Considering how important mass air flow sensors are in several applications, keeping these sensors in proper working order is vital to reduce the chances of equipment failure.
Poorly performing or malfunctioning sensors can cause additional complications - for example, poor efficiency of fuel on combustion engines, or the compromising of patient safety where mass air flow sensors are utilized in artificial ventilation systems.
The most frequent failure reasons are contamination or sensor aging, the former of which can often be corrected by cleaning the sensor. The deterioration or aging of components, can cause false resistivity readings, resulting in inaccurate measurements.
Download the Full eBook Here
Identifying Faulty Sensors
For mass air flow sensors on engines, the first warning sign of a failing sensor is commonly problems with the performance of the engine. Typical signs include slow acceleration, reduced engine power, or reduced fuel efficiency. This is because the data from the mass air flow sensor is utilized by the onboard electronic control unit to control the level of fuel being injected to the engine.
An optimal air/fuel ratio needs to be maintained to optimize combustion efficiency, so if the sensor is under-reading the mass air flow, not enough fuel is injected, which causes decreased engine power.2
The contrasting scenario, where the mass air flow sensor overestimates the amount of air intake into the engine, can cause engine flooding because too much fuel has been injected.This is challenging because fuel can enter the cylinders which means they cannot seal and therefore the fuel volume in the engine begins to become greater the upper explosion limit. As such, ignition cannot take place.3 If ignition cannot take place, the engine cannot start.
A visual observation of the mass air flow sensor can assist in determining whether contamination is the source of the problem or not. Especially in environments like car engines, dirt or debris frequently come into contact with the sensor and this adjusts the thermal resistivity of the wires.
As the majority of hot-wire type mass air flow sensors work on the resistivity distance between the sensor wire and a reference wire, this will create readings that are inaccurate. These kinds of sensors often have built-in self-cleaning circuits to correct this problem, but these can error alongside the sensor.
An additional choice is to use a multimeter to read either the frequency or voltage signals (according to the sensor type). This can be employed to verify that the voltage is similar to what the battery generates and that the electrical connection problems with the supply are not the error source.
The resistance of the sensor wires can then be evaluated. Infinite or high resistance values can be a signal that one of the sensing wires in the air flow sensor is physically burnt or broken.
For building ventilation and medical use, mass air flow sensors are expected to give continuous monitoring and recording of mass flow rates. For heating ventilation and air conditioning (HVAC) applications, sensors may not always be placed in regions that are easy to access, therefore making the visual verification of sensor issues more difficult.
This is also true for mass air flow sensors employed on remote pipelines for leak detection. For medical applications, sensors faults must be detected efficiently and as soon as possible, as any problems with mechanical ventilation could possibly harm human life. One solution to this issue is to use diagnostic systems and automated fault detection.
In building ventilation systems, an example of this is the use of decision tree software, where metrics are employed to evaluate correct and incorrect air-handling performance.4
If the performance is too different from a set of assigned values, this can signify that there are possible issues. The software is able to identify the issue by drawing on various different diagnostics.
These problems could entail the ventilation being on incorrect settings, the output being too low, or a more critical issue, for example the mass air flow sensors incorrectly determining the flow rate through the ventilator compared to the output flow rate.
Due to the increasing complexity and amount of engine sensors, automated detection is also becoming common for engines. Here, neural network-based diagnostics can either utilize the mass air flow sensor as a diagnostic tool or find issues with the sensor itself.5
The benefit of automatic detection and the use of sensor networks with either model-based or algorithmic approaches6 is that they can perform in a continual manner and can be configured with alarm systems to quickly trigger alerts to explain that a sensor has failed or malfunctioned.
For all sensors, ideally, regular diagnostics, maintenance, and inspection should be performed to keep sensors in the best working order. The ideal approach to diagnostics is somewhat related to the application, however there are several options available now that can simply be adapted for various types of mass air flow sensors.
Download the Full eBook Here
References and Further Reading
- Mass Air Flow Sensors (2013), https://www.azosensors.com/article.aspx?ArticleID=156
- Fleming, W. J. (2008). New Automotive Sensors — A Review. IEEE Sensors Journal, 8(11), 1900–1921.
- Paasi, J., Kalliohaka, T., & Glor, M. (2009). Chargeability of ethanol-petrol biofuels. Journal of Electrostatics, 67(2–3), 247–250. https://doi.org/10.1016/j.elstat.2009.01.027
- Katipamula, S., Pratt, R. G., Chassin, D. P., Taylor, Z. T., Gowri, K., & Brambley, M. R. (1999). Automated fault detection and diagnostics for outdoor-air ventilation systems and economizers: methodology and results from field testing. ASHRAE Transactions, 105.
- Kimmich, F., Schwarte, A., & Isermann, R. (2005). Fault detection for modern Diesel engines using signal- and process model-based methods. Control Engineering Practice, 13(2), 189–203. https://doi.org/10.1016/j.conengprac.2004.03.002
- Isermann, R. (2005). Model-based fault-detection and diagnosis - Status and applications. Annual Reviews in Control, 29(1), 71–85. https://doi.org/10.1016/j.arcontrol.2004.12.002