To enable higher computational power and speed with a smaller footprint, an increased number of semiconductor chips are packaged into smaller areas within electronic packages resulting in high heat flux densities. To dissipate heat efficiently and keep the junction level temperatures below a certain threshold required for reliability, thermal interface materials (TIMs) are leveraged at various interfaces within the electronics package. However, most TIMs, in particular thermal greases, degrade with time due to frequent powering on and off of the device and the associated thermal and thermomechanical cycling. Specifically, the material either flows out from the interfacial region (i.e, pump out failure) or the "phases'' separate (for example, the high thermal conductivity filler particles separate from the polymer matrix or matrix dries out). Typical power cycling experiments designed to mimic operational conditions require very long experiment times. Inspired by previous accelerated test systems, in this work, we develop an accelerated testing facility with forced mechanical cycling using a high-precision linear stage, in addition to heat flux control, to induce degradation at faster rates. Specifically, we designed and built an experimental setup capable of squeezing thermal greases at a constant rate until a fixed pressure is achieved. This facilitates squeezing the TIM at controlled rates to specified pressures to achieve different bondline thicknesses (BLTs). Then for mechanical cycling, the BLT is perturbed with controlled amplitudes and rates of displacement. Throughout the cycling, controlled temperature gradients are established using a custom-designed heater (that enables simultaneous heating and thermal imaging) and a forced convection heat sink. Here, we evaluate the degradation of DOWSIL TC-5550 thermal grease subjected to input heat flux of 6.1 W cm^{-2} and average squeezing rate of 60 µm s^{-1}. To quantify the evolution of the thermal resistance and local variations in the thermal resistance, steady-state infrared thermal maps are taken at fixed intervals during the mechanical cycling with controlled temperature gradients across the system. The initial experimental results suggest that DOWSIL TC-5550 degrades due to void formation within 6 hours when mechanically cycled, whereas a traditional power cycling test takes several days to induce degradation. More research is needed to understand how the mechanisms that lead to failure in accelerated testing are related to degradation observed in application.