Someone told me yesterday that they thought that gasoline pumps were rigged to pump less than what is indicated on the display. That piqued my interest for two reasons, one because I don't like getting cheated, and the second, because in a past life, I was a water meter guy.

Between 1985 and 1993, I worked for Schlumberger Industries, Water and Gas Group, Technical Services Division. I traveled the U.S. installing, testing and calibrating water meters ranging in size from 5/8" - 1.5" (residential), 2" to 12" commercial and industrial, all the way to 60" behemoths that you would find feeding mejor cities.

In that line of work, when calibrating a meter, the acceptable range depended on the type of meter. Smaller meters, 5/8" through 2" are almost always positive displacement types which used either oscillating or nutating discs. They were very good at capturing low flows and were considered accurate if they registered 95%-101.5% at the lowest flow rating and 98.5 - 101.5 at the highest flow rating. My assumption is that the meters in gasoline pumps are also of the positive displacement variety but I am not sure exactly what the technology is. If anyone knows, please provide details.

Anyway, I googled gasoline pump accuracy and found many articles on the subject. I was able to find that at least in a few states, the acceptable accuracy is 6 cubic inches per five gallons. That translates into +/- one half of one percent. So with $2/gallon gasoline, the acceptable accuracy is +/- one cent.

According to the articles that I read, gas pumps are checked by the state as frequently as once per year, and as infrequently as once every four years. It does look like improvements have been made over the past several years however, with failure rates dropping significantly. It was also interesting to note that there was no trend toward cheating the customer. Failures were evenly distributed between overcalibration and undercalibration.

Go figure!

It's a Wonderful Day for Science!