Don’t Let Sensor Drift Lead to Aircraft Drift
There are many considerations that need to balanced when designing or purchasing systems for aircraft. The most important factor, assuming the specified functionality is met, is typically flight safety, or at least ensuring a system meets a minimum safety requirement. After this cost can dominate decisions – however, the cost of a component goes beyond the purchase cost. All systems require maintenance, for example, and maintenance can be expensive.
When it comes to an air data computer, the primary concern is they deliver high accuracy information without ever failing. Their design assurance level (DAL) A rating helps to ensure that reliable operation is guaranteed, however their accuracy is not. All common air data computers currently rely on air pressure readings to feed raw parameters to equations. These parameters are generated by sensors and all sensors are subject to changing over time, leading to a phenomena known as sensor drift.
Sensor drift leads to inaccuracy. In the case of air data computers, this inaccuracy leads to uncertainty about the aircraft’s altitude and speed. If such uncertainty exceeds a certain limit then a maintenance procedure to recalibrate or replace may be required. This would be necessary where consistently high accuracy is a requirement in order to meet initiatives such as Reduced Vertical Separation Minima (RVSM). As the skies have become busier and the accuracy of air data computers has improved, measures such as RVSM have been implemented to allow a greater density of aircraft to fly in RVSM airspace without compromising the required high levels of safety.
Aircraft need to have RVSM approved equipment in order to fly freely in these zones. This requires that the air data computers on board can meet accuracy requirements and that such is maintained. This is because, in general, the distance allowed between aircraft has been halved in RVSM. Failure to maintain such systems could result in situations where aircraft operate in close proximity due to misleading position data that wouldn’t have occurred previously. In an extreme situation, two aircraft could have sensors falsely reporting altitudes so that a collision occurs.
Some systems are inherently more sensitive to drift due their physical and material design. A good example is comparing different temperature sensors – thermocouples are cheap and relatively robust compared to resistance temperature detectors (RTD) but they are more subject to atmospheric effects that can impact the reliability of their measurements. In air data computers, silicon based sensors are common and have the advantages of being small, light and able to yield highly accurate results. Vibrating cylinders by comparison are somewhat larger and heavier but importantly their physical and material properties mean they are not as susceptible to drift (typically <0.01% FS per year) as their silicon counter parts.
Curtiss-Wright conducted research on this and found significant differences in accuracy readings after a year had passed. Recalibration can be conducted on both silicon and vibrating cylinders to ensure the systems produce good results, but the data shows the likely frequency for such maintenance activity is significantly lower for vibrating cylinder technology. This in turn implies that the frequency of a maintenance action, for the air data system at least, is reduced.
At the end of the day, there are several factors to consider when choosing any system. When it comes to a system as important to an aircraft’s safe operation as an air data computer, reliable accuracy should be a high priority. This is not only because of the possible impact on flight safety but also because sensor drift has a long term cost implication.
Reduce your program cost, development time and certification risk