An infrared (IR) device is like a camera, in that it has a limited field of view. It might be able to see, for example, a 1 -deg cone or a 100-deg cone. When measuring a surface temperature, be sure the surface completely fills the field of view.  If the object you are measuring is small, select an IR device with a narrow cone that has a tightly focused field of view.   Alternatively, move the sensor close to the surface so it fills the field of view.   If it is impractical to perform either of these steps for the target to fill the field of view, you often can compromise and measure relative temperatures.  Simply read the temperature of a small target against a cool, unchanging background. The displayed temperature will go up and down as the target's temperature changes, allowing you to judge when a process is getting out of control.

In most cases an IR instrument receives a combination of reflected energy and emitted energy. The relationship of these energies is how we designate whether the object being measured is an efficient radiator.  The best radiator is one where emissivity is more than 90% of the total and reflected energy is less than 10%.  A poor emitter-one with, perhaps, 20% radiating efficiency provides 80% of the energy to the sensor from reflected sources.  In other words, a target surface with 20% percent emissivity might provide most of the energy to a sensor by reflection, and a nearby furnace's reflection might completely mask the temperature of the target.

For accurate infrared readings, then, determine the emissivity level of the surface material you are measuring and avoid looking at hot, spurious reflected targets. If you don't know the emissivity level, you still can measure temperature with infrared by "forcing" the emissivity to a known high level. That is, cover the surface with masking tape (which has an emissivity of 95%) or a highly emissive paint. You also can look up your target's emissivity in an emissivity reference table. However, tables cannot account for localized conditions, such as oxidation and surface roughness, so the information may not be entirely accurate for your situation.

If you are unsure whether an IR sensor or a thermocouple is the appropriate instrument to use for a specific measuring situation, the 10 questions below can help you decide. If you answer "Yes" to any of these questions about the surface you are measuring, you probably should use IR detection.

  Is the surface too hot to be measured with a thermocouple?
  Is it too large to be measured without many thermocouples?
  Is it moving so much it might break the thermocouple's lead wire?
  Is its electrical potential so high that using a thermocouple might be dangerous?
  Is it so low in mass that the thermocouple itself might affect the surface temperature?
  Is it too fragile or wet to accommodate a thermocouple contact?
  Is it too chemically active to accept a thermocouple or its probe?
  Is its atmosphere too hostile for a thermocouple
  Is it inaccessible to a thermocouple or its instrumentation?
  Is it near noise-producing electric or magnetic fields?