Temperature Measurement · Non-Contact Sensing · Instrument History
What Is an Optical Pyrometer? How It Reads Temperature by Eye, Not by Touch
Long before electronics existed, engineers measured molten steel by matching the color of light with their own eyes. This guide explains what an optical pyrometer is, how it actually works, the real physics behind it, and a working calculator you can try yourself.
What Is an Optical Pyrometer?
An optical pyrometer measures the temperature of something too hot to touch, just by looking at the light it gives off. No wires touch the object, no probe goes near the heat, the whole reading happens through a lens and a human eye.
Before there were sensors and circuit boards, potters and metalworkers still needed to know how hot their kilns and furnaces were. In the 1700s, the English potter Josiah Wedgwood built one of the earliest heat measuring devices, simply to know if his pottery kiln was hot enough. He couldn't stick a thermometer inside molten clay, so he found another way, watching how clay shrank at different heats.
That same basic idea, judge the heat by watching what it does, eventually led to something far more precise. In 1901, two scientists in Germany, Ludwig Holborn and Ferdinand Kurlbaum, and a separate inventor in the United States, Everett Fleet Morse, all built the same kind of device independently, at almost the same time. It was called the disappearing filament pyrometer, and it was the first instrument that could reliably measure temperatures above 1000°C. That single invention opened the door to modern steelmaking, since suddenly engineers could actually measure furnace heat instead of guessing at it.
For decades afterward, companies like Leeds & Northrup in the United States and the Cambridge Instrument Company in England built thousands of these devices, many still surviving today in university physics collections and industrial museums. A typical unit from that era, the Leeds & Northrup Catalogue 8622 for example, could read anywhere from about 775°C up to 2800°C, all from a battery powered box roughly the size of a small toolbox. Skilled operators using these instruments could reach a resolution of about 10°C, even at readings close to 2000°C, a genuinely impressive result for a device with no electronics at all.
Eventually, engineers found ways to automate the brightness matching itself, replacing the human eye with a photocell, and later a microprocessor, that could perform the same comparison electronically and far more consistently. Those automatic versions still relied on the same basic filament comparison trick, just executed by a machine instead of a person. From there, it was a short step to the fully electronic spectral-band and two-color pyrometers that dominate industrial temperature measurement today.
How the Disappearing Filament Trick Actually Works
The classic optical pyrometer uses a simple comparison. It doesn't measure temperature directly at all, it matches brightness, and brightness happens to be a reliable stand-in for temperature.
If the filament looks brighter than the object, it's too hot. Turn the current down. If it looks darker, a visible dark line against the glow, it's too cool. Turn the current up. Only at the exact match does the wire seem to vanish completely, blending perfectly into the glow behind it. A dial connected to that current, calibrated in advance against known temperatures, gives the final reading.
Watch: Disappearing Filament Optical Pyrometer in Action
This video shows the filament matching process described above, which is much easier to see than to read about.
Why a Hot Object Changes Color as It Heats Up
Every heated object gives off light across a whole range of wavelengths, not just one color. What changes with temperature is which wavelength carries the most energy. A cooler object, say around 600°C, glows mostly in deep red. Push the temperature higher, and that peak shifts toward orange, then yellow, then eventually a blue-white glow at extremely high temperatures, the same reason a welding arc looks so much whiter and harsher than a bar of steel glowing dull red in a forge.
This shift is described by Wien's displacement law, a genuine physical relationship between temperature and the wavelength where radiation peaks. It's part of the same family of physics, blackbody radiation, that Planck's law describes in full. An optical pyrometer doesn't need to solve any of that math directly, the disappearing filament trick sidesteps it entirely by comparing brightness at a single fixed wavelength instead of trying to measure the whole spectrum. But understanding why the color shifts at all helps explain why a red filter, isolating one narrow, predictable slice of that shifting spectrum, makes the whole brightness comparison far more reliable.
Key Terms You Need to Know First
The temperature an optical pyrometer directly reads off its dial. It assumes the object behaves like a perfect radiator. Real objects usually don't, so this number is often slightly wrong.
The object's actual temperature, corrected for how well or poorly it radiates heat compared to a perfect radiator. Almost always equal to or higher than the brightness temperature.
A number between 0 and 1 describing how efficiently a surface gives off heat radiation compared to a perfect radiator. Polished metal might sit around 0.2 to 0.3. Rough, oxidized, or painted surfaces sit much closer to 1.
A theoretical, perfect radiator with an emissivity of exactly 1. Nothing real is quite a true blackbody, but many rough, oxidized surfaces come close enough for practical purposes.
Why the Dial Reading Isn't Always the Real Temperature
Here's the catch with brightness matching. A perfect blackbody radiator glows a certain way at a certain temperature, and the pyrometer is calibrated against that perfect case. Most real objects, shiny metal especially, don't radiate quite that efficiently. They look dimmer than a blackbody at the same actual temperature. So the pyrometer, fooled by that dimness, reads a temperature that's too low.
This gap between what the dial says and what's actually true can be corrected mathematically, using a formula built on Wien's approximation of Planck's radiation law.
Where:
T = true temperature (kelvin)
Tb = brightness temperature, the dial reading (kelvin)
λ = effective wavelength, commonly 650 nm for a red filter
C2 = second radiation constant, 0.014388 m·K
ε = emissivity of the surface (0 to 1)
When emissivity equals 1, a perfect blackbody, ln(1) is zero and the true temperature equals the brightness temperature exactly, no correction needed. Below 1, the correction always pushes the true temperature higher than what the dial shows.
A polished metal surface and a rough oxidized surface at the same real temperature will show the same reading on an optical pyrometer.
They won't. The polished surface, with lower emissivity, radiates less light and reads noticeably cooler on the dial than the rough surface, even though both are genuinely the same temperature.
Using a red filter just makes the image easier on the eyes, nothing more.
The red filter narrows the comparison to one known wavelength band, which is exactly what makes the emissivity correction formula usable in the first place. Without it, the instrument would be comparing a messy mix of wavelengths with no single, defined value to calculate against.
True Temperature Calculator
Enter the brightness temperature shown on the dial, along with the surface emissivity, to estimate the real temperature.
Optical Pyrometer vs Other Non-Contact Temperature Tools
| Type | How It Measures | Typical Range | Emissivity Sensitive? |
|---|---|---|---|
| Optical (disappearing filament) | Brightness match at one wavelength, usually red | 600°C to 3000°C | Yes, significantly |
| Two-color (ratio) pyrometer | Compares brightness at two wavelengths | 250°C and above | Much less, ratio partially cancels emissivity error |
| Radiation (broadband IR) pyrometer | Total radiated energy onto a thermocouple sensor | 500°C and up, effectively no upper limit | Yes, requires manual emissivity setting |
Table data cross-referenced from ScienceDirect Engineering Topics and Optris technical resources.
How These Instruments Actually Get Calibrated
An optical pyrometer is only as good as its reference lamp, and that lamp doesn't calibrate itself. The process usually starts against a known temperature standard, historically the melting points of pure metals like gold or copper, which occur at exact, repeatable temperatures every single time.
The operator points the pyrometer at the reference standard while it sits precisely at its known melting point, then adjusts the internal calibration so the dial reads that exact value. This gets repeated at a second known temperature further along the scale, giving two fixed points to define the relationship between filament current and true temperature across the whole range. Everything in between gets interpolated from those two calibration points. This is also exactly why historic optical pyrometers were trusted as reference instruments themselves, other temperature sensors were often checked against a properly calibrated optical pyrometer, not the other way around.
Advantages and Disadvantages of the Optical Pyrometer
- No physical contact needed with the hot object
- Distance from the target doesn't matter, only the image size in the eyepiece
- Simple, rugged, mostly mechanical design with few parts to fail
- Reasonably accurate, around ±5°C under ideal conditions
- Needs a skilled human operator, cannot run unattended
- Only works above roughly 600 to 700°C, when the object glows visibly
- Reading depends on emissivity, requiring correction for accurate results
- Operator judgment can add another ±10°C of error on top of the instrument's own accuracy
Do's and Don'ts When Using an Optical Pyrometer
- Make sure the target fills enough of the eyepiece image before taking a reading
- Apply an emissivity correction when measuring polished or unusual surfaces
- Recalibrate the reference lamp regularly against a known temperature standard
- Use protective filters for extremely high temperature targets to protect the operator's eyes
- Don't trust the raw dial reading on shiny or polished metal without correcting for emissivity
- Don't attempt readings below the instrument's minimum visible glow temperature
- Don't expect continuous automatic monitoring, this is a manual, one reading at a time tool
- Don't view an unfiltered extremely hot source directly without proper eye protection
Where Optical Pyrometers Are Actually Used
Even though electronic instruments have replaced most everyday use, the specific conditions that originally made optical pyrometers valuable, extreme heat, no room for a contact probe, and a need for a trusted reference reading, still show up in a handful of genuine applications today.
Reading molten metal temperature without risking a contact probe.
Checking kiln heat where contact sensors would melt or foul.
Historically used as a reference standard for other instruments.
Measuring filament temperature in early incandescent lamp research.
Reading exhaust or combustion chamber temperatures from a safe distance.
Confirming workpiece temperature reaches the correct range before shaping.
Quick FAQs: Optical Pyrometer
- Thermocouple vs RTD: 7 Critical Differences That Decide Which Sensor You Actually Need
- Thermistor vs Thermocouple: 6 Key Differences and Which One Wins for Your Application
- How Does a Gas Filled Thermometer Work?
- What Is a Thermowell? Types, Selection and Design Explained
- How to Calibrate a Temperature Transmitter
External References
- Wikipedia: Disappearing-Filament Pyrometer
- ScienceDirect: Optical Pyrometer, Engineering Topics Overview
- Science Museum Group: Disappearing Filament Pyrometer, Cambridge Instrument Co.
What we learn today
- An optical pyrometer reads temperature by matching brightness, not by touching anything, using a filament that visually disappears when it matches the hot object's glow exactly.
- The instrument was invented independently in 1901 in both Germany and the United States, and became the first practical way to measure temperatures above 1000°C.
- The dial reading, called brightness temperature, needs a correction for emissivity to find the object's true temperature, since most real surfaces don't radiate as efficiently as a perfect blackbody.
- Modern electronic pyrometers have mostly replaced the manual optical type, but the underlying brightness comparison principle still shapes how non-contact temperature measurement works today.