Area navigation, or RNAV, is one of the cornerstones of the future air traffic system as envisaged by both the FAA and the Joint Program Development Office's Next Generation Air Transportation System (NGATS). RNAV enables aircraft to fly point to point without reference to ground-based navigation aids. It also is the principle upon which required navigation performance (RNP) is based.
For the last five years, the FAA and the air carrier industry have been working together to introduce RNAV procedures in several of the major terminal areas around the country, including Atlanta, Dallas-Fort Worth, Las Vegas, Los Angeles, Miami/Fort Lauderdale and Memphis. What most of us thought would be a straightforward task turned out to include some very interesting challenges. That we have identified many of the anomalies and come up with solutions is a tribute to the cooperation and perseverance of the FAA, controller work force and industry.
This column focuses on the following differences, which have been identified in the large and diverse population of flight management computer (FMC) systems:
- Lack of standardized turn performance when transitioning between two straight legs;
- Lack of a standard means by which to apply initial segments of RNAV departure procedures and the resulting disconnect with procedure design criteria; and
- Variability between FMCs when transitioning to different phases of flight.
The issues discussed here, however, do not in any way reflect unfavorably on any FMC avionics vendor or airframe manufacturer.
Standardized Turn Performance
At Las Vegas, FAA evaluated track keeping performance by plotting radar tracks off one of the RNAV standard instrument departures (SIDs). The data showed three separate groupings of tracks between two track-to-fix (TF) legs. The TF leg type is the most common means of defining a straight segment of an RNAV path. One airframe data set was outside the depicted track, one remained very close to the desired track and one flew slightly inside the desired track.
One cause of this dispersion was the different bank angle limits that are designed into FMC systems. The track angle change did not seem to have a significant impact on the results.
The simplest solution to this problem is to use the radius-to-a-fix (RF) leg for all terminal operations where a curved path is required. That would resolve nearly all of the turn performance differences, but not all current FMC systems have the RF leg capability, so we have to continue to use the TF legs instead of utilizing an RF leg to facilitate going around a curve.
RNAV Departure Procedures
There are several leg types and path terminators that could be used for the initial segments of a SID, and while some are very similar in performance, there is enough difference to create path variances. As a result, there have been a number of events on parallel runway departures using "RNAV off the Runway." To address this scenario, most facilities with parallel runways are now using radar vectors to the initial RNAV fix until a better solution can be found.
Transitioning to Different Phases of Flight
Different airframe manufacturers design their systems to transition between phases of flight, e.g., takeoff, departure, en route, terminal and approach, differently. One airplane may use a 28-degree bank angle at 15,000 feet, while another may use 15-degree bank angle at the same altitude.
So where does that leave us? Right now there is not much we can do to change the situation due to the very high cost of the software modifications that would be required to make all FMCs consistent. Again, I want to emphasize that what I have described here is not a safety issue but one of path consistency and repeatability. We have not observed any of these systems operating outside the designated RNAV performance values for the procedures.
For the future, however, the FAA's Performance Based Aviation Rulemaking Committee (PARC) has created a flight management system (FMS) standards action team to address these issues and recommend changes. The PARC group also is studying the creation of a document that will specify certain requirements-starting at some point in the future-that all FMCs will be required to do in exactly the same way.
For now we will accept these differences with the recognition that, as industry upgrades their fleets, a greater percentage of the aircraft will be have similar capabilities, such as the RF leg, that will help resolve these differences.
Capt. Frank Alexander, Northwest Airlines, chairs the PARC's Arrival & Departure Working Group.