As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.
For most rooftop hotspot work, start with P1 (Thermal Master P3). Its listing states a 40 by 30 degree field of view that supports rooftop solar inspection up to about 10 m, a manual focus lens, and sensitivity of 35 mK or better, which is the figure that decides whether a small cell delta shows up at all. If you want a handheld with a built-in screen and no phone tether, P6 (Raythink EX10 Pro) publishes 320 by 240 super resolution and 40 mK sensitivity with a laser pointer for targeting. When the thermal image suggests a problem but you need proof of lost power, P9 (FLIR PV48) is the one pick here that traces an I-V curve and measures panel temperature directly, up to 800 W per panel.
Why a hotspot is a thermal problem before it is an electrical one
A hotspot forms when one cell, one bypass diode or one solder joint carries more current than it should, or when shading and soiling force the rest of the string to push current through a weak cell. The result is local heat. That heat is the first measurable symptom, and it appears before the panel’s total output drops enough for anyone to notice on an inverter readout.
This is why the primary tool for hotspot work is a thermal camera rather than a multimeter. The camera does not tell you how many watts you have lost. It tells you where to look, and it does so across a whole array in the time it takes to walk the roof or pan from the ground.
The second tool is an I-V curve tracer. Once the camera has flagged a module, the tracer answers the harder question: is this module actually producing less than its neighbours, or is it just running warm because it is working harder in full sun? A warm module in full irradiance is normal. A warm module at reduced power is a fault. You can read more on that distinction in how irradiance and output relate.
The specification that decides whether you see the hotspot at all
Thermal sensitivity, quoted in millikelvin, is the smallest temperature difference the sensor can resolve. A camera rated at 35 mK can distinguish a 0.035 C delta; one rated at 40 mK needs a slightly larger delta before the difference appears on screen. For hotspot work this matters because a mildly failing cell may only run a few tenths of a degree above its neighbours at low irradiance, and a camera with poor sensitivity will render that cell the same colour as everything around it.
Resolution matters for a different reason. A 128 by 128 native sensor spread across a 40 degree field of view gives each pixel a coarse footprint on a module that is several metres away. Super resolution, which several listings here quote, interpolates and sharpens the native data into a larger display grid. It genuinely improves the image, but it does not create detail that the sensor never captured. Treat a quoted 240 by 240 or 320 by 240 super resolution figure as a display and processing claim, and the native figure, where published, as the optical limit.
| Pick | Stated thermal resolution | Stated sensitivity | Stated temperature range | Stated accuracy |
|---|---|---|---|---|
| P1 Thermal Master P3 | 256 x 192 native, 512 x 384 processed | 35 mK or better | -20 C to 600 C | plus or minus 2 percent |
| P6 Raythink EX10 Pro | 320 x 240 super resolution | 40 mK | -4 F to 1022 F | plus or minus 2 C or 2 percent |
| P4 UNI-T UTi320E | 320 x 240 IR resolution | Not stated in the listing bullets | -40 C to 400 C | Not stated in the listing bullets |
| P3 Flagfront YXI96 | 240 x 240 super resolution | Not stated in the listing bullets | -4 F to 1022 F | within 2 percent |
| P9 FLIR PV48 | Not applicable, I-V tracer | Not applicable | Temperature measurement, range not stated | Not stated in the listing bullets |
Note that the two FLIR and PROVA entries elsewhere in this category are electrical testers, not imagers, and are excluded from the thermal comparison above. The thermal camera roundup covers the imaging side in more depth.
Emissivity, angle and the mistake that produces false hotspots
Every surface radiates infrared at a rate set by its emissivity. Glass, the outer surface of most modules, has a high emissivity and behaves predictably. Metal frames, junction boxes and bare connectors do not, and they will read low or wildly variable unless you set the camera’s emissivity value to match. A shiny connector that reads 20 C cooler than the glass beside it is not a cold fault; it is a reflection problem.
The second common mistake is angle. Thermal cameras read best when you are roughly perpendicular to the surface. At a steep angle, the glass reflects the sky, and on a bright day that reflected sky is cold, which drags the apparent temperature of the module down and can hide a genuine hotspot. Walk the array rather than scanning it from one corner.
The third mistake is timing. A hotspot is most visible when the module is under load and the sun is on it. Early morning, late evening and heavily overcast conditions reduce the temperature delta between a failing cell and a healthy one. If you are inspecting for hotspots, do it in strong irradiance with the array producing. The daily output curve explains why the same array can look very different at two times of day.
The picks for hotspot work
Thermal Master P3
Phone-attached thermal camera with a manual focus lens and a stated 10 m rooftop inspection range, which is the combination that matters when you are scanning modules from ground level.
Raythink EX10 Pro
Handheld with its own 2.8 inch screen, 320 by 240 super resolution and 40 mK sensitivity, plus a laser pointer so you can mark the exact module you flagged.
UNI-T UTi320E
Native 320 by 240 IR resolution on a 3.5 inch screen, with the widest published low-end temperature range here at minus 40 C, useful for cold-morning checks.
FLIR PV48
Not a camera. An I-V curve tracer and panel tester that measures maximum power and temperature, so you can prove a suspected hotspot is costing output.
Flagfront YXI96
240 by 240 super resolution handheld with a temperature alarm and a stated 6.6 foot drop rating and IP54 sealing, for sites where the meter gets knocked about.
When the camera says hotspot but the panel is fine
A thermal image alone cannot convict a module. Several benign conditions produce a warm patch: a leaf or bird dropping that has shaded one cell and forced it into reverse bias, a module that is simply in a hotter part of the roof, or a string that is running at a slightly different operating point. The camera shows you the pattern; it does not tell you the cause.
This is where an I-V curve tracer earns its place. The FLIR PV48 measures maximum power, temperature and the full I-V curve, and the listing states it handles up to 800 W per panel. If the flagged module’s curve shows a depressed maximum power point or a stepped shape, you have electrical evidence of a fault. If the curve is clean and the module matches its neighbours, the warm patch is environmental and you can move on.
For string-level work rather than single panels, the PROVA 218 PV Analyzer and the HT Instruments Solar I-Ve both publish I-V curve tracing, with the HTI unit stating coverage up to 1500 V and 40 A, which puts it in string and utility territory rather than single-module work. The 1500 V tester page covers that class.
Maintenance and handling that nobody mentions
Thermal lenses are delicate and usually coated. Wiping one with a sleeve or a paper towel is how you turn a sharp camera into a permanently soft one. Use a blower first, then a lens cloth, and keep the cap on between inspections. On the phone-attached units, the connector is the weak point: repeated plugging into a phone in a pocket stresses the plug and the phone port. The extension cable included with P1 exists precisely so you can leave the phone in a pocket and hold only the camera head.
Battery behaviour is the other thing that catches people out. P5, the AMPBANK H128, publishes up to 24 hours of continuous working time, which is unusual and useful for a full day on a commercial roof. The others publish no runtime figure, so plan a charging break or carry a power bank. Cold weather reduces lithium battery capacity regardless of the stated figure, so a winter inspection day is the one to plan for.
Finally, treat a thermal camera as a locating tool, not a record. If you need a defensible report, export the images and write down the module identifier, the irradiance at the time, and the camera settings. A thermal image with no context is hard to compare against next season’s inspection.
Where thermal imaging stops being the right answer
Thermal cameras find heat. They do not measure insulation resistance, they do not verify that a string is correctly earthed, and they do not tell you whether a module passes its safety tests. If your job is a commissioning check or a warranty dispute, a thermal image is supporting evidence, not the test itself.
They also struggle with very small or deeply buried faults. A failing bypass diode inside a junction box may not produce a surface signature you can see from the ground, especially on a framed module with a metal backsheet. For that work you need to be on the roof with the camera close, or you need electrical testing at the junction box.
And for anything involving live DC at string voltages, the thermal camera is the safe first step precisely because it is non-contact. You spot the suspect module from a distance, then isolate and test. The solar panel tester roundup covers the electrical side, and the ceiling leak camera page covers the building-envelope use of the same instruments.
Recommended Tools
Products shown because their documented specifications fit this guide. Price and availability are on Amazon and can change.
Frequently Asked Questions
Do I need a thermal camera to find a solar hotspot?
For locating one, yes, or at least a non-contact way to see heat. A hotspot is defined by a temperature difference, so an instrument that maps temperature across the module is the direct tool. You can sometimes infer a fault from string current measurements, but that tells you a string is underperforming, not which module is responsible.
What sensitivity do I need for solar panel work?
The listings here publish 35 mK for P1 and 40 mK for P6 and P5. Lower is better, because a small cell delta is easier to see. For rooftop work at distance, sensitivity and resolution together decide whether a single hot cell stands out from its neighbours. If a listing does not publish a sensitivity figure, as with P3 and P4, you cannot compare it on that basis.
Can I use a phone-attached camera for a commercial rooftop?
Yes, and P1 is built for it, with a stated field of view supporting rooftop inspection up to about 10 m and a 19 inch extension cable. The limitation is that your phone is the display, so battery life and screen brightness in direct sun become the constraint. A handheld with its own screen, such as P6 or P4, avoids that.
Is a thermal camera enough to prove a panel is faulty?
No. It shows a temperature anomaly. Confirming that the anomaly costs output requires an electrical measurement, which is what the FLIR PV48 does with an I-V curve and maximum power reading up to 800 W per panel. A warm module in full sun can be entirely healthy.
Why do listings quote a super resolution figure and a native figure?
Super resolution is processing that upscales and sharpens the native sensor data into a larger display grid. P1 states 256 by 192 native converted to 512 by 384, and P7 states 80 by 60 native upscaled to 240 by 180. The native number is the optical limit; the processed number describes the image you see.
What temperature range do I need for solar modules?
Modules under load in strong sun typically sit well within the range of every camera here. The published ranges run from minus 40 C on P4 to 600 C on P1 and 1022 F on P6 and P3. The wide high ends are for electrical and mechanical work, not for the modules themselves.
Affiliate disclosure: Akermin earns a commission from qualifying Amazon purchases made through links on this page. Our editorial picks are based on documented specifications and owner feedback, not commissions.
Related Guides
Solar Testing Tools for Commercial Rooftops: 5 Picks for Different Jobs
Five solar tools for commercial rooftop work, sorted by the measurement each one is built to make.
Read More→
Solar Testing Tools for O&M Technicians: 4 Picks for Field Work
Four solar O&M tools compared by what they measure, what they connect to, and where each one runs out of range.
Read More→PV Testing Kits for System Commissioning: 5 Picks
A comparison of five PV testing kits for commissioning, from a full IEC 62446-1 test set down to a single MC4 adapter lead.
Read More→