In the last chapter we looked at all the wonderful radio aids to navigation that have been developed over the decades.
Well guess what, once you get about 150 miles offshore you don't have any of those. You're also outside of radar coverage. Get lost out over the Pacific and you're really lost.
While people were flying the oceans prior to World War II, it didn't really become commonplace until driven by the necessity of wartime.
So let's take a look at how we can get across a large unfriendly body of water.
First off, when we fly a long distance we have to use what's called a "Great Circle Route". Since the Earth is a globe and not a flat surface the shortest distance between two points is not a straight line.
Great Circle Route - San Francisco to Tokyo
As you can see in this picture, the shortest route from San Francisco to Tokyo goes quite a ways North. Almost to Alaska. Likewise going from New York to London will take you close to Greenland and Iceland.
Now that we know what route we need to fly, we need some way to actually navigate it.
The most basic form of navigation is called Dead Reckoning. Why is it called that? Heck if I know. Maybe because if you reckon wrong you're dead. If your navigation systems all crap out, you're left with dead reckoning.
Lindbergh managed to cross the Atlantic using dead reckoning but he was only trying to hit Western Europe, which is a big target.
It works simply enough. From a known position, fly a predetermined heading and speed for a certain amount of time. For example, if we start out over Chicago and fly due west for an hour at 300 knots (nautical miles per hour) we should be 300 nautical miles due west of Chicago at the end of that hour.
At this point you may be wondering why we use nautical miles. Much of aerial navigation was copied over from the nautical world. A nautical mile is 6000 feet (versus 5280) and nautical miles divide evenly into lines of latitude. This makes the calculations a lot easier.
The problems with dead reckoning should be pretty obvious. We're basing everything on forecast winds. If the actual winds are different, and they likely will be, we could be quite a ways off from where we think we are. Since each navigational leg is based on the previous point, the errors will be cumulative as we go along.
The next step up the navigational food chain is Celestial Navigation. The same way sailors have been navigating for centuries. In the KC-135 we still practiced this as recently as the late 1990s.
I've never actually been trained in celestial navigation, that's what navigators did in the Air Force. I have only the most basic knowledge of how it works and I'm sure there are people here who know more about it than I do. Here's the Wikipedia definition:
Celestial navigation is the use of angular measurements (sights) between celestial bodies and the visible horizon to locate one's position on the globe, on land as well as at sea. At a given time, any celestial body is located directly over one point on the Earth's surface. The latitude and longitude of that point is known as the celestial body’s geographic position (GP), the location of which can be determined from tables in the Nautical or Air Almanac for that year.
Simply put, by using a sextant and an almanac a navigator can take "shots" on multiple celestial objects (sun, moon, stars) and plot the "lines of position" on a chart. Where those lines intersect is the aircraft's position. The accuracy of that position is largely dependent on the skill level of the navigator.
Repeat the process again in a half hour or so and you now have two points over a known time period. That lets you know all kinds of good stuff like how fast you're actually going (groundspeed) and what course you're actually flying.
An aircraft sextant is a bit different from a nautical sextant. It looks more like a periscope and can be extended through a small port in the top of the aircraft. It has a small bubble floating in liquid that acts as a horizon. That's all I know, never having actually used one.
On the KC-135 the Boom Operator would stand on a stool and take the sextant shots while the Navigator worked the plots. One Nav described it to me as being like "doing his taxes". The pilot's main task was to fly a very steady heading and speed and to try not to move the throttles too much. The small changes in pitch from the pod-mounted engines could make sextant shots difficult.
The claimed accuracy of celestial navigation is 3 miles which is actually pretty good. I made several ocean crossings in the tanker with only a single INS (don't worry I'll talk about those in a minute) and celestial for backup.
Of course, you actually have to be able to see something for this to work. A high overcast deck can make celestial navigation difficult.
The next step up from there is the Doppler Navigation System (DNS). These systems mount four small radar transmitters under the aircraft. By measuring the frequency shift in the beams (Doppler Effect) it can fairly accurately measure the aircraft's motion across the ground. By comparing this to the aircraft's indicated speed and heading, the actual groundspeed and drift can be accurately calculated.
Basically, the DNS always knows how fast it's going and in what direction. If we load our starting position into it before we take off, it can use this information to continuously estimate our position. Think of it as a very accurate form of dead reckoning.
The nice thing about them is they're totally self contained. They don't require a signal from a ground based transmitter or a satellite.
They don't always work well over the ocean, however. The water itself is actually moving, so that can cause errors in measuring groundspeed. If the ocean is particularly smooth, the Doppler may not work at all because the radar signals may not reflect off the surface.
A better setup is the Inertial Navigation System (INS). These were originally developed for missiles and started finding their way into aircraft in the 1960s. How does it work? Science! Originally they used mechanical gyroscopes, which today have been replaced with ring-laser gyros. Since gyroscopes resist movement, they allow acceleration to be measured. Hence the "inertial" part. Acceleration over time lets you calculate speed. Speed over time lets you calculate distance. Add a few more gyros (one for each axis of movement) and you can also measure lateral and vertical acceleration.
Delco Carousel INS - 1970s Technology
An INS, just like it's Doppler predecessor, is just a very accurate form of dead reckoning. It knows where it started out (we load present position on the ground) and it knows how far it's traveled and in what direction. That allows it to very accurately estimate its current position. Small errors will accumulate over time, so they will "drift" roughly 1/2 mile for each hour of flying time. Spend 4 hours over the ocean and you may be 2 miles off course at the other end. Not bad at all.
Even with GPS we still use these. A typical airliner will have 3 inertial units which are constantly compared to each other. In the 757 we call them Inertial Reference Units (IRU) because we use them for more than just navigation. They also provide attitude information and vertical speed to our flight instruments.
The navigation computer is constantly monitoring the INS units and comparing them to each other as well as to the GPS signal (if it has one). If one INS starts to differ significantly from the other two it will be "voted off the island". The navigation computer takes the best information it has (usually the GPS) to fix its position. If we lose the GPS signal for some reason we still have a pretty good INS position. That's good because navigators are long gone and I don't know how to use a sextant.
757 Triple INS
Let's suppose we've made it across the pond and our INS has drifted off by a couple miles. Most airliners can update their INS position based off a VOR signal. In fact the 757 is always automatically looking for VORs and will use the signal to update the INS.
So even if your position drifts a bit, once you get within 150 miles or so of a VOR you're back in business.
Now let's put this all together and cross the Atlantic. If we're going from say New York to London, our great circle route will take us up near Gander Newfoundland. That's where we'll get our oceanic clearance and "coast out".
Since the North Atlantic is actually a pretty busy place, they have various "tracks" set up for the crossing. Gander will assign you a certain track plus an altitude and mach number to fly to keep you separated from everyone else that is crossing at the same time. The tracks vary from day to day to make best use of available tailwinds (or to minimize headwinds). The tracks run west to east at night, when most airlines are flying to Europe. They reverse them in the daytime when most flights are coming from Europe.
Eastbound North Atlantic Tracks
(these change from day to day)
Let's suppose for tonight we're assigned the "Y" track, which is the southernmost one depicted. It's defined by the waypoints VIXUN LOGSU 49N50W 50N40W 50N30W 50N20W SOMAX and KENUK
Since there isn't any radar coverage out over the ocean, the only way Gander will know where we are is by us telling them. When we cross the various waypoints on the track we'll have to call them up and make a position report.
Since we're outside of normal VHF radio range, we have to use an HF (High Frequency) radio, basically a shortwave radio. Normally they're annoying to listen to because they're full of static, pops and hisses. Fortunately we've got something called SELCAL which lets us squelch out all that garbage until somebody calls us (by our plane's unique identifier) or we need to talk to somebody.
Let's suppose we just crossed 49 North 50 West at 0300Z (GMT) and we need to make a position report. What we need to tell Gander is:
What point we just crossed.
When we got there.
Our altitude.
When we estimate getting to the next point (our navigation computer knows this).
Our next point after that.
So, key the mike and wait for the radio to channelize.
"static"
"Gander, Gander this is Chickenhauler 51 with position"
"more static"
"Chickenhauler 51 go with your position"
"snap, crackle, pop, squeak"
"Chickenhauler 51.........49 North 50 West 0300Z Flight Level 360"
"Estimate 50 North 40 West 0355Z"
"50 North 30 West Next"
Once you do a few of them they're easy. I don't know exactly how oceanic control tracks us, but I suspect it's something very low tech like marking on a board with a grease pencil.
If Gander needs to call us for some reason they just transmit our unique SELCAL code and we'll get a "Beep! Boop!" over the radio that tells us somebody's calling. Much better than the old days of listening to static the whole way across the ocean.
Somewhere around the halfway point we'll stop talking to Gander and start talking to Shanwick (Ireland). We'll make position reports to Shanwick until we come into radar coverage somewhere near Scotland.
Some planes even have a fancy little gizmo called ADS-B, which I've never even seen. It uses a datalink to automatically transmit position reports for you. I never minded doing position reports because it broke up the mind numbing monotony of flying across the ocean. Sometimes the radios can be so busy that it's hard to get a work in edgewise, so I guess ADS-B is a good thing.
Now you know about as much as I do about ocean crossings. Oh, and one more thing. Lindbergh was crazy to cross the Atlantic in that thing. They didn't call him "Lucky" for nothing.