Effects of Temperature on Dimensional Measurement
A common phenomenon well-known to engineers and metrologists alike, is that everything changes when temperatures change. Although these subtle changes are usually invisible to the naked eye, the need for expansion joints when constructing bridges presents a more noticeable example, as a bridge’s span length increases and decreases due to the current seasonal climate. In industrial measurement where tolerances are low, and a small difference in size becomes critical, then these small, subtle changes in temperature can become very important.
Whenever a material is subjected to heat, then the distance between the individual atoms will change. This distance is dependable on the material used, although the most common change to occur is an increase in distance. Since this heat effects all atoms in the material equally, then the change in length is proportional to the original length. So, a constant of proportionality can be observed as the length change is proportional to the temperature change, thus coining “The Coefficient of Thermal Expansion”, though over large temperature ranges, problems can still occur.
Because of the above coefficient, to define the size of a piece of material, we must also state the temperature reading at the stated size. This is where standardization comes into effect, as ISO 1 states that all dimensional measurements should take place when the material is 20˚C (68˚F), unless otherwise stated.
Above is a table showing a number of common materials, along with how they respond to changes in temperature. The CTE numbers in the second column are averages, as many different material types can still use the same names (e.g Stainless, and High Carbon Steel). The third and final column represents an estimate regarding the importance of temperature in measurements. A very high accuracy measurement would be a length measurement that has an uncertainty of around 1ppm.
100mm Samples & 5C Expansion
The above graph illustrates another example of how to think of thermal expansion. If all samples shown have a length of 100mm, then raising a temperature by 5˚C increments will result in the lengths changing as shown. This provides a rough template for how close a metrology laboratory should be to 20˚C, depending on what type of material is being measured. Also, the graph shows two types of steel gage blocks (25nm & 500nm), this is due to the peculiarity within steel gage blocks, in that the CTE is dependent on the length of the block.
Ultimately then, an understanding of thermal expansion is needed to provide highly accurate measurement results. Here at QCI, our laboratory is set at a constant 20˚C resulting in the perfect environment to provide customers guaranteed precise metrology solutions. Our Midlands based metrology centre uses the latest technology, offering a full range of measurement, reporting and programming services. Customers receive the measurement information they need for product verification and quality control to the highest degree of accuracy. We operate a temperature controlled clean room environment to help minimise factors that contribute to measurement uncertainty, such as thermal expansion which is covered above.