The Case of the Flea on the Elephant’s Back

Synopsis:
A hydraulic cylinder with 275-ton lifting capacity is destroyed by a failed component weighing less than 2 grams.

Background:
One of two 17’ long main lift cylinders from the largest of truck cranes was in routine service for repair of a leaking gland. After repair the cylinder was connected to a test bench for leak testing and oil was pumped into the cylinder, cycling the cylinder in and out to eject air. On the third cycle extending the cylinder, it chattered and then jammed.

Assignment:
TI was requested to determine the exact cause of cylinder failure and the possibility of repair.

Analysis:
The rod was tubular rather than solid in order to save substantial weight. It was determined that the 1” thick wall of the rod had collapsed inward 4.6” at a cylinder extension of 10’.

The cylinder bore was found to be out of round and deeply scored from the distorted rod and thus the cylinder was a total loss. A new cylinder was in stock in Germany at a cost of $100k and air freight was necessary at an additional $20k. The piston diameter was 14.8” and the rod diameter 13.4” and thus in extension the cylinder could develop almost six times the force as when retracting. The cylinder was connected to hydraulic hoses of the test bench with quick connect fittings. The male and female fittings employ spring loaded poppet valves to prevent loss of oil when disconnected. When the fittings are connected, the poppets meet nose-to-nose and each pushes the other partially open against the spring pressure. It was discovered that the bench-side rod end fitting had its poppet cocked and jammed well inward of the seat after an oil-hammer event that occurred when a pocket of air escaped through the fitting.

This allowed the cylinder-side fitting poppet valve to close, thus preventing further flow of oil out of the cylinder on the rod end. While it took almost no oil pressure to extend and retract the cylinder since it was not connected to any load, when the poppet valve closed pressure rapidly escalated. With the almost six times greater area of the piston on the extend side, the oil pressure on the retract side was almost six times greater than the oil pressure delivered by the test bench. The test bench was capable of delivering 3,200 PSI and thus the pressure on the retract side of the cylinder could have reached 18,000 PSI. Tubes have a far greater ability to resist pressure on the inside than on the outside and this is why the tubular rod collapsed inward rather than the cylinder bursting outward.