Fatigue Testing Rig

I designed this testing rig as a TA for a graduate-level mechanical design course at Carnegie Mellon University. The rig cycles 9 student parts for 1 million cycles at 64 ft-lbs of torque each for a total of 576 ft-lbs of torque on the rig itself.

Student teams were tasked with designing a component capable of withstanding 64 ft-lbs of torque for 1 million cycles of loading. One end of their part was clamped onto a smooth 2” shaft while the other end of their part was attached to a spring. My rig would oscillate the shaft through 23 degrees of travel, which resulted in the students’ parts stretching and relaxing the spring, generating the required torque.

The components that bear most of the load are fabricated from steel for its superior fatigue resistance compared to aluminum. The 80/20 frame that surrounds the rig is for support and does not sustain heavy loading. The rig was designed to be robust while being relatively easy to fabricate and adhering to the budget.

We borrowed a 208 V, three-phase motor to use for the test which was capable of generating 30 ft-lbs of torque. The rig uses a two-stage rope-and-drum transmission with a total gear ratio of 32:1 with 1:2 in the first stage and 1:16 in the second stage. A robust, commercially available gearbox would have been easier, but was out of our budget.

The rope used in the transmission was 3mm Vectran rated to 1900 lbs, which should have been strong enough for the project. However, during the testing of the students’ parts, the rope in the second stage of the transmission kept breaking. I tried ropes of larger diameter and different materials, but those too, kept breaking. Eventually I discovered that as the ropes were winding and unwinding around the pulley, they were squishing and unsquishing, resulting in the ropes heating and eventually failing. After switching to flat nylon webbing, which minimized such squishing, the rig was able to complete the requisite 1 million cycles.