As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.
Because an infrared instrument cannot separate the radiation a surface emits from the radiation it merely reflects, and only the operator can tell it how warm the surroundings are. Emissivity sets the split between the two parts; the reflected ambient temperature sets what the reflected part is worth. Get emissivity right and leave the reflected figure at a default that does not match the room, and the instrument subtracts the wrong quantity of borrowed infrared. On a painted wall that error is a fraction of a kelvin. On bright metal it runs to tens of kelvin, in either direction.
What the detector receives from a shiny surface
Radiation arriving at the detector from one patch of surface has two origins. The first is thermal emission from the surface itself, scaled by how efficiently that surface radiates. The second is radiation from everything else in the hemisphere the surface faces, arriving at the surface and bouncing off it. The reflected share is one minus the emissivity, because an opaque surface either emits or reflects whatever it interacts with.
A matte painted surface with an emissivity of 0.95 therefore hands the detector a signal that is 95 percent its own and 5 percent borrowed. Bright rolled metal with an emissivity of 0.15 hands over a signal that is 85 percent borrowed. The detector cannot tell the two apart, because infrared carries no label saying where it came from. The instrument is solving one measurement for two unknowns, and it closes the gap by being handed the second unknown: the apparent temperature of whatever the target is facing.
The arithmetic happens in radiant power rather than in degrees, which is why the effect is larger than intuition suggests. Power scales with the fourth power of absolute temperature, so swapping a surroundings value by twenty kelvin does not shift the subtraction by twenty units of anything. It shifts it by the difference between two fourth powers.
Where the reflected ambient figure comes from
- A uniform indoor room. When walls, floor and ceiling sit within a kelvin or two of the air, air temperature is a serviceable stand-in and costs nothing to obtain.
- The crumpled foil method. Hold crumpled and reflattened aluminium foil against the target, set emissivity to 1.0 and read the foil. The value you get is the figure to enter.
- Outdoors. An upward facing surface reflects the sky, which under clear conditions behaves like a body well below air temperature. A downward or sideways facing surface reflects ground and buildings instead.
- Near a dominant hot or cold object. An oven door, a cold single glazed window or a lighting bank overwhelms the room average, and a single entered figure becomes a compromise that should be recorded alongside the reading.
For outdoor work the surrounding conditions are worth capturing rather than estimating. The Ambient Weather WS-2902 states that its sensor array measures wind speed and direction, temperature, humidity, rainfall, UV intensity and solar radiation, which is the whole context a wall or roof survey happens inside; the WS-2902 review covers what that array does and does not report. On photovoltaic work the ambient figure belongs with the electrical one, and the Fluke 393FC lists solar irradiance, ambient temperature and PV module temperature among its measurements alongside its clamp functions, so the surroundings value and the module value come from one instrument.
The same plate entered three different ways
Take a steel plate whose true surface temperature is 60 degrees Celsius, in a workshop whose air and walls sit at 20 degrees Celsius. Emissivity is entered correctly in every case. The only thing changing is the reflected ambient temperature.
| Reflected ambient entered | Reported at emissivity 0.30 | Reported at emissivity 0.95 |
|---|---|---|
| 0 degrees Celsius | About 85.6 degrees Celsius | About 60.7 degrees Celsius |
| 20 degrees Celsius, the true value | 60.0 degrees Celsius | 60.0 degrees Celsius |
| 40 degrees Celsius | About 17.3 degrees Celsius | About 59.2 degrees Celsius |
Every figure in that table is the radiative balance worked through: emissivity times the fourth power of the object temperature in kelvin, plus one minus emissivity times the fourth power of the surroundings in kelvin, then inverted using whatever surroundings value the operator typed. At emissivity 0.30 a twenty kelvin error in one direction adds more than twenty five kelvin to the answer, and the same error in the other direction removes more than forty. The instrument is not malfunctioning. It is obeying the number it was given.
Why the setting does almost nothing on brick and paint
The right hand column of that table is the reason building thermography can be relaxed about the parameter. Brick, plaster, timber, roofing felt and ordinary paint all sit around 0.90 to 0.95 on published emissivity tables, so one minus emissivity is five to ten percent. A twenty kelvin mistake in the surroundings is attenuated by that factor before it reaches the answer, and what arrives is under a kelvin.
Electrical and mechanical work is the opposite case. Bare copper busbar, plated lugs, polished shafts and bright sheet metal sit low on the same tables, which is where one minus emissivity becomes the dominant term. The same discipline that makes a reading reproducible on metal makes it almost unnecessary on masonry, and knowing which situation you are in is most of the skill. At the other extreme, instruments built for furnace and kiln work benefit from the fourth power law doing the job for them, as the pages on high temperature infrared thermometers describe: when the target radiates seventy times as strongly as its surroundings, the reflected share is noise.
Logging the surroundings so the number is defensible
A surface temperature quoted without the two parameters behind it cannot be reproduced, which matters the moment the reading goes into a report or a comparison over time. Recording the surroundings as a logged series rather than a remembered glance at a display is the cheap version of rigour. A USB temperature and humidity data logger of the kind used for environmental records states intervals from one second to eighteen hours, a range of minus 4 to 158 degrees Fahrenheit, and an accuracy of plus or minus 3.5 percent relative humidity between 25 and 85 percent relative humidity. Whether a given logger also reports a radiant or globe temperature is a separate question, and on most listings of this type it is not stated in the listing bullets.
Fixed room monitors do the same job for repeat surveys of the same space, and the options are gathered in temperature and humidity monitors. Where the record has to leave the building with you, the wireless logger options remove the walk back to the sensor.
Entering a reflected temperature you can justify
Four habits turn the parameter from a guess into a defensible entry. Measure the surroundings with the foil method whenever the target is metallic, rather than typing the air temperature from memory. Re-measure when you move, because a plant room and a corridor two doors away are different radiant environments. Record both parameters next to every reading, so a future comparison is a comparison and not a coincidence. And where the surface allows it, cover a patch with matte tape or paint and read that instead, because a surface with an emissivity above 0.9 makes the whole question nearly irrelevant.
Frequently Asked Questions
Is reflected ambient temperature the same as air temperature?
Not quite. It is the apparent temperature of everything radiating onto the target, which indoors in a uniform room is close to air temperature. Near an oven, a cold window or a bank of lamps, and outdoors under a clear sky, the two figures separate sharply.
How do I find the reflected temperature with the instrument itself?
Crumple a sheet of aluminium foil, flatten it again and hold it flat against the target with the shiny side out. Set emissivity to 1.0 and read the foil. Crumpled foil reflects diffusely and emits very little, so the reading is the apparent temperature of the surroundings as the target sees them.
What if my instrument has no reflected temperature field?
It is assuming one, usually a room temperature value near 20 degrees Celsius. That assumption is fine on paint and plaster and unreliable on bright metal, so the honest fix on metal is to tape or paint a patch rather than to adjust parameters you cannot reach.
Does getting it wrong always make the target look cooler?
No. Entering a reflected temperature above the true surroundings subtracts too much borrowed radiation and pulls the answer down. Entering one below the true surroundings subtracts too little and pushes the answer up. The direction follows the sign of your error.
Why does a clear sky matter outdoors?
A clear sky radiates like a body far colder than the air beneath it, so a shiny roof, flashing or module frame facing upwards is reflecting something much colder than the air temperature you might otherwise type in. Overcast skies behave much more like the air.
Does the parameter matter at furnace temperatures?
Far less. Radiant power rises with the fourth power of absolute temperature, so a surface at 600 degrees Celsius emits roughly seventy times as much as its surroundings at room temperature. The reflected share becomes small enough that a rough entry costs little.