Non-Linear Finite Element Analysis of a Pipe Denting Problem During a Pipe Transfer Operation

This paper investigates the problem of a pipe rolling down a transfer skid and becoming permanently dented as it hits the stops at the end of the skid. See Figure 1. The transfer skid stops under consideration had been successfully used in other applications. However, in this application there was an increase in distance between stations connected by the transfer skid. Thus, as the pipe rolled down the skid it built up significant kinetic energy, which needed to be dissipated upon impact with the skid stop. Unfortunately in this case, the skid stops did not always absorb enough energy to ensure integrity of the pipe. Consequently, certain sizes and grades of pipe experienced denting as they would impact the stops. The skid stop and the pipe must absorb the total kinetic energy possessed by the pipe just before impact. The energy absorbing characteristics of both the skid stop and pipe were developed using a static method. For the skid stop a physical test was performed to obtain the contact force versus deflection curve. For the pipe a finite element analysis was conducted to determine the contact force versus deflection curve. In the finite element model the effect of local yielding of the pipe material was incorporated into the analysis. The energy absorbed by each component was estimated as the area under the contact force-deflection curve and for each component the energy absorbed versus contact force curve was developed. Combining these two results gives the total system energy absorbed by both the pipe and skid stop as a function of contact force. This is compared to the total energy in the pipe just before contact to determine the actual maximum contact force and the actual energy absorbed by each component. The energy that could be elastically given back to the system was also obtained from the model. These results were compared to actual field measurements of dent size and pipe rebound height. The comparisons proved the validity of the model.