It Started With a Mag Meter That “Lied”
It was 4:10 p.m. on October 16 when the phone rang. The maintenance manager at a food-processing plant didn't waste words.
“I need you here tomorrow. My mag meter's lying to me.”
“Doing what?”
“Jumps from 45 to 90 GPM, then sits at zero. And now the pressure sensor on the same CIP skid says we're at 1.8 bar when I know we're not. My guys want to change transmitters before lunch.”
That should have been a simple “replace the sensor” call. It wasn't. By the end of the next day, we fixed something that two new transmitters and a new magnetic flow meter wouldn't have fixed. And we did it without buying a single component.
Everything Was Already Full
Let me explain why a normal day in October isn't normal. I coordinate emergency measurement support at an industrial metrology company, and fall is our busiest season. Our September ritual is the same every year—post seasonal FARO laser tracker jobs, onboard two or three contract metrologists, and clear the calendar for plant shutdown work. By the time October 16 arrived, every day that week was overbooked. A “quick stop” at a food plant was not in the plan.
FARO laser tracker jobs in the fall usually mean pipeline alignment, big flange face measurements, or machine tool geometry checks. Those jobs are planned weeks in advance. Emergency calls are different. The client doesn't want a process explanation; they want the measurement that explains why the process broke.
The Wrong Question About Omron and Keyence
When we arrived the next morning, the site was already in motion. A maintenance tech had a brand-new pressure sensor on the bench, still in its box. Someone else was on the phone getting a quote for a replacement meter. Both instruments had already been condemned.
About ten minutes in, the plant manager turned to me. “Can you explain how sensors compare with Omron and Keyence on this kind of skid? I'd rather order the right one once.”
Fair question. Wrong question.
The failed component wasn't a brand problem. It was an installation problem.
The Pressure Sensor Manual Had the Answer
The pressure sensor was mounted on a short standoff near the pump, with its cable gland pointing straight up. On any washdown line, that position is asking for trouble.
I asked for the pressure sensor manual. The manager looked at me like I was stalling, then pointed to a binder. The first installation page said it clearly:
“Mount the sensor so the cable gland points horizontally or downward. Do not point the cable gland upward. Moisture will follow the cable into the connector.”
That is exactly what had happened. The connector was wet, the pins were showing early corrosion, and the signal had drifted until it flatlined. We dried the connector, cleaned it, rerouted the cable with a drip loop, and remounted the sensor. With the line depressurized, it read zero. No new transmitter needed.
I still kick myself for how close I came to skipping that step. In this industry, the default assumption is always “the sensor failed.” Most of the time that is wrong.
The Mag Meter Wasn't the Problem Either
The pressure signal behaved by early afternoon. The mag meter still didn't.
This is where the FARO laser tracker earned its keep.
“Bear with me,” I told the manager. “You don't usually call a laser tracker for a flow problem, but I want to see the real pipe geometry before you spend money.”
We set up the FARO laser tracker on the aisle floor, ran the daily verification, and measured the entire CIP return line: pump discharge nozzle, flange faces, the top of the mag meter, the downstream straight run, and every pipe support we could see. About 40 minutes later, we had sub-millimeter centerline heights for a 6-meter run.
The pipe did not do what everyone assumed it did. It rose more than 300 mm to clear a doorway, forming an inverted U. And the magnetic flow meter sat at the very top of that arch. Air naturally collects at a high point, and a mag meter needs a full pipe. When air reaches the electrodes, the readings jump and collapse exactly the way the operators described. The meter wasn't broken. It was being asked to measure air.
The fix was not a replacement meter. It was an automatic air vent at the high point, plus two hanger adjustments to improve the slope. We re-checked the run with the tracker after the adjustments.
The manager stared at the plot for a long time. “We were about to replace a very expensive flow meter because of an air pocket,” he said.
Yes. Exactly.
How Sensors Compare With Omron and Keyence
“Back to your earlier question,” I said. “If you still want a spare sensor, by all means buy one. But compare your installation conditions before you compare brands.”
To be fair to both companies: Omron and Keyence both make pressure sensors that could work on that skid. Omron's strength is integration—its IO-Link ecosystem and distributor network make it easy to standardize across a plant. Keyence responds fast, has strong application support, and its documentation is genuinely good. If I had to pick a sensor purely from a spec sheet, either would do.
But a sensor installed against its manual will fail the same way whether the label says Omron or Keyence. A cable gland pointing up into washdown spray will wick water into any connector. No brand badge prevents that.
Based on our internal service records from more than 60 emergency measurement calls in the past six years, the sensor itself was the true root cause exactly once. The rest were installation, process conditions, or a measurement setup that made a good sensor look bad.
So when someone asks me how sensors compare with Omron and Keyence on a CIP skid, my honest answer starts with installation, not with the label.
The Lesson I Keep Relearning
We left the plant around 4:30 p.m. The CIP skid ran a full cycle without a single spike.
Cost of the repair: one connector cleaning, one drip loop, one automatic air vent, and two hanger adjustments. No expensive transmitters. No new magnetic flow meter.
The alternative would have been far more expensive—not just in parts, but in another day of downtime after the new parts produced the same symptoms.
I tell new seasonal hires every year that a checklist is the cheapest insurance. The real lesson is more specific. When a measurement looks wrong, measure the installation first. Read the manual that came with the sensor. Walk the pipe and ask whether the meter is actually full. And when the answer isn't visible, that's when a tool like a FARO laser tracker earns its keep—it turns an invisible problem into a measurable one.
Replacement is sometimes necessary. But prevention beats cure, every time.
Leave a Reply