Tests of beams made of portland cement concrete have shown that such a beam can be broken by frequent and rapid application of a load barely more than half as great as the load required to cause rupture in single application. So, it is known that concrete is a material which, like many others, suffers fatigue under repeated stress. Concrete pavements undoubtedly suffer such fatigue under the repeated application of vehicular loads. Without question, the load that a pavement will withstand in frequently repeated application is not as great as the heaviest load it would support once or infrequently applied. Precisely the degree of such fatigue of the road slab, i. e., what load repeated in what frequency on a slab of given dimension will rupture the slab, has not been determined, and the desirable eventual determination will be extremely difficult. Meanwhile, the great variety of conditions under which concrete roads exist and are built precludes even an approximate generalization of the effects of fatigue.
Cracking of concrete
It is a matter of common observation that cracks form progressively in concrete pavements throughout their life. Many of these cracks result from causes other than vehicular loads. Some, undoubtedly, are caused by loads, but a load-induced crack is rarely noticed and may be unnoticeable at the moment of its formation. Hence it is practically impossible to impute observed cracking to particular load causes. The most that can be said with assurance is that concrete pavements of presently designed dimensions are cracked by vehicular loads, that they are probably cracked by loads less than those which theoretically they should support in single application, and that ordinary prudence requires safe allowance for the possible effects of repeated application in the limitation of axle loading.
Joint pumping
Another phenomenon associated with concrete pavement, to which the axle loading of vehicles is known to contribute, is the occurrence known as joint pumping. Some jointing of concrete pavements is unavoidable. If the joints are omitted, cracks form as a result of temperature contraction. Where either joints or cracks exist, surface water may penetrate to the soil subgrade, or water may reach the sub- grade at the joints in other ways. When a vehicle passes over the pavement at a joint the edges of the slabs forming the joint are bent slightly downward by the load. If, then, there is free water in the vicinity of the bottom of the joint, the downward pressure of the bending slabs tends to force it upward and out of the joint onto the surface of the pavement. If the subgrade soil is of such character that it enters into suspension in the water that collects, some of the soil will be pumped out with the water. By such pumping action, frequently repeated, portions of the subgrade may be gradually removed, leaving the pavement slab without subgrade support in the vicinity of the joint. Lacking such support, the pavement may crack near the joint under a passing load heavy enough to overtax its unsupported strength.
Joint pumping, as will be seen from this description, is not caused by vehicle load alone. Detrimental pumping occurs only in the
