Absolute pressure sensors
What are absolute pressure sensors?
With a wide variety of pressure sensors available, it can be challenging to know which one is best suited to a specific task. When it comes to measuring air pressure, specifically for applications such as barometric measurements for weather or in altimeters, an absolute pressure sensor is the device of choice. However, your possible application usage isn’t limited just to air or gas.
The absolute measurement is made possible by measuring the target pressure relative to the known pressure of an absolute vacuum (see diagram below). This can be compared with measuring temperature in Kelvin, where the lowest possible temperature is 0 °K.
By using a vacuum as the reference against which everything is measured, all measurements will deliver a value larger than the absolute minimum as defined by the reference. This is essential to accurate measurement, since Boyle’s Law states that the pressure of a gas is inversely proportional to its volume at a constant temperature. Thus, anything other than a perfect vacuum will result in an absolute pressure sensor whose measurement varies with temperature.
An absolute pressure sensor measures pressure relative to a sealed vacuum reference in a hermetically seal chamber
A perfect vacuum is, however, highly challenging to achieve, especially if the sensor is to remain within an acceptable price range. Thus, absolute sensors typically have to make do with an approximate vacuum, typically in the range of 35 microbar (0.0005 PSI).
How do absolute pressure sensors work?
With a sealed vessel as the reference point, a sensing technology is then applied to the surface of the vessel whose electrical characteristic varies with changes in strain. There are many different approaches to this.
One common method is the piezoresistive strain gauge. These embed a resistor, (whose value changes with respect to mechanical strain) into a material such as silicon, polysilicon, metal foil, or as sputtered metal onto a thin film. In order to maximise the output signal and reduce errors, the sensor typically uses four resistors in the Wheatstone bridge configuration.
| |
Wheatstone bridge structure of an absolute pressure sensor |
With today’s high levels of integration, it is not uncommon for your piezoresistive sensor to also include compensation circuitry, such as resistors, all on a single substrate (see right).
Other sensing technologies also make use of a component’s value variation when deformed. For example, capacitors vary in capacitance when placed under strain. Sometimes a change in an inductor’s inductance can be caused by a locally placed diaphragm that moves in reaction to pressure changes.
The piezoelectric effect is also a common pressure measuring technology. This makes use of the fact that some materials, such as quartz, generate a voltage dependent on applied pressure.
As a result of the massive advances in silicon manufacturing technology in recent years, some mechanical elements are being machined into silicon chips known as microelectromechanical systems or MEMS devices.
They mostly utilise the same physical properties of the electronic components already mentioned, but leverage some moving parts machined into the semiconductor material. Such devices rarely provide the sensing element’s output signal; instead they precondition the signal electronically before outputting it via a package pin.
How do I integrate an absolute pressure sensor into my circuit?
With so many different pressure-sensing technologies available, there is no single way to integrate a sensor into your circuit. In most cases, you will be looking to connect your absolute pressure sensor to a microcontroller.
Some sensors are so simple they require a significant amount of circuitry to condition the signal for use with a microcontroller. A sensor that simply provides access to a Wheatstone bridge circuit will require significant amplification to deliver an output large enough for a typical microcontroller’s analogue-to-digital converter (ADC) to measure.
One method to generate an analogue output from a Wheatstone bridge sensor, with adjustments for both offset and span
A quad op-amp configuration as shown in the diagram above provides an example for such an amplification circuit. Such circuitry needs to be carefully designed and may also require proper screening and low-noise design techniques to guarantee a reliable output signal.
Due to process variation and component tolerances, your circuit may also need individual calibration for each circuit board you produce. Temperature compensation may also need to be considered.
Sensors with digital outputs are much easier to connect to a microcontroller. These will include all the signal conditioning, amplification and temperature compensation. The measurements are then converted into a digital value and stored in an internal register.
The interfaces offered to the microcontroller are typically I2C or SPI. Some sensors may support both, allowing you to select between the one that suits your application best. This is the case with the example shown below.
The integrated circuitry inside the sensor can perform amplification, conditioning and digitisation of the sensor measurement
A sensor with a digital interface may not meet your needs if you need to measure rapid variations in pressure. An SPI or I2C interface only supports a certain number of data transfers per second. With more than one device on the bus, the available bandwidth drops with the increase of devices hanging on the bus.
For measurement of pressure that varies quickly, it is likely that you will need to invest time in the development of your own analogue front-end, coupled with an ADC with a suitable conversion time.
Can I use an absolute pressure sensor in my design?
Many absolute pressure sensors are provided in a small housing suitable for fixing through-hole or surface-mount to a printed circuit board (PCB). These are known as board level sensors. This makes them ideal for a consumer application where sensing can be undertaken on the PCB, e.g. in an altimeter or sports watch.
However, such sensors are not suited to the high temperature of liquids or gases. Neither are they suitably protected against dust, moisture or the chemicals often used for cleaning in industrial environments.
Industrial sensors are typically robustly packaged. They are likely to be made of a non-corroding material, such as stainless steel, and are threaded, allowing them to be fitted to pipes and storage tanks.
Industrial engineers typically want to select their hardware and link it all together. They are not so interested in building custom circuitry to handle the sensor output. As a result, industrial sensors are grouped into three main types: sensors, transducers, and transmitters. We touch on these briefly below, but for more information on these sensor types read chapter 5.
The term ‘sensor’ typically indicates a device that generates a ratiometric output. This means that your sensor’s output will be dependent on the sensor’s supply voltage. Thus, a sensor with a 10mV/V output will generate a 0 – 50 mV output for a 5.0 VDC supply. Such devices can be quite raw in their packaging, with pads or legs suitable for soldering to a circuit board or cabling.
A ‘transducer’ is a complete sensor, including signal conditioning, designed for you to fit directly into an industrial environment. Your output signal will typically be a voltage that relates to pressure, generally lying in the 0 – 10V range. However, some transducers generate an alternating signal in the 1 – 6 kHz range.
Some older transducers do not have a “live zero” when the sensor is at its lowest measuring point. This makes it impossible for you to determine the difference between a minimum pressure measurement and a broken sensor or connecting cable. This is something to be considered for systems with a high-level safety requirement.
A pressure ‘transmitter’ typically indicates a sensor that uses a 4 – 20 mA output signal rather than a voltage output. These devices often only require a two-wire interface (supply and ground), and offer good electrical noise immunity (EMI/RFI). Supply voltage for such sensors lies in the range of 8 – 24 VDC.
Such sensors are designed for use with other industrial equipment, such as a programmable logic controller (PLC). These then communicate via digital buses with one another and other industrial systems. Buses include Fieldbus, standardised as IEC 61158, IO-Link, PROFIBUS and CANopen. With more intelligence being built into industrial sensors, it is becoming increasingly common to find that they are ready to be directly attached to such industrial networks.
What applications are absolute pressure sensors used in?
With the rise in smart watches and navigation systems, absolute pressure sensors find homes wherever elevation above sea level (altimeter measurement) is required. Weather stations also use them for barometric pressure measurements.
Petrol and diesel vehicles make use of them too, measuring the pressure in the engine manifold. Such sensors are known as manifold absolute pressure sensors, or MAP sensors for short. The engine’s electronic control unit (ECU) uses this information to determine optimal combustion of the air-fuel mix and ignition timing.
In industrial applications, it is often necessary to develop a partial vacuum. This is the case in food packing, where the residual pressure determines the shelf life of the produce. The absolute pressure sensor can ensure that the pressure in each package is the same.
At the other extreme, industrial absolute pressure sensors are available that support measurements of more than 300 bar (4400 PSI).
If you'd like to read more on other pressure measurement types then click the links below:
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.
