Evaluation of Runway Bearing Capacity: In-Situ Measurements and Laboratory Tests A. Graziani 1 , F. Cardone 1 , E. Santagata 2 , S. Barbati 2 1 Dipartimento di Idraulica, Strade, Ambiente e Chimica Università Politecnica elle Marche 2 Dipartimento di Idraulica, Trasporti e Infrastrutture Civili Politecnico di Torino Eighth International Conference on the Bearing Capacity of Roads, Railways, and Airfields. Graziani, Cardone, Santagata, Barbati (CIRS) Evaluation of Runway Bearing Capacity BCR2A Conference, 6/29–7/2, 2009 1 / 35
37
Embed
Evaluation of Runway Bearing Capacity: In-Situ ... · Evaluation of Runway Bearing Capacity: In-Situ Measurements and Laboratory Tests A. Graziani 1, F. Cardone , E. Santagata 2,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Evaluation of Runway Bearing Capacity: In-SituMeasurements and Laboratory Tests
A. Graziani1, F. Cardone1, E. Santagata2, S. Barbati2
1Dipartimento di Idraulica, Strade, Ambiente e ChimicaUniversità Politecnica elle Marche
2Dipartimento di Idraulica, Trasporti e Infrastrutture CiviliPolitecnico di Torino
Eighth International Conference on the Bearing Capacity of Roads,Railways, and Airfields.
The structural design for the full-depth reconstruction was based on the CBR method as de-scribed by Federal Aviation Administration AC 150/5320-6D (FAA, 2004). The design traffic fleet, over a 20 years pavement life, was converted to 1778 yearly departures of a Boeing 747-200 aircraft (Design Aircraft) with a Takeoff Weight of 368,000kg (810,000lb). Using routine geotechnical tests the subgrade soil was classified as a silty clay of variable compressibility (CL-CH), with a design CBR of 7 (obtained as the 85th percentile of the values coming from DCP tests results). A lime stabilization was adopted to improve subgrade strength, assuming a design CBR value of 20. A crushed aggregate subbase layer was placed over the stabilized sub-grade, underneath a Cement Treated Base course. Considering that the construction operations had to proceed as rapidly as possible, this intermediate granular layer was deemed necessary to separate two layers where pozzolanic reactions were (supposedly) developing in their peak phase.
The spreadsheet F806FAA.xls (FAA, 2005) was used for the calculations (Fig. 1). The final structure (Fig 1.b) was designed considering the locally available materials and construction techniques. The asphalt surfacing was further subdivided in a Stone Mastic Asphalt wearing layer (30mm), a binder layer (40mm) and a base layer (80mm). All asphalt mixtures were speci-fied according to the current European Norm (EN 13108), and the use of a modified asphalt was required. The Cement Treated Base course was a “low-stiffness mixture”, typically used in Ital-ian motorways, with an unconfined compression strength (UCS) between 2.5 and 4.5MPa.
The full-depth reconstruction project was completed in 23 days, during this period the run-way was used with a displaced threshold. The residual field length (1805m) allowed to safely manage operations of the B737 aircraft, as requested by Aerdorica. New non-precision instru-ment approach procedures were developed since the ILS glide slope was not available during the working period. Temporary horizontal signs were provided and a Precision Approach Path Indicator (PAPI) was installed to assure a more accurate approach to the touchdown zone.
The working area was divided in 2 zones (construction lots). In the “Green Zone” located 350m apart from the displaced threshold all the construction activities were carried out during the daylight, without disturbing the traffic flow. In the “Red Zone”, located 160m to 350m apart from the displaced threshold, the construction activities were carried out, from 11:30 pm to 6:00 am. During this period the runway was closed to avoid penetration of the obstacle limi-tation surfaces.
3 MATERIALS CHARACTERIZATION
3.1 General remarks
An extensive experimental program was carried out to achieve a careful characterization of ma-terial properties, particularly for the stabilized subgrade. The following tests were part of con-struction quality control: ! Dynamic Cone Penetrometer tests (DCP); ! Dynamic Plate Bearing tests (DPBT); ! Static Plate Load tests (SPT). Cores obtained from compacted asphalt layers were tested for composition and voids. In addi-tion a detailed laboratory characterization of the asphalt mixtures, including dynamic modulus estimation and fatigue performance, was performed using both laboratory compacted specimens and pavement cores.
Area A: 520mfull-depth pavementreconstruction;area NOT USABLE forTake-off/Landing;2.085 m of pavement stillavailable;after enforcement of all safetyrequirements (strip, RESA)residual field length allowsoperations up to the B737airplane;airport fully operative.
non-standard (FAA) constructionstandard Italian materials and construction techniques
2 PROJECT OUTLINE
The structural design for the full-depth reconstruction was based on the CBR method, as de-scribed by the Federal Aviation Administration AC 150/5320-6D (FAA, 2004). The design traf-fic fleet (Tab. 2), was converted to 1778 yearly departures of a Boeing 747-200 aircraft (Design Aircraft) with a Takeoff Weight of 368,000kg (810,000lb). The spreadsheet F806FAA.xls (FAA 2005a) was used for calculations.
The final structure was designed considering locally available materials and construction techniques (Tab. 3). Lime stabilization was adopted to improve subgrade strength (3% hydrated lime). A crushed aggregate subbase was placed over the stabilized layer and below a Cement Treated Base (CTB) course. The CTB was a “low-stiffness” mixture, typically used in Italian motorways. Considering that construction had to proceed as rapidly as possible, the intermediate unbound granular layer was deemed necessary to separate two layers where pozzolanic reac-tions were (supposedly) developing in their peak phase. The asphalt surface course was further subdivided into a Stone Mastic Asphalt wearing layer (30mm), a binder layer (40mm) and a base layer (80mm). All asphalt mixtures were specified according to the current European Norm (EN 13108), and the use of a modified asphalt was required.
The full-depth reconstruction project was completed in 23 days, during this period the run-way was used with a displaced threshold. The residual field length (1805m) allowed the opera-tions of the B737 aircraft to be safely managed, as requested by Aerdorica. New non-precision instrument approach procedures were developed as the ILS glide slope was not available during the working period. Temporary horizontal signs were provided and a Precision Approach Path Indicator (PAPI) was installed to assure a more accurate approach to the touchdown zone.
The working area was divided in 2 zones (Figure 1). In the “Green Zone” located 350m from the displaced threshold all the construction activities were carried out during the daylight, with-out disturbing traffic flow. In the “Red Zone”, located 160m to 350m from the displaced thresh-old, construction activities were carried out from 11:30 pm to 6:00 am. During this period the runway was closed to avoid penetration of the obstacle limitation surfaces.
Table 3. Summary of pavement design and specifications __________________________________________________________________________________________________________
Layer Thickness Aggregate Air Asphalt Strength* Modulus** Dmax Voids content mm mm % % MPa __________________________________________________________________________________________________________
AC Stone Mastic Asphalt 30 10 2÷6 6.5÷7.5 AC Binder Course 40 20 4÷6 4.5÷5.5 AC Base Course 80 31.5 4÷6 4.0÷5.0 Cement Treated Base 200 2.5!UCS!4.5 " 100 Crushed Aggr. Subbase 100 CBR " 40 " 60 Lime Stabilization 400 CBR " 30 " 50 Natural Subgrade (CL-CH) CBR = 7 __________________________________________________________________________________________________________
*UCS: Unconfined Compression Strength (MPa); CBR on 4 days soaked samples ** Measured in Static Plate Load Test, 1
Figure 3. Experimental Vs literature fatigue lines (for EA=5000MPa)
4 STRUCTURAL RESPONSE
4.1 General Remarks
The structural response of the “as-built” pavement was evaluated using an Heavy-Falling Weight Deflectometer. Tests were carried out with a 30cm loading plate, at a nominal vertical force of 160kN. Deflections were collected using a 9-sensor configuration, with a constant spac-ing of 300mm. The HWD survey, carried out 72 days after construction end, comprised a total of 42 measurement points, positioned on 4 parallel alignments (±3 and ± 6m offsets from the Center Line). The pavement temperature was measured at three different depths and a mean value of 8°C was calculated.
The shape of the measured deflection basins were initially examined and 3 homogeneous sec-tions were found, closely matching the 3 construction lots. The back-calculation analysis was performed with the linear-elastic based program BACKFAA (FAA 2006), using the 85%-reliability deflection profiles of each homogeneous section. The asphalt stiffness modulus were referred to a reference temperature of 20°C using the stiffness-temperature relationship obtained in the laboratory, in particular the 20Hz isochrone curve was used (Eq. 3 & Tab. 6).
not only “fatigue”. . . still very good performance;modified asphalt does his job.
The actual section was converted into an equivalent “standard” FAA section using appropriate equivalency factors. A total thickness of 90mm (32in) was determined for the equivalent section and the B747 aircraft design curves were used to estimate the bearing capacity. The key parame-ter of the evaluation procedure was the subgrade CBR value. The DCP results (Tab. 4) indicated an average CBR of 6 and a design value as low as 4 (85% percentile). As a consequence, a bear-ing capacity of 172,000kg (380,000lb) was estimated. It was an exceedingly low value if com-pared with the MTOW of the design aircraft. This mainly derived from the fact that the pave-ment was built using materials that differed, somewhat considerably, from the FAA standard.
5.3 Analytical-empirical approach
An elastic multi-layered model was used to calculate stresses and strains produced by the design traffic loads. These were then compared to the correspondent “critical values” to compute pavement damage. Asphalt concrete fatigue cracking and subgrade rutting criteria were consid-ered. Damage calculations were performed following the procedure described by Monismith (Monismith et al. 1987). With this approach it was possible to directly take into account the lat-eral traffic wander.
5.3.1 Pavement design temperature and traffic Using a 30 years database of temperature data, 4 design periods were selected (Tab. 10). For each season two different design air temperatures were determined (USACE, 2001), respec-tively for the analysis of the vertical strain at the top of subgrade (!v,SG) and the horizontal ten-sile strain at the bottom of asphalt concrete surface (!h,A). The corresponding design pavement temperatures were obtained with the relationship developed by Witczak (Witczak 1972).
The design traffic previously used for design (Tab. 2) was considered and departures were distributed between the four design periods according to the actual airport schedule. The traffic load frequency is normally assumed at 10Hz for runways and 2Hz for taxiways. At the airport, because of the absence of a parallel taxiway, aircrafts normally use the runway to taxi from/to the apron, thus a “midway” design frequency of 5Hz, was selected. According to HoSang (Ho-Sang 1978), the lateral wander was characterized with a standard deviation of 2,4m.
5.3.2 Pavement structures and material properties As previously noted, the weakest pavement structure was selected for evaluation (Tab. 9). For each design pavement temperature, asphalt concrete design stiffness values were calculated cor-recting the back-calculated stiffness for frequency and temperature (Tab. 10). The correction factors were determined using the same stiffness-temperature relationship previously used to re-fer stiffness values to the reference temperature (20Hz isochrone). A total of 8 structures were considered in the analysis.
Transfer functions adopted by the FAA were used to compute the allowable number of strain repetitions for asphalt concrete (NA) and subgrade (NSG) failure criteria (FAA 2005b).
Table 10. Design periods characterization ____________________________________________________________________________________
Design Period AC Design Modulus Percent of !v,SG analysis !h,A analysis total traffic MPa MPa % ____________________________________________________________________________________
1 June, July, August & September 2263 3005 45 2 May & October 3841 4653 15 3 March, April & November 5591 6340 20 4 December, January & February 7101 7721 20 ____________________________________________________________________________________
traffic: same as in design;� load frequency: 5Hz; lateral wander: 2.4 m
cumulative damage calculation (Miner’s rule) for each airplane:
ni =�
k
ni,k
Ni,k
� ni = actual number of strain repetitions of magnitude �i,k =;� Ni = corresponding number of allowable repetitions, computed using the
5.3.3 Structural analysis and damage computations A linear-elastic multi-layer computer program (BISAR) was used for the (forward) calculation of stresses and strains. For each landing gear the maximum strain locations were initially deter-mined and a critical cross section identified. Along this section the strain distribution was com-puted using a total of 81 points with a 0.25m spacing. At each point, the damage produced by each single aircraft was computed using Miner’s accumulation law (Monismith et al. 1987).
The damage calculations were repeated for all the design conditions described above and the results were summed to obtain the damage produced by each single aircraft. The cumulative damage was then calculated summing the effects of all the aircrafts (Fig. 4). Maximum values of 0.165 and 0.002 were found, respectively for the asphalt and subgrade strain criteria.
Cumulative damage values are extremely low and fatigue in the asphalt layer is the critical condition, yielding a residual life of 121 years. This result is mainly a consequence of the direct evaluation of aircraft wander. In addition, a major contribution in limiting strain values can be credited to the stiffness properties of the modified asphalt layers obtained using HWD and labo-ratory testing.
Figure 4. Cumulative damage for asphalt strain criterion
6 CONCLUSION
The paper presents a case history regarding the structural evaluation of a runway pavement. The project involved the structural rehabilitation of a 520m long asphalt concrete section through full-depth reconstruction.
A careful characterization of the pavement materials was achieved through an extensive ex-perimental program. Static and dynamic plate tests showed that the lime stabilized soil and the unbound granular subbase had similar stiffness, while a higher figure could be estimated for the cement treated base. Moreover, the combined use of DCP and plate tests allowed to identify the depth of the stabilized layer and define a reference stiffness range for the interpretation of de-flection tests.
The properties of the asphalt concrete layers were measured both on cores and laboratory compacted specimens. In particular, stiffness characteristics have been assessed using cyclic uniaxial compression tests. Even if laboratory compacted specimens showed somewhat higher stiffness values than cored samples, similar rheological behavior could be predicted and a single master curve was used. In addition 4 point bending tests performed in controlled stress condi-tion showed a good fatigue performance of the mixtures.
Data from in situ and laboratory tests was used to assist the back-calculation of pavement de-flections. This was of invaluable importance in identifying a consistent representation of the
keys to obey the tight construction schedule:� realistic specifications (“ad-hoc” specs);� “on-line” quality control;� close coordination between “pavement team” and the Operative personnel.
integration between NDT and Destructive tests (quality control);
use of polymer modified binder:� requires the Mastrer Curve to be measured;� enhance fatigue life
analytical evaluation, necessary when using non–standard section andnon–standard materials.