About Temperature Sensors
Site Web

Site Links:
Environment & Weather

Other links:

Note: The E-trail has needed updating for some time. We are getting to it now, so please be patient if you see pages with different formats. They'll all look great soon!
The Emissivity Trail: Stop 1
Total or Spectral Emissivity?
(Does it matter? You bet it does!)

Where do you get your emissivity values?
Do you look them up in a table?

Does the table specify spectral emissivity values?
Great !

You've got a head start. Better still, if you had the emissivity measured for you by a laboratory that knew their beans, you have a better emissivity value to use to correct your Infrared or IR Radiation Thermometer, about 2/3 of the time.

If the term"Spectral Emissivity" is foreign to you, you could have temperature measurement problems you don't even know about.

Why is spectral emissivity important?

Because it is the emissivity that one needs to correct the reading of an IR Thermometer, or a quantitative Thermal Imager, in cases where the object being measured is substantially hotter than anything else around and is not transparent

If you use the total emissivity found in many tables you are not likely to be using the correct values for your instrument, unless you have a total radiation thermometer! There are a whole bunch of them on the market at really low prices just like in the 1930's through 1950's except using modern detectors. But the low cost, simple gadgets don't seem to be used in places where measuring accuracy is more important than price, unless someone made a mistake.

Our focus in all these pages is not on barely adequate temperature sensors, we're thinking about ones that are used to obtain measurements accurate and repeatable to 1% of temperature or better. Otherwise, why bother? It's easy to get poor measurement quality, It takes a bit more effort to do things right! That's the key to repeatable and accurate measurements.

Most modern, high accuracy IR Radiation Thermometers have a carefully selected, relatively narrow spectral response wavelength passband. That means that they depend upon the emission properties of the object of measurement in that same spectral passband.

Most metals, especially really clean surfaces, and all oxides and dielectric materials have emissivity that varies with wavelength. If you use the total emissivity instead of the spectral emissivity for that fancy sensor that cost you a zillion pesos, you are likely to get some serious measuring errors with that value.

Even if you use a "modern" Total Radiation Thermometer, you may need to know the correct "total" emissivity setting to get the best possible accuracy because even modern "totals" are't truely total i.e. very few, if any respond with a Temperature raised to the 4th power response function despite what you may hear. The proof is in the testing, but that's a whole seperate book on testing Infrared Radiation Thermometers and not necesary for understanding the few points made here!

Realize, too that there are cases when you do not need to correct the sensor for emissivity: e.g.
1. when the object of measurement is immersed in an environment at about the same temperature, and,
2 when the object is immersed in an environment that is significantly hotter than the object.
In both of these cases one makes the errors larger by using an emissivity correction. There's more to the story than this, but that's coming at later stops along the E-Trail.

So, what good is the total emissivity?
It is very valuable in the engineering discipline of radiation heat transfer. Most of the emissivity tables published are for that discipline. Even some vendors of non-contact temperature sensors reference such tables (incorrectly). We have found this fact to be a great screening tool to seperate the solid technical suppliers of such devices from those who are a bit squishy.

If this doesn't make sense to you, try some faith, faith in simple rote, like:

Repeat the E-trail mantra #1 200 times before lunch for the next week:

Emissivity, no mystery. Total for heat transfer, Spectral for temperature.

Got It? Good!

Does the total emissivity help at you at all? Well....,

Not really. (Check the logic here, it's pretty hard to beat!)

It helps, but only for some materials which have relatively flat spectral emissivity curves. However, you have to know the spectral emissivity properties in order to judge how flat the spectral emissivity curve is (otherwise you are guessing or doing "handwaving" science or engineering e.g "it doesn't vary much, not to worry" {followed by a wave of a hand or display of thumb and forefinger very close together-not quite quantitative}).

If you need the spectral emissivity curves in order to be able to evaluate the total emissivity values, why do you need the total emissivity values at all? (Duh! Why didn't they tell us that in the first place? Gooood question, lots of possible answers that we don't repeat in proper company or even around the campfire on the E-Trail when we're funnin'.)

Bottom line: get the spectral emissivity values, they are the only things you will need. There are several vendors that will measure the spectral emissivity of representative samples of your materials, especially industrial solid materials- some do it at very low cost, sometimes free if you are a regular customer of other products and services. Sometimes there is a problem getting truely representative samples, but that gets into a lot more detail than we want to just now. Hang in, we'll get there eventually!

If you come away from the E-Trail with only one saddle sore, let it be this one:
Spectral e-missivity is it!

One final, thought-inviting question on this part of the trail.

If you use a Two Color Pyrometer or Ratio Radiation Thermometer, how do you select an e-slope or non-greyness correction or an emissivity ratio setting for it?

Did someone offer you a pair of Total Emissivity values for the device or just tell you that you can forget about emissivity with a ratio pyro? On many materials the e-slope is a critical setting because ratio devices are more sensitive to errors in emissivity ratio correction than comparable single waveband IR thermometers are to errors in emissivity correction. We'll get to the math a little later to prove it, but be careful when someone tells you that the ratio units are emissivity-free. It's all a matter of how sensitive they are to errors in setting the emissivity ratio and the actual value for your material. Ask for the numbers or numerical estimates, if in doubt. If the salesman can't back claims with facts and numbers then perhaps you need to find a more knowledgeable salesperson, one that can.

If you have some other inputs or questions, drop us a feedback. We'd really like to understand how all the people using Two Color devices are coping (just click on our highlighted link below and you'll get a feedback form).

(NOTE: One of these pages along the E-missivity trail will get to explaining what emissivity is, besides just a number you need to correct a temperature reading. The concept is important, but so, too, is some of the background.

Hang in there! Check out the rest of the trail while you wait.

The Emissivity Trail-Stop 2 or, the trail next followed.

References to Read or, the best trail to follow.

Go back to the Start of the E-trail.

Feedback on your experiences with emissivity and/or ratio thermometers

P.S. Interested in a listing on our vendor directories. It's self service at TempSensor.net, click here.




Home | Applications | Sensors | Vendors | Displays & Electronics | Calibration
E-missivity | Links | References | Meetings

Copyright © Temperatures.com, Inc. 2003-2004. All rights reserved.
The URL for this page is: http://www.temperatures.com/etrail1.html