Dealing with unfavorable test results

Article By : John Dunn

This story may disturb the squeamish, and engineers who have integrity when it comes to unfavorable test results.

This post is a continuation of a 2011 writeup that focused on a project to devise a customized infrared sensor for remote temperature measurements.

Once you’ve been properly shocked and revulsed by that story, please look at the drawing below, which is modified from the above essay to show some of the test results that I obtained when I was working on this thing.

thermometry design and experiment resultsFigure 1 Here are some results from the infrared thermometry experiment.

The goal was to measure the responses of a thermistor and a thermopile to the arrival of infrared from a “hot plate” temperature reference. In brief, the thermistor responses were not really definitive, but the further from the hot plate, the lower the thermopile output signal would become, dropping approximately 3% between the two positional extremes.

This was almost 20 years ago and I’m afraid that the succeeding multitude of sunrises and sunsets have diminished my memory of what the text fixture details exactly were, but the trend was clear. No matter how I modified things, there was a drop of approximately 3% in thermopile output versus distance.

My boss with the ear wax didn’t want to hear that. He showed me a sheet of paper that my predecessor had left behind, ostensibly bearing test data taken on a similar set-up where the thermopile output was shown as virtually unchanged versus position, and wondered why then was I telling him anything different?

The ear wax man was NOT open to any discussion that there might be something wrong with his expectations so I pressed on with the efforts. I built two different versions of this set-up for myself in hopes of correcting some sneaky little error that perhaps I had made, but the end results were always the same. I had no choice but to conclude that my predecessor had falsified his report in order to placate the ear wax man and had then quit.

It is a sad thing to contemplate that not every engineer is of absolute integrity.

John Dunn is an electronics consultant, and a graduate of The Polytechnic Institute of Brooklyn (BSEE) and of New York University (MSEE).

This article was originally published on EDN.

Related articles:

Leave a comment