Piezoresistive pressure sensors
What are piezoresistive strain gauge pressure sensors
Piezoresistive strain gauges are among the most common types of pressure sensors. They use the change in electrical resistance of a material when stretched to measure the pressure.
These sensors are suitable for a variety of applications because of their simplicity and robustness. They can be used for absolute, gauge, relative and differential pressure measurement, in both high- and low-pressure applications.
In this article we’ll discuss the various types of piezoresistive pressure sensors available, how they work, and their relative merits.
Working principle
The basic principle of the piezoresistive pressure sensor is to use a strain gauge made from a conductive material that changes its electrical resistance when it is stretched. The strain gauge can be attached to a diaphragm that recognises a change in resistance when the sensor element is deformed. The change in resistance is converted to an output signal
There are three separate effects that contribute to the change in resistance of a conductor. These are:
- The resistance of a conductor is proportional to its length so stretching increases the resistance
- As the conductor is stretched, its cross-sectional area is reduced, which also increases the resistance
- The inherent resistivity of some materials increases when it is stretched
The last of these, the piezoresistive effect, varies greatly between materials. The sensitivity is specified by the gauge factor, which is defined as the relative resistance change divided by the strain:
Where strain is defined as the relative change in length:
Pressure sensing elements
Strain gauge elements can be made of metal or a semiconducting material.
The resistance change in metal strain gauges is mainly due to the change in geometry (length and cross-section area) of the material. In some metals, for example platinum alloys, the piezoresistive effect can increase the sensitivity by a factor of two or more.
In semiconducting materials, the piezoresistive effect dominates, typically being orders of magnitude larger than the contribution from geometry.
| |
Piezoresistive strain gauge measurements are made using a Wheatstone bridge circuit |
Function
The change in resistance in the sensor is usually measured using a Wheatstone bridge circuit (as shown below). This allows small changes in the resistance of the sensor to be converted to an output voltage.
Piezoresistive strain gauge measurements are made using a Wheatstone bridge circuit
An excitation voltage needs to be provided to the bridge. When there is no strain and all the resistors in the bridge are balanced then the output will be zero volts. A change in pressure will cause a change in resistances in the bridge resulting in a corresponding output voltage or current. How this is calculated is shown in the formula below.
Performance can be improved by using two or four sensing elements in the bridge, with the elements in each pair being subject to equal and opposite strain. This increases the output signal and can minimise the effects of temperature on the sensor elements.
Construction
Metal sensing elements
One or more strain gauge sensors made from a length of wire can be attached to the surface of a diaphragm.
Pressure on the diaphragm will stretch the wires and change the resistance. The sensor elements can be bonded on to the surface with adhesive or the conductor can be directly deposited on the diaphragm by sputtering. The latter method removes potential problems with adhesives failing at high temperatures and also makes it easier to construct small devices.
A metal wire sensor can also be made by wrapping a wire between posts that are displaced by changing pressure. This construction can also work at higher temperatures because no adhesive is needed to attach the wire to the posts.
Semiconductor sensing elements
Semiconducting materials, most commonly silicon, can also be used to make strain gauge pressure sensors. The characteristics of the sensing element, particularly the size of the piezoresistive effect, can be adjusted by doping; in other words by adding carefully controlled amounts of impurities (dopants) to the semiconductor.
More lightly doped silicon results in a higher resistivity and a higher gauge factor. However, this also increases the thermal sensitivity of both the resistance and gauge factor.
Fabrication process
Semiconductor sensors can be constructed in a similar way to metal wire sensors, by depositing the silicon strain gauge elements on to a diaphragm.
They can also be constructed directly on a silicon surface by using the same manufacturing methods used for making electronic semiconductor devices. This allows very small sensors to be manufactured cheaply with precisely controlled properties such as sensitivity, linearity and temperature response.
Electronic components can also be fabricated on the same silicon chip to provide signal conditioning and simplify the electrical interface. Sensors based on these micro-electronic mechanical systems (MEMS) are described in more detail in [LINK: MEMS Pressure Sensors].
Design
To ensure the highest accuracy, you’ll need to consider several factors that could affect the output. Any variation or noise in the excitation voltage will cause a corresponding change in the sensor output. You will need to ensure that this is less than the required measurement accuracy.
You may need to provide an adjustable calibration resistor in the bridge circuit to set the output voltage to zero when there is no pressure.
You’ll need to keep the resistance of the wires to the sensor small to avoid introducing an offset to the measurement and reducing sensitivity. Also, the temperature coefficient of the copper wires may be greater than that of the sensor, which can introduce undesirable thermal sensitivity.
Longer wires are also more likely to pick up noise. This can be minimised by using twisted pairs and shielding.
Using a higher excitation voltage increases the sensor output and improves the signal to noise ratio. However, the higher current can cause heating of the sensing element, which will change the resistivity and sensitivity of the sensor.
This self-heating can also affect the adhesive bonding the strain gauge to the diaphragm, which can introduce errors and cause accuracy to degrade over time. The self-heating effects can be reduced by using a higher-resistance strain gauge.
The optimum supply voltage is a balance between minimising self-heating and obtaining a good signal. You can determine this experimentally. For example, with no pressure and the sensor output zero, you can increase the excitation voltage until the output is seen to change (because of self-heating). The excitation voltage should then be reduced until the output error disappears.
If possible, you should use an amplifier circuit close to the sensor to minimise connection lengths, boost the output signal and improve the signal-to-noise ratio. This can also do some filtering of the sensor output to remove external noise.
You can minimise the effects of any changes in the excitation voltage, such as a voltage drop caused by long wires, by monitoring the excitation voltage at the sensor and either subtracting that from the sensor output or using it as a reference voltage for the analogue to digital converter (ADC).
Specifications
Typical metal strain gauge sensors have a gauge factor of around 2 to 4. With a typical maximum strain of a few parts per thousand, this means a change in output of around 1mV for each volt of excitation.
Silicon-based sensors are usually doped to provide a gauge factor of around 100 to 200, which gives a good compromise between sensitivity and thermal characteristics. The output from a silicon sensor can be around 10 mV/V.
Advantages and disadvantages
Piezoresistive strain gauge pressure sensors have the advantage of being robust. Their performance and calibration is also stable over time.
One disadvantage of these sensors is that they consume more power than some other types of pressure sensor. This may mean they are not suitable for battery powered or portable systems.
Metal film sensing elements have the advantage of simple construction and durability. They also have a higher maximum operating temperature (up to about 200°C) than silicon strain gauges, which are limited to below 100°C.
Silicon strain gauges provide a much larger output signal, making them well-suited to low-pressure applications, down to around 2 kPa.
MEMS pressure sensors can be made much smaller than metal wire sensors and can be integrated with electronics for signal processing, which can control for non-linearity and temperature dependence.
Want to learn more about the other core technologies used in pressure sensors? Click the links below to jump to the section you're interested in.
Looking for more on pressure sensor technology? Check out the further chapters of this guide below, or if you're pressed for time you can download it in a PDF format here.
