In many modern uncooled thermal imaging systems, the FPA is an array of microbolometer detector elements positioned at the focal plane of the infrared optical system. Understanding what happens at this stage makes it much easier to interpret the specifications that appear later in a thermal product datasheet.<\/p>\n\n\n\n
For a broader introduction to thermal image formation and basic specifications, see our thermal imaging for beginners<\/strong><\/a> guide.<\/p>\n\n\n\n A Focal Plane Array is a two-dimensional detector array positioned where the infrared optical system forms an image.<\/p>\n\n\n\n Instead of scanning the scene with one detector element, an FPA contains many detector elements arranged in rows and columns. Each element samples infrared energy from a small part of the scene.<\/p>\n\n\n\n In an uncooled microbolometer FPA, incoming infrared radiation is absorbed by very small thermally isolated detector structures. The absorbed energy causes a small temperature change, which changes an electrical property of the sensing material. Readout circuitry then converts those changes into electrical signals that can be processed into a thermal image. NIST describes the basic microbolometer process as infrared absorption heating a microbridge, followed by a temperature-dependent resistance change that is converted into an electrical signal.<\/p>\n\n\n\n A typical microbolometer is not simply a miniature visible-light camera pixel.<\/p>\n\n\n\n Its sensing structure is designed to absorb infrared radiation while remaining thermally isolated from the underlying readout electronics. Modern uncooled devices commonly use a suspended membrane connected to the readout circuit through narrow support structures.<\/p>\n\n\n\n The simplified process is:<\/p>\n\n\n\n LYNRED similarly describes uncooled LWIR microbolometers as suspended structures containing thermoresistive material, with readout circuitry beneath the detector array.<\/p>\n\n\n\n The FPA therefore provides the thermal measurements that form the basis of the image. It does not, by itself, determine how the final display will look.<\/p>\n\n\n\n Thermal detector resolution is normally written as a horizontal and vertical pixel count.<\/p>\n\n\n\n For example:<\/p>\n\n\n\n A 640 \u00d7 512 detector therefore contains almost three times as many sensing elements as a 384 \u00d7 288 detector.<\/p>\n\n\n\n If two systems use comparable optics and cover the same field of view, the higher-resolution detector can sample the scene at more spatial locations.<\/p>\n\n\n\n That can provide more information about small structures and target boundaries.<\/p>\n\n\n\n But resolution alone does not determine final image quality or practical range. Lens quality, field of view, pixel pitch, thermal sensitivity, focus, atmospheric conditions and image processing still matter.<\/p>\n\n\n\n This distinction is important.<\/p>\n\n\n\n A device may have:<\/p>\n\n\n\n For example, displaying a 384 \u00d7 288 detector image on a higher-resolution OLED does not turn the thermal detector into a higher-resolution FPA.<\/p>\n\n\n\n The display can present the image more smoothly or provide additional interface detail, but the original thermal measurement still comes from the native detector array.<\/p>\n\n\n\n Likewise, interpolation or super-resolution does not change the physical detector pixel count.<\/p>\n\n\n\n Resolution tells you how many detector elements exist.<\/p>\n\n\n\n Pixel pitch tells you how far apart those elements are physically.<\/p>\n\n\n\n Those two specifications should always be read separately.<\/p>\n\n\n\n A 640 \u00d7 512 detector with 12\u03bcm pitch and a 640 \u00d7 512 detector with 17\u03bcm pitch contain the same number of detector elements, but their physical array dimensions are different.<\/p>\n\n\n\n Pixel pitch also interacts with lens focal length to determine angular sampling, or IFOV.<\/p>\n\n\n\n That is why comparing only detector resolution can miss an important part of the imaging geometry.<\/p>\n\n\n\n For the full relationship between 12\u03bcm, 17\u03bcm, detector dimensions and IFOV, see our guide to 12\u03bcm pixel pitch in thermal imaging<\/a><\/strong>.<\/p>\n\n\n\n Refresh rate, frame rate or image frequency describes how often the thermal imaging system produces a new image frame.<\/p>\n\n\n\n A system operating at:<\/p>\n\n\n\n Expressed as time between frames, that is approximately:<\/p>\n\n\n\n A higher frame rate provides more temporal samples of a moving scene.<\/p>\n\n\n\n That generally makes motion appear more continuous and can make panning or tracking moving objects easier.<\/p>\n\n\n\n It does not<\/strong> add spatial resolution.<\/p>\n\n\n\n A 384 \u00d7 288 detector running at 60 Hz still has 384 \u00d7 288 native detector elements.<\/p>\n\n\n\n Likewise, a 640 \u00d7 512 detector running at 30 Hz still contains more spatial samples than a 384 \u00d7 288 detector running at 60 Hz.<\/p>\n\n\n\n Resolution and frame rate describe different dimensions of performance.<\/p>\n\n\n\n Another important distinction is the difference between frame rate<\/strong> and detector response time<\/strong>.<\/p>\n\n\n\n A microbolometer does not respond instantaneously to a change in incoming infrared radiation. Its thermal structure needs time to heat or cool toward the new condition.<\/p>\n\n\n\n This behavior is often described by a thermal time constant.<\/p>\n\n\n\n FLIR notes that the thermal time constant in some of its microbolometer cameras is approximately 7\u201312 ms and specifically warns that this should not be confused with detector integration time.<\/p>\n\n\n\n That number is not a universal value for every thermal detector, but it illustrates the principle:<\/p>\n\n\n\n how frequently frames are read out and how quickly each detector element responds are related system characteristics, but they are not the same specification.<\/p>\n<\/blockquote>\n\n\n\n Imagine a thermal system producing 60 frames per second.<\/p>\n\n\n\n The frame interval is approximately 16.7 ms.<\/p>\n\n\n\n That does not automatically mean the detector can perfectly resolve every thermal event occurring within that interval.<\/p>\n\n\n\n If the detector’s thermal response is slow relative to scene motion or temperature change, the resulting image can still show temporal blur or reduced contrast on rapidly changing objects.<\/p>\n\n\n\n High frame rate therefore cannot be evaluated completely independently from detector response characteristics.<\/p>\n\n\n\n A third specification can create confusion: display refresh rate<\/strong>.<\/p>\n\n\n\n The thermal detector may produce images at one frequency while the OLED or LCD display updates at another.<\/p>\n\n\n\n For example, a display capable of refreshing at 60 Hz does not automatically mean the thermal detector is producing 60 unique thermal frames each second.<\/p>\n\n\n\n If the detector pipeline supplies only 30 thermal frames per second, a faster display can show those frames smoothly, but it cannot create additional physical thermal measurements.<\/p>\n\n\n\n When comparing products, distinguish between:<\/p>\n\n\n\n Whenever possible, the specification that matters for scene motion is the rate at which genuinely new thermal image data are available.<\/p>\n\n\n\n It is useful to think about resolution and refresh rate along two different axes.<\/p>\n\n\n\n This is why one specification should not be used as a substitute for another.<\/p>\n\n\n\n A higher-resolution detector does not automatically have a higher refresh rate.<\/p>\n\n\n\n A higher refresh rate does not automatically produce more spatial detail.<\/p>\n\n\n\n And a smaller pixel pitch does not automatically guarantee lower NETD.<\/p>\n\n\n\n NETD, or Noise Equivalent Temperature Difference, is another detector-related specification that is frequently mixed together with resolution.<\/p>\n\n\n\n NETD describes the ability of a thermal imaging system to distinguish small differences in thermal signal under defined test conditions.<\/p>\n\n\n\n Lower NETD generally indicates greater thermal sensitivity under those conditions.<\/p>\n\n\n\n But NETD is not another form of resolution.<\/p>\n\n\n\n Two detectors can have the same resolution and different NETD values.<\/p>\n\n\n\n Two detectors can also have similar NETD specifications but different resolutions, pixel pitches or optical configurations.<\/p>\n\n\n\n A useful way to think about the distinction is:<\/p>\n\n\n\n All three can affect the experience of using a thermal imaging device, but they measure different things.<\/p>\n\n\n\n NETD should also not be interpreted as proof that a device can \u201csee through\u201d fog, rain, vegetation or other physical obstructions. Atmospheric transmission and target-to-background contrast still affect the infrared signal reaching the detector.<\/p>\n\n\n\n The detector output is only the beginning of the imaging pipeline.<\/p>\n\n\n\n Raw FPA data normally requires processing before it becomes the thermal image shown on the display.<\/p>\n\n\n\n Depending on the device, the processing chain may include:<\/p>\n\n\n\n This is why two devices using sensors with similar headline specifications can still produce visibly different output.<\/p>\n\n\n\n The detector defines the available measurements. The optics determine how infrared energy reaches that detector. Processing determines how those measurements are corrected and presented.<\/p>\n\n\n\nWhat Is a Focal Plane Array?<\/h2>\n\n\n\n
How a Microbolometer Pixel Works<\/h3>\n\n\n\n
\n
![\"Diagram](\"https:\/\/www.yubeen.com\/wp-content\/uploads\/2026\/09\/microbolometer-pixel-working-principle-1024x640.webp\"){kind=tracking}
<\/figure>\n\n\n\nFPA Resolution: How Many Detector Elements Are There?<\/h2>\n\n\n\n
\n
Detector Resolution Is Not Display Resolution<\/h3>\n\n\n\n
\n
Pixel Pitch and FPA Resolution Work Together<\/h2>\n\n\n\n
What Does Refresh Rate Mean in Thermal Imaging?<\/h2>\n\n\n\n
\n
\n
![\"Diagram](\"https:\/\/www.yubeen.com\/wp-content\/uploads\/2026\/09\/thermal-imaging-30hz-50hz-60hz-frame-rate-1024x640.webp\")
<\/figure>\n\n\n\nRefresh Rate Is Not the Same as Microbolometer Response Time<\/h2>\n\n\n\n
\n
Why the Difference Matters<\/h3>\n\n\n\n
Sensor Frame Rate and Display Refresh Rate Are Also Different<\/h2>\n\n\n\n
\n
Resolution and Refresh Rate Solve Different Problems<\/h2>\n\n\n\n
Specification<\/th> Primarily describes<\/th> It does not directly tell you<\/th><\/tr><\/thead> Detector resolution<\/td> Spatial sampling<\/td> Motion update rate<\/td><\/tr> Pixel pitch<\/td> Detector geometry \/ angular sampling with the lens<\/td> Thermal sensitivity by itself<\/td><\/tr> NETD<\/td> Sensitivity to small temperature differences under defined conditions<\/td> Detector resolution<\/td><\/tr> Image frequency \/ frame rate<\/td> Temporal sampling<\/td> Spatial detail<\/td><\/tr> Thermal time constant<\/td> Detector response dynamics<\/td> Number of detector pixels<\/td><\/tr> Display resolution<\/td> How the processed image is presented<\/td> Native FPA resolution<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n What Role Does NETD Play?<\/h2>\n\n\n\n
\n
From Raw FPA Data to the Displayed Image<\/h2>\n\n\n\n
\n