparable that of the Munson Improved Gravel road, primarily because the grades are sustained for such long durations with milder transitions at the top of the grades. The IRI values of Churchville C are also comparable to Munson Improved Gravel road. The typical operational speed limit is 56 km/hr (35 mph), with a reduced speed of 32 km/hr (20 mph) in the turnarounds.

The WNS of Churchville C Course is presented in Figure 10.

Figure 10. Churchville C Course spectrum.

5.1.12 Belgian Block.

a. This is a paved course with unevenly laid granite blocks imbedded in concrete and mortar, forming an undulating surface. It duplicates a rough cobblestone road such as is found in many parts of the world. The course, which is about 1.2 km (0.75 mi) long, does not form a complete loop. Endurance testing is typically done by forming a loop by using a section of Munson Improved Gravel Road. The Belgian Block course contains two turns on the north side, an approximately 560 m (1,800 ft) straight section, followed by more turns on the south side.

The course is useful as a standard rough road for accelerated tests of wheeled vehicles. It is generally included in cycles of courses used for vibration studies. The motion imparted to a vehicle is a combination of roll and pitch and high frequency vibrations.

b. The WNS of the Belgian Block Course is presented in Figure 11. The peak at

0.269 cycle/m (0.082 cycle/ft), a wavelength of 3.7 m (12.2 ft), indicates periodic undulation at this frequency. The typical operational speed limit is 40 km/hr (25 mph), with the expectation that speeds may need to be reduced in the turns.

Figure 11. Belgian Block spectrum.

5.1.13 ATEF Paved.

ATEF Paved road is a generally level paved course consisting of some straight sections and some broad turns to form a 7.2-km (5.2 mi) long loop. The road has a wear surface similar to highway surfaces in good condition. The purpose of the course is to enable testing at sustained high speeds. The maximum test course speed is 112 km/hr (70 mph). The WNS of ATEF Paved is presented in Figure 12.

Figure 12. ATEF Paved spectrum.

5.1.14 ATEF Gravel.

The ATEF Gravel course runs alongside ATEF Paved (exterior to the loop), separated by a small grass strip. ATEF Gravel has no significant crowning and has a wear surface of “Bank Run” gravel. The purpose of the course is to enable testing at sustained high speeds on a gravel wear surface. The maximum test course speed is 80 km/hr (50 mph). The WNS of ATEF Gravel is presented in Figure 13.

Figure 13. ATEF Gravel spectrum.

5.1.15 Overlay of ATC Test Courses.

The average wave number spectra for most of the ATC courses described above are combined and presented in Figure 14.

Figure 14. Overlay of ATC Endurance Test Courses.

5.2. Yuma Test Center (YTC) Courses.

5.2.1 Summary.

The statistics for the surface profile terrain severity were calculated for each endurance course at YTC and are presented in this section. Five values are presented for course RMS: The average over the entire course (from approximately 10 data samples); the standard deviation of those data samples; and, to show the variability of RMS within the courses, the minimum, maximum, and standard deviation of the RMS for 1,024-ft sections of the course (using Welch’s method with a 256-foot block-size and 252-foot overlap). Note that some of the added statistics for YTC’s courses are presented due to the generally higher variability present in YTC’s automotive test courses. The statistics are defined as follows:

a. RMS (Table 5 and Figure 15).

TABLE 5. YTC COURSE RMS STATISTICS

COURSE

NAME

RMS (in.) 1,024-FT SECTION RMS (in.)

Average Std. Dev. Min Max Std. Dev.

Desert March 1.13 0.03 0.23 3.15 0.56 Kofa High-Speed Gravel 0.11 0.04 0.08 0.27 0.04 Kofa Level Gravel 0.17 0.01 0.09 0.30 0.05 Laguna High-Speed Paved 0.04 0.00 0.02 0.08 0.01 Laguna Hilly Trails 0.56 0.09 0.30 0.91 0.15 Laguna Hilly Trails Short Course 0.50 0.06 0.28 0.71 0.12 Laguna Level Trails East 0.41 0.01 0.27 0.68 0.09 Laguna Level Trails West 0.52 0.02 0.22 0.90 0.15 Laguna Paved 0.07 0.00 0.03 0.18 0.03 Middle East 2.01 0.09 0.25 4.70 1.01 Patton Hilly Gravel 0.51 0.02 0.31 0.85 0.11 Patton Hilly Trails 1.70 0.05 0.56 3.09 0.63 Patton Level Gravel 0.47 0.09 0.21 0.91 0.16 Patton Level Trails 1.02 0.05 0.28 2.08 0.40 Rock Ledge 1.56 0.02 0.42 2.77 0.60

Figure 15. YTC Course RMS values (average, minimum, and maximum from 1,024-foot sections).

(1) Average normal displacement RMS over the entire course (taken over approximately a dozen profiles).

(2) Standard deviation of the course RMS, a measure of the variance over time (taken over approximately a dozen profiles).

(3) Minimum sectional RMS of the course, when broken up into 1,024-foot sections (averaged over at least three profiles).

(4) Maximum sectional RMS of the course, when broken up into 1,024-foot sections;

maximum RMS is significant in that this is the roughest expected terrain (averaged over at least three profiles).

(5) Standard deviation of the 1,024-foot sectional RMS values (a measure of the variance in roughness over the course) (averaged over at least three profiles).

b. International Roughness Index (IRI) (Table 6).

TABLE 6. YTC COURSE IRI STATISTICS

AVERAGE IRI

(inches / mile)

SECTION IRI, 50 MPH

(inches / mile)

50 mph 25 mph 15 mph Min Max Std Dev Desert March 1054 1347 1527 425 2445 361 Kofa High-Speed Gravel 205 496 579 143 332 44 Kofa Level Gravel 318 708 928 206 560 78 Laguna High-Speed Paved 47 61 76 32 64 8 Laguna Hilly Trails 871 1161 1272 461 987 132 Laguna Hilly Trails Short Course 692 915 1011 412 873 116 Laguna Level Trails East 560 661 753 478 797 64 Laguna Level Trails West 623 797 879 409 1146 157 Laguna Paved 73 83 92 38 169 28 Middle East 1664 1759 1905 316 3422 622 Patton Hilly Gravel 747 1226 1518 437 1132 127 Patton Hilly Trails 1423 1190 1118 826 2541 416 Patton Level Gravel 659 811 850 292 1007 150 Patton Level Trails 1049 1198 1156 379 2548 375 Rock Ledge 2200 2942 3296 675 4378 1010

(1) IRI, being a velocity-dependent parameter, is presented at three speeds: the standard IRI speed of 80 km/hr (49.7 mph), a speed of 40 km/hr (24.9) mph, and a speed of 25 km/hr (15.5 mph).

(2) Additionally, the course was broken up into 1,024-foot sections to observe how IRI varied throughout each course. The minimum, maximum, and standard deviation of the 50 mph IRI is presented for the 1,024-foot sectional course.

c. Grade (Table 7).

TABLE 7. YTC COURSE GRADE STATISTICS

GRADE (%)

Max Std. Dev.

Desert March 24.1 3.4 Kofa High-Speed Gravel 2.3 0.9 Kofa Level Gravel 1.8 0.9 Laguna High-Speed Paved 1.3 0.8 Laguna Hilly Trails 23.7 8.3 Laguna Hilly Trails Short Course 11.3 3.6 Laguna Level Trails East 9.5 3.1 Laguna Level Trails West 9.8 2.9 Laguna Paved 1.6 0.8 Middle East 30.8 5.0 Patton Hilly Gravel 26.0 8.0 Patton Hilly Trails 31.7 11.3 Patton Level Gravel 7.2 1.6 Patton Level Trails 16.2 3.0 Rock Ledge 15.4 4.0

(1) Maximum grade in either direction of the course is presented as it represents the most difficult grade a vehicle would negotiate on the course. Since maximum grade is often presented as a system requirement and/or performance limitation, test planning should include consideration that this parameter does not exceed the capabilities of the vehicle system.

(2) Standard deviation of the grade can be considered a measure of the average grade on the course. A higher average standard deviation indicates a higher proportion of hilly terrain.

Hilly courses at YTC are those that are considered to have a standard deviation of the grade greater or equal to 5%. Note that the standard deviation is equal to RMS of the grade, since all courses start and end at the same elevation (zero mean grade).

5.2.1 Laguna Paved (LP).

The Laguna Paved Course is a 3.6 km (2.25-mile), smooth, near-level (0.8-percent grade), single-lane roadway with 152-meter (500-ft) radius turnarounds at each end, surfaced with high-strength asphalt. Performance and durability tests of vehicles are performed on this course.

The wave number spectrum is presented in Figure 16.

Figure 16. YTC Laguna Paved Course WNS.

5.2.2 Laguna High-Speed Paved (LHSG).

The Laguna High-Speed Paved Course is a two-lane, 7.3 km (4.5 mi), smooth, near-level (0.8-percent grade), oval roadway with 2.4 km (1.5 mi) straightaways and 366-meter (1200-ft) banked radius turns at each end, surfaced with high-strength asphalt. Performance and durability tests of vehicles are performed on this course. The wave number spectrum is presented in Figure 17.

Figure 17. YTC Laguna High-Speed Paved Course WNS.

5.2.3 Kofa Level Gravel (KLG).

This course is designed to represent an improved secondary gravel road. The course is a nearly level (~1.0% grade on the straightaways) oval loop 5 km (3.1 mi) long and 12 m (40 ft) wide, with 1.25-mile straightaways and 500-ft radius turns. The course is composed of quarried road construction grade gravel that has been compacted and graded. KLG, as a high-speed secondary road, exhibits a strong washboard effect at wave numbers of 0.3 to 0.5 cycles per foot; the washboard is graded when the wave number spectrum amplitude approaches 5.0 x 10-

3 feet2/(cycles/foot). The wave number spectrum is presented in Figure 18.

Figure 18. Kofa Level Gravel Course WNS.

5.2.4 Kofa High-Speed Gravel (KHSG).

Kofa High-Speed Gravel is an enlargement of Kofa Level Gravel intended for higher speed operations, with 1.3-mile straightaways and 1,300-foot radius turns. KHSG also presents a strong washboard effect during the peak of the washboard cycle. The wave number spectrum is presented in Figure 19.

Figure 19. Kofa High-Speed Gravel Course WNS.

5.2.5 Patton Level Gravel (PLG).

Patton Level Gravel is a 5.8 km (3.6 mi) loop with a surface of quarried road construction grade gravel that has been compacted and graded. The four landforms that the Patton Level Gravel course crosses are dissected fan, alluvial fan, alluvial terrace, and sand dune. The course is used for testing track-laying vehicles and heavy trucks under conditions similar to a secondary graveled road. The course consists of short, straight sections and curves of varying radii, which provide a test of steering mechanisms at medium vehicle speeds. (Note that the WNS spikes at high wave numbers correspond to track imprints in the sandy sections of the course, and do not effect terrain roughness.) The wave number spectrum is presented in Figure 20.

Figure 20. Patton Level Gravel Course WNS.

5.2.6 Patton Hilly Gravel (PHG).

The Patton Hilly Gravel course is an 8.0 km (5.0 mi) course containing many steep slopes.

Course surfaces vary from sand and gravel to exposed bedrock. The two prominent landforms that the Patton Hilly Gravel course crosses are alluvial wash and fan, and other less common landforms consisting of bedrock, dissected fan, and alluvial terrace. The surface cover of the upper 5 cm (2 in.) of all these landforms are mostly subrounded to angular gravel that range from poorly-graded gravel with either silt, sand or clay to well-graded gravel with sand (Unified Soil Classification System (USCS): poorly graded gravel (GP), poorly graded gravel - silty gravel (GP-GM), and clayey gravel to well graded gravel (GC to GW)). The two dominant landforms exhibit Cristobol-Gunsight soil found on the Qf2 alluvial fan surface and Carrizo soil developed in alluvial washes. The wave number spectrum is presented in Figure 21.

Figure 21. Patton Hilly Gravel Course WNS.

5.2.7 Patton Level Trails (PLT).

The Patton Level Trails is a 10.5 km (6.5 mi) loop that traverses level sandy terrain with many bumps to provide a severe test of vehicle suspensions with moderate loads on the drive train.

The three prominent landforms that the Patton Level Trails course crosses are alluvial fans, alluvial plains, and washes. The surface cover of the upper 5 cm (2 in.) of these landforms are mostly subrounded to angular gravel that range from poorly-graded gravel with either silt, sand or clay to well-graded gravel with silt (USCS: GP, GP-GM, clayey gravel to silty sand (GC to SM), poorly graded sand - silty sand (SP-SM), and SP). These landforms exhibit three different soil types, which include Cristobol-Gunsight found on the alluvial fan surface, Superstition- Rositas developed on alluvial plains, and Riverbend developed on alluvial terraces and washes.

The wave number spectrum is presented in Figure 22.

Figure 22. Patton Level…