|
John C. Glennon, D. Engr., P.E.
January 2006 (copyright)
See companion papers
Roadway Hydroplaning – Measuring Pavement Wheel Rut Depths to Determine Maximum Water Depths
Roadway Hydroplaning – A Framework to Determine Critical Pavement Wheel Rut Depths
 Hydroplaning by highway vehicles is a phenomenon characterized by complete loss of directional control when a tire is moving fast enough that it rides up on a film of water and thereby loses contact with the pavement. Although several vehicle, roadway, and environmental factors affect the probability of hydroplaning, a general rule of thumb for rural highways is that hydroplaning can be expected for speeds above 45 mph where water ponds to a depth of one-tenth inch or greater over a distance of 30 feet or greater. In other words, for any set of driver inputs and tire conditions, hydroplaning is a function of water depth.
Roadway collisions involving hydroplaning became more apparent as roadway speeds increased, wider pavements were built, asphalt pavements became more widespread, and greater pavement wear occurred because of greater traffic and heavier loads. Although hydroplaning is a very complex phenomenon, it is known to be associated with several factors. The likelihood of hydroplaning on wet pavements increases with roadway and environmental factors that increase water depth and with driver and vehicle factors that increase the sensitivity to water depth as follows:
Roadway Factors (affecting water depth)
- Depth of Compacted Wheel Tracks
- Pavement Microtexture
- Pavement Macrotexture
- Pavement Cross-slope
- Grade
- Width of Pavement
- Roadway Curvature
- Longitudinal Depressions
Environmental Factors (affecting water depth)
- Rainfall Intensity
- Rainfall Duration
Driver Factors (affecting sensitivity to water depth)
- Speed
- Acceleration
- Braking
- Steering
Vehicle Factors (affecting sensitivity to water depth)
- Tire Tread Wear
- Ratio of Tire Load to Inflation Pressure
- Vehicle Type
The intention of this paper is not to minimize the importance of factors such as tire tread depth, tire pressure, and vehicle speed in contributing to the probability of hydroplaning but rather is to point out to highway designers that a significant portion of the driving population either drives vehicles with barely legal tire tread depth, or is hardpressed to maintain optimum tire pressure, and/or is often uninformed about reasonable reductions in normal speed during wet weather, particularly on isolated roadway sections that are prone to hydroplaning.
 The highway engineering community showed great concern about hydroplaning in the 1960's and early 1970's. With 70-mph nationwide speed limits and with wider pavements being built, particularly on Interstate highways, hydroplaning accidents increased dramatically. As a partial reaction to these trends, research projects were initiated to study the problem. Some of the more significant work on highway hydroplaning was done by Bob M. Galloway and others1,2,3. The results of this research appear to have been largely overlooked since 1973, when nationwide speed limits were reduced to 55 mph and hydroplaning became less of an issue. Now with nationwide speed limits back at 70 mph, hydroplaning accidents seem on the rise and new research 4 seems to reiterate the earlier recommendations.
A highway agency concerned with minimizing the occurrence of hydroplaning can only exert minimal control over driver, vehicle, and environmental factors. It can, however, identify and correct highway locations that show a propensity for hydroplaning. The major factor that can be controlled through roadway and pavement design is pavement water depth.
Although several kinds of highway situations produce the opportunity for hydroplaning, the most prevalent places are where water is allowed to pond such as at the bottom of sag vertical curves, at horizontal curve transitions, and on pavements with noticeably compacted wheel tracks or ruts. These locations are usually corrected by asphalt mix overlays designed and constructed to increase the cross-slope, eliminate rutting, improve the texture of the surface, and provide a surface resistant to wear and compaction.
When hydroplaning occurs, more than one factor often contributes to the loss of control. More than not, hydroplaning occurs during a moderate to heavy rainfall on roadways with 50-mph speeds or higher. But other factors also need to be present either to create a sufficient water depth or to reduce the resistance to lesser water depths. Combinations of roadway factors which are particularly susceptible to hydroplaning are:
- Higher grades draining downhill in wheel ruts to a sag with little or no cross slope.
- Wide pavements on grade with little or no cross slope.
- Pavement surfaces with little texture and little or no cross slope.
- Roadway curve transitions on wide pavements with little or no cross slope.
- Roadway curve transitions on steeper downhill grades with little or no cross slope.
- Deeper wheel ruts with little or no cross slope.
- Long downhill grades where water is dammed along an overgrown turf shoulder and builds up until it reaches a highway curve transition where it flows in sheets across the roadway.
- Other combinations of the above.
 Research indicates that the potential for hydroplaning is sensitive to water depth. The major roadway factors affecting water depth are pavement texture, roadway cross slope, roadway grade, pavement depressions, rutted wheel tracks, and inadequate roadside drainage facilities.
The pioneering research was first published in 1971 by Gallaway and Rose1 and further expanded in 1979 by Gallaway, et. al, 2. Although much of their focus was on defining adequate pavement texture, one of their major findings was that pavement cross slope was a dominant factor in draining water from the pavement surface. They recommended that 1.5% should be the minimum admissible cross slope, and that most pavements, particularly wider ones, should have a 2% cross slope. This research also emphasize the importance of drainage path length as a function of both cross slope and longitudinal grade as defined by the following equation:
L2F = (W2C)[ 1+ (SG/SC)2]
Where
LF= Drainage path length, in feet
Wc= Pavement drainage width, in feet
Sc= Cross slope, in ft. / ft.
Sg= Longitudinal grade, in ft. / ft.
As an example, consider a 12-foot drainage width, W, with a zero longitudinal grade, Sg. For this roadway, with any cross slope, Sc, the drainage path length is only 12 feet. But, if the grade is 5% and the cross slope is 1% on this 12-foot pavement width, the drainage path length now becomes 60 feet. Now if the pavement drainage width is increased to 36 feet, the drainage path length becomes 184 feet. With the 36-foot drainage width and the 5% grade, if imprecise construction tolerance allows a section of roadway to be built with only a 0.5% cross slope, the drainage length now becomes 362 feet. It is these longer drainage path lengths, sometimes combined with pavement depressions or wheel ruts that are very susceptible to hydroplaning on higher speed roadways.
In 1999, Huebner, Anderson, and Warner4 gave further definition to the relationship between drainage path length and water depth. They proposed design guidelines based on exercising the PAVDRN 5 computer program They stated that for a given pavement geometry, in a steady-state condition, the longest drainage path length produces the deepest water film thickness and, hence, the critical path for hydroplaning. In other words, by minimizing the the drainage path length, the potential for hydroplaning can be minimized, all other factors being constant. Therefore, for any combination of rainfall intensity, roadway grade, and pavement texture, the potential for hydroplaning can be controlled through the proper design of pavement cross slope. The subject research shows that for nominal texture depths and moderate rainfall intesity, hydroplaning can occur at only 50 mph if the cross slope is only 1%. The minimum cross slopes recommended by this research ranged from 1.5% for 45-mph roadways to 8.0% for 65-mph roadways.
From the above discussion, lack of adequate roadway cross slope appears to play a major role when roadway conditions are significant in creating hydroplaning. Therefore, it would seem informative to look at roadway cross slope design standards.
The American Association for State Highway and Transportation Official (AASHTO), formerly named the American Association of State Highway Officials (AASHO) has been recognized for over 60 years as the authoritative source for highway design standard. But not until 1954, did AASHO6 ever give any specification to cross slope. The following is the complete discussion of cross slope in the 1954 AASHTO Policy (similar language is found in later AASHTO Policies7:
Normal Cross Slope
Two-lane and wider undivided pavements on tangents or on flat curves have a crown or high point in the middle and slope downward toward both edges. The downward cross slope may be a plane or curved section or a combination of curve and plane. With plane cross slopes there is a cross slope break at the crown line and a uniform slope on each side. Curved cross sections usually are parabolic, with a flat but rounded surface at the crown line, and increasing cross slope toward the pavement edge. Since the rate of cross slope is variable, the parabolic section is described by the "crown height", the vertical drop from the center crown line to the pavement edge.
On divided highways each one-way pavement may be crowned separately, as on 2-lane highways, or it may have a unidirectional slope across the entire width of pavement, almost always downward to the outer edge.
Since a large percentage of highways are on tangent or flat curve alignment, the rate of cross slope for this condition is an important element in cross section design. Pavement superelevation on curves is determined by speed-curvature relations as given in chapter III, but the cross slope or crown on tangents or on flat curves is complicated by two contradictory controls. A reasonably steep lateral slope is desirable to minimize water ponding on flat sections of uncurbed pavements due to pavement imperfections or unequal settlement [emphasis added]; with curved pavements a steep cross slope is desirable to contain the flow of water adjacent to the curb. On the other hand, pavements with steep cross slopes are objectionable in appearance and may be annoying and uncomfortable in operation. Hazard may attend driving on steep cross slopes on tangents due to the tendency of vehicles to veer toward the low edge of pavement. Vehicles are built with symmetrical caster-camber arrangement for pavements that are level transversely, and cross slopes up to 1/4-inch per foot are barely perceptible as far as effect on vehicle steering is concerned. Cross slopes of 1/4 to 1 inch are noticeable in steering but not significant in effect. The latter rate requires a conscious effort in steering and would increase the proneness of lateral skidding when vehicles brake on icy or wet pavements, and even on dry pavements when stops are made under emergency conditions.
The presence of high winds may significantly alter the effect of pavement crown upon steering. In rolling terrain with alternate cut and fill sections or in areas alternately forested and cleared, any substantial cross wind produces an intermittent impact on a vehicle moving along the highway and affects its steering. In areas where such conditions are likely, it is desirable to avoid high rates of cross slope.
On high-type 2-lane pavements, crowned at the center, the rate of cross slope for each lane normally should be 1/8- to 1/4-inch per foot (or 0.01- to 0.02-foot per foot). Where two or more lanes are inclined in the same direction on multilane pavements, each successive lane outward from the crown line preferably should have an increased slope. The lane adjacent to the crown line should be pitched at the normal minimum slope and, on each successive lane outward, the rate should be increased by about 1/16-inch per foot (0.005-foot per foot). Cross slopes greater than 1/4-inch per foot should be avoided on high-type surfaces.
Use of cross slope steeper than 1/4-inch per foot on high-type, high-speed pavements with a central crown line might be questioned for operational reasons. In passing maneuvers, drivers must cross and recross the crown line and negotiate a total "roll-over" of more than 1/2-inch per foot, or a cross slope change of over 4 percent. The reverse-curve path of travel of the passing vehicle causes a reversal in the direction of centrifugal force, which force is further exaggerated by the effect of the reversing cross slopes. Trucks with high body loads are caused to sway from side to side when traveling at high-speed, at which time steering control may be difficult. For this reason crown lines on such pavements determined by cross slope rates in excess of 1/4-inch per foot are noticeable and might be hazardous at high speed.
On pavements designed with rounded crown section, the "roll-over" is not apt to be objectionable. In an operational sense, rounded crown sections are good designs for 2-lane tangent pavements. In a design and construction sense, they cause difficulty in warping from a tangent section to a superelevated section. On pavements of 3 or more lanes the parabolic or circular crown section should be used with caution. Any such curved section determined to have a desired slope across the center lane, or lanes, usually results in an undesirably steep slope across the outer lanes. For example, an undivided 4-lane highway has a parabolic or circular crown section designed so that the inner lane edge is 0.12 foot below the centerline. If lanes are 12 feet wide the average rate of cross slope on the outer lane is 0.03, an amount questionably high in an operational sense.
In some cases on multilane highways a rounded crown section is used for the central two lanes, with a plane section on the outer lanes having a cross slope rate of about 0.015 to 0.02. Here the outer lane cross slope is made the same or slightly steeper than that at the end of the curved section.
The above cross slope rates largely pertain to high-type surfaces. Without curbs these types qualify for minimum rates of cross slopes on tangents due to the excellent control upon construction methods and smoothness of the finished surface. For intermediate and low types of pavement, a greater cross slope should be considered. Table IV-1 shows a range of values applicable to each type of surface.
Table IV-1 On intermediate-type surfaces the running speeds may not be much less than those on high-type pavements because of the generally lower volumes and fewer freight vehicles. A somewhat greater cross slope rate for intermediate-type surfaces is used because of the likelihood of less accuracy in construction procedures and greater proneness to settlement and warping of the cross section. On these surfaces any longitudinal grade will assist drainage in the event of a "rutting'" type of settlement; otherwise a greater than normal cross slope should be used. Also, the intermediate-type surfaces frequently are of coarser texture, which tends to retard the runoff of water.
Low type surfaces---loose earth, broken stone, or gravel---require a yet greater cross slope on tangents to prevent the absorption of water into the surface, and due to greater surface irregularities. On highways with these surfaces, vehicle speeds generally are low so that, in an operational sense, no sacrifice is made.
Where pavements are designed with outer curbs the minimum rates in table IV-1 are questionably low because of the likelihood of a sheet of water over a substantial part of the traveled way adjacent to the curb. For any rate of rainfall the width of traveled way inundated varies with the rate of cross slope, roughness of gutter, frequency of discharge points, and longitudinal grade. A cross slope rate greater than 1/8-inch almost always is desirable and, in some cases, more than 3/16-inch is needed to limit inundation to about half of the outer traffic lane. A cross slope of 3/16-inch per foot is suggested as a practical minimum for curbed high-type pavements and 1/4-inch per foot for intermediate-type pavements. Curbs with adjacent steeper gutter sections permit the use of lesser rates of pavement cross slope.
In summary, small cross slopes are needed on uncurbed pavements to assist drainage in event of uneven settlement. Somewhat greater cross slopes are needed on intermediate- and low-type surfaces. Curbed pavements require greater slopes to reduce water sheeting on the traffic lane adjacent to the curb. Satisfactory design for curbed pavements requires careful analysis of all the elements pertinent to surface water drainage on the particular highway.
expert, worktIn the 1984 AASHTO Policy and later additions7, a change was made to 1.5% as the minimum recommended cross slope on high-type pavements
In analyzing several roadway paving projects where lack of adequate cross slope became an issue in litigation, my observations are that neither the paving contractors nor the resident engineers seem to have any distinct appreciation for the importance of cross slope. As a result, either for an initially-constructed pavement or for one that's been overlayed one or more times, defective sections of less than 1% cross slope have been built. And, as discussed earlier, the hazard of these sections is greatly amplified by contiguous roadway grades, roadway curve transitions, pavement wheel ruts, drainage onto the pavement, and/or sag vertical curves. For some repaving projects, the contract often calls for matching the existing cross slopes, regardless of whether cross-slope problems already exist. Other repaving contracts have no mention whatsoever of cross slope. If a paving contractor is simply asked to overlay with 2 inches of asphalt, but is not given any cross-slope control requirements, that contractor will generally lay 2 inches of asphalt without any regard for existing deficiencies in the cross slope. Then too, resident engineers or paving inspectors are not generally inclined to challenge paving contractors when the cross slope is found insufficient because the contract lacks any specifications with which to require work stoppage or repaving.
Although AASHTO appears to be concerned about pavement drainage, that concern is not strongly expressed, particularly in the cross slope specifictions where it allows a mere 1.5% of cross slope for high-type pavements. This designation of high-type is not directly defined by AASHTO, but connotes a pavement appropriate for high-speed roadways; a strong pavement with close construction tolerances on cross slope. But the achievment of this kind of pavement may be something of a pipe dream. Not uncommon is to find a high-speed roadway, particulary one with an asphalt pavement, that was designed for a 1% cross slope, but was built with closer to a 0.5% slope. As is emphasized in the previously referenced research, these lower cross slopes can be critical in creating water depths susceptible to hydroplaning.
Obviously, based on research findings and in consideration of pavement irregularities (settlements, wheel ruts, etc.) that seem all too common, AASHTO should consider recommending 2-2.5% minimum cross slopes to minimize the propensity for hydroplaning, particularly for high-speed roadways.
For repaving contracts, cross-slope specifications should be written into the contract and paving inspectors should be required to ensure that the cross slope is built to specification. When an existing cross-slope deficiency could create difficulty in satisfying the contract as bid, then special contract provisions should be included to mill or patch that section in order to create a reasonable cross-slope underlayment. When existing cross-slope deficiencies are found during paving, change orders should be instituted to correct the problem.
1. Gallaway, B.M., and Rose, Jerry G., The Effects of Rainfall Intensity, Pavement Cross Slope, Surface Texture, and Drainage Length on Pavement Water Depths, Texas Transportation Institute, Research Report No. 138-5, 1971.
2. Balmer, G.G., and Gallaway, B.M., Pavement Design and Controls for Minimizing Automotive Hydroplaning and Increasing Traction, American Society of Testing and Materials, ASTM STP 793, 1983.
3. Gallaway, B.M., et. al., Pavement and Geometric Design Criteria for Minimizing Hydroplaning, Federal Highway Administration, Report No. FHWA-RD-79-31, 1979.
4. Heubner, R. Scott, Anderson, David A., and Warner, John C., Proposed Design Guidelines for Reducing Hydroplaning on New and Rehabilitated Pavements, National Cooperative Highway Research Program, Research Results Digest Number 243, September 1999.
5. Heubner, R.S., PAVDRN Computer Program with User's Guide, NCHRP Project 1-29, " Improved Surface Drainage of Pavements", Transportation Research Board, June 1998.
6. American Association of State Highway Officials, A Policy on Geometric Design of Rural Highways, 1954.
7. American Association of State Highway and Transportation Officials, A Policy on Geometric Design of Highways and Streets, 1984,1990,1994, and 2001.
About the Author
Dr. John C. Glennon is a traffic engineer with over 45 years experience. He has over 120 publications. He is the author of the book "Roadway Safety and Tort Liability" and is frequently called to testify both about roadway defects and as a crash reconstructionist.
RESUME OF JOHN C. GLENNON
Book , books,pavement edge drop expert, pavement edge drop off expert, guardrail expert, work zone safety expert, construction zone safety expert, roadway hydroplning expert, traffic engineering expert, traffic sign expert, traffic signal expert, pavement marking expert, highway safety expert
|
