As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.
Because the terminal is smaller than the circle the instrument averages, and the rest of that circle is cooler metal. A single displayed temperature describes the whole circle, so a ten millimetre lug occupying four percent of a fifty millimetre circle contributes roughly four percent of the answer. A lug genuinely at 90 degrees Celsius in a panel at 35 degrees can be reported below 40. Standing closer, using tighter optics or switching to an imager are the fixes; turning the emissivity dial is not.
A reading is an average over everything in the circle
Non contact thermometers have no spatial resolution at all. One detector receives radiation from a cone, converts the total to a temperature and displays it. The display therefore answers a question nobody asked: what single temperature would a uniform surface have to be at to radiate as much as this mixed scene does. When the scene is uniform that answer is exactly right. When the scene is a small hot object surrounded by cool busbar, trunking and air, the answer is dominated by the cool part because the cool part owns most of the area.
Electrical terminals are close to a worst case. The thing that overheats is usually a joint, a crimp or a lug a few millimetres across, bolted to a conductor designed to carry heat away from it, inside an enclosure at ambient. Everything about that arrangement works to dilute the one reading you want.
Working the dilution arithmetic on a lug
Take a 12 to 1 instrument held 600 millimetres back, which is a natural distance when a panel is open and you would rather not lean in. The circle is 600 divided by 12, or 50 millimetres. Put a 10 millimetre hot zone at 90 degrees Celsius in the middle of it, with the surrounding metal at 35 degrees. Area scales with the square of diameter, so the hot zone is ten over fifty squared, which is four percent of the circle.
| Width of the hot zone | Share of a 50 mm circle | Temperature the instrument reports |
|---|---|---|
| 10 mm | 4 percent | About 37.8 degrees Celsius |
| 20 mm | 16 percent | About 45.9 degrees Celsius |
| 30 mm | 36 percent | About 58.0 degrees Celsius |
| 40 mm | 64 percent | About 73.1 degrees Celsius |
| 50 mm, filling the circle | 100 percent | 90.0 degrees Celsius |
Those figures assume a hot zone at 90 degrees Celsius, surroundings at 35, and the same emissivity across both. The first row is the one to remember: a terminal hot enough to warrant an immediate shutdown is reported as three degrees above the air around it.
Radiance averages before temperature does
A simple area weighted average of temperature would give 37.2 degrees Celsius for that first row. The instrument reports nearer 37.8, and the reason is worth understanding because it works in your favour. The detector sums radiant power, and radiant power rises with the fourth power of absolute temperature, so the hot four percent contributes more power than its area share implies. Converting the summed power back to a temperature therefore lands slightly above the naive average.
The bonus grows with the temperature difference. It is the mechanism that lets a spot thermometer find a severely overheating joint at all, because a very hot small target eventually drags the average up far enough to notice. It is also why a modest fault, the kind worth catching before it becomes a failure, hides so effectively: a joint twenty degrees above its neighbours barely moves a diluted average.
Three other reasons the terminal reads low
- Plated and bright metal. Tinned lugs, bare copper and stainless fixings radiate poorly, so even the four percent of the circle that is the terminal hands over less signal than a matte surface at the same temperature would.
- Reflections inside the enclosure. A steel box reflects infrared onto every surface in it, so part of what arrives from a shiny lug is a picture of cooler metal elsewhere in the panel rather than the lug itself.
- The standoff you chose for safety. Keeping a sensible distance from live parts is correct, and it enlarges the circle in direct proportion, which is the trade off that makes the first row of the table so common in practice.
Getting a terminal temperature you can act on
The geometry gives you three honest routes. Shrink the circle by closing the distance where it is safe to do so, since the diameter falls in direct proportion and the area falls with its square. Shrink it further with tighter optics: a 16 to 1 infrared gun states a 0.5 second response and an adjustable emissivity from 0.1 to 1.0, and at 600 millimetres its circle is 38 millimetres rather than 50, which lifts a 10 millimetre lug from four percent of the average to seven. Its sibling 12 to 1 model publishes the same accuracy figure of plus or minus 1.5 percent with a response under 500 milliseconds, which makes the ratio the deciding difference between them; whether that 1.5 percent is of reading or of full scale is not stated in the listing bullets, and on a terminal barely above ambient that distinction is worth more than the digit it changes.
Use the instrument for what dilution cannot spoil, which is comparison. Three phases read identically from the same distance will still rank themselves correctly, and a warm outlier is a reason to look closer regardless of its absolute value. Record the distance alongside every reading, because a reading without its standoff cannot be repeated. Adjusting emissivity, by contrast, is the wrong lever here: it corrects the surface property and does nothing about area, so an instrument like the dual laser gun with emissivity from 0.1 to 1.0 still reports an average of whatever its 12 to 1 circle covers. Where absolute terminal temperatures genuinely matter, the array based alternative is the right tool, and the trade is laid out in thermal camera against infrared thermometer. Published optics for individual models appear in the infrared thermometer comparison, the TempPro review and how to choose an infrared thermometer.
Frequently Asked Questions
How big can the error be?
On a ten millimetre lug inside a fifty millimetre circle, a terminal genuinely at 90 degrees Celsius against 35 degree surroundings is reported near 38 degrees. The instrument is reporting an honest average of what it was pointed at.
Does standing closer fix it?
It is the single most effective fix, because the circle shrinks in direct proportion to the distance. Halving the standoff halves the circle diameter and quarters its area, which raises the share of the average that belongs to the terminal.
Why is the real reading higher than a simple area average?
Because the instrument averages radiant power rather than temperature, and power rises with the fourth power of absolute temperature. The hot fraction therefore contributes more than its share of area, which is a small mercy rather than a rescue.
Does a higher ratio instrument solve it?
It shrinks the circle at a given distance, so a 16 to 1 instrument measures a smaller area than a 12 to 1 one from the same place. It does not change the fact that the circle must sit inside the hot part to report it.
Can I trust the reading as a screening tool?
Yes, for comparisons between identical points in the same enclosure, which is how most electrical screening is done. A phase reading several degrees above its two neighbours is meaningful even when none of the three absolute values is correct.
What does the panel door do to the reading?
A metal enclosure reflects infrared from everything inside it, including the hot part you are trying to measure and the cooler metal around it, so part of what you receive from a shiny lug is a picture of the box rather than the lug.