Method uses a soil and hammer model and  
 calculates the force using the measured  
 velocity and then compares the calculated  
 force with the measured force; if the values  
 do not match, the soil model is changed,  
 and the analysis repeated. 
 When piles  are driven,  soil  resistance  
 that opposes the penetration of piles consists  
 of  two  components,  such as  static  
 resistance  and  dynamic  resistance  
 (Rausche et. al, 1985). CAPWAP computes  
 Bridge, Bent  
 No 
 Pile Size &  
 MFSR (kip) 
 Tip Elevation  
 (ft)/  
 Length (ft) 
 From the Table 1 results, it is evident  
 that total capacity of each pile exceeded  
 both corresponding NDR and MFSR when  
 piles were founded on hard PWR, with  
 N values ranging between 50/5" to 50/1".  
 Therefore,  it  may  be  concluded  that  geotechnical  
 axial capacities of these rock-like  
 materials are high and comparable to rock.  
 the total capacity and then subtracts the  
 dynamic component to determine the  
 static resistance.  For  formulation  of  Case  
 Method, FHWA-NHI-16-003, Volume 2 can  
 be referred. 
 5. PDA test results and discussions 
 The following five PDA test results (Table  
 1) were obtained from a project site in  
 the Atlanta metro area. All dynamic tests  
 were  conducted  using  an  ICE  Model  
 NDR 
 (kip) 
 Table 2 summarizes the penetration  
 depths  into  the  PWR  strata,  NDR,  maximum  
 stress and blow counts (blows/inch). 
 TECHNICAL 
 Table 1: PDA and CAPWAP results 
 Table 2: Summary of PDA/CAPWAP results 
 Bearing  
 Materials-  
 PWR 
 Top of PWR  
 El. 
 I30v2 open-ended diesel impact hammer.  
 Hammer cushions consisted of Conbest  
 II (Nylon) and aluminum plates (a total of  
 four-inch thickness). All capacity results  
 are static pile capacities for various sizes  
 of steel H-piles driven to hard PWR. The  
 stresses due to pile driving did not exceed  
 the maximum permissible stress of 45 kips  
 per square inch corresponding to 0.9 Fy. 
 Auger  
 Refusal  
 Elevation 
 1CAPWAP Results (kip) 
 Side Shear End Bearing 2Total  
 Capacity 
 Bridge A,  
 Bent 10 
 HP14x117,  
 860 
 940.45/  
 25.30 
 802.00 PWR,  
 N=50/1" 
 957.86 936.26 228.00 822.00 1050.00 
 Bridge B,  
 Bent 1 
 HP14x117,  
 860 
 1062.02/  
 24.10 
 818.00 PWR,  
 N=50/5" 
 to 50/4" 
 1082.86 1047.11 251.20 626.9 878.10 
 Bridge C,  
 Bent 3 
 HP14x73,  
 520 
 938.67/  
 42.42 
 459.00 PWR,  
 N=50/5" 
 945.84 938.34 241.00 294.00 535.00 
 Bridge D,  
 Bent 
 20 
 HP14x117,  
 860 
 919.75/  
 17.80 
 852.00 PWR, 
 N=50/5" 
 920.56 909.06 225.90 718.50 944.40 
 Bridge E,  
 Bent 1 
 HP12x53,  
 385 
 979.26/  
 47.98 
 378.00 PWR,  
 N=50/3" 
 984.73 957.23 180.20 400.00 582.20 
 1The CAPWAP analyses were conducted using Jc Method based on End Of Initial Drive (EOID) Tests 
 2 These capacities are not ultimate capacities; tests were terminated depending on the required minimum tip elevations and NDR=Maximum  
 Factored Stregth Limit State Load/0.65, where, jdyn = 0.65 (AASHTO Table 10.5.5.2.3-1) 
 Bridge/Bent No. MFSLSL (kip) Maximum Stroke  
 (ft) 
 Maximum  
 Compressive  
 Stress (ksi) 
 Blows per inch Penetration into  
 PWR (ft) 
 Beta Integrity  
 Factor (BTA,  
 100%) 
 Bridge A, Bent 10 521 9.5 28.4 16 17.41 100 
 Bridge B-Bent 1 532 10.3 32.1 12 20.84 100 
 Bridge C-Bent 3 276 8.7 30.1 10 7.5 100 
 Bridge D-Bent 20 553.3 11.79 30.83 20 0.81 100 
 Bridge E-Bent 1 246 11.1 37.3 20 5.47 100 
 122  |  EDITION 6 2019  www.piledrivers.org 
 
				
/www.piledrivers.org