Area Navigation Systems (RNAV) — Page 282, Lesson 278

I want to walk you through the heart of how a modern airliner actually navigates itself — the Area Navigation System, or RNAV. We're looking at the architecture that ties the inertial systems and the flight management computers together, and I want you to understand it exactly the way you'd need to on the line. Let's start with the overall picture. On this aircraft we have a triple IRS — three Inertial Reference Systems — and a twin FMC, two Flight Management Computers. The key idea is redundancy and cross-checking. The information from one of the FMCs feeds the other systems. Now, here's a subtle but critical point: there will be a difference in position between the two FMCs, and there will also be a difference between the EFIS — the Electronic Flight Instrument System — and the non-selected FMC. In other words, the two computers don't always agree perfectly, and the displays don't always agree with the backup computer. That's normal, and it's something you need to be aware of. Now, what happens if one FMC fails? If one FMC is inoperative, the functions can be carried out by the serviceable FMC — the good one takes over all the work. But if both FMCs are inoperative, then the IRS information is used directly in the EFIS. The displays still work, you still get your attitude and heading, but the automatic performance functions are not available. That's a big deal — no automatic thrust management, no performance calculations. So you have degraded capability, but you still have basic navigation. Let's go deeper into the principle of operation. The title here is "Triple IRS, Twin FMC." Positional information and heading from the triple INS/IRS — the Inertial Navigation System / Inertial Reference System — is fed into the FMC. Inside the FMC, that information is compared to check for any system having gross errors, and then it's averaged. So the FMC looks at all three inertial positions, throws out any one that's wildly wrong, and averages the rest. That averaged position may then be compared with an external reference. The external reference can be DME/DME — Distance Measuring Equipment pairs — or VOR/DME, or GNSS, the Global Navigation Satellite System. So the FMC has two sources of truth: the inertial system and the external radio or satellite references. Now here's where the clever mathematics comes in. The FMC uses Kalman filtering to produce position and velocity. Let me explain what that means. A Kalman filter is an algorithm that takes noisy measurements from multiple sources and blends them to produce the best estimate of the true state — in this case, your position and your velocity. It's not just averaging; it's a statistically optimal way of combining data that accounts for the uncertainty in each source. This filtering may be done purely using the IRS information, or it may be done using a combination of IRS and external reference. So the FMC can navigate on inertial alone, or it can blend inertial with DME/DME, VOR/DME, or GNSS to get a better fix. Now I want to talk about a very specific operational limit that you absolutely need to know. When operating at latitudes in excess of 84 degrees — that's 84 degrees north or south, very high polar latitudes — the FMC will de-couple the IRS. Here's how it works: the left FMC uses the IRS in the order left, centre, right. The right FMC uses the IRS in the order right, centre, left. So each FMC picks a different inertial unit as its primary, and they cross over in the middle. Over a short period of time, each FMC will change its FMC position to the appropriate IRS position. So the FMC essentially hands over navigation authority to the inertial system. Why does this happen? This is the mathematical reason, and I want you to really understand it. The calculation of change of longitude from departure is a function of the secant of latitude. Secant is the reciprocal of cosine — one over cosine. At latitudes approaching 90 degrees, the cosine of latitude approaches zero, so the secant approaches infinity. That means the secant is increasing very rapidly. Let me give you the exact numbers from the manual. At 86 degrees exactly, sec 86°00' equals 14.3356. At 86 degrees and one minute — just one minute of arc higher — sec 86°01' equals 14.3955. So in that single minute of latitude, the secant jumps by about 0.06. What does that mean physically? It means a small error in latitude will result in a large error in the calculation of change of longitude. Because you're multiplying by that huge secant value, any tiny latitude error gets amplified enormously. This would give an apparent large divergence between the IRS positions in terms of the calculated longitude — the three inertial units would appear to disagree wildly on longitude, even though in fact the actual difference between them is small. The units aren't actually far apart; the mathematics of longitude at high latitude makes them look far apart. So the de-coupling is a protective measure. By having each FMC lock onto its own IRS and not trying to compare or average them at those latitudes, the system avoids chasing phantom errors that are really just artifacts of the secant function. It's a brilliant piece of engineering, and it's exactly the kind of thing you'd be tested on. Let me make sure you have the key terms locked in. RNAV — Area Navigation — is the overall system that lets you navigate along any desired path, not just to and from ground stations. FMC is the Flight Management Computer, the brain. IRS is the Inertial Reference System, which uses accelerometers and gyroscopes to track position and heading without any external signals. EFIS is the Electronic Flight Instrument System, your primary flight displays. DME is Distance Measuring Equipment, which gives you slant range to a ground beacon. VOR is VHF Omnidirectional Range, which gives you a bearing to a ground station. GNSS is the Global Navigation Satellite System — GPS and its equivalents. And Kalman filtering is that statistical blending algorithm I described. So the whole picture is this: three inertial units feed the two flight management computers. The FMCs compare, average, and blend with external references using Kalman filtering to get the best possible position and velocity. If one FMC dies, the other handles everything. If both die, you fall back to raw IRS data on the displays, but you lose automatic performance. And above 84 degrees latitude, the system deliberately decouples the inertial units to avoid the mathematical trap of longitude calculation near the poles. That's the architecture of the RNAV system in a nutshell. You now understand not just what the system does, but why it does it — and that "why" is what separates a pilot who operates the system from one who truly understands it.