FLIR Thermal Camera Images: A 6-Step Field Checklist for Emergency Inspections

When a critical motor starts overheating, or a panel trips for the third time in a week, the last thing you need is a 40-page manual. You need a sequence of actions that gets you from symptom to decision in the shortest possible time.

I coordinate emergency field inspections for industrial clients. In the past seven years, I've handled roughly 200 rush callouts—ranging from a $500 single-panel check to a $15,000 whole-plant thermographic survey. In March 2024, a client called at 4:30 PM with a scheduled plant audit the next morning. The infrared survey hadn't been done. Normal turnaround was three days. We deployed a technician with a FLIR AX8 that was already configured for that site, captured data overnight, and delivered the report at 6:55 AM. The audit passed. That result wasn't luck—it was the 6-step checklist I'm about to walk you through.

Step 1: Choose the Right Imaging Tool for the Job

This is where I see the most confusion. Contractors ask, "Should I buy a FLIR AX8 or a handheld thermal camera?" In my experience, that's the wrong question. The FLIR AX8 is a fixed-mount thermal camera—it stays installed and feeds continuous temperature data over your network. It's excellent for monitoring switchgear, motor bearing housings, and critical process equipment over time. A handheld unit is what you bring to a new or intermittent problem; it's portable and more versatile for moving through a plant.

The two tools don't really replace each other. If you already have an AX8 installed on a piece of equipment, you have a baseline—you know what "normal" looks like at different load conditions. That baseline is invaluable when something changes. If you don't have that, you're guessing normal from memory or from the one data point you have right now.

For emergency callouts, more expensive is not automatically better. A $1,500 to $1,900 AX8 (as of January 2025, verify current pricing at authorized distributors) is more than sufficient for most industrial monitoring. The camera is a tool, but the outcome depends on the process you follow with it.

Step 2: Set the Camera Manually Before Capturing FLIR Thermal Camera Images

From the outside, thermal imaging looks like photography. The reality is that the camera's automatic settings can mask the exact fault you're hunting for. If you leave level and span on auto, the camera rescales to each scene's hottest and coldest points—a 40°C hotspot and a 4°C hotspot look identical on screen because both appear as "white on the scale."

Before you take your first FLIR thermal camera image, check these three settings:

• Emissivity. Set it to the material-specific value: 0.95 for most electrical insulation, 0.80 for oxidized copper, 0.96 for painted surfaces. A shiny metal surface might read 0.30—get this wrong and your temperature could be off by 10°C or more.
• Reflected apparent temperature. Usually close to ambient, but if you're near a radiant heat source, a wrong value shifts every reading you take.
• Level and span. Lock them manually once you have a useful range. This is the step people skip, and it's the first thing I check when reviewing another inspector's work.

Step 3: Capture Thermal Images That Hold Up in a Report

FLIR thermal camera images are only useful if someone who wasn't there can interpret them. A thermal image of a connector at 80°C means nothing by itself. What was the load? What was the ambient temperature? Which phase was it on, and what were the other phases reading?

Per ASTM E1934—the standard practice for infrared examination of electrical and mechanical equipment—the operator should document the equipment, the settings used, and the conditions under which the inspection was performed. We follow that requirement and add one rule of our own:

1. A thermal image with a visible temperature readout on the screen.
2. A standard visual photo from the same position (plus a second angle if possible).
3. Load conditions and ambient temperature noted on the work order.

It sounds simple, but we've reviewed reports from other firms where the "thermal team" left no visual reference at all. Without a visual counterpart, the reader has no idea where the hotspot is, which makes clients and insurance adjusters—understandably—skeptical. Our reports get accepted on the first review because we treat every image pair as part of a defensible record.

Step 4: Validate Electrical Findings with a Multimeter

A thermal anomaly is almost always a symptom. The root cause could be a loose connection, a phase imbalance, or a failing load down the line. To diagnose properly, you need voltage and current readings—otherwise, you're just looking at a temperature map and calling it an answer.

Every few weeks, a client asks me: "Which Fluke multimeter do I buy without overspending?" Here's the short version:

• Fluke 87V — the industrial workhorse. $440 to $520. It handles motor drives, phase checks, and high-energy environments where a cheaper meter might not. If you do one type of electrical troubleshooting, this is the one.
• Fluke 117 — $170 to $210. Good for general commercial and light industrial work. Non-contact voltage detection is handy for quick go/no-go checks.
• Fluke 289 — $600 and up. Data logging, trend graphs, and detail for long-term monitoring. Worth it if you're doing before-and-after documentation across multiple visits.

You don't need the most expensive model to verify a thermal hotspot. The multimeter confirms what the thermal image suggests; the report is stronger when both measurements line up.

Step 5: If the Anomaly Is Mechanical, Switch to the Right Measurement Tool

Many hot spots on rotating equipment point to mechanical wear—a bearing that has lost clearance, a coupling that's misaligned, or measurable wear on a sliding surface. Temperature alone won't tell you the cause.

A blade micrometer is the fastest way to measure groove wear, keyway dimensions, or shaft play in the field. It reads to 0.001 inches, and for a go/no-go decision under time pressure, that's enough generally. They cost around $75 to $150, and they're one of the most underused tools in a quick-turnaround kit.

For higher-precision dimensional work, a metrology CMM (coordinate measuring machine) is the right instrument—if—and this is a lesson learned from a dozen over-engineered callouts—if you have a part drawing with a specified tolerance. If you don't know the acceptable limits, a CMM generates precise numbers that answer no questions. Use it when the decision involves micron-level geometry, not for a quick bearing check. That's overkill that burns time the client is paying for.

Step 6: Deliver the Report Within 24 Hours—Or Faster

I should add that we learned this one the expensive way. In 2023, we lost a $12,000 annual contract because we delivered the full report four days after an inspection the client expected within 24 hours. Every spreadsheet calculation said a 48-hour turnaround was safely within what we'd promised. My gut said this particular client was the type to read the deadline literally. I went with the spreadsheet. He sent next year's contract to a competitor.

Now, our policy is a standard report format with three severity levels: Critical, Watch, Normal. Each finding includes the image pairs, the electrical and mechanical data collected, and a specific recommended action. When a client reads the report, they immediately know what we're asking them to do—no ambiguity, no phone tag.

Common Mistakes to Avoid

These are the mistakes that cost money, regardless of how disciplined you think you are:

• Leaving emissivity at 0.95 for every shot. Metallic and reflective surfaces can change the reading dramatically. Set it based on the material, or your temperatures are fiction.
• Not recording load conditions. A motor at 50% load displays a different thermal profile than one at 100%. Your image is meaningless without that context.
• Thermal images with no visual counterpart. It seems faster in the field, but it forces a return visit when a client or auditor asks for the visual reference. Always pair the images.
• Swapping a bearing without checking the lubricant first. A bearing housing running hot is often a lubrication issue, not a mechanical-wear issue. Pull a sample before you condemn the part.
• Measuring coated surfaces with a blade micrometer. You're measuring the coating, not the part. Take a second reading after verifying coating thickness, or you'll make decisions based on the wrong dimension.
• Assuming a CMM is in calibration. In the last five vendor audits we've participated in, roughly 15% of CMMs on production lines were operating outside the manufacturer's specification. Ask for the calibration certificate before you trust the numbers.

One final thought. The quality of your inspection output is the brand your clients remember. It took me about three years and 200 inspections to fully understand that a defensible report matters more than the brand name on the camera. When I compare the reports we produced in our first year to what we deliver now, the hardware hasn't changed dramatically—what changed was the process. Follow a checklist like this one, document everything, and the work will speak for itself.