A new EU research project is proposing the use of circular runways as an alternative to conventional airports. BILL READ FRAeS looks at the pros and cons of the circular argument


The quickest way to get from A to B is in a straight line. Or should it be going around in a circle? A consortium of five European research centres led by The Netherlands Aerospace Research (NLR) has published a report looking into the potential advantages of building an airport with a circular runway. Headed by NLR’s Senior R&D Manager, Henk Hesselink, the Endless Runway project was carried out by NLR, DLR (Germany), ONERA (France), INTA (Spain) and ILOT (Poland) and received funding through the European Commission 7th Framework Programme ( FP7).

An idea before its time?

Popular Science Monthly published a circular runway concept in 1919 showing a circular track in Manhattan.

The idea of the circular runway is not new. In 1919 Popular Science Monthly published a concept illustration of a banked circular iron runway located on top of city skyscrapers. The idea was that aircraft would circulate clockwise and take off and land on a particular area indicated by a signal where there was a headwind. Aircraft would enter and exit the runway via a lift on the roof of one of the buildings.

In 1938 a stuntman named Jean Roche demonstrated circular flat take-off at Riverdale airfield in Maryland using a hub, spindle and release gear. During WW2 a series of straight runways which connected at 45deg were used for training purposes to practice cross wind and tail wind landings. Circular flat take-offs were tested on ice in March 1955 on Lake Kegonsa, Wisconsin, using a Ercoupe light aircraft attached to 400ft rope connected to a steel barrel frozen into the ice

The US Navy conducted tests on a General Motors banked test track between 1964 and 1965. (USN)

More thorough trials took place between 1964 and 1965 by the US Navy NWEF (Naval Weapons Evaluation Facility) which conducted take-off and landings on the General Motors Desert Proving Grounds circular banked test track near Mesa, Arizona. The 13.7m wide track increased in angle from just over 0° on the inside to 22° on the outside and had a 1,281m diameter and a circumference of 8,047m. Flight tests began in March 1964 using a T-28C Trojan propeller aircraft which had a large ground clearance between the propeller and its wing-tips and was fitted with tricycle landing gear. Pilots of the T-28 reported good stability during take-off and landings with winds have little influence on angle of bank, control forces, and control displacement.

The tests of the T-28 were followed by take-off and landing trials using an A-4B jet, an A-1E single engine propeller plane and a C-54 transport aircraft. After a few trials, pilots were able to land aircraft successfully, although they described landing on the circular track as ‘flying into a hole’, as they had to approach the runway with the left wing banked down at an angle of 15deg. During the first landings, they tended to touch down on one wheel. Once the aircraft had touched down, pilots experienced a tendency to level the wings as on a straight runway, causing the aircraft to drift towards the outside of the track. After a few trials pilots reported an exceptional lateral stability, such that cross winds were no longer a factor. Margins of errors for landing speed, touchdown or degree of bank were also not so critical as the runway tended to correct them.

One future concept has proposed an airport on the sea with rotating runways. (NLR)

The concept was again highlighted during the ‘Fentress Global Challenge: Airport of the Future’ in 2011 in which two students from Stanford University and Malaysia's University of Science both proposed both a circular runway concept. (Thor Yi Chun, the University of Science in Malaysia under the name of the Aero-Loop). Other recent proposals have included a floatable ‘cruise home port’ and a floating airport.

Devil in the details

The Endless Runways Final Report (NLR)

The report on the Endless Runway concept is very comprehensive and goes into considerable detail looking at all the different aspects of how such a unique design of airport could be designed, built and operated. These include take-off and landing performance for different existing aircraft and future aircraft designs, 3D modelling of the ideal runway angle, airport infrastructure, noise contours, security, floods, ice, wildlife, birds, pollution, water quality and ATC issues. However, the report is also not shy in admitting that there would be significant challenges to be overcome in certain areas, including that of safety, which might only be solved through the use of future technology yet to be developed.

Hub and spoke airport

All the airport infrastructure would have to fit within the runway diameter. (NLR)

A circular runway would need a minimum diameter of 3.5km with an inner runway radius of 1,500m. The total runway length would be around 10,000m, comparable in length to three straight runways and long enough to allow multiple simultaneous operations on the runway and to build the airport infrastructure inside the middle of the circle. The runway width would be around 140m - much wider than a conventional straight runway.

The endless runway would be connected to the terminal apron via a taxiway system consisting of an outer and an inner taxiway ring. The outer taxiway would operate in the same direction as the runway and would be connected to the runway through highspeed exit taxiways where one aircraft can hold if needed. The inner taxiway would operate in the opposite direction to the outer one. Taxiways between the airport’s buildings would link the inner circular ring to the inner airfield area. Because the terminal would be much closer to the runway, the report estimates that aircraft taxi times could also be reduced by between 40% to 95%.

Space beneath the banked runway could be used for parking. (NLR)

The terminal buildings would be located in the centre of the circular site around which the aircraft would park. To allow room for more aircraft, future designs could include aircraft with folding wings. There would be also be considerable underground infrastructure, including access tunnels to get in and out of the airport, as well as parking and storage spaces beneath the angled runway. Much use would be made of automated transport systems, including trains and people movers under the ground and automatic taxiing for aircraft going to and from the runway. The authors of the report also envisage the inclusion of vertical take-offs and landings within the airport.

Landing sideways

Landing a large commercial commercial aircraft on a banked circular runway would require a new set of pilot skills. (NLR)

Because aircraft will be required to take-off and land in a circle with centrifugal force pulling them sideways, the circular runway would need to be banked in a similar way to curves on a motor racing circuit. As an aircraft accelerates for take-off, it would move from the flat inner part of the runway toward the outer banked part until it reaches the lateral position where the bank angle fits its lift-off speed. The same would apply the other way around for landings. Junctions linking the runway to the outer taxiway would be level or only have a slight slope but would need to be sufficiently large to allow aircraft to turn.

Not surprising, changing a runway from a straight level surface to a curved banked formation creates a number of challenges - not the least of which is retraining pilots on how to land and take-off. However, the trials from the 1960s demonstrated that a circular runway track can correct pilot errors when landing too high with a slow speed or too low with a high speed. The report also noted that ‘for the comfort of passengers, the centrifugal forces must not exceed values that will cause them to feel uncomfortable’. There would also need to be new procedures for aircraft making missed approaches or incidents in which an aircraft or other obstacle was blocking the runway.

Wingtip and engine nacelle clearance for large commercial jets would be less on a circular runway. (NLR)

Another problem with a banked runway is the issue of clearance between the runway surface and the aircraft engines, wing tips and undercarriage. Many modern commercial jets have very low slung engines which would come even closer to the ground on a curved surface. The report considers the cross sections of a number of large commercial jets (747-100, A340-600 and A380-800) and concludes that the clearance space would still be sufficient. It also looks to the future and proposes a new ERAC (Endless Runway Aircraft Concept) aircraft design tailored to the requirements of the circular runway featuring a double width fuselage to permit a wider undercarriage and engines mounted above the wings to allow for extra ground clearance. The one future aircraft design which the report concluded would not be suitable for endless runway operations was the blended wing body. Other issues to be tackled included surface contamination from rain, snow, ice and how these might affect the runway friction coefficient. The threats posed by FOD and birds have also to be considered.

The report details three different alternative ‘pavements’ that could be used for runway construction - rigid, flexible and semi-rigid - but does not suggest which might be most suitable for a circular angled runway. A rigid pavement is made of concrete with limited bending capacity. Rigid pavements are  slow and expensive to construct and can suffer from fissures caused by excess loads or temperature variations. A flexible pavement is made of granular layers of asphalt, tar, resins, lime, cement, vinyl or rubber which can absorb and recover from loads but can also be deformed if the elastic limit is surpassed. It has the advantage that it is adaptable to different types of ground and can be easily repaired. However, flexible pavements can be adversely affected by high temperatures and exposure to aviation fuel and de-icing substances. Semi-rigid pavements are a combination of the first two categories.

Kiss those crosswinds goodbye

The principal advantage of a circular runway, argues the report, is that it would no longer be necessary to change runway approaches depending on which way the wind is blowing. In low wind conditions, aircraft could approach the airport from different directions and land immediately. In stronger wind conditions, aircraft would fly in sequence towards the Endless Runway to land into the headwind.

Every time the wind changed, there would be a different point on the endless runway for optimal take-off and landings. Using two take-off and landings positions, the designers estimate that an endless runway could have a sustainable capacity of 80 movements per hour - compared to 115 movements for a three or four runway airport. However, the maximum capacity for an endless runway could be as much as 110-146 movements per hour.


Rethinking air traffic management

Traffic could approach and leave an endless runway from any direction. (NLR)

Regarding air traffic management (ATM) in and out of a circular airport, the report proposes an increased use of automatic systems. ‘ATM actors,’ as the report refers to them, would need, ‘to become managers rather than operators.’ It then goes into considerable detail about the different ATC and navigation systems which can be used at airports, including ILS (instrument landing system), MLS (microwave landing system), GPS (Global Positioning System) or space-based augmentation (SBSAS).

At a conventional airport, aircraft are assigned standard departure routes (SIDs) when taking off from a specific runway towards a specific TMA (terminal control area) exit point. For approaches a pilot can use navigational beacons like VORs/DMEs/NDBs to navigate his or her aircraft towards the final approach. An air traffic controller can actively separate inbound and outbound traffic flows by using radar vectoring, flight level and speed control. With radar vectoring, aircraft can be put in the right sequence and separation between subsequent aircraft can be established quickly. Vertical separation has the advantage that it can keep aircraft safely away from each other without the need to precisely know their positions in the TMA. For aircraft following the same route, ATCs can coupling the speeds of subsequent aircraft to ensures that they will not overtake each other. Approach controllers can also influence the time between subsequent aircraft on final approach by adjusting the moment an aircraft is turning to base to intercept the final approach path - a action known as ‘tromboning’. When descending towards an airport, aircraft adopt a continuous descent operation (CDO using low thrust settings to save fuel. At the ILS (instrument landing system) intercept altitude the aircraft will level off, intercept the ILS localiser signal and then the ILS glideslope, after which a regular final approach is flown.

Because of the very different layout of the endless runway, the report admits that current ATC operations, procedures and systems would have to be changed. Aircraft will be able to approach the runway from any direction, so will only need lateral and vertical separation to keep them apart in the TMA rather than speed control. Instead of using an ILS approach which aims at a single point on the runway, aircraft could continue to use a CDO until touchdown. The report recommends that an Endless Runway airport should have a landing navigation system that is independent from a chosen touchdown point and which can handle all visibility conditions. However, the report then admits that, currently, such a landing navigation system is not available but anticipates that future satellite navigation systems with local area augmentation could provide the needed flexibility, reliability, and accuracy.

It also concludes that the future development of 4D-operation and the future ATM concept of performance based navigation (PBN) which would allow a more flexible choice of routes would greatly assist ATM operations around a circular airport. Pilots using the endless runway would be able to automatically determine their optimised flight paths including descents and glide slopes. However, as pilots would not be flying standard routes, the system would have to ensure that they had good situational awareness with respect to other air traffic and their expected take-off and landing routes.

Circular advantages

A concept image of the ERAC future aircraft customised for circular runway operation. (NLR)

The report lists the advantages of the endless runways as follows:

  • Such an airport would take up only a third of the space of a conventional airport.
  • Difficult manoeuvers in harsh weather conditions, such as crosswinds, will be unnecessary as pilots can land in whatever direction is most favourable.
  • Aircraft can land and take off at any point in the circle
  • The risk of wake turbulence from following the paths of other aircraft could be avoided.
  • Flight times could be reduced as aircraft could fly straight in and out of the airport
  • Aircraft would have to circle the airport less and use less fuel.
  • The route between the terminal and the runway would be shorter, reducing taxi times
  • Reduction in noise pollution as aircraft could avoid flying over some residential areas.

Circular disadvantages

However, there have also been a number of concerns raised, particularly from pilots with regard to safety issues when taking off and landing. These include:

  • All pilots using the airport would need to be trained in new landing procedures, perhaps even having to have a new license with a circular rating.
  • Although aircraft will be able to land into the wind, they would sill still be subject to crosswinds during deceleration.
  • Although the optimal take-off position could be determined, it is not possible to know exactly when the aircraft will rotate.
  • The designers propose that the runway can be used by more than one aircraft but, if winds are strong, there will only be one optimal take-off position
  • Aircraft will not save fuel by landing directly because they will have to manoeuvre around the airport so that they land in into the wind
  • As long as the optimal position to land on the runway remains steady, aircraft will have to line up to use it with a safe distance between them, meaning extra challenges for ATC
  • Aircraft will have to land very precisely at the optimal position into the wind - a position that would be different every time. This variable position would have to clearly visible to pilots in both day and night conditions
  • Aircraft will have to turn during landing while flying slowly leading to a loss of lift and the risk of the aircraft stalling close to the ground
  • If aircraft approached and departed from the airport in different directions, then noise pollution would be spread out all over a greater area instead of being contained in two directions
  • If the airport relied on automatic systems for assist take=off and landings, what would happen in the event of these systems failing
  • What would happen in the event of an emergency in which an aircraft became immobile on the runway or had to conduct a landing without undercarriage? A belly landing on a curved runway would be much more likely to lead to the disintegration of the aircraft

Other concerns include building costs. It is not practical to rebuild an existing airport with a circular runway, so a new one would have to be built from scratch. Because of the longer and larger runway, construction of such an airport could be 110% - 160% higher than for a conventional airport. There is also a lack of flexibility in the inner infrastructure, as all the airport facilities must be located within the circle where only limited space is available. This infrastructure could also not be expanded if demand increases.

An endless future?

The concept image used on the project’s main publicity flyer does not feature the banked runway and includes an illustration of a blended wing body which the report concluded was not suitable for endless runway operation. (NLR)

The authors of the report freely admit the existence of such problems and in fact highlight several of them in the report. They conclude that much work still needs to be done on such details as which circle size would allow sufficient building activity inside the circle, how would passengers and airport personnel get access to the infrastructure inside the circle, what is the best bank angle, how should aircraft approach the runway, how to indicate the touchdown point to the pilot and how to avoid overflying densely crowded areas.

The report estimates that development of the endless runway concept could take another 20 years to come to fruition, by which time aircraft and ATC systems would be sufficiently matured to enable computers to control the landings to precisely line up with the runway.

The researchers have also performed simulations using a PC-based flight simulator and an air traffic control simulator to investigate how current operations and procedures would need to be changed and what new developments will be necessary. The next stage of development would be to fly drones on an off a curved track, after which flight tests could be conducted using larger aircraft. These would include tests on the standard behaviour of the aircraft on the endless eunway during take-off and landing, different operational circumstances, different weather conditions and emergency procedures during each phase of flight.

Will the concept of an endless runway forever spin around in circles or will it rotate and spiral up to success?

Wait around and see ...


Bill Read
4 April 2017

Comment title