learn more

Temperature influence

Correct the temperature influence

For larger workpieces and higher accuracy requirements, the influence of temperature must be estimated or corrected[13]. If there is no temperature compensation, measurement is usually only useful under measuring room conditions. In this context, it is particularly important for the user to estimate the temperature influences. The influencing parameters in the measuring machine, in particular the expansion behaviour of the scales, must be taken into account. As a rule, these are not known exactly to the user.

Estimating the influence of temperature is important

A special case exists if the material of the machine scales largely corresponds to that of the measuring object with regard to the coefficient of expansion. This is roughly the case with steel scales and steel workpieces, for example. If it is ensured that the temperatures on the scales and on the workpiece largely match, a temperature compensation is not necessarily required in order to be able to measure at temperatures deviating from 20 °C. However, in the case of larger temperature deviations, a temperature compensation is not necessary. However, in the case of larger temperature deviations, other influences, e.g. due to changes in the machine geometry, can also lead to measurement errors.

Temperature compensation reduces the influence

If the measuring machine has integrated temperature compensation and the workpiece temperature is taken into account, measurements can be carried out largely insensitive to temperature. With temperature-compensated machines, various measures are taken to reduce the influence of temperature on the measuring process. These measures can include

  • Temperature measurement on the scales and internal length correction or, alternatively, scales with thermal expansion close to zero without correction
  • Capturing the workpiece temperature and correcting the temperature-related expansion of the measuring object, taking into account its coefficient of expansion
  • Use of thermally particularly suitable materials for the machine structure, e.g. with good thermal conductivity or low thermal expansion
  • Correction of the thermally induced machine distortion
  • Thermally insulated structure

Measuring under production conditions

These measures can reduce the influence of temperature to such an extent that measurements can be taken with sufficiently low thermally induced measurement uncertainty even under production conditions. It should be noted that machines with temperature-stable scales (coefficient of expansion close to zero) can produce blatant measurement errors in the order of magnitude of the linear thermal expansion of the measuring object if the workpiece expansion is not corrected correctly.

Expansion coefficient must be known

The expansion coefficient of the workpiece required for the temperature compensation is usually taken from tables. With these values, a deviation in the order of 10 % of the nominal value is usually to be expected. If this is not sufficient for the temperature compensation, the coefficient must be calibrated on the workpiece at great expense. Calibration uncertainties in the order of 0.1 % of the nominal value are achieved here. The measurement error AT of the temperature T can be between 0.5 K and 0.05 K, depending on the quality and type of temperature measuring system.

Uncertainty considerations for temperature compensation[13] and the estimation of possible maximum measurement errors of the length measurement with deviating workpiece or measuring room temperature show that the influence of the expansion coefficient is very dominant (Fig. 68). The effects of different procedures on the expected measurement errors are discussed below on the basis of four cases:

  • Without temperature compensation, the difference Aαm between the expansion coefficients of the machine scales αm and the workpiece αw has an effect. It is common to use mchine scales made of steel or glass. The measurement error AL for the reference length L0 is calculated as follows:

    AL = L0 × Aαm × ΔT, with ΔT = T – 20 °C

  • When measuring steel, aluminum or plastic parts, for example, this results in very different measurement errors. Due to the small difference between the expansion coefficients of appliance scales and steel parts, relatively small measurement errors can be expected for the latter over a relatively wide temperature range. With plastic parts, on the other hand, unacceptable measurement errors in the order of 0.1 mm occur even at low temperature deviations. A usable measurement result can practically only be achieved at a reference temperature of 20 °C ± 1 K.
Expansion coefficient must be known
<p>Fig. 68: Temperature-related maximum measurement error in relation to the measurement length for the materials steel (ordinate blue), aluminum (ordinate green) and the plastic POM (ordinate red) with different methods and parameters of the temperature compensation: a-c) without temperature compensation, device scales made of glass; d-f) with temperature compensation for the materials corresponding to the respective ordinates, machine scales with zero expansion</p>

Machine behaviour becomes negligible

  • With temperature compensation, the scale behaviour is corrected very well and can usually be neglected. In conjunction with machine scales made of e.g. special ceramics (coefficient of expansion close to zero), the influence of the scale is almost completely eliminated. The maximum measurement errors result practically from the deviations in the temperature measurement of the workpiece AT and the determination of the expansion coefficient Aαw depending on the deviation from the reference temperature (ΔT = T – 20 °C) and the workpiece expansion coefficient αw:

    AL = L0 × (αw × AT + ΔT × A αw)

Simple temperature compensation brings benefits

Even with relatively coarse temperature measurement, the workpieces can be measured relatively independently of the temperature, but only with relatively large remaining measurement errors. The deviations increase in proportion to the expansion coefficient (Fig. 68d). An additional calibration of the expansion coefficient does not bring any significant improvement and can therefore be omitted. In practice, this procedure is sufficient in many cases and is usually a clear benefit compared to measuring without temperature compensation.

Accurate temperature compensation is optimal

  • If a very accurate temperature measurement is used, the remaining measurement errors are reduced considerably (Fig. 68e). However, if the expansion coefficient is not known very precisely, this effect is partially cancelled out if there are large deviations from the reference temperature. This procedure is generally recommended for calibration standard temperature conditions and accuracy requirements.

Calibration of the expansion coefficient is usually not necessary

  • The measurement error is only reliably very low if the expansion coefficient is well known and the temperature is measured accurately, even if the temperature of the measuring object deviates very strongly from the reference temperature (Fig. 68f). However, this procedure is very time-consuming due to the calibration of the expansion coefficient and is only used in special cases.

Useful measures

For accurate measurements or to reduce the influence of strong temperature fluctuations, the following additional measures should be applied according to need:

  • Enclosure in case of strong temperature fluctuations
  • No draughts or direct blowing on the coordinate measuring machine
  • Few heat sources in the immediate vicinity
  • As great a distance as possible from the walls
  • Thermal insulation of floor and walls
  • No direct sunlight or lighting
  • Electrical equipment of the coordinate measuring machine and lighting in 24-hour operation
  • Heat equalisation of the measuring objects before measurement (air shower)
  • Styli and extensions made of thermally insensitive materials
  • Short measuring times for low drift, alternatively repeated qualification of the datum system

Despite all the above measures, climatic chambers with a constant temperature (in terms of space and time) cannot be dispensed with for special accuracy requirements. The corresponding requirements for compliance with the machine specification can be found in the data sheets of the coordinate measuring machines.