A simple flow control loop of a mass flowmeter and a shut-off value was setup to fill a railcar. Operations would enter a number of pounds of liquid to dump into the railcar and the system would run until this setpoint was reached. Nothing complicated about it at all. Not being complicated doesn't mean you won't have problems though.
When the railcars were weighed later, a difference was detected between what the control system said had been dumped into the railcars and what they actually weighed. The flowmetering system had measured more liquid being dumped into the railcar then what the scale indicated. As the instrument technician for the area, I was asked to check out the flowmeter to make sure it was operating correctly. Not having been involved in the installation of this particular system, this would be my first look at it.
Well, I took a look and I observed that it had a 3-in. pipe making a 90º bend down into the flowmeter, into the valve, and then into flex hose attached to the other end of the valve. Knowing how these flowmeters work and how they really dislike trying to measure a liquid flow with entrapped air, my educated guess was the entrapped air was causing an erratic flow reading and indicating an error towards the high side. My opinion was based on encountering similar situations in the past. I told operations that the meter had been installed incorrectly and needed to be relocated to a different spot to give an accurate flow reading. This turned out to be my first mistake.
"I took him out to the meter. He took a quick look at, said it was mounted incorrectly, then left."
The plant's engineer was the one that spec'ed out the installation of the flowmeter in the first place and he did not like being told that the installation was incorrect. His argued that he had read the manual and because it was installed according to the manual, it would be fine. I then made my second mistake.
I told him that according to the manual it was installed incorrectly and that furthermore, after spending three days in training at the factory where these meters were built, I had gained even more insight into its proper installation as was absolutely sure this meter was installed incorrectly. Guess who won that argument? Correct, not the instrument technician with factory training and hands-on experience, but the engineer with little time or practical experience on the job. As many technicians know, this happens often.
I then spent most of the day troubleshooting the meter, verifying every setting I could, even swapping out cards looking for any problem at all. The only thing I did not do was run a flow though the meter and verify the reading against a calibrated scale. After all of this, I could not turn up a single verifiable problem and dutifully reported my lack of findings back to operations, again with the advice the meter needed to be moved to a better spot.
The next day I got a call from another technician. He had been asked to look at a flowmeter and couldn't find a problem with it either He was wondering if I could take a look at the meter in question. Three guesses as to what flowmeter he was looking at and the first two don't count.
I mentioned to him to check a few things and also told him that in my humble opinion, the problem was with the installation, not the meter. Now, you would think that after two technicians examined the meter and could not find a problem, that maybe someone would start listening to us. Of course, that someone didn't.
A couple of days later I get a call from the factory representative from the flowmeter's manufacturer. This generally isn't too unusual as he is often at our plant about once a week and tends to touch base with me on any issues when he is in the vicinity. This time he was wondering if I could show him where a certain flowmeter was located--he had come in because he had been asked to check out. So, without offering any comments on what I thought the problem was, I took him out to the meter. He took a quick look at, said it was mounted incorrectly, then left.
A week later, I stopped in and asked the engineer whether or not the issue with the flowmeter had ever been resolved. It had, I was told, so I then inquired as to what the problem actually was.
"Oh we needed to move it, it works great now." I avoided making any smug comments and just remarked that I was glad to know that the issue had been resolved.
On the plus side, I never had a disagreement with the engineer on an instrumentation question again.