At 2:17 AM in March 2024, my phone lit up with a number I didn't recognize. Normally I let those go to voicemail. But when you're the only field service engineer on call for a twelve-county territory, you answer.
"Conveyor line three is down," said the plant manager at Maple Ridge Foods. "Motor contactor keeps tripping. If we don't have product moving by 6 AM, we miss a delivery window worth about $50,000 in penalties."
I grab my bag, mentally checking what's inside: FLIR One Gen 3 thermal camera, a digital multimeter, a Mitutoyo digital micrometer, and a set of calipers I'd borrowed from a coworker weeks earlier. No formal checklist. That last detail would cost me.
I had a 45-minute drive ahead, which gave me time to think through the possibilities: bad contactor, shorted motor, failed overload relay, or something dumber like a tripped E-stop. In that order.
The First Pass
When I got there, the maintenance guy had already swapped the contactor once. He shrugged and said, "May be the motor." A motor on that line costs about $4,000 to replace, and the nearest spare was three hours away. The plant had already lost two hours of production.
My first instinct was to grab the multimeter and start measuring phase-to-phase resistance. That's what a 2020 me would have done. But I'd been testing a theory for a few months: thermal should come before electrical.
Here's the thing—a regular multimeter tells you what's happening now, at the exact probes you touch. It doesn't show you the pattern. An infrared camera shows the whole panel at once. I pulled out my FLIR One Gen 3, plugged it into my phone, and spent ninety seconds scanning the starter bucket. The thermal image lit up like a Christmas tree. On the load side of the contactor, one terminal was running at 187°F. The other two sat around 105°F. That's a classic loose connection, not a dying motor.
I've read plenty of FLIR One Gen 3 thermal imaging camera reviews before buying it, but the real test is whether it holds up at 3 AM in a dusty MCC room. It did.
FLIR Sensors and How They Work
I'm not going to rehash the full physics lesson (and honestly, I can only explain it to a point). The FLIR sensors in that little dongle are microbolometers—they detect infrared radiation and translate it into temperature values. No external light needed. That's why it works in a dark MCC room at 3 AM when even the task lights flicker.
According to FLIR's product page (accessed January 2025), the One Gen 3 has a thermal resolution of 256 × 192 pixels (about 49,000 points of thermal data). That's not a high-end research camera, but for electrical panels and motor boxes, it's plenty. I've used it to find overheated breakers, bad terminations, and failing bearings.
The Micrometer Detour
Here's where things got embarrassing. After deciding it was a loose lug, I wanted to verify the motor shaft wasn't damaged before recommending a simple re-torque. That meant checking the shaft diameter wasn't worn. I dug for my calipers—and realized I'd left them in another truck. All I had was the Mitutoyo digital micrometer.
Fine. Micrometer it is.
Except the battery was dead. Or nearly dead. The display was flickering like a dying neon sign.
I spent five minutes fumbling in the dark, and at one point I Googled 'how to turn off Mitutoyo digital micrometer' with one bar of signal. The answer, eventually, was to hold the "SET" button for two seconds. Should've known that before. (Note to self: read the damn manual.)
That decision—checking the shaft despite the dead battery—felt risky. The upside was verification. The downside was wasting time and looking incompetent. I kept asking myself: is this worth the production pressure? And honestly, yes. Replacing a $4,000 motor because I skipped a two-minute measurement would've been a far bigger error.
The Fix
I re-torqued the lug on that contactor to spec—about 25 inch-pounds, which I know because my torque screwdriver was still where it belonged, in the truck—and ran a thermal scan again. The hot spot dropped from 187°F to 112°F under load. Not perfect, but in acceptable range for that connection. Then I used the Mitutoyo micrometer (after replacing the battery with one from the plant's spare drawer) to confirm the shaft was within 0.0002 inches of spec. It was.
We hit the restart button. Conveyor line three rumbled back to life at 5:44 AM, sixteen minutes before the deadline. The plant manager's shoulders dropped about four inches.
But here's the thing about recovering from a near-miss: after the line ran for an hour, I still felt that knot in my stomach. What if the heat was coming from something deeper? I didn't relax until the day shift reported zero issues at 8 AM. That post-decision doubt doesn't go away just because you're good at your job.
The thermal scan after the repair wasn't just about numbers. It gave me a visual confirmation that my work actually solved the problem. That kind of feedback loop is hard to get with a multimeter alone.
What I'd Change Next Time
This job worked out. But it exposed at least three gaps in my own setup.
First, no tool checklist. We didn't have a formal process for verifying battery levels and tool presence before leaving the shop. That's how I ended up with a dead micrometer in the middle of nowhere. The third time something like that happens, you finally create a checklist. I've now started keeping a laminated card in every truck.
Second, the misconception that thermal imaging is a nice-to-have. It's not. If you're doing industrial troubleshooting, a compact thermal camera is as essential as a multimeter. I'd argue it's more essential for the first pass. What was best practice in 2020—multimeter first, everything else later—may not apply in 2025. The industry has evolved. A cheap thermal camera can reduce a two-hour diagnosis to a two-minute scan.
Third, know your tools. I had a micrometer and barely knew how to turn it off. That's on me. Now I keep the manual for every test instrument in a Google Drive folder on my phone.
One disclaimer: my experience is based on about 150 emergency calls at mid-sized food and beverage plants in the upper Midwest. If you're in a different industry—oil and gas, aerospace, utility-scale energy—your priorities might differ. I can't speak to how these principles apply to every environment. But the fundamentals haven't changed: measure twice, cut once, and pack a spare battery.
If you're building out your own kit, the pieces that served me best last year were the FLIR One Gen 3, a good digital multimeter, and calipers for the jobs that need more range than a micrometer. And when you find yourself searching 'where to buy calipers' at 3 AM on a service call, just remember: my credit card has been through that exact search too.
It's January 2025 now, and I still carry the same kit—minus the dead batteries. And yes, the calipers are in the truck.
