The case of the weird stilling well

By Jake Brodsky

I was recently promoted from field technician to engineer at a water and sewer company. My boss brought me over to look at an effluent flow meter for a waste-water treatment plant. These meters are critical, as they are the basis for the permit for the plant’s operation.

The problem was that the meter wouldn’t stay calibrated. It was a rather unusual setup, to say the least. The meter was an open channel flow meter with an 84-inch wide rectangular weir. We couldn’t use any other types of weirs or flumes because the channel had to remain intact while
construction took place on a replacement structure. The weir was wide because too much head would cause the flow to back up to places where we couldn’t allow it to go. This was a temporary setup, my boss advised. “There is nothing more permanent than ‘temporary’ around here.” I
replied. He smiled knowingly.

Given the range of flows the plant put out, we needed to accurately assess the head behind the weir to with an accuracy of around 0.05 inches. The engineers had thought of that and provided a stilling well where we could accurately determine the head behind the weir.  The stilling well had an ultrasonic level gauge. Usually the gauge seemed to work well.

However we often encountered times when the gauge would be significantly out of calibration. This was a problem because it meant that an operator had to come by and read a very small staff gauge on the side of this stilling well every 15 minutes until we could get an instrumentation
technician to recalibrate the meter. At some times of the year, we’d recalibrate this thing as much as two or three times a week.

I looked at this setup and pondered a bit. Knowing that the speed of sound through air can change significantly with temperature, I grew suspicious of the temperature measurement. Looking at a meter and horn from spare stocks I ran the temperature up and down a bit and didn’t
find anything wrong. The measurements were fine.

Then I went back to the site again to see if there was anything unusual about the horn’s mounting. The horn was made of a huge piece of stainless steel mounted on top of the stilling well. Where was the temperature measurement device? Why, it was embedded in horn, near the transducer. In a mass of steel. In the weather, not in the stilling well.

Take a cold winter morning: the effluent was typically about 45 or 50 degrees F. The outside air temperature? Well below freezing on most mornings. There was our error! The air in the stilling well was often as much as 30 degrees off. I called the manufacturer of the meter and explained my predicament. They sent me a thermistor like the one they had used in the horn. I coated it with epoxy and placed it in the stilling well’s air space instead of using the horn’s thermistor. The
performance was rock steady.

It remained in service that way for about two years. Some “temporary” fix, eh?

That was back around 1990. I am a senior staff engineer, registered in the State of Maryland. I have built a career at the Washington Suburban Sanitary Commission for over 24 years.