Circular Runways?

Here’s an interesting article about a Dutch concept advocating circular runways for airports.  Here’s the underlying website with the original research.

Ostensibly, there is some sense to the idea inasmuch as it would no longer be necessary to build runways that are very much longer than required for an airplane’s take-off or landing roll.  If a plane needs more runway, it just goes around the loop another time or two.

Another possible benefit is that no matter where the prevailing winds are coming from, the runway will be operational – planes can fly directly into the wind while taking off or landing just by choosing the appropriate point on the loop to touch down or lift off from.

The concept’s authors also claim that a circular runway could handle many times more planes an hour than a normal runway – indeed, they are suggesting it could handle four times as many planes.

They go on to say that this would also allow for multiple routes to and from an airport, so that noise issues could be spread fairly and widely around the airport.

And this is probably the point where we need to start to introduce the authors to the real world.

Noise Issues

First of all, currently airports have narrow tightly defined flight paths in and out of airports, so as to concentrate airplane noise in narrow regions.  The airport will buy up houses in those high noise zones, and will pay for soundproofing of houses in medium noise zones.

Now let’s think what happens if planes start arriving and departing from every point on the compass.  All of a sudden, everyone for miles around will become unhappy, rather than just an unfortunate few, and the mitigation costs will soar.  No-one will say ‘Oh, great, I only have planes flying overhead four times an hour instead of forty times an hour’ and neither will anyone say ‘I didn’t formerly have any plane noise at all, but I’m not complaining, it is only fair that I should share in the noise effects of the airport’.

And talking about multiple routes, let’s think about that.

Multiple Routes

This image is taken from their final paper.  Look at it – planes taking off and landing in all directions, and all going neatly around the circular runway in an anti-clockwise direction.  Nice, right?  And very efficient, indeed.

Now, please, look at it a second time.  Do you notice – the planes are indeed taking off and landing in all directions, and directly into each other’s paths!  Not so nice.

The complexity of routings and traffic management is mind-boggling in such a system.  Sure, something a computer could probably handle – until something went wrong.  And as for flying VFR – forget it!  What about a missed approach or aborted take-off – how would the system respond to sudden emergencies that caused airplanes to need to be where they shouldn’t be?

We see this as a disaster waiting to happen.

More Planes Per Hour

In theory, if everything goes perfectly, you could have a perfectly synchronized glorious graceful ballet of planes landing and taking off, all sharing the same circuit.  The authors of this study say ‘the circumference of our circular runway is three times the length of a typical straight runway, and with some small operational efficiencies, that means we can have four times as many planes on the same equivalent length of runway’.

In reality, as soon as something stops going perfectly, you have planes tripping over each other, except that when we say ‘tripping over each other’ we mean ‘crashing into each other and exploding in spectacular fireballs’.

With three traditional runways, each is a separate system and is operated semi-independently of each other – and typically with all planes approaching and departing in the same directions.  You never have a landing and taking off plane both on the same runway at the same time.  With one circular runway, you would have four different active takeoff/landing events all on the one runway.  A timing or other problem with one then risks disaster with the other three sharing the same runway.

If one thing is certain, Mr Murphy and his law are still alive and well, and there is awesome potential for unexpected things to happen, with appalling results.  The worst airplane disaster ever remains the 1977 Tenerife disaster when two 747s collided on the runway, killing 583 people.  It is essential that all airplanes have a huge amount of safety space and time around them to prevent such things.

Prevailing Winds

The circular design is supposed to mean that no matter where the wind is blowing from, planes can still take off and land by choosing the right spot on the loop.

But most airports are chosen to be located in places with fairly reliable prevailing winds.  The interesting thing about prevailing winds is that they are, well, prevailing and it is acceptable to design most airports in most locations with a single runway in one direction, or possibly with a cross runway for the few times it is needed.  Few airports are troubled for more than very small percentages of their operating time by impossible cross-winds.

The bigger problem isn’t so much cross-winds per se as it is gusting winds (in any direction).  Steady wind is moderately easy to fly through (even if ‘crabbing’ down to land looks spectacular from the ground), but gusting winds make life ‘interesting’ for pilots and passengers alike.

So the circular runway is first solving a problem that doesn’t really exist, and secondly not addressing the actual problem that does exist.  Plus – oh yes, creating some new problems, too.  Let’s now look at some new problems.

Navigation Aids

Have you ever noticed any of the navigation aids that are dotted around airports?  Some indeed are visual – the Visual Approach Slope Indicator and the lights leading to the runway.  Others are electronic, helping the pilot pre-position his plane way before reaching the runway.

How many sets of such things would need to be added to a circular runway?

Here’s an interesting article about nav aids and other related and runway type things.

Aiming Point

This, we feel, is one of the two really big issues.  Part of the reason why a regular runway is so long is that it allows for a measure of imprecision as to exactly where on the runway the pilot lands the plane.  Most of the time, they (or their auto pilots) land within 100 ft of their ‘aiming point’, but sometimes an unexpected event (suddenly rising or sinking air, sudden head or tail winds, or just plain miscalculation) can see them landing significantly before or after their aiming point, and for that reason, the aiming point is set a way down the runway, not right at the very end.

Where would the aiming point be on a circle and how large a zone of imprecision could be offered in such a case?

Remember also that a pilot wouldn’t simply be flying in a straight line to land on a straight runway.  He’d be looping in, perhaps in some sort of spiral, so that the plane was already flying in a circular pattern like the runway.

Which brings up a second complication.  With a wide flat runway, you can land to the side of the center-line if you’re blown a bit off course, or for any other reason, and as long as the wheels are on the runway, it doesn’t really matter.  But this circular banked runway has varying angles of bank from the innermost point (no bank) to the outermost point (maximum bank) and it becomes necessary to land the plane at the exact right spot in two dimensions rather than only in one.  Too far up or down the loop, or the wrong amount of matching bank on the plane, and you’re risking hitting the ground with a wing tip and all sorts of other nasty things.

Due to the complexity of the path to the aiming point and the lack of margin, this is almost certainly the type of landing that could only be reliably done by auto-pilots.

Centrifugal/Centripetal Force

And now for our piece de resistance.  Astonishingly, it seems the authors of this concept completely failed to consider this.  Think about what happens when you spin something in a circle around you, on a piece of string.  It pulls on the string, doesn’t it, trying to fly off and out in a straight line.

This is what is termed a centrifugal force (wanting to fly out) balanced by a centripetal force (provided by the string) to keep the object in the circle.

The relevance of this to a circular runway is that the plane will probably be experiencing a force in the order of about 0.25g acting on it due to its angular velocity as it spins around the circle before taking off.  This is a function of the radius/diameter of the circle and the speed it is zooming around.

Because the runway will necessarily be banked, this force will be pushing the plane down onto the ground.  It is extra force that needs to be countered by more powerful engines, and/or larger wings, less weight, more speed (because the faster the plane goes, the more lift it generates) and a larger radius/diameter for the circular runway.

Except – ooops – the centrifugal force is proportional to the square of the speed.  So you actually don’t want to increase the speed at all.  You want to make it as slow as possible, and the runway as large as possible.

So if you’re going to try to take-off at a lower speed, but needing more ‘power’ to do so, that will mean much larger wings and much more powerful engines, and probably reduced payloads.  Airplanes will have to be totally redesigned.

Which means we now have the tail wagging the dog.  Completely changing the currently optimized set of trade-offs involved in designing an airplane, purely to ‘solve’ a ‘problem’, which may or may not be a problem to start with, and which may or may not be a solution either.

So, will a circular runway be coming to your local airport any time soon?  We don’t think so.  But here’s an interesting article about earlier attempts to float the idea, including one very imaginative idea – spinning planes around in a circle then flinging them out.  Perhaps ‘float’ is not the best term to use?