Академический Документы
Профессиональный Документы
Культура Документы
4 Dec 2015
PDA is a commonly used test which shows exactly what is happening as a pile is being driven. It is cheap and quick to set up and shows a vast amount of informaEon about the driving
hammer and soil resistance in real Eme as the pile is being driven.
However, the PDA test has earned a bad reputaEon because of poor quality interpretaEons by incompetent operators, compounded by the fact that very few civil engineers in charge of
the project understand adequately how to read the PDA results.
This presentaEon is for engineers to understand the basics of what is the PDA test and what the shape of the stress wave tells us about the pile driving system.
Introduc7on
PDA is a powerful tool to observe and measure what is occurring during pile driving. It can tell us what is happening
when the hammer strikes the pile, but also more importantly, it can help us to evaluate the soil capacity and the
distribuEon of soil resistance as well as pile integrity along the enEre length of the pile.
Compared to a full staEc load test, PDAs are many Emes cheaper and more convenient. Therefore, PDAs can be done on
a large number of piles (up to 100% of the driven piles in some projects, giving the best insurance against pile failure).
For driven piles, PDA provides instantaneous informaEon on what is happening during driving, and allows immediate
recEcaEon if the pile is not able to achieve its required capacity.
PDAs can be done on non-driven piles, but requires the pile to be driven for the PDA test to mobilize soil resistance.
For raked piles, the soil resistance may be very dierent from verEcal piles, and this cannot be seen from verEcal
borehole data or verEcal pile load tests. PDA can measure this eect easily.
In order to derive full benets from the PDA, we need to study the dierent porEons of the stress wave, and understand
what is causing the changes in shape of the stress wave. This arEcle provides a guide to interpreEng what the PDA test
data is telling us.
Wave Theory
During pile driving, the pile behaves like a long elas7c spring. The
impact from the hammer does not act on the enEre pile
instantaneously, but rather takes a nite Eme to travel down the pile.
The impact creates a force pulse which travels down the pile at a
speed c = (E/) = 5122m/s for steel. Behaviour of the force pulse is
governed by one dimensional wave theory.
A long pile actually behaves like a damper, rather than a rigid body. .
A rigid body behaves completely dierently, and in certain respects
does the opposite of what the correct theory shows.
PDA measures the wave pulse and the reecEons caused by soil
resistance during pile driving and is a much more correct way of
predicEng pile capacity.
*Likins, Fellenius & Holtz, (2012) Pile Driving Formulas, PILEDRIVER, Q2 2012 | Vol. 9, No.2
Pile driving
formulas assume
pile to be a rigid
body WRONG!
PDA Instrumenta7on
PDA instrumenta7on consists of (2)strain and (2)accelera7on sensors mounted on the pile. The sensors
are located usually below the hammer and above ground level.
Strain and acceleraEon are measured and converted into force and velocity for analysis.
The sensors are usually mounted on an exposed porEon of the pile not less than 2D from the top.
Two sets of sensors are used; mounted on opposite sides of the pile and averaged in order to
isolate the eects of bending of the pile.
Sensors
Output from the sensors can be sent wirelessly to a nearby PC for analysis or even directly over the
internet to an engineer sifng in his oce.
Just by looking at the shape of the stress wave, much informaEon can be deduced about the eecEveness
of the pile driving equipment, the magnitude and distribuEon of skin fricEon end bearing, as well as the
integrity of the pile. Therefore, it is useful to learn about what exactly the wave shape is revealing.
Soil resistance during driving can be esEmated directly on site from the PDA output using the Case
Method, but this is inaccurate. To properly obtain the soil capacity, it is necessary to do a wave matching
analysis such as CAPWAP, iCAP. ALC uses Full Wave Matching (FWM) which gives all the pile driving
parameters as well as the soil parameters.
The PDA sensors measure strain and accelera7on, which translate to force and velocity respecEvely. The velocity is
mulEplied by the impedance Z of the pile to give the result in force units. These are then ploked against Eme. (blue
Force; red Vel*Z)
It is omen more convenient to work with the downward and upward components of the stress wave, Wdn and Wup.
The downward wave, Wdn = (Force + Vel*Z) / 2 is generated by the hammer. It is convenient to remember
the downward wave as the average of the force and velocity traces.
The upward wave, Wup = (Force Vel*Z) / 2 is the reected wave due to soil resistances and changes in pile
cross-secEon. The upward wave is the dierence between the force and velocity traces divided by 2.
Wdn (brown) and Wup (green) are calculated from the PDA Force (blue) and Velocity (red) data and can be ploked
together on the same graph for easy reference.
According to wave theory, Wdn and Wup travel in opposite direcEons and do not aect one another.
Force and Velocity can also be back-calculated from the Wdn and Wup waves:-
A downward stress wave is generated by the impact of the hammer ram on the cushion. The wave passes through
the cap block and is transmiked down the pile.
The downward wave is modied and also parEally reected upwards by the soil resistance and changes in pile crosssecEon along the pile.
When the downward wave reaches the bokom end of the pile (at t = L/c), it reverses direcEon and becomes an
upward wave. The amount and type of reecEon depends on the soil condiEon at the Ep of the pile.
The upward wave then travels up the pile and becomes a downward wave again when it reaches the top end of the
pile (at t = 2L/c). On the way up, it is again modied and parEally reected downwards by the soil resistance along
the pile.
SWA plot
800
600
Force
400
Vel*Z
Wave
up
200
Wave
down
0
-200
-400
0#
0
0#
1
1#
2
2#
3
3#
(t = L/c) t/(L/c)
4
4#
5
5#
6
6#
Basic#wave#dn##
Basic#wave#up#
PDA#level#!#Y#
Sha=#reec@on#up#!#X#
!1#
Wave%Reec)ons%
End#bearing#reec@on#
up#!#X#
Wave#Peak#dn#!#X#
By analyzing the shape and magnitude of the wave reecEons, we can deduce the pile cross secEon, the soil
resistance and pile integrity at each point along the pile.
Force$1$
Force$4$
Force$7$
600$
Force$10$
Force$13$
VelZ$1$
400$
VelZ$4$
VelZ$7$
VelZ$10$
200$
Tons%
VelZ$13$
Wave$up$1$
0$
Wave$up$4$
Wave$up$7$
Wave$up$10$
!200$
Wave$up$13$
Wave$dn$1$
Wave$dn$4$
!400$
Wave$dn$7$
Wave$dn$10$
Wave$dn$13$
!600$
0$
1$
2$
3$
4$
5$
6$
7$
t/(L/c)%
This plot shows the eect of placing the sensors at dierent locaEons along the pile. The darkest lines are for the
sensors located at the pile top, and the progressively lighter lines are for sensors located further and further down
the pile, but sEll above the mud line.
The WaveDown and WaveUp curves (brown & green) do not change in shape with dierent sensor loca7ons. They
only are shiZed to the right and leZ side respec7vely (phase shiZ in 7me).
The Force and Velocity curves (blue & red) do change shape as the are the sum and dierence respecEvely of
the WaveDown and WaveUp curves which are shimed in opposite direcEons.
This makes it easier to work using the WaveDown and WaveUp curves, rather than using Force and Velocity.
The lower down the sensor is, the longer the lag Eme before the WaveDown pulse arrives at the sensor, thus
shiZing the curve to the right. The amount of shim is the Eme it takes for the wave to travel from the pile top
to the sensor locaEon.
The WaveUp pulse travelling in the opposite direcEon will reach the lower sensor earlier, hence the shiZ of the
curve to the leZ by the Eme required for the wave to travel between the sensor and pile top. This 7me shiZ
must be accounted for to accurately locate the pile 7p reec7on.
Wave Reec7ons
SWA plot
800
600
Force
400
Vel*Z
Wave
up
200
Wave
down
0
-200
-400
0#
Sensor level
0#
0
Wave reaches
sensors
1#
1
skin fricEon
reecEon
reaches
sensors
2#
2
3#
3
4#
4
5#
5
6#
6
Basic#wave#dn##
2nd Ep
Basic#wave#up#
reaches sensors
PDA#level#!#Y#
t/(L/c)
reecEon
Sha=#reec@on#up#!#X#
!1#
Wave%Reec)ons%
End#bearing#reec@on#
up#!#X#
Wave#Peak#dn#!#X#
A study of the wave reecEons is key to understanding the stress wave. Signicant changes in the wave shape occur when
the Fdown or Fup wave passes through the PDA measurement level shown by the red dashed line.
400
200
Force
total soil
Skin FricEon
600
Vel*Z
Wave
up
Wave
down
Part 1
Hammer
-200
Part 2
Sham
-400
0
0#
0#
1
1#
2
2#
3
3#
Part 3 Tip
t/(L/c)
ReecEon
4
4#
5
5#
6
6#
Basic#wave#dn##
Basic#wave#up#
PDA#level#!#Y#
Sha=#reec@on#up#!#X#
!1#
Wave%Reec)ons%
End#bearing#reec@on#
up#!#X#
Wave#Peak#dn#!#X#
In interpreEng the stress wave the 3 parts of the stress wave are:-
Part 1 Hammer
WaveDown is aected only by the hammer and pile top properEes. Not yet aected by soil.
The downward wave shows the characteris7cs of the hammer and pile top.
Part 2 ShaZ
WaveUp is modied by the skin fricEon along the side of the pile and by changes in pile cross secEon.
The dierence in separa7on of the F and V curves gives the soil skin fric7on.
Part 3 Tip
WaveUp is addiEonally modied by the end bearing resistance as well as the pile Ep reecEon.
The maximum height of the WaveUp curve equals half the total soil resistance.
HAMMER
Part 1 - Hammer
600
Force
800
400
At this part of the wave, the Force, Velocity and Wave
Down curves should coincide. If they do not, the
results are not properly calibrated and should be
rejected.
This is called propor7onality.
Part 1 Hammer
Dura7on = 2Ls/c;
Ls = length to mud line
200
t = 0
t = L/c
t = 2 L/c
Vo
W1
Vo
W1
K1
Simple theoreEcal
formula.
Results in an
exponenEally
decaying stress wave
W2
Vo
W1
More complicated
theoreEcal formula.
Rounds o the
exponenEal curve.
SEll does not look
like the PDA data
W2
K1
Very complicated
theoreEcal formula.
Shape looks more
realisEc.
SEll insucient to
match actual
measured wave
Vo
W1
K1
W2
K2
W3
W1
K1
K2
BEST MODEL
800
Data from: StresswaveCOMPARISON1-plots.xlsx
700
600
Force (T)
500
400
300
200
100
0
0
0.005
0.01
0.015
0.02
time (sec)
0.025
0.03
0.035
0.04
0.045
1000
K=
Eect of K1
900
900
800
800
600
500
600
500
300
150
Increasing K1
300
200
4
5
50
550
100
150
200
250
300
50
100
600
500
400
250
300
700
600
Increasing W2
500
400
2000
300
300
5000
10000
200
200
Eect of W2
800
700
150
wh =
W=
Eect of W1
800
Increasing W1
200
20000
40000
100
1000
1500
2000
2500
3000
100
5000
10000
0
0
50
100
150
200
250
300
Z=
Eect of Z
900
50
100
150
200
250
300
20000
K2 =
Eect of K2
700
800
600
700
600
500
500
400
400
100
300
150
300
200
200
200
150
Decreasing W2
250
350
250
100
300
Increasing Z
0
0
50
100
150
200
250
300
450
100
Increasing Vo
0
K2
350
100
0
K1
250
200
W1
W2
400
400
Vo
700
700
450
100
550
0
0
50
100
150
200
250
300
Vo
W1
K1
W2
K2
Z
K1+
K1-
K2 -
Vo.Z +
W1+
K1+
(W /
1 Z) +
1.
2.
The raEo Z/W1 i.e. the pile impedance / the ram weight
determines the rate of decay of the funcEon, i.e. how
much it slopes downward.
3.
4.
W2 determines how far the rst dip is from the rst peak.
SHAFT FRICTION
600
Force
Sham FricEon
400
200
Sham FricEon / 2
800
t = 0
t = L/c
t = 2 L/c
wh
K2
V
Z
2.1
484
3.7
182
Set
Set_Qav
0
0.00
Qt=15,Rtotal=901
TargetRs%
TargetRtotal
50%
901
Qs
Qt
2
15
Js
Jt
2
2
wh
K2
V
Z
2.1
484
3.7
182
Set
Set_Qav
0
0.01
Examples
Qt=20,Rtotal=809
TargetRs%
TargetRtotal
50%
809
SWA plot
20.5
K1
145
2.1
K2
484
Set
TargetRs%
50%
Set
TargetRs%
0%
Set_Qav
TargetRtotal
925
Set_Qav
0.48
TargetRtotal
1131
plot
SWA plot
SWA plot
1000
1200
1000
800
800
800
800
600
600
Vel*Z
200
Wave
up
Wave
down
800
600
800
Force
600
Force
400
-200
200
Wave
up
Wave
down
400
-600
Wave
1
down
t/(L/c)
-200
$#
)#
*+,-.#/+01#23##
$#
*+,-.#/+01#45#
678#91019#!#:#
;<+=#>1?1.@A3#45#!#B#
C32#D1+>-3E#>1?1.@A3#
45#!#B#
F+01#61+G#23#!#B#
'$()*+,%
"#
%#
&#
'#
(#
*+,-.#/+01#23##
)#
inFric,on
2
W
K1
wh
K2
Set
Set_Qav
!"#$%&$'$()*+,%
W
K1
wh
K2
Set
Set_Qav
5
145
2.1
484
0
0
2
145
2.1
484
$#
0
0
TargetRs%
TargetRtotal
TargetRs%
TargetRtotal
;<+=#>1?1.@A3#45#!#B#
C32#D1+>-3E#>1?1.@A3#
45#!#B#
F+01#61+G#23#!#B#
"#
%#
2
145
2.1
484
Set
Set_Qav
SWA plot
SWA plot
W
K1
wh
K2
0
0.17
Set
Set_Qav
Wave
down
200
100
300
100
Wave
up
Wave
up
Wave
down
Wave
down
-100
'#
)#
(# *+,-.#/+01#23## )#
*+,-.#/+01#45#
678#91019#!#:#
;<+=#>1?1.@A3#45#!#B#
C32#D1+>-3E#>1?1.@A3#
45#!#B#
F+01#61+G#23#!#B#
$'$()*+,%
*+,-.#/+01#23##
*+,-.#/+01#45#
678#91019#!#:#
;<+=#>1?1.@A3#45#!#B#
C32#D1+>-3E#>1?1.@A3#
45#!#B#
F+01#61+G#23#!#B#
$#
$#
$#
$#
"#
"#
%#
%#
t/(L/c)
t/(L/c)
&#
&#
'#
'#
(#
(#
*+,-.#/+01#23##
)#
)#
*+,-.#/+01#45#
-100
10
145
2.1
484
Set
Set_Qav
0
0
200
100
-100
*+,-.#/+01#45#
C32#D1+>-3E#>1?1.@A3#
C32#D1+>-3E#>1?1.@A3#
45#!#B#
F+01#61+G#23#!#B# 45#!#B#
F+01#61+G#23#!#B#
1 -400
"#
%#
$#
TargetRs%
TargetRtotal
50%
696
200
200
Wave
up
100
Wave
down
-100
Wave
up
-600 4
20
t/(L/c)
"#
&#
%#
31
-400
42
53
t/(L/c)
'#
(#
&#
-600
)#
t/(L/c)
!"#
t/(L/c)
Set
Set_Qav
*+,-.#/+01#23##
$#
"#
W
K1
wh
K2
V
Z
20.5
145
2.1
484
3.7
182
Set
Set_Qav
0
0.03
%#
&#
'#
(#
$#
*+,-.#/+01#45#
;<+=#>1?1.@A3#45#!#B#
C32#D1+>-3E#>1?1.@A3#
45#!#B#
F+01#61+G#23#!#B#
!"#$%&$'$()*+,%
0
0.00
TargetRs%
TargetRtotal
50%
597
Qs
Qt
2
40
Js
Jt
2
2
!"#
!"#$%&$'$()*+,%
50%SkinFric,on
Qt=50,Rtotal=503
TargetRs%
TargetRtotal
50%
503
Qs
Qt
SWA plot
1000
800
'#
!"#
*+,-.#/+01#45#
64
(#
678#91019#!#:#
)#
10
145
2.1
484
Set
Set_Qav
0
0
Wave
down
TargetRs%
TargetRtotal
"#
%#
&#
C32#D1+>-3E#>1?1.@A3#
45#!#B#
F+01#61+G#23#!#B# '#
Force
Vel*Z
200
Wave
up
Wave
down
-200
Wave
up
Wave
down
400
200
-200
-200
-400
-400
-600
0
(#
t/(L/c)
)#
*+,-.#/+01#23##
*+,-.#/+01#45#
;<+=#>1?1.@A3#45#!#B#
!"#$%&$'$()*+,%
t/(L/c)
200
-600
678#91019#!#:#
0%
965!"#
400
Vel*Z
;<+=#>1?1.@A3#45#!#B#
0%SkinFric,on
withW=10
Wave
up
Wave
down
*+,-.#/+01#45#
400
600
Force
678#91019#!#:#
C32#D1+>-3E#>1?1.@A3#
45#!#B#
F+01#61+G#23#!#B#
$#
W
K1
wh
K2
600
*+,-.#/+01#23##
-600
0
t/(L/c)
*+,-.#/+01#23##
-400
*+,-.#/+01#23##
Vel*Z
-300
)#
)#
W
20.5
50%SkinFric,on
Thin pile secEon subsEtuted.
K1
145
wh
2.1
Qt=40,Rtotal=597
V rises above F and ve Wup.
K2
484
V
3.7
Z
182
F peak amer 2L/c is very low, => poor drivability
2
2
Wave
up
Wave
down
-400
(#
PDA Wdn
1
Vel*Z
Wave
up
Wave
down
200
-200
'#
t/(L/c)
SWA plot
Wave
down
&#
Force
600
Vel*Z
%#
678#91019#!#:#
Js
Jt
Vel*Z
500
300
"#
800
800
Vel*Z
PDA Wup
Force
600
Wave
up
2
30
Force
400
$#
700
Force
-600
5
PDA Force
1000
SWA plot
400
$#
-200
SWA plot
Force
$#
Qs
Qt
$#
50%
804
PDA Vel*Z
*+,-.#/+01#23##
50%SkinFric,on
Qt=30,Rtotal=696
;<+=#>1?1.@A3#45#!#B#
!"#
Force
-400
678#91019#!#:#
;<+=#>1?1.@A3#45#!#B#
;<+=#>1?1.@A3#45#!#B#
50%SkinFric,on
withW=10
TargetRs%
TargetRtotal
$#
$#
t/(L/c)
C32#D1+>-3E#>1?1.@A3# 0
45#!#B#
F+01#61+G#23#!#B#
Vel*Z
400
-200
*+,-.#/+01#23##
SWA plot
600
678#91019#!#:#
!"#
W
K1
wh
K2
-600
1000
600
-400
1
!"#
)#
-400
Wave down
200
800
-200
-200
800
-300
0
0.01
SWA plot
-300
-300
Wave up
SWA plot
0%
800
1000
100
-200
-300
4
Set
300
Vel*Z
-200
(#
!"#$%&$'$()*+,%
20.5
678#91019#!#:#
145
;<+=#>1?1.@A3#45#!#B#
2.1
C32#D1+>-3E#>1?1.@A3#
484
45#!#B#
3.7
F+01#61+G#23#!#B#
182
Set_Qav
0%SkinFric,on
withW=5
TargetRs%
TargetRtotal
Force
Vel*Z
-200
0%
421
400
-100
L/c)
2.1
TargetRs%484
TargetRtotal
0
0.47
500
200
Force
Wave
down
Wave
up
Wave
up
W
K1
wh
K2
V
Z
SWA plot
200
200
-400
*+,-.#/+01#23##
)#
400
300
Vel*Z
'#
(#
0%SkinFric,on
5
withW=2
145
600
Force
Vel*Z
'#
!"#$%&$'$()*+,%
W
K1
wh
K2
300
400
400
*+,-.#/+01#45#
600
Force
&#
!"#
50%
427
50%
658
400
No skin fric7on
F and V curves close together before t = 2L/c
678#91019#!#:#
50%SkinFric,on
withW=2
50%SkinFric,on
withW=5
A plot
500
&#
!"#
$#
*+,-.#/+01#45#
!"#
t/(L/c)
$#
Force
;<+=#>1?1.@A3#45#!#B#
0
(#
Wave
down
Piletop force
678#91019#!#:#
t/(L/c)
'#
%#
-800
0
SWA plot
0
-400
*+,-.#/+01#45# -200
"#
$#
-600
-600
200
Wave
up
Vel*Z
Wave
up
200
-400
-400
Qs
Qt
400
Vel*Z
-200
L/c)
Vel*Z
Force
-400
Wave
down
Vel*Z
Force & Vel*Z (Tons)
Wave
up
400
50%
747
600
Force
400
-200
Force
600
Vel*Z
TargetRs%
TargetRtotal
SWA plot
1000
0%SkinFric,on
wh
0
0.01
50%SkinFric,on
2.1
484
Set
Set_Qav
145
2
2
Qt=25,Rtotal=747
SWA plot
20.5
K2
Js
Jt
1000
W
K1
wh
2
20
2.1
484
3.7
182
1000
inFric,on
Qs
Qt
wh
K2
V
Z
C32#D1+>-3E#>1?1.@A3#
45#!#B#
F+01#61+G#23#!#B#
$#
"#
%#
&#
'#
$#
!"#
t/(L/c)
!"#$%&$'$()*+,%
(#
)#
*+,-.#/+01#23##
*+,-.#/+01#45#
$#
"#
%#
&#
'#
$#
678#91019#!#:#
;<+=#>1?1.@A3#45#!#B#
C32#D1+>-3E#>1?1.@A3#
45#!#B#
F+01#61+G#23#!#B#
!"#
!"#$%&$'$()*+,%
(#
Part 3
TIP REFLECTION
400
200
t = 0
Force
Wdn
Wup
~ End Bearing / 2
600
800
Velocity
t = 2 L/c
t = 3 L/c
PDA records the actual instantaneous soil resistance during driving. This may be very dierent from the soil resistance
available long-term to support the design pile load.
To measure soil setup, the usual procedure is to perform a second PDA test a few days amer the pile is driven. The soil
resistance is measured for the rst few blows before the soil has Eme to somen. This re-strike test when compared with
the results of the original PDA test will show the change of soil resistance over Eme.
It is important to do a re-strike test on a number of selected piles to determine the eect of soil setup.
WAVE MATCHING
Wave matching is using the computer model to recreate the PDA test results.
The hammer, pile and soil parameters are varied in the computer model such that the computer generated stress
wave match the actual PDA measurements.
Computer Modeling
Computer programs for pile driving analysis using the wave equaEon have been around since the 1960s. The pioneering work
was done by E.A.L. Smith[1][2], who proposed modeling the pile as a series of lumped masses and springs. Smith also
formulated computer model representaEons for the pile hammer and cushion, as well as the soil model taking into account
the soil quake Q and damping parameter J.
The Method of CharacterisEcs (MOC) was developed by Van Hamme et al[7] in the 1970s as an alternaEve method of
calculaEng the wave propagaEon down the pile. This method represents the pile as a transmifng conduit with impedance Z
and tracks what happens to the stress wave when it encounters soil resistance or changes in pile impedance.
In 2004, ALC developed a spreadsheet implementaEon of the Smith model. The spreadsheet interface was considered to be
the most expediEous for development and visualizaEon of the results. Experiments in full wave matching led to a hybrid
model using lumped mass springs to represent the hammer and the Method of CharacterisEcs to track the stress wave in the
pile.
As a result, WEAP, PDA analysis and CAPWAP can all be done by a single spreadsheet program. The use of the spreadsheet
makes wave matching, plofng and comparison of results and changes to the computer model easy and convenient.
However, the Smith method has the disadvantage that the force is
calculated at the spring locaEon, while the velocity is at the mass
locaEon, which is not the same point. (half a segment length
dierence)
Since PDA signals are taken at the same point in the pile, there is an
inherent inaccuracy in the Smith simulaEon when used for wave
matching.
The stress wave is separated into a downward and an upward travelling component.
Each component is propagated independently of the other. Tracking the up and down
waves is the basis of the method of characteris7cs. The procedure inherently shows
directly how the wave is being transmiked and reected and helps in understanding
what factors aect the pile driving system.
The hammer and pile are divided into segments, each with uniform cross secEon, and
the soil resistance is assumed to act as concentrated loads in between segments. The
upward and downward components of the stress wave are assumed to be transmiked
unchanged through the individual segment and is modied by the soil resistance /
change in pile secEon at the juncEon between segments.
This method has the advantage that the force and velocity are calculated at the same
point in the pile, and is thus beker suited for wave matching. It also gives twice the
number of data points along the pile compared to the Smith model for the same Eme
interval.
Screenshots
Fdn, Fup
HAMMER MODEL
FORCE
VELOCITY
PDA DATA
Start from lem to right. Try to match all the peaks of all the curves, but most importantly, try to match Wup as closely as possible, as this
gives the pile capacity. Past 3L/c, the matching becomes less and less important to achieve.
1. Match rst impulse wave peak using pile hammer and pile Ep parameters.
- Pile cushion sEness K1 determines the width of the rst pulse.
- Hammer eciency and stroke have the same eect of raising the rst peak.
- Weight of hammer W1 Elts the right side of all plots upwards.
- Weight of helmet W2 moves the posiEon of the second peak to the right.
- Top of pile spring K2 determines the height of the second peak.
- The ram length and number of ram segments aects the detail shape of the curve.
- The coe of resEtuEon for the cushions e1 and e2 are usually best set between 0.9 to 1.0.
- Can try dierent values of e1, e2 as well as changing the cushion damping for beker match.
2. MOST IMPORTANT. Match the Wup curve at t < 2L/c using individual sham resistances to match the PDA Wup plot.
- Start from the top soil segment and work towards the pile toe.
3. Match the plateau of Wup using end bearing, keeping skin fricEon constant.
4. Match the peaks near 3L/c using Jsham to suppress the peaks. May need to readjust the sham resistance and sham Q.
- Sham J decreases sham resistance and Ep J decreases Ep resistance. Not easy to disEnguish between the two.
5. Match the Wup peak amer 2L/c using JEp. May need to readjust end bearing and QEp every Eme you adjust JEp.
- Q side moves the sham porEon of the curves to the right. QEp moves the Wup curve near 2L/c to the right.
6. PILE SET must be > 0. Can try matching pile set using total soil resistance. This is not always accurate.
7. Soil Unloading R and Q have a small eect on the right side of the chart > 3L/c. Mostly for cosmeEcs.
Remember the only parameters that maker are the SHAFT RESISTANCE and the TOTAL RESISTANCE. All other parameters are of likle
use except for matching the plots in order to get the soil resistances.
SWA plot
800
Piletop force
Force
600
Vel*Z
400
Wave up
Wave down
200
PDA Force
0
PDA Vel*Z
PDA Wup
-200
PDA Wdn
-400
0
t/(L/c)
Conclusions
PDA is like an x-ray insight into the innards of the pile. The shape of the PDA curves shows a great deal of informaEon on
the status of the pile driving, and if engineers learn how to read it, it contains much useful informaEon which is not
available by other means.
1. The rst part of the PDA wave trace reveals the driving characterisEcs of the hammer used.
2. The second part of the PDA trace shows the skin fricEon encountered by the pile as well as the condiEon and
integrity of the pile itself.
3. The third porEon of the trace shows the pile end bearing capacity and soil reacEon characterisEcs at the pile Ep.
PDA is done in real-Eme during driving, and thus allows immediate remedial acEon to be taken to recEfy inadequate pile
driving or pile damage during driving.
Boreholes only show the verEcal soil prole at one point locaEon, and the number of boreholes is limited. StaEc load tests
are extremely cumbersome and costly, and can only be done in very limited numbers. The low cost of the PDA test allows
widespread monitoring on a large number of piles, and shows much more than a convenEonal load test. In parEcular, for
raked piles PDA shows the eect of the pile rake, which may be dierent depending on rake direcEon in certain soil
condiEons.
Ideally, we should do a PDA test for every pile, and this has been done on some of our marine projects, where piles are
expensive to install and very dicult to do staEc load tests on. With modern electronics, the test data can be sent directly
from the sensors over the interne. Modern somware like iCAP can do wave matching automaEcally and immediately, giving
the assurance of the structural and geotechnical capacity of each pile.
THANK YOU
References
Smith Method
1.
Smith, E.A.L., (1955), Impact and Longitudinal Wave Transmission, ASME, Transac=ons of the ASME.
2.
Smith, E.A.L., (1960), Pile Driving Analysis by the Wave EquaEon, ASCE, Journal of the Soil Mechanics and
Founda=ons Division, 86.
3.
Hussein, M., and Likins, G., (1995), Dynamic TesEng of Pile FoundaEons During ConstrucEon, ASCE Structural
Division, Proceeding of Structures Congress XIII.
Vo
W
Increasing Vo
Increasing Vo makes the stress pulse
higher but retains the same shape
Increasing W
V0 = 2gH
Ref:- Accuracy in Numerical Analysis for Pile Driving Dynamics Deeks & Randolf, 1992
Vo
EFFECT OF K
1.2
0.9
0.8
Vo
Drop%height%aects%only%the%magnitude,%
but%not%the%shape%of%the%pulse%
0.6
H = 0.6
FORCE
0.5
0.4
1.2
0.3
1.5
0.2
Pulse&width&varies&with&so0er&K&
0.6
Decreasing K1
K=1
2
0.4
5
10
Increasing Vo
0.1
0.2
0.5
0.2
0
0
10
TIME
20
15
25
30
35
0.1
40
10
15
20
TIME
25
30
35
40
EFFECT OF W1
1
0.9
0.8
0.7
FORCE
0.6
Keeping
Z=1,K=,
vary W
2
Heavy&hammer&*&force&follows&
exponen4al&decay&curve.&&
Light&hammer&*&short&force&pulse.&
0.5
0.4
5
10
0.3
0.2
0.5
Increasing W1
0.1
0.2
0
0
10
15
20
TIME
25
30
35
40
EFFECT OF Z
2
1.8
1.6
1
2
1.4
FORCE
1.2
10
1
0.8
0.6
0.5
0.2
0.1
Increasing Z
0.4
0.2
0
0
10
15
20
TIME
25
30
35
40
W
K
0.8
0.8
FORCE
0.7
REFERENCES
2. G. G. Goble and F. Rausche, WEAP86 program documentation in 4 Vols. Federal Highway Administration,Office of
Implementation, Washington DC, 1986.
3. F. Rausche, G. G. Goble and G. E. Likins, Recent WEAP developments, Proc. 3rd Int. Con$ on Application of
Sires- Wave Theory to Piles, Ottawa, 1988, pp. 164-173.
4. P. Middendorp and A. F. van Weele, Applicationof characteristicstress wave method in offshore practice, Proc. 3rd
Int. Con$ on Numerical Methods in Ofshore Piling, Nantes, Supplement, 1986, pp. 6-18.
5. M. F. Randolph, Analysisof the dynamics of pile driving, in P. K. Banerjee and R. Buttefield (eds.),Developments in
Soil Mechanics I V: Advanced Geotechnical Analyses, Elsevier Applied Science, also University of Western Australia
Research Report No. G1001, 1991.
299
Differentiatingequation (70), substituting into equation (71), and simplifying, the spring force can
be expressed as
Vo
where
kr
For most combinations of rn: and kr, p2 is greater than zero, and the solution presented above
applies. For a small range of combinations, p2 is less than or equal to zero, and a different
solution must be used. When this is the case, the cubic denominator has three real roots, and
equation (58) can be written in the following way:
298
A. J. DEEKS A N D M. F. RANDOLPH
0 + 271
a0
a,
+ a,s + u2s2 + s3
b 2 = ~3 + 2 J Q c o s - 3
Letting
b3 = 7+ ~JQcosa2
0 f 471
3
and
if p2 is greater than zero, the cubic denominator of equation (58) has one real root and two
imaginary roots. The inverse transform can be found readily if the following substitutions are
made:
0 = arccos
(6)
= a.
+
This allows the cubic denominator to be factorized, and equation (58) becomes
Llif =
which can be expanded to
Llif =
w2
+ b i ) [ ( s+ b2)2+ 0'1
[ '{
+ (b2 - b1)2 s + bl
a0
(S
((s
(S f
a0
(S
+b2
+sb2)2
+ w2 + + b2)' + w2
(S
300
Performing the inverse Laplace transform, the anvil velocity can be found.
af =
m2
a0
(b2 - bl)'
e-bif'
- e-bzf'
(coswt*
+ b,--b,sinut*)]
0
c2 = b2 - bl,
4 = arctan-,c20
The anvil velocity and the force on the pile head are then
f,' = rn:ii:
+ fif
kc
ma
Z
L g=
mr
a0
F, = ___
w2 cz'
with
Fs
= (b
a2 - b2 b3 - bz
As = ______ ,
a2 - bl b3 - bl
ao(a2 - b l )
- bl)(b3 - bl)
Bs=--
a2 - b3 bz - bl
a2 - b, b2 - b3
(86)
However, if two of the roots are the same, the solution is different again. When B is zero and b3 is
equal to b Z ,the solution becomes
f t
= ti: = F,e-clr*[l
f:
= FSe-lr*[l
- e-c2r(l + b,t*)]
- e-c2*(1+ bzAst*)]
PDA Nota7on
From PDA-W Manual of OperaEon[4]
1.14.6 Short Recommended List of Useful Result Quan::es
FMX MAX FORCE
VMX MAX VELOCITY
DMX MAX DISPLACEMENT
DFN FINAL DISPLACEMENT
EMX MAX ENERGY
ETR ENERGY TRANSFER RATIO (= EMX / RATING) BPM BLOWS PER MINUTE
STK STROKE - O.E.DIESELS ONLY
CSX MAX COMPR-STRESS
CSI MAX INDIV.C-STRESS
CSB MAX TOE C-STRESS
TSX MAX TENSION STRESS
TSN MAX T-STRESS; WU ONLY
BTA INTEGRITY FACTOR
LTD LENGTH TO DAMAGE
PILE CAPACITY METHODS: RA2, [ RMX, RSP ] - JC; RX4..RX5..
Sham Resistance: SFR - funcEon of JC, or SF4, SF5, etc.
End Bearing: EBR - funcEon of JC, or EB4, EB5, etc.