Featured Video
This Week in Quality Digest Live
Metrology Features
Dean Solberg
If you can think about it and model it, you can 3D-print it
Gary Confalone

More Features

Metrology News
Four times faster data acquisition, enabling precise measurement at the submicron level
Caters to the growing need for sophisticated scanning in an entry-level solution that is budget friendly
Topics for the Go Forth and Measure project are virtually unlimited
Ask questions, exchange ideas and best practices, share product tips, discuss challenges in quality improvement initiatives

More News

Fred Mason


Compensating for Temperature

What’s it all mean?

Published: Wednesday, December 13, 2006 - 22:00

Continuing last month’s topic of temperature effects on measurements, what about temperature compensation? Some measuring systems claim increased accuracy with the usually optional temperature compensation. What’s that? What’s it supposed to do? And does it do what it claims to do?

Remember that parts, especially metals, expand when heated. When the length of that part is being measured, it’s important to take temperature-related influences into account. A common example is handling gauge blocks with tweezers to avoid transferring heat from your hands to the blocks.

Temperature-induced changes in the length of a part depend on the magnitude of the change in temperature and the rate of change. You probably recall that the temperature of a gas changes faster than that of a liquid, which changes faster than that of a solid. In other words, it takes longer to raise or lower the temperature of anything solid than it does liquids or gas. This means that if you measure the air near a part, you don’t really know the temperature of the part. We’ll get to why this matters a little later.

The temperature of a closed system
Let’s start with a common, nonmeasuring example where temperature management is crucial. The engine of your car is a solid (the metal block) that uses a liquid coolant, that itself is cooled by moving air (gas) from the fan. Internal combustion and friction heat up the engine as it operates. Coolant (and oil) circulates throughout the engine block to dissipate some of that heat. Coolant also circulates through the radiator where cool air dissipates some of the coolant heat as it passes around the tiny channels in the radiator. The refrigerated coolant returns to the block where this continuous process maintains the engine’s temperature within a specific range. Failure of any of these cooling mechanisms shows what temperature-induced expansion can do to your car.

You notice that your dashboard temperature gauge points to a certain position when you drive your warmed-up car. Performance of the engine, its horsepower, fuel consumption, and overall efficiency are based on it operating at or near that point. If the water pump fails, coolant doesn’t circulate through the radiator, heat builds up (the temperature gauge reading rises) in the engine, and the engine starts to sputter and stammer. The higher-than-normal heating causes engine components to expand so they no longer have the clearances they were designed to have. That, in turn, increases friction, creating more heat, and, if you don’t shut off the car in time, causes the engine to seize and stop operating—not a pleasant thought.

What does this have to do with measurement? We’re getting there. The explosion of the air-fuel mixture in each cylinder of an internal combustion engine takes place at thousands of degrees. The coolant in the radiator of an operating automobile is typically around 170 degrees. The air that blows through the radiator is at the current air temperature, which depends on the season, locale, etc. The flowrate of the air over the radiator also matters. In any case, it’s much cooler than the temperature of the operating coolant in the radiator, which too is much cooler than the cylinders where the sparkplugs are firing. The $64,000 question: Where is the temperature sensor and what is it measuring?

Inference from the temperature
The engine temperature sensor is near one of the lines to the radiator where it measures the coolant temperature. In other words, the temperature gauge on your dashboard provides a visual indication of the temperature of the coolant. This reading, if normal, implies that the combustion temperature is also normal, and that the entire system is normal. Obviously, if the temperature sensor were in one of the cylinders, that reading would be much, much higher due to the temperature of the combusting air-fuel mixture. If the sensor was outside the radiator it might give the outdoor temperature. The temperature of an automobile engine is measured in one location, and from that reading, it’s inferred that the rest of the system is also at the appropriate temperature. If the coolant temperature reading is OK, there’s no need to worry about the combustion temperature or the air temperature outside; you can assume all systems are normal.

Compensating for temperature
So what does an overheating engine have to do with measurement? Some coordinate measuring machines (CMMs) have temperature compensation systems that measure environmental attributes (temperature being the largest contributor) in or on the machine, preferably near the measuring area. And that’s the point of the overheating engine example: Where the compensation system measures the temperature is important.

Because the size of a part can change with temperature, the temperature compensation system tries to measure the temperature at or near the part, so its measurements can be corrected for those changes. That raises several issues. Exactly where are those temperature measurements taking place? You can put a sensor on the part itself. You can put more sensors on the measuring machine. Where on the machine? How many do you need? Do they measure the air or the solid structure of the measuring machine? Are they accurate? Do they track the environmental changes? What do their readings do to part measurements? Yes, temperature compensation is meant to remove one of the variables that affect measurements. But adding temperature sensors to collect data that will be used to correct part measurements increases the potential for measurement confusion. It doesn’t reduce it.

Why compensate?
Why use temperature compensation at all? The cynical answer is that the measuring machine is poorly constructed of thermally sensitive materials, especially combinations of materials that all have different temperature coefficients—that shrink and expand at different rates—they need a temperature compensation system to minimize the effects of those variations. No matter how well the measurement system is constructed, the location of the temperature compensation sensors can be critical. How good are those sensors?

On the other hand, a measuring machine can be built of engineered materials that exhibit minimal thermal sensitivity—materials such as granite and cast iron that react to thermal changes much more slowly than lighter weight, perhaps less expensive, materials. Such solidly built systems, when used in work processes that follow appropriate part-handling practices, can avoid the need to infer part measurements from a temperature compensation mechanism.

Think about it—why measure something else to improve the accuracy of what you actually want to measure? Because every measurement is prone to error, measure directly and avoid adjusting measurements based on temperature sensors that may or may not be measuring conditions that affect your parts.

What is it really compensating for?
Temperature compensation is sold as a way to improve measurement accuracy. When you consider the design of measurement systems, take into account the structural design, and the quality and integrity of the materials of which the system is built. Does temperature compensation improve measurement accuracy, or does it try to make up for design shortcuts and cost reductions? Buy a well-designed measurement system and put a thermometer in the window.

Until next time, yes, measurement matters.


About The Author

Fred Mason’s picture

Fred Mason

Frederick Mason has more than 20 years of experience in metrology in engineering and in domestic and international marketing roles. He has a broad range of experience, including holography, laser and white-light interferometry, microscopy, and video and multisensor metrology. He’s the vice president, marketing communications, for Quality Vision International, parent company of Optical Gaging Products, RAM Optical Instrumentation, VIEW Micro-Metrology, and Quality Vision Services.