Pavement discussions

A report from this year’s European Pavement Workshop from the organisers.

Prof. André Molenaar opened the workshop by noting that the current workshop is, in fact, a European continuation of the three former Dutch workshops organised by CROW. Interest in this workshop was high, considering that just over 100 people from all over the world were participating.

The most interesting topics will be briefly described in this article. The first keynote speech was by Ray Rollings, a military pavement engineer with over 30 years of experience on the (un)expected deterioration of military and civil airport pavements. Ray pointed out that problems with materials and construction are more profound than those related to pavement design. Nowadays 30-ton wheel loads with 2.4 MPa tyre pressure are no longer the exception.

Cyril Fabre of Airbus started by showing a video of the A380-800 maiden flight cycle. The images were quite impressive. The distance of the runway needed for take-off was between 1 and 1.5 km, although the plane was not at max. take-off weight.

Airbus invested a lot in the Pavement Evaluation Program, which started in 1998 and, the analysis of modelling and test results will be finished in 2005. Airbus cooperated with the French STBA, LCPC and ADP (Airport De Paris). As a result, the landing gear has been optimised to inflict minimum damage to the pavements. One of the conclusions is also that design methods should better accommodate modern gear loads.

Subgrade criteria

In the first technical session, Greg White, from Australia, elaborated on the development of APSDS and its validation with the US Corps of Engineers S77-1 pavement. The S77-1 pavement design method is based on the classical CBR-method. In his study, Greg focused on strain related subgrade criteria. He investigated the various input parameters and their effect on the calculated asphalt thickness. An important aspect is the lateral wander of aircraft over the pavement width. In APSDS the contact pressure on the pavement is equal to the tyre pressure of the aircraft’s wheels.

Up to now, a constant value for lateral wander is used. However we know, for example, that the A380 is quite different in lateral wander from a B777, so in the future a variable value should be inputted. Also, a calibration for a 6-wheel gear could be necessary for new versions as well. The sensitivity study showed that of all parameters, the subgrade modulus is the most influential on total pavement thickness, with aircraft mass also being important.

Alur Nataraj introduced a new method of settlement by vacuum techniques, used on the fifth and latest runway at Schiphol airport, which lies4.5 m below sea level. Not having crushed rock material available for stone set bases, a typical feature of Schiphol’s 5P-pavements is the use of a Cement Treated Bases (CTB). In this case, the compressive strength of the CTB on site should be more than 90 percent of the strength obtained in laboratory tests. The presentation ended with the statement, ‘If you build a runway at Schiphol airport without major problems, you can build runways everywhere’.

Miss Sharda Bramatewari recently graduated at Delft University on a research study for alternative pavement structures on taxiways at Schiphol airport. She compared the damaging effect of the B747, B777 and A380 on a reference structure at Schiphol Airport to an alternative one with the use of specific asphalt mixes.

Based on her study, she concluded that the structural damage of a B777 is 10 to 15 times less than the other two aircraft and as a consequence, the pavement lasts longer. In this case, the pavement life is governed by the CTB. The reference structure comprises 200 mm asphalt on 500 mm CTB, of which the asphalt package can be decreased somewhat when using the alternative materials. These materials are an anti-reflective cracking layer (20 mm), a binder course that is resistant to permanent deformation and an asphalt mix with a high elastic modulus, which could reduce thickness by 25%.

Miss Michaela Schubert launched the new German Guideline for the design and evaluation of military airports in Germany. Three reference aircraft were used: a Tornado fighter jet, C130 and A310, because those are stationed on each military airfield. For example: for a Tornado load, the pavement consists of 290 mm concrete slabs on 350 mm unbound granular base layer.

ACN-PCN method

In the afternoon session, the key words of the first section were cement concrete, design and bearing capacity.

The limits of ICAO’s mandatory ACN-PCN load classification method was recognised and addressed by George Charonites, who reported with Prof. Loizos on the ACN-PCN method of flexible and rigid pavements. An inventory was made and a variety of methods was discussed: FAA (US), STBA (France), NTUA (Greece) to name a few. The principle drawback is the use of the Derived Single Wheel Load in the ACN calculation. As an alternative, the Aircraft Loading Index was introduced. This analytical based index is used next to the ACN-PCN-method as an Airfield Pavement Capacity Index. The concept is, in fact, an improvement of the ACN-method used for classifying aircraft loads.

Marc Stet also addressed the limits of the current ACN-PCN method and introduced a new Dutch CROW-guideline on PCN assignment. He pointed out that the ACN-calculation procedure for classifying aircraft loads is strict and prescribed by ICAO, and that the underlying criteria do not relate with the analytical methods used in modern pavement design. The need for a method truly describing pavement damage and strength was pointed out. The assignment of the PCN (bearing capacity pavement) is left up to the airport.

This gives the opportunity to base the PCN on the same analytical criteria that are currently used in structural pavement design. By using a reversed design approach, the Dutch guideline takes into account realistic pavement life and damage. The guideline takes the ACN-index as a relative scale only, whereas the PCN is based on structural pavement life and true bearing capacity.

Chairman Janpiet Verbeek challenged the audience by proposing a vote on the necessity of the classical ACN-PCN system. The majority favoured developing a better classification method for aircraft damage, describing true pavement bearing capacity. Based on the results of large scale pavement tests, new criteria and advanced Finite Element models are being developed. This can be the basis for an updated design and classification method. However, for the time being, the use of the ACN-PCN method is mandatory and prescribed as the only method to use.

The keywords of the second part of the afternoon session were: asphalt, lab-testing and polymer modified asphalt (PMA) and polymer modified bitumen (PMB).

Arian de Bondt reported on PMB in asphalt. One of the questions was: What amount of PMB should be added to find any effect in the properties of the asphalt mix? This varies between 5 and 10 percent, and depends on the deterioration that you want to prevent: cracking or rutting. There is no linear range for adding PMB’s; normally an amount of 5% is used.

Marc Stet presented the results of a CROW-study on how to choose the right PMB for a specific PMA. This CROW-research was initiated by the Dutch market parties. The conclusion was that in general the use of PMA is beneficial and that use depends on loads, the possibility of failure and the importance of the pavement.

Nadim Haddad ended day 1 presentations with the experiences of a field project from Queen Alia International Airport, where the existing pavement needed rehabilitation. Concrete slabs (6*6 m, 400 mm thickness) were repaired and overlaid with a 180 mm asphalt layer. Compared to other options, he managed to save 50 percent of the cost compared to a complete reconstruction and 50 percent of the time needed for construction.

Pavement modelling

In the morning session on day two, the topics were pavement modelling related to design, practical aspects for airport pavement performance, performance of overlaid pavements, and special condition modelling and analysis of pavement performance.

Julien Bilal presented a new French design method, ELSA (Etat Limité de Service Admissible or Acceptable Service Limit Level). One of the reasons behind ELSA is that the CBR-method is no longer sufficient for today’s airplane and pavements. ELSA incorporates dynamic loading and uses Cesar 3D FEM software. The model was calibrated with experimental pavements by means of monitoring the pavement for three years.

Wouter van Bijsterveld presented the reconstruction works on the northern runway of Frankfurt Airport, where equipment on the runway was only allowed from 22.30 till 06.00 hrs. This was the reason that only sections of 15 m per night could be totally reconstructed. The old concrete and base layer were removed and the 600 mm gap filled with a full-depth asphalt pavement. This package needs a considerable time frame to cool down. This was the reason for searching alternative solutions such as crack and seating of the old concrete and putting an asphaltic overlay system on top.

Halil Ceylan produced a software application called I-Pave, based on Artificial Neural Network technology (ANN). The predictions from this tool correlate almost 100 percent with the results from 2D or 3D FEM programs. Where I-Pave can do 2 million calculations in one minute, FEM programs would need more than 10 days. The tool can be used within the boundaries of the ANN-setup and could be implemented as quickly as tomorrow.

Jean-Maurice Balay combined two presentations about the Airbus Pavement Evaluation Program, rigid phase. He described the set-up of the rigid pavement program and explained the structures. Pavements with 420 mm (without dowels) and 310 mm (with dowels) concrete slabs (5 * 5 and 7.5 * 7.5 m) on a 150 mm lean concrete or 430 mm crushed stone on various subgrade conditions have been tested.

The pavements were fully instrumented. The static loads used were the main landing gear of the A380-800, B747-600, MD-11, A380 (with 22 wheel landing gear!) and B777-300ER. The development of cracks with number of passes was recorded and presented in drawings. The smaller concrete slabs performed better. Another conclusion was that non-dowelled concrete slabs performed better.

Analysing the pavement structure was done with the César FEM software of LCPC. The results of the static loading tests were used to calibrate and validate the model. However, the top-down cracking (full thickness) and the cracking of the non-trafficked slabs could not be explained.

Ad van Leest presented results of a CROW working group that was seeking alternatives for the current surface coatings. Surface coatings must meet statutory requirements and should have sufficient resistance to FOD and chemical substances. The steel brush test turned out to be the most discriminating test for this purpose. There is still a challenge for contractors and suppliers to develop new products that have the same properties as the tar bearing antiskid surface coating.

Prof. Molenaar concluded that given the available computer programs and fast computers, complex design analyses are not a problem anymore. The problem, according to Prof Molenaar, is much more in proper material characterisation, in taking into account the effects of e.g. moisture and temperature, and in quantifying the effects of the way in which we construct our pavements on the quality of the end product. Defect types like ravelling, asphalt rutting, scaling and surface cracking are very much related to the quality of the material of the surface layer and to contact pressure distributions and environmental conditions. If these characteristics are not modelled to a sufficient depth, then we will remain unable to predict and explain important defects.

Furthermore, calibration of design methods and tools is essential in gaining the necessary confidence in them. Therefore, the work done by Airbus in their Pavement Experimental Program and the work done by FAA in Atlantic City with their NAPTF is of great importance.

Adrian J. van Leest

Mr Adrian J. van Leest is Project Manager at CROW and is currently dealing with topics in the field of airfield pavements, road pavements and railway constructions.

Marc J.A. Stet

Marc J.A. Stet is a senior consultant at KOAC-NPC, the Dutch Road Research Laboratories specialising in the design and analysis of heavy duty pavements. He is a member of several CROW-working groups and has written several publications on airport pavements and concrete pavements.