Electrical thermography is about finding a connection that is running hotter than the identical one beside it, which needs enough resolution to separate adjacent lugs and enough temperature range to cope with a badly overloaded terminal. This hub compares nine imagers on exactly those figures and flags where the listings say nothing.
Compare Phases
Load It First
Never Open Live
As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.
Nine imagers judged on what electrical work actually demands: resolution to separate adjacent terminals, a temperature range that covers an overheating lug, fusion imaging so a hot spot can be named, and a way to get the image into a report. Thermal imaging is a screening technique and never a substitute for de-energizing before touching anything. We ran no tests.
The widest stated temperature range here, on a 3.5 inch industrial screen.
The highest stated resolution here plus storage and PC transfer.
A 9 gram module that rides in a shirt pocket to every call.
No pick matches that combination yet. Try a wider budget or a different measurement.
| Product | Format | Stated resolution | Stated temperature range | Fusion or dual light | Export stated | |
|---|---|---|---|---|---|---|
| PHOEBUS LINK thermal camera | Standalone | 320 x 240 | Not stated | Not stated | JPG export, 512 MB, PC transfer | View |
| Thermal Master Thor001 | Standalone | Not stated | -4 F to 1022 F | Not stated | Not stated | View |
| VEVOR USB-C thermal camera | Phone, Android and iOS | 256 x 192 native, 512 x 384 with TISR | Not stated | Not stated | Not stated | View |
| BSIDE HX1 | Standalone | 240 x 240 | Not stated | Not stated | Not stated | View |
| GOYOJO GH192 | Standalone | 240 x 240 super IR | Not stated | Yes, 2 MP visible camera fusion | Not stated | View |
| Yukicam dual-light camera | Standalone | 240 x 240 with TISR | Not stated | Yes, dual-light fusion | Not stated | View |
| AMPBANK H128 | Standalone | 128 x 128 native, 240 x 240 with TISR | Not stated | Not stated | Not stated | View |
| AMPBANK M192 | Phone, USB-C | Not stated | Not stated | Not stated | Not stated | View |
| InfiRay P2 Pro | Phone, USB-C Android | Not stated | Not stated | Not stated | Not stated | View |
Sorted by how much each listing publishes. Only one states a temperature range and only one states an export path, which is a fair picture of what this tier of the market documents. Figures come from the manufacturer listings.
Electrical targets are small, hot and adjacent, which changes the priorities:
Breaker poles and terminal lugs sit close together. A native 320 x 240 detector distinguishes one from the next at working distance; an upscaled 240 x 240 from a 128 x 128 sensor often does not.
A failing connection can run hundreds of degrees above ambient. A camera that tops out where building work ends will simply saturate. Only one listing here publishes a range.
The finding is usually one phase warmer than the other two under the same load. That comparison survives emissivity error, which is why electrical thermography is more forgiving than it looks.
Predictive maintenance means comparing this quarter with last quarter. Storage, JPG export and PC transfer decide whether that is possible.
Scanning through an open enclosure means standing in front of live equipment. Arc flash boundaries, PPE and permissions come from your safety program, not from the tool.
The usual approach is comparative. A connection running a few degrees above its identical neighbours under the same load is worth watching and re-scanning; a difference of tens of degrees is usually treated as needing action soon. Published criteria vary by standard and by asset owner, so follow the criteria your program specifies rather than a number from a listing.
Yes. A loose or corroded connection heats because current is flowing through the extra resistance. Scan a panel with no load and a serious fault looks identical to a healthy one. Commonly the target is a substantial fraction of normal operating load before the scan means anything.
Not usefully. Steel covers conduct and spread heat, so by the time a fault is visible on the outside it is severe. That is why thermography is done with covers removed or through infrared windows, and why the work sits inside an electrical safety program rather than beside it.
It does for absolute readings. Bare copper and aluminium have low emissivity and reflect their surroundings, so they read far cooler than they are. The practical answers are to compare identical surfaces against each other, and to fit emissivity targets or tape on assets you scan routinely.
It is enough to find an obvious problem on a routine visit, which is worth a lot given the alternative is not scanning at all. It is not enough to document a finding with a temperature value, and none of the phone modules here publish resolution or range.
Temperature range is the specification electrical work exposes fastest, and the Thor001 is the only camera on this page that publishes one. Thermal Master states -4 F to 1022 F, which comfortably covers a lug running at two hundred degrees as well as the ambient reference beside it. Cameras that top out low are fine on a building envelope and useless the moment a connection is genuinely failing.
The screen is the other reason it leads. A 3.5 inch IPS panel described as industrial grade and scratch resistant is a meaningful specification in a plant room, where the alternative is a phone screen you cannot read and would rather not have near an open enclosure. The listing frames the camera around fast, efficient field inspections, which is the right job description.
What is absent is resolution. Thermal Master states no detector figure at all on this listing, so whether it can separate two adjacent terminals is unknown. If your work is dense panel scanning where the lugs are an inch apart, confirm the detector size with the manufacturer first, or choose the stated 320 x 240 camera below.
Predictive maintenance lives or dies on records, and this is the only listing here that describes a route from camera to document. PHOEBUS LINK states 512 MB of onboard storage, JPG export and PC data transfer for report generation, which is the unglamorous specification that determines whether last quarter's scan can be compared with this one.
It also carries the highest stated resolution on the page at 320 x 240, which is 76,800 detector elements. On a distribution panel that is the difference between a warm blur across three breakers and a clearly identified middle pole. The listing explicitly names bearing wear, motor overheating and pump failure, which tells you the intended application is rotating plant as much as switchgear.
The listing states no temperature range, no thermal sensitivity, no screen size and no battery figure. For a maintenance program, get the range in particular from the manufacturer, since a motor terminal box can exceed the ceiling of a camera built for building work.
Most electrical thermography never happens because the camera is in the van and the panel is on the third floor. A stated 9 gram module in a shirt pocket changes that arithmetic: it comes out during a routine call, gets pointed at a panel for thirty seconds, and either finds nothing or finds the thing that turns into a return visit.
The magnetic macro lens InfiRay includes is more useful in electrical work than in building work, because the targets are small. Getting close enough to resolve one terminal from its neighbour is a real constraint with a wide lens, and the macro attachment is the answer for control gear and terminal blocks rather than for scanning a whole board.
Use it with clear eyes about what it is not. The listing publishes no resolution, no sensitivity and no temperature range, so it is a finding tool. A reading from it should not go into a report as a measured value, and it should never encourage you to open an enclosure you would otherwise have left alone.
Short answer: the Thermal Master Thor001 is the electrical inspection camera to buy on the strength of the only stated temperature range here, -4 F to 1022 F, on an industrial 3.5 inch screen. The PHOEBUS LINK camera is the pick when findings go into a maintenance report, with a stated 320 x 240 detector, 512 MB of storage and PC transfer, and the InfiRay P2 Pro at a stated 9 grams is the module that actually gets carried to every call.
The single most useful thing to understand before buying is that the finding is almost always relative. Three phases feeding the same load should sit at the same temperature; one that is warmer has extra resistance somewhere, which means a loose lug, a corroded crimp or a failing breaker. Two identical motors on the same duty should run alike. Because the judgement is made between two surfaces of the same material in the same scene, emissivity error largely cancels out, and a camera that would give a poor absolute reading can still give a reliable comparison. That is why resolution and range matter more here than radiometric accuracy: you need to see both objects clearly, and you need the hot one to stay on scale.
Building envelope work images large areas, so a modest detector is enough. Electrical work images a terminal block where the lugs are an inch apart, viewed from a safe distance. The number of detector elements landing on each terminal decides whether you get one warm object or three distinguishable ones. The PHOEBUS LINK listing states 320 x 240, which is 76,800 measurements and the most on this page. The VEVOR module states 256 x 192 native, then upscales to 512 x 384 by interpolation, which produces a larger picture with the same 49,152 measurements behind it. The AMPBANK H128 is honest that its 240 x 240 comes from a 128 x 128 sensor. On a panel, that difference is visible.
Most consumer thermal cameras are specified around human and building temperatures. A genuinely failing electrical connection is not in that band: a badly loose lug carrying load can reach several hundred degrees, and a camera that saturates below that shows a white patch without telling you how bad it is. Only the Thermal Master Thor001 publishes a range here, at -4 F to 1022 F, and it is the reason that camera leads the page despite not stating a detector size. For every other camera in the table the ceiling is unknown, which for a maintenance program is a question to settle with the manufacturer before purchase rather than after.
Two practical conditions decide whether a scan means anything. The equipment has to be carrying current, because the heating is resistive and a de-energized or lightly loaded panel hides everything. And the covers have to be off or an infrared window fitted, because steel spreads and hides heat so effectively that a fault visible through a closed cover is already advanced. Both conditions put the technician in front of energized equipment, which is exactly why this work belongs inside an electrical safety program with arc flash boundaries, PPE and permissions defined before the camera comes out. The camera is a screening instrument; it changes nothing about the hazard. Our guide to choosing a thermal camera for electrical work goes further into the working procedure.
Bare copper busbar, aluminium lugs and new galvanised enclosures all have low emissivity, meaning they emit little of their own infrared and reflect a great deal of their surroundings. Point a camera at polished busbar and you will often read the temperature of the room, or of yourself. Three practical answers exist. Compare like with like, since two identical bars in the same scene misread by the same amount. Image the insulation or the boot adjacent to the connection, which has high emissivity and is heated by conduction. And on assets scanned routinely, fit emissivity targets or a small piece of matte tape at the measurement point so every future scan reads the same surface.
An electrical scan is worth most when it is repeated. A connection that was two degrees warm last quarter and is eight degrees warm now is a maintenance decision; the same eight degrees with no history is a guess. That requires storing images, exporting them and filing them against the asset, which is why the PHOEBUS LINK listing stands out as the only one stating 512 MB of storage, JPG export and PC transfer. Everything else on this page states nothing about storage or export, so check before committing to a camera for a program rather than for a toolbox.
Imagers were compared on manufacturer listing text: stated detector resolution and any upscaled figure, stated temperature range, fusion or dual-light imaging, screen, storage and export. Two macro lens accessories and an automotive night-vision device present in the source data were excluded so all nine rows are usable inspection cameras. We ran no thermal tests and publish no measurements. Amazon data supplies images, availability and links only.
Ten borescopes for engine and cylinder work compared on probe diameter, reach, articulation and screen size.
Read Guide→
Nine steerable borescopes compared on articulation direction and angle, probe diameter, screen size and lens count.
Read Guide→
A compact combination meter with a pinless sensor for scanning and fold-out pins for confirmation, aimed at inspectors who need both in one pocket.
Read Guide→