A Policy on Grade Separations for Intersecting Highways was the logical sequel to the two policies on intersections. It was the final one of the seven geometric design policies, and it was completed in 1944.
The title of this policy is somewhat misleading because it treated not only the separating structure, but actually covered in great detail the design of grade separated traffic interchanges. Design data were grouped in three categories: (1) Structures and approaches, (2) ramp arrangements, and (3) ramp design.
One important design element not treated in detail by these several design policies was the relation between curvature, superelevation, and design speed.
Curves
By 1930 most States were superelevating all curves except those of very long radius; however, there was little uniformity as to superelevation rates, just as there was little consistency in minimum radii for curves. Empirical controls were generally applied, with maximum cross slopes of about 10 percent.
A BPR study in 1920 focused attention on the necessity of a tire-pavement friction factor for superelevation design and also noted that a spiral curve or transition length was needed at each end of a curve section for the change from normal pavement crown to a superelevated section.[1]
One of several conclusions resulting from a long series of studies on tire-pavement skidding relationships conducted by the Iowa State University was that the maximum permissible speed as used in designing curves should not exceed that for which a useful side-friction coefficient of 0.30 was required to counteract centrifugal force. This conclusion was reached in 1934.[2]
In 1935, the BPR collected data from drivers across the country operating their own vehicles on curves of known radius and superelevation by asking them to report the speed at which they began to feel a side pitch outward. Analysis of these data resulted in a new design premise that safe operation on curves would be attained when the superelevation was sufficient to counteract centrifugal force for three-quarters of the expected speed, relying on side friction to supply the remaining horizontal resistance up to a maximum side friction factor of +0.16 at 60 m.p.h.[3] The speed concerned was advocated to be the “assumed design speed,” which was to be used as a basis for coordination of all alinement and geometric design values. The side friction factor to be used in calculating the minimum radius or the maximum rate superelevation did not enjoy the same degree of finality as the speed criteria, although the values used were apparently on the safe side. Research is continuing at the present time in an endeavor to discover pavement materials and construction methods to improve the skid resistant qualities of pavements.
In 1937 the BPR completed a highway curve design manual (later published as Transition Curves for Highways, 1940) embodying the above proposals. The manual presented data for 10 m.p.h. increments in design speed for all curve design features—curve radii, superelevation, curve widening and transition (spiral) curves. These concepts and design details were greatly needed and soon gained wide use, thus stabilizing, to a large extent, curve design practices throughout the country and nullifying for the time being any necessity for the AASHO Committee to concentrate its efforts on this subject.
Gradients
Common logic has always dictated that, from the traveler’s point of view, the most desirable route between two points is the one that is straightest and has the least rise and fall. In the days of wagon roadbuilding, circuity and indirection of alinement often had to be substituted for directness in order to obtain a suitable profile. The wagon roads were later converted to motor highways, frequently with little change in alinement and grade despite the fact that automobiles and trucks could negotiate grades steeper than the 4 to 6 percent commonly used for horsedrawn vehicles. This was done in the interest of economy, although the roadbuilders would have preferred a better alinement.
As the highway network expanded and more roads were built on new locations, particularly during the late 1920’s and the 1930’s, advantages were taken of the better gradeability of motor vehicles, and grades as steep as 9 percent were used sometimes to provide a straight alinement. Design with long tangents became commonplace and road distances were shortened by hundreds of miles in the aggregate. A BPR summary of practice in 1929 stated:
On main-line highways it is customary to adopt a maximum grade of 5 percent in gently rolling country and 7 percent in rough country, but it is no longer considered good practice to resort to sharp curvature in order to avoid grades steeper than 7 percent. If local conditions permit either a 7 percent grade with a sharp curve or a short 9 percent grade with a wider curve, the latter design is thought to be the better practice because it is safer for modern motor traffic.[4]
In the rolling terrain commonly encountered in the midwest and far west where roads were developed on section line locations, this type of design resulted in many hundreds of miles of “roller coaster” highway profiles. Design of profiles with frequent grades over 5 percent tended to minimize earthwork quantities, with only shallow cuts and fills. As traffic volumes, speeds, and truck loadings increased, the deficiencies of short sight distances and high-downgrade speeds proved that this type of profile was somewhat hazardous.
The alternative was a profile design of a railroad grade type, that is, long, easy grades with long flat vertical curves in conjunction with long horizontal tangents connected by gentle curves. Prior to 1930, any extensive mileage of such highway construction would have been out of the question because of the large earthwork quantities and attendant high costs.
The rapid mechanization of earthmoving equipment that began in the early 1930’s revolutionized construction methods and made feasible the construction of highways of a type that had heretofore existed only in the fanciful minds of design engineers. Tractor-drawn self-loading scrapers with capacities of 12 cubic yards came upon the scene for the first time. Pneumatic tires had been improved to such an extent that they could be used on heavy trucks, thus affording sufficient flotation to operate on newly placed
394
- ↑ A. Luedke & J. Harrison, Superelevation and Easement as Applied to Highway Curves, Public Roads, Vol. 3, No. 31, Nov. 1920, pp. 3–12.
- ↑ R. Moyer, Further Skidding Tests With Particular Reference to Curves, Proceedings, 14th Annual Meeting, Vol. 14 (Highway Research Board, Washington, D.C., 1934) p. 129.
- ↑ J. Barnett, Safe Side Friction Factors and Superelevation Design, Proceedings, 16th Annual Meeting, Vol. 16 (Highway Research Board, Washington, D.C., 1936) pp. 72, 73.
- ↑ A. Bruce, The Effect of Increased Speed of Vehicles on the Design of Highways, Public Roads, Vol. 10, No. 1, Mar. 1929, pp. 15, 16.
