Fluke Thermal Imager for PCB Analysis: Hotspot Detection and Power Component Diagnostics

A PCB can look completely normal and still have a thermal problem.

A MOSFET may pass the electrical test. An LDO may produce the correct output voltage. A board may even pass final inspection. Put it under continuous load, however, and one small area can become much hotter than the surrounding circuit.

This is where infrared thermography becomes useful.

For PCB work, I see a thermal camera as a troubleshooting tool rather than a replacement for electrical testing. It showswhere heat is being generated. The next step is finding out why.

What Can a Thermal Camera Actually Tell You?

A thermal image shows the apparent surface temperature pattern of the board.

That sounds simple, but measuring the exact temperature of electronic components is not always straightforward. Bare metals such as copper have low emissivity and can reflect infrared radiation from surrounding objects. Fluke specifically warns that low-emissivity electrical surfaces can make absolute temperature measurements unreliable. For this reason, comparing similar components under the same load is often more useful than trusting one temperature number.

This changes the way I use aFluke Thermal Imager.

I am usually looking for questions such as:

Why is this MOSFET hotter than the other three?

Why is one side of an IC much hotter than expected?

Where does the temperature rise begin?

The shape of the hotspot can be as useful as the maximum temperature.

Why Emissivity Matters on a PCB

A black plastic IC package and a shiny copper pad do not emit infrared energy in the same way.

If the camera is set up correctly for a plastic package but pointed at exposed copper, the temperature reading can be misleading.

For PCB thermal analysis, I prefer to compare areas with similar surface finishes. When an absolute reading is important, the emissivity setting and measurement setup need to be controlled carefully.

Fluke notes that emissivity varies with surface condition, viewing angle, temperature and wavelength. Its guidance also points out that polished copper can have extremely low emissivity, making direct infrared measurement difficult.

So if a copper pad looks cooler than an IC package, that does not automatically mean the copper is actually cooler.

Macro Lenses Become Important With Small SMD Parts

A standard thermal camera may show a hotspot, but that does not necessarily tell you which component is causing it.

This becomes a problem with 0402, 0603, SOT-23, and other small packages sitting close together.

A macro lens can make a major difference here. Fluke offers macro infrared lenses specifically for R&D and PCB inspection. Its 25-micron macro lens is designed to reveal thermal details on very small PCB targets, and Fluke gives PCB heat dissipation and faulty-component inspection as specific use cases.

That is useful when a single thermal hotspot contains several components.

A camera may show one warm area. A close-up image may show that only one IC, resistor, or section of the package is actually generating the heat.

I would not choose the lens from package size alone, though. The camera resolution, lens, working distance, and field of view all affect what can actually be resolved.

Finding an Overheated LDO

LDOs are easy to diagnose thermally when the basic power loss is known.

For a linear regulator, a rough estimate is:

P ≈ (VIN − VOUT) × IOUT

Suppose a regulator converts 12V to 5V at 300mA.

The simplified power loss is:

(12 − 5) × 0.3 = 2.1W

That is a lot of heat for a small package.

If the thermal image shows the LDO becoming the main hotspot, the next step is not simply "use a bigger copper area." Check whether the input voltage is unnecessarily high, whether the load current is correct, and whether the regulator package can dissipate the heat.

Thermal imaging makes this problem visible very quickly.

Checking MOSFET Heating

For a MOSFET, conduction loss is roughly related to:

P = I² × RDS(on)

But switching losses also matter when the device operates at higher frequency.

This creates a useful diagnostic situation.

Suppose two MOSFETs on the same PCB carry similar current, but one is much hotter. That does not automatically prove the MOSFET itself is defective.

Possible causes include:

· Different gate-drive voltage

· Higher RDS(on)

· Poor soldering

· Uneven copper area

· Excessive switching loss

· Higher current through that particular device

This is wherethermal performance testing for TO-252 DPAK MOSFETsbecomes useful. A DPAK device depends heavily on its PCB thermal path, so the thermal image can show whether heat is spreading properly into the surrounding copper.

Can Thermal Imaging Find a PCB Micro Short?

Sometimes.

A short circuit or partial short can create localized heating if sufficient current is flowing through the fault.

The thermal pattern can help locate the area before the exact fault is found with electrical testing.

But I would not claim that a thermal camera can always find a micro short circuit.

A high-resistance fault may generate very little heat. An internal short between layers may also be difficult to see from the surface if the heat is weak or spreads through the board.

ForPCB micro short circuit detection with a thermal camera, I normally treat the thermal image as a locator.

Once the suspicious area is identified, use resistance measurements, current injection, voltage measurements, or other fault-isolation methods to confirm the fault.

What Does Abnormal MOSFET Heating Look Like?

One useful case is comparing identical components under the same operating conditions.

If four MOSFETs are expected to share a similar load and three show similar thermal patterns while one has a noticeably stronger hotspot, that deserves investigation.

This is also useful forhow to detect thermal runaway in power MOSFETs.

I would look for a temperature rise that continues to accelerate as the device warms, rather than simply seeing that one device is hotter than another.

The exact cause could be increasing conduction loss, insufficient cooling, excessive current, or another circuit condition. Thermal imaging shows the symptom; electrical measurements are needed to identify the cause.

Using Thermal Imaging for QA Before Mass Production

Thermal imaging can also be used before a product enters volume production.

I prefer to establish a baseline board first.

Run a known-good PCB at defined:

· Input voltage

· Load current

· Ambient temperature

· Operating time

Then save the thermal image.

During later production testing, unusual hotspots can be compared with that reference.

This can help identify differences caused by a wrong component, poor solder joint, insufficient copper, unexpected current paths, or assembly variation.

For QA teams, this is more useful than simply setting a rule such as "nothing can exceed 80°C." A component's temperature should make sense for its role in the circuit.

My View on PCB Thermal Inspection

A thermal camera should not be the only instrument on the bench.

A multimeter tells you what the circuit is doing electrically. An oscilloscope shows voltage and switching behavior. A thermal imager shows where the electrical losses are turning into heat.

Those three views complement each other.

For aFluke Thermal Camera for Circuit Boardinspection, I would start with a stable test condition, compare similar components, pay attention to emissivity, and use a macro lens when the target is too small for the standard lens.

That approach is much more useful than simply searching for the hottest pixel.

A PCB thermal image is not a diagnosis by itself. It is a map. The real value comes from using that map to decide which component or section needs a closer electrical inspection.