Causes for blowback.

Fig. 1 shows a typical extruder screw. Numerous extruder-shaft failures, broken gear teeth, drive spline fretting and other failures had occurred on this motor-gear-extruder combination over the years. Different causes were thought to explain these failures, and the unit was rebuilt and put back into service many times. Usually, improper startup techniques with cold product left in the extruder and “bumping” the motor to free the product were given as the primary failure causes.

Fig. 1. Extruder blowback model.

At the request of the operating site, an analysis was done to determine what effect blowback may have had on the system torque and, thus, the loads acting on the extruder screw, gears and bearings. An analysis of the spring–back of the screw during blowback was specifically requested.

Model.

The model was developed on the idealized system, as shown in Fig. 1.¹ As the motor turns the screw, the shaft torque, T_s (in-lb), winds up the long shaft of length, L_s (in.), and diameter, D_s (in.), and twists the screw u radians. The shaft stays twisted until the blowback occurs.

Since not much was understood on what occurs during blowback, it was assumed that, under the worst-case scenario, the torque twisting the shaft goes to zero instantaneously, meaning that no processing is being done even though the shaft is still revolving. The product just slips without friction in the barrel housing enclosing the screw. This is shown in Fig. 1. The shaft snaps back due to the unleashing of the potential energy of the wound-up shaft when there is no torque to keep it twisted. The shaft then unwinds until the potential energy is used up. It then cycles at the fundamental torsional natural frequency of the screw until the product starts to process again, meaning torque is reinstated. The buildup takes place over a much longer period, and it is assumed that the blowback occurs in only one cycle.

Root-cause possibilities.

This effect was calculated to be 25% of the mean torque value of 150,000 in.-lb. Since it is not a long-term cyclic event, it is probably not the root cause of the failures.¹ Because the extruder is not producing product during the event, it won’t be allowed to continue the cycling without a correction. Many repeated blowbacks could eventually cause a problem if allowed to continue. It was, therefore, recommended to install a continuous torque monitoring device to capture the blowback effect and the torque it produced.

Although these calculations did not point out the precise cause of the failures, the logical thought processes eliminated a number of possible causes. It helped provide a plan forward if failures continued. One solution was to rewrite the operating procedures to eliminate the possibility of blowback. HP

LITERATURE CITED

¹ Sofronas, A., Case Histories in Vibration Analysis and Metal Fatigue for the Practicing Engineer, John Wiley & Sons, to be published in late 2012.