Every device in this catalog, from the €30 module to the FLIR E96 at several thousand euros, shares the same detector technology: the uncooled microbolometer. It's a smart industrial choice, one that fits in your hand, runs on battery, and needs no special maintenance. But it isn't the only technology out there, and it has physical limits that even the best microbolometer will never cross.
Above this market lies an entire world of scientific cameras, built for research, aerospace, combustion studies, or precision electronics. It's worth knowing about, even though it has no place in our comparator.
The fundamental difference: the detector is cooled
A microbolometer measures infrared radiation by observing the heating it causes on a microscopic membrane. It's simple and rugged, but the detector's own thermal noise limits sensitivity, no matter how well it's manufactured.
Scientific detectors avoid this problem by actively cooling themselves, often to a temperature near -200°C, using an onboard Stirling engine or, more rarely today, liquid nitrogen. Cooling the detector reduces its own thermal noise to almost nothing, unlocking a level of sensitivity uncooled electronics can never reach.
A field microbolometer rarely goes below 30 mK of sensitivity. A cooled scientific detector routinely reaches under 20 mK, and some specialized modes detect differences on the order of a millikelvin.
The materials change too. Instead of the amorphous silicon or vanadium oxide used in microbolometers, these sensors use indium antimonide or mercury cadmium telluride, semiconductors far more complex to manufacture and far more expensive.
A speed field equipment can't reach
The best field device in our catalog reaches 60 frames per second. Scientific cameras operate on an entirely different scale: some series advertise over 100 frames per second at full resolution, and up to several thousand frames per second in windowed mode, where only a small area of the sensor is read out to go faster.
This speed is used to film phenomena the eye never sees: combustion, a mechanical impact, an electrical discharge, an electronic component heating up in microseconds. No conventional building or industrial diagnostics use case needs this kind of speed, which is why the consumer market doesn't offer it.
Comparable resolutions, a radically different use
Some scientific models reach a resolution close to a million pixels, a figure that seems comparable to the 640 × 480 of the best field cameras in our catalog but doesn't carry the same meaning. On a high-end scientific camera, that resolution is used to observe a microscopic phenomenon with specialized optics capable of resolving a few microns per pixel, for analyzing integrated circuits or fine-scale materials. It's not about covering a wide scene, but zooming in extremely close on a tiny detail.
Features that don't exist anywhere in our catalog
External synchronization triggers image capture to within a tenth of a microsecond, in response to an outside electrical signal, to capture exactly the instant of a fast, repeatable event.
So-called "lock-in" mode isolates a periodic thermal signal buried in noise, by comparing thousands of images synchronized to a known frequency. This technique can detect temperature variations finer than the sensor's raw sensitivity would suggest, used in particular to spot an invisible crack in a material subjected to repeated stress.
Dedicated microscopic optics make it possible to observe an electronic circuit at the micron scale, a use case unrelated to the macro lenses already in our catalog for board repair, which remain at the millimeter scale.
The price, and why it's almost never public
Most of these cameras are sold by quote. The relevant lines, notably at FLIR, Telops, or Xenics, start around $15,000 to $20,000 for entry models, and comfortably exceed $100,000 for the most advanced configurations with specialized optics. A public datasheet exists, but the price requires direct contact with sales, unlike field equipment sold online at a catalog price.
Why this site doesn't compare them
Three reasons combine. The first is audience: this comparator is built for professionals in building work, electrical work, industry, and electronics repair, not lab researchers. The selection criteria barely overlap: a thermographer looks for NETD and IFOV, a researcher looks for frame rate in kilohertz and compatibility with external triggering.
The second is data availability. Without a public price sheet and with specs sometimes customized on request, it's difficult to build product pages as rigorous and comparable as the rest of the catalog.
The third is regulatory: several of these cameras, due to their sensitivity and their ability to detect fine thermal signatures, fall under particularly strict export controls, sometimes even more restrictive than the 9 Hz limits already covered for field equipment in our other articles.
This article exists to provide context, not to guide a purchase: if your need falls at this level, you're probably already in touch with a specialized integrator rather than searching a consumer comparator.
All values cited come from manufacturer datasheets and the sources listed on each product page. No data is estimated.