Measuring position in a laboratory is normally performed at a consistent temperature to maintain accuracy, but certain specifications demand precise position measurement across a broad range of operating temperatures, which is more challenging.
This article outlines the main issues and offers 10 informative tips for design engineers.
Understanding the Problem
Being clear about what is required is the first stage in planning for precise position measurement across broad temperature ranges. The key data requirements are:
- Maximum and minimum storage temperatures
- Maximum and minimum operating temperatures
- Maximum permissible measurement error at the various operating temperatures
Tip #1 – Be Clear About What the Real Requirements are.
A cost budget and an error budget are the two realistic budgets to consider when planning a technical solution to any requirement.
The Error Budget
The variation between the measured and true position will include multiple different error sources, such as:
- Thermal drift in the output of the sensor
- Errors from the measurement performance of the position sensor
- Thermal effects in the mechanical structure of the host, mostly from differential thermal expansion
- Mechanical effects as a result of clearances in bearings or couplings, for example, backlash in gears
Tip #2 – Prepare an Error Budget and Ensure all Contributing Factors are Considered.
Understanding Measurement Performance
The following is a summary of position sensor measurement performance:
- Resolution refers to the smallest measurable variation in position
- Accuracy describes the maximum amount of deviation from the true position
- Linearity describes how effectively the output, over a range, matches a straight line. In many cases, accuracy and linearity are the same if no offset is present.
- Repeatability describes the level of reproducibility
Tip #3 – Understand What Aspects of Measurement Performance are Important in Your Application and Ensure the Sensing System Aligns with These Requirements.
Whenever measurement performance parameters are outlined, they should be linked to a temperature as well as a temperature coefficient. This coefficient outlines the variation in the output of the sensor as the temperature changes.
A device that is thermally stable will have a small temperature coefficient. Temperature coefficients are usually expressed in parts per million per Kelvin. This is a challenging unit as it is normally a minute number and is frequently ignored. This number must be multiplied by the temperature differential.
Tip #4 – Make Sure Your Error Budget Includes the Sensor Temperature Coefficient.
Differential Thermal and Mechanical Effects
It is usually not the sensor elements that are the main focus, but instead the positioning of the host elements, for example the angle of a shaft. These will also contribute to the error budget as a result of factors like thermal expansion, clearances, backlash, and mechanical tolerances.
Thermal expansion is a natural occurrence that should not be overlooked. Differential thermal expansion is more challenging as it can result in large measurement errors if its influence is not reduced by proper material selection or mechanical design.
The impact of thermal expansion can be negated if the mounting structure of the sensor contracts or expands by the same amount as the components being analyzed.
The thermal coefficient of the sensor should ideally match the effects of thermal expansion. In many examples, the mechanical configuration of the sensor in relation to the host equipment means that differential thermal expansion must be considered in the error budget.
In cases where this effect takes up the majority of the error budget, temperature compensation is always an option. This is where the local temperature is measured and the sensor’s output is compensated in relation to this.
This method is ineffective as the temperature is not likely to be uniform, there is an increase in complexity and cost, and there will be temperature or time lags which all contribute to a reduction in reliability.
Tip #5 – Mechanical Effects and Differential Thermal Expansion Must be Included in any Error Budget. Wherever Possible, These Should be Minimized Through Careful Mechanical Design and Material Selection.
Choosing the Right Sensor
The basic physics of a position sensor normally explains how large its temperature coefficient is. Selecting the correct sensor for the application is made simpler with a simple understanding of this. The following are some of the main concepts that are utilized to measure position:
- Potentiometers: They quantify the resistance of a material that is electrically conductive. The coefficients are likely to be bigger because conductivity varies with temperature.
- Optical: While light transmission is independent of temperature, alignment can be influenced by thermal drift which reduces accuracy.
- Magnetic: Magnetic sensors analyze field strength which changes with temperature, so thermal coefficients may be high. Tight installation tolerances are required for precision magnetic sensors, so further thermal effects may be significant.
- Capacitive: As capacitance varies with temperature, a great number of capacitive devices have significant temperature coefficients, which are increased more by variations in humidity. Capacitive sensors are also prone to foreign matter and condensation.
- Inductive: Inductance changes with temperature but the majority of precision inductive sensors utilize a ratiometric method founded on the ratio of two inductances at minimum. As the values of both will vary by equivalent amounts, thermal coefficients are normally low.
Tip #6 – Design the Host System to Minimize Differential Thermal Expansion.
Tip #7 – Select a Sensor with a Small Thermal Coefficient or One Whose Temperature Coefficient Matches the Host System.
Inductive sensors (resolvers) utilize transformer methods with precision wound spools. They have become the first preference in the aerospace, oil and gas, and military industries, where there is frequently a broad range of temperatures.
The fundamental physics means that they are perfectly tailored to challenging operating environments but they are not commonly utilized because of their high cost, bulk, and weight.
Tip #8 – Resolvers are Often an Automatic Choice for High Temperature Applications.
New Generation Inductive Sensors
The inductive encoder is the new generation of inductive sensor. It utilizes the same fundamental physics as the resolver, but instead of the complex analog electronics and the bulky transformer constructions, the inductive encoder employs digital electronics and printed circuit boards.
This development opens up a host of applications for inductive sensors, including 2D and 3D sensors, short throw linear devices (less than 1 mm), high precision angle encoders, and curvilinear geometries.
Tip #9 – If a Traditional Inductive Sensor is Too Big, Bulky, Expensive or Not Sufficiently Accurate, Consider One of the New Generation of Inductive Sensors.
Along with being lightweight and compact, the inductive encoder provides incredibly stable measurements over a broad range of temperatures.
Incoders have minute temperature coefficients of less than 0,25 ppm/K, which translates to a change that is smaller than one fifth of an arc-second per Celsius.
Tip #10 – Inductive Encoders Offer Especially Low Thermal Coefficients.
This information has been sourced, reviewed and adapted from materials provided by Celera Motion.
For more information on this source, please visit Celera Motion.