The Fatigue Phenomenon

Abstract

During testing, we noted the influence of cyclic loads on tie-rod load losses. In a number of tests, we measured the deformations induced in the tie rods by the loads of railway trains. During direct tensile tests, we observed strand failures in tie rods subjected to these loads.

The fatigue phenomenon is well known in mechanics, but less so in civil engineering.

A simplified presentation could be the following: a steel bar that breaks in tension for a force of one ton will break under a cyclic force of 100 kg performed 10E6 or 10E7 times.

During a static tensile test, a video made for the operators showed the effects of the load transmitted by the axles during the passage of a train.

For each axle, measuring devices recorded a load-induced deformation (Fig. 13.1).

Fig. 13.1

Source Rincent BTP—Recife

Transmission of the load.

Depending on the position of the applied load relative to the wall, the second row of tie rods generally has the lowest force values, this is especially true for transmitted dynamic loads. These overloads are usually applied for short periods of time (Figs. 13.2, 13.3 and 13.4).

Fig. 13.2

Measured tension forces (tons). Force in tons deformations in mm

Fig. 13.3

Static test on a 40 years old tie rod not subjected to dynamic overload. Force in tons deformations in mm

Fig. 13.4

Static test of a tie rod receiving a dynamic overload

The behavior changes at 32.5 tons and above, and the tie rod in the fourth row does not support a higher load.

The main difference is that the tie rod in Fig. 96 only receives static loads from the ground, whereas the tie rod in Fig. 97 received dynamic stresses from the passing trains in addition to the ground thrust.

The induced mechanical fatigue is a very important factor, and the behavior of the tie rod under the loads cannot resist a strong mechanical action.

For an assumption of 10 trains of 100 wagons per day with 4-axles for 200 days per year for 40 years, the result is 3,2E7 dynamic overloads.

This calculation, which is an evaluation, must be compared with the Wölher curve which is used in mechanics courses on fatigue (Fig. 13.5).

Fig. 13.5

Wölher curve

The displacement versus load curve of a railway train shows that the induced vertical deformations are mil8imetric (Fig. 13.6).

Fig. 13.6

Source Laurent Soyez Thesis

Vertical displacements, railway convoy.

Harbors are another location where cyclic loads are applied daily on anchors.

Deep-water tie rod tests are often conducted during high tides, as the low tide level allows access to the usually submerged tie rod heads.

At this location, the tidal range was 6 m.

Tests carried out on the same day at different times on the same anchor show a difference in force of 15 tons for an average force of 100 tons. This phenomenon, related to the thrust of the water on the sheet pile curtain, is well known. These cyclic loads of varying amplitudes are part of the anchors loads.