Date of Award

Spring 2016

Project Type


Program or Major

Mechanical Engineering

Degree Name

Master of Science

First Advisor

Chris M. White

Second Advisor

Diane Foster

Third Advisor

May-Win Thein


This study pertains to the design and evaluation of a force-platform type instrument comprised of three strain gauge load cells, intended to measure the rate of change in mass and centroid position of an object undergoing rapid ablation by heated flow of air. The strain gauges are affixed to a steel frame, oriented in a triangle beneath a 25 X 20 cm rectangular platform, into which is machined a 9 X 9 regular grid for the purpose of static calibration. An algorithm derived from first principles is used to convert the distribution of the applied load among the strain gauges into a measurement of the location of the centroid of the applied load. The fundamental concepts of image correction are applied to generate a set of continuous functions (referred to as "mapping functions") that remove inherent bias error in centroid measurements, resulting in an average reduction in RMS error of 96% for a stationary load. The bias error, while repeatable for a constant load, varies with load, thus each mapping function is unique. Loads moving along a 25 cm straight track exhibited an average linearity error of 0.08%, however the bias is shown to increase after correction in some locations, likely due to the spatial variability of repeatability in position measurements.

Frequency response of the measurement system was observed by applying a pulse at fixed frequencies by means of a rotating camshaft affixed to the frame of the balance, and powered by a DC motor. Motor speed is regulated via speed controller, and measured with a rotary encoder. The dominant frequency in the energy spectrum matched well with the applied frequency measured by the encoder, with an RMS error of only 0.15 Hz. The frequency response reveals that the sensor system is acceptable for low frequency measurements, making it appropriate for ablation measurement, as ablation is expected to be a low frequency phenomenon in the intended experiment.

Two experiments were conducted to evaluate the performance of the instrument under real test conditions. The first employed a heat gun to induce melting and flow of a block of wax placed upon the balance, but resulted in significant creep, even with foam insulation between the wax and the surface of the balance. The final experiment emulated a shift in centroid by measuring load distribution of a box filled with sand while sand was removed with a vacuum. The results qualitatively matched the expected centroid trajectory, but the theoretical centroid position proved difficult to model, so the measurement error is not quantified.