Citation
Field energy measurements of standard penetration testing

Material Information

Title:
Field energy measurements of standard penetration testing
Creator:
Farrar, Jeffrey Ayers
Publication Date:
Language:
English
Physical Description:
ix, 400 leaves : illustrations ; 29 cm

Subjects

Subjects / Keywords:
Drill pipe -- Testing ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references.
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Science, Department of Civil Engineering.
Statement of Responsibility:
Jeffrey Ayers Farrar.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
24526506 ( OCLC )
ocm24526506
Classification:
LD1190.E53 1991m .F37 ( lcc )

Full Text
FIELD ENERGY MEASUREMENTS OF STANDARD
PENETRATION TESTING
by
Jeffrey Ayers Farrar
B.S., Kansas State University, 1978
A Thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver in partial fulfillment
of the requirements for the degree of
Master of Science
Department of Civil Engineering
Geotechnical Engineering
1991
5 i'\ s


This thesis for the Master of Engineering
degree by
Jeffrey Ayers Farrar
has been approved for the
Civil Engineering Department
by
G. (T. Gobi e


ABSTRACT
Farrar, Jeffrey Ayers (M.S., Engineering)
Field Energy Measurements of Standard Penetration Testing
Thesis directed by Professor Nien-Yin Chang
Field energy measurements of the Standard Penetration Test (SPT) have
allowed for correction of variation in the mechanical aspects of the
test. SPT is used for a majority of explorations in geotechnical
engineering practice. From 1977 to 1985 a considerable amount of
data was compiled by researchers regarding stress wave energy
transmission in the drill rods during SPT. In this study, field
energy measurements were performed on 12 drill rigs with 16 hammer
systems using the Binary Instruments SPT calibrator. Of the hammer
systems tested, there were 4 hydraulic automatic and 4 mechanical
trip hammers along with conventional hammers operated by rope and
cathead method. Testing was performed in accordance with American
Society for Testing and Materials procedure D-4633-86. During
testing there was considerable difficulty in obtaining reliable
stress wave travel times in large N size drill rods. Some data were
questionable due to lack of knowledge of drill rod area. A critical
review of the existing test procedure indicates that correction for
wave speed is not required. The short wave travel times were caused
by reflections from drill rod joints combined with zero load shift of
the piezoelectric load cells used. Additional recommendations are
presented to improve the existing testing procedures. Results of


this study add significant data to our knowledge of energy
transmission in SPT. Measured average energy delivered by the
frequently used safety hammer was 76 percent. This energy is up to
16 percent higher than average of data compiled by other researchers.
This is attributed to possible short stress wave travel time data in
the earlier database and the use of different types of load cells.
Additional research is recommended on hammers of different
manufacture, drill rod joints, load cells, and kinetic energy
methods.
The form and content of this abstract are approved. I recommend its
Publication.
Signed


Contents
Figures...................................................iv
Tables....................................................vi
Acknowledgements.........................................vii
Symbols and abbreviations ............................. ix
CHAPTER
1. INTRODUCTION
Purpose and Scope .................................. 1
Description and History of SPT.......................2
Use of SPT in Geotechnical Engineering.............7
Variability in SPT Testing........................ 10
Thesis approach ................................... 16
2. THEORY OF ENERGY TRANSFER IN THE SPT................19
One Dimensional Wave Equation Theory................19
Hammer-Rod Impact Conditions........................22
Wave Transmission at Changes in Area................27
Energy Transmission Along the Rods..................28
3. DEVELOPMENT OF SPT ENERGY MEASUREMENT METHODS. ... 35
Possible Methods to Compute Energy or Efficiency 35
Development of Stress Wave Energy Measurements . 39
Application of the Binary Instruments Device ... 45
Standardization of Stress Energy Measurement ... 57


CONTENTS
CHAPTER Page
4. REVIEW OF EXISTING STRESS WAVE MEASUREMENT IN SPT . 61
Initial Study of Palacios ......................... 61
Summary Studies by the National Bureau
of Standards........................................63
Measurements of Automatic Hammers ................. 67
Energy Data from Other Countries....................73
Recommendations for Energy Corrections ............ 78
5. PRESENTATION OF FIELD ENERGY MEASUREMENT DATA ... 81
Summary of Testing Program ........................ 81
Format of Data Presentation ........................82
Initial Checks with Pilcon and Safety Hammer ... 89
Corps of Engineers Automatic and Trip Hammers . 91
Safety Hammer Measurements Jackson Lake .... 97
Automatic Hammer Measurements Jackson Lake . .100
Safety Hammer Measurements Gus Pech Drill Rig . 104
6. ANALYSIS OF FIELD DATA...............................105
Evaluation of Kc Correction Factor.................105
Evaluation of Short Wave Travel Times ............ 106
General Hammer Performance.........................115
Comparison of ER1- Data to Others..................117
7. CONCLUSIONS..........................................120
Summary of Major Findings ........................ 120
Recommendations for Test Method Improvement . 122
Recommendations for Future Research .............. 123
i i


CONTENTS
REFERENCES....................................................125
APPENDIX
A. Standard Test Method for Stress Wave Energy
Measurement for Dynamic Penetrometer Systems
- ASTM Method D-4633-86 ........................... 131
B. Example of Field Data Sheets.........................136
C. Field Data Sheets and Summary Graphs..................139
i i i


FIGURES
Figure Page
1.1. Equipment used to perform SPT..............................5
1.2. SPT safety hammer......................................... 5
1.3. SPT donut hammer............................................6
2.1. Mechanism of impact of a cylindrical hammer on a
finite cylindrical rod ................................... 25
2.2. Effect of impedance ratio on the theoretical wave
shape for the same hammer mass and drop...................26
2.3. Wave transmission and reflection due to and abrupt
area change................................................29
2.4. Illustration of drill rod wall types........................32
3.1. Schematic illustration of energy transfer in the SPT
test and configuration for stress wave testing ........... 36
3.2. Variation of Ei/E* with rod length for AW rod-safety
hammer combination ....................................... 43
3.3. Schematic showing key elements of the Binary
Instruments SPT energy measurement device ................ 46
3.4. Illustration of compressive wave return force-time
history and trigger duration time ........................ 54
3.5. Example of electronic shorting of load cells during
impact.....................................................55
4.1. Summary of variation of E^E* with rod length for
different rod-hammer combinations..........................62
4.2. Summary of ERf data for safety hammers......................68
4.3. Summary of ER{ data for donut hammers.......................69
4.4. The CME 140-lb Automatic SPT Hammer during performance
of the SPT.................................................71
4.5. Frequency Diagram of ER1- results Japan study.............76
iv


FIGURES
Figure Page
4.6. ER, vs number of rope turns around the cathead
- Japan study ........................................... 77
5.1. Schematic Diagram of the Pilcon Trip Hammer...........89
5.2. Frequency diagram for series 3 Pilcon trip hammer . . .90
5.3. Damco SPT automatic hammer .............................. 94
5.4. Frequency diagram, series 8, Damco #2.....................96
5.5. Frequency diagram, series 15, Damco #3 .............. 97
5.6. Frequency diagram, series 21, safety hammer, Jackson
Lake Dam..................................................99
5.7. Anvil for CME automatic hammer...........................101
5.8. Frequency diagram, series CME-6, CME automatic hammer . 103
6.1. Comparison of ER{ and travel time, Damco #2 hammer . . .109
6.2. Comparison of ERf and travel time, Vicksburg
automatic hammer........................................110
6.3. Comparison of ERi and travel time, Damco #3 hammer . . .111
6.4. Comparison of ERi and travel time, Soiltest hammer . . .112
6.5. Comparison of ERi and travel time, safety hammer NW
rods.....................................................113
6.6. Comparison of ERi and travel time, safety hammer AW
rods.....................................................113
6.7. Frequency diagram, Comparison of safety hammer data . . 118
v


TABLES
Table Page
1.1. Traditional N Value Su Correlation......................10
1.2. Traditional N Value Correlations for Sand.................10
1.3. SPT Equipment Variables...................................13
1.4. SPT Procedural/Operator Variables ....................... 13
1.5 Illustration of Variation of N at Differing Energies . .15
2.1. Summary of Typical SPT Drill Rod Information from
Manufacturers Catalogs .................................. 30
3.1. K2 Factors................................................41
3.2. K1 Factors................................................48
3.3. Tabulation of Calibrator Display Numbers for Drill
Rod Areas.................................................52
4.1. Tabulation of Average ERf.................................65
4.2. Summary of Energy Ratios for SPT Procedures...............80
5.1. Summary of SPT Energy Measurements........................86
5.2. Summary of Hammer Dimensions Corps of Engineers Study .93
6.1. Summary of Kc Correction Factors.........................107
6.2. ' Summary of Hammer Performance .......................... 116
vi


ACKNOWLEDGEMENTS
The author is deeply indebt to Dr. N.Y. Chang who as my thesis
director, helped me through the Masters program.
The author wishes to acknowledge all of those which assisted him with
data collection and interpretation. Don Stelma and Brad Buehler,
geologists of the Bureau of Reclamation, Boise, performed data
collection at Jackson Lake and Gus Pech drill rigs. Joe Gatz and
Mark Vispi, U.S. Army Corps of Engineers, Waterways Experiment
Station assisted with measurements at Waterways.
The author is indebted to Drs. William Kovacs, George Goble and John
Schmertmann who pioneered energy measurement methods. They provided
useful comment and review of data collected and provided me with the
opportunity to participate in this area of endeavor.
The Bureau of Reclamation provided support to cover field costs both
from research funding and project funding. Research funding was
provided under Project Engineering and Scientific Studies (PRESS)
program DF-20 "Drilling, Sampling and Insitu Testing Research". The
findings presented herein are those of the author and do not
necessarily reflect those of Reclamation.


SYMBOLS AND ABBREVIATIONS
a = cross sectional area of drill rod
ab = cross sectional area of rod or bar
ah = cross sectional area of hammer
A = cross sectional area
BPM = blows per minute
c = Velocity of strain wave propagation
ca = actual compression wave velocity measured
c5 = value of compression wave velocity assumed during integration
E = Young modulus
Ecal = Rod size correction factor for Binary Instruments device
Efp =.Default rod size correction factor, Ecal, for matched load cell
used with Binary Instruments device
E- = incident stress wave energy (strain energy) in drill rod from
first compressive wave that is produced by hammer impact
E* = Nominal kinetic energy of SPT from 140 lbm hammer dropped freely
from 30-inches, 4200 in-lbf.
ERj = E/E* = drill rod stress wave energy ratio
ERjc = Reference drill rod stress wave energy ratio
ERim = Measured drill rod stress wave energy ratio
F = Normal force in rod
Fcal = Force calibration factor for Binary Instruments device
K = kinetic energy
K1 = Correction factor to account for location of load cell distance
a1 from the impact surface
K2 = Correction factor to account for early termination of energy
delivery to drill rod due to reflected tensile wave in short
rod length


SYMBOLS AND ABBREVIATIONS
1 = length of drill rods
a1 = Distance from impact surface to load cell
1' = Distance from the load cell to bottom of rods
Mh = Mass of hammer
N = SPT blow count value, number of blows
Nc = Corrected blow count for selected energy ratio, ERic
Nm = Measured raw N value
a = Normal stress in the rod due to wave travel
o.,ar,ot = Normal stresses in the rod due to the incident, reflected,
and transmitted waves
p = Mass density of the material
r = Characteristic impedance of hammer-rod system = ab/ah
SPT = abbreviation for Standard Penetration Test ASTM D-1586
t = time
U = Strain energy
u = Displacement of a bar from its undisturbed condition
vbo = Spatial particle velocity of rod/bar after initial impact pulse
transfer
vho = Spatial particle velocity of hammer after initial impact pulse
transfer
Vhj = Initial impact velocity of hammer
z = Characteristic impedance ratio E*a/c
IX


CHAPTER 1
INTRODUCTION
Purpose and Scope
The purpose of this study is to present the results of field
energy measurements of standard penetration testing (SPT) obtained
from 1986 to 1987 by the author while employed by the Bureau of
Reclamation. A Binary Instruments Model 102 SPT calibrator, on loan
from the National Bureau of Standards, was used to perform
measurements of SPT drill rod energy. Drill rod energy data was
collected for both Reclamation and the Corps of Engineers to use in
important SPT studies of sites subject to potential earthquake
induced ground liquefaction. Studies of SPT testing performed from
1975 to 1985 showed the importance of energy efficiency in SPT
testing and they culminated in development of the Binary Instruments
Device. Based on data collected during the 1975 to 1985 time period,
design recommendations were made for correcting SPT data for energy
effects and a test procedure was developed by the American Society
for Testing and Materials (ASTM). Since 1985 there has been little
additional data collected with the Binary Instruments Device.
During data collection with the Binary Instruments device, some
important difficulties were encountered. When testing larger size
drill rods, premature tensile wave cutoff time was experienced. The
use of new piezoelectric load cells resulted in different energies


for typical hammers systems than those collected from strain gauge
load cells. Data with the new load cells have resulted in higher
average energies for groups of hammer systems than those recommended
for design and analysis procedures issued in 1985.
Energy measurements were performed on 12 drill rigs and 16
hammer systems. Of the hammer systems tested, there were 4 hydraulic
automatic hammers and 4 mechanical trip hammers along with
conventional safety hammers operated by rope and cathead methods.
The publication of this data will make an important contribution to
our knowledge of penetration testing.
Description and History of SPT
In 1902 Col. Charles Gow developed a method of securing a soil
sample in an open ended one inch pipe at the bottom of a drill hole
by hammering the pipe into the soil. This method of sampling was an
important development in civil engineering as it allowed a more
economical method of securing a relatively undisturbed soil sample
for classification purposes. Previous to this time period, wash
borings and crude penetration tests were used to classify the ground
characteristics. The only method of observing undisturbed soil
structure was via expensive and dangerous hand excavated test pits.
In 1922, the Charles R. Gow Company merged with the Raymond Concrete
Pile Company. L. Hort and G.A. Fletcher devised a 2-inch diameter
split barrel sampler and changed the drop weight mass from 110 to 140
2


lb. while working at Raymond. Early references to its inception are
references by Mohr (1943) and Hvorslev (1949). With the inclusion of
the SPT extensively in the classic text Soil Mechanics in Engineering
Practice by Terzaghi and Peck (1967) the test found widespread use
among the world. In the U. S it is estimated that up to 80 to 90
percent of routine foundation designs are accomplished using the SPT
( Kovacs, 1981).
The test evolved into the SPT with modifications over 50 years.
An interesting review of the evolution of SPT testing in the United
States was presented by Fletcher (1965). SPT practice surveys in the
U.S. were conducted by Ireland et al, (1961) for and American Society
of Civil Engineers task committee report on static and dynamic
penetration test methods and by Kovacs, (1981). In April 1958,
standardization was initiated by ASTM when a tentative method was
published. The ASTM test method D-1586 was finally adopted as a
standard method in 1967
The method for performing the test now consists of driving a
"standard sampler" into the soil by means of dropping a 140 lbm drive
weight dropped from 30-inches onto the drill rods Figure 1.1 shows a
typical equipment arrangement used for SPT testing The test is
performed by dropping a 140 lbm hammer from a height of 30-inches
Many types of hammers are used for performing the SPT The two most
common hammers in use in the U. S. are the safety and.donut hammers.
Figure 1.2 illustrates the safety hammer which consists of a long
weight which encloses an internal impact anvil for obvious safety
3


advantages. The donut hammer, shown on Figure 1.3 is a short wide
weight centered on a guide pipe. The donut hammer impacts an
external anvil attached to the drill rods. The hammers are raised
with a rope wrapped around a rotating pulley (cathead). After the
hammer is raised 30-inches, the rope is thrown smartly toward the
cathead to release friction and allow impact. The configuration and
size of the anvil is important to energy transfer. Use of a
cushioning block and larger anvil sizes generally result in lower
energy transfer. In 1980, automatic hammers consisting of
hydraulically powered chain lift devices were introduced in the U.S.
(Riggs, 1983). The automatic hammer systems are beginning to find
widespread use.
There are a wide variation of hammers in use around the world.
Some counties such as the United Kingdom and China (Shi-Ming, 1982)
use donut hammers with a mechanical tripping mechanism which
activates at the top of the stroke allowing hammer free fall. The
Pilcon trip hammer is used extensively in the United Kingdom. Most
all countries in the world conform to the hammer weight and drop
height standards. Excellent practice surveys of the use of SPT in
different countries are presented in the First European Symposium on
Penetration Testing, (1974) and by Thornburn, et al, (1985).
To perform the SPT, the sampler is "seated" with hammer blows 6-
inches into the soil at the base of the borehole and then the number
of hammer blows required to advance the sampler through 1 foot of
penetration is counted. The number of hammer blows, N, can be
4


Figure 1.1. Equipment used to perform SPT (Kovacs et al.,1983}
Figure 1.2. SPT safety hammer (Kovacs et al.,1983)
5


3.375 in.
(85.7 mm)
Figure 1.3. SPT donut hammer (Kovacs et al,1983)
6


correlated to engineering parameters. The SPT sampling tube has been
standardized as a 2-inch outside diameter and 1-3/8-inch inside
diameter barrel which typically ranges in length from 21 to 27-
inches. The sampler is equipped with a ball check device to improve
sample recovery. The sample barrel is split for easy inspection of
the soil sample. After the sample is recovered the boring is
advanced by cleaning to the next test depth and the test is repeated.
Use of SPT in Geotechnical Engineering
The success of the SPT test can be attributed to the combination
of penetration testing and the recovery of a soil sample. The
penetration data has been correlated to numerous engineering
parameters and localized correlations by practicing geotechnical
engineers allow refinement of these correlations. The amount of
expensive laboratory tests can be reduced when these correlations are
developed. The recovery of the soil sample allows a positive
confirmation of subsurface conditions.
For this report it is only necessary to understand the range of
penetration resistance values for a range of soils where the data is
useful and how important is accuracy of the data for estimating
engineering behavior of the foundation investigated. A review of
ranges of N values and engineering correlations for sands and clays
will be presented.
7


The SPT can be used to estimate the undrained strength of clays
as shown on Table 1.1. These correlations should be used with
caution because the primary component of penetration is side friction
governed by the remolded strength. In clays of only minimal
sensitivity the undrained strength from the correlation will be lower
than actual. Therefore the correlation is of little value except in
relatively stiff insensitive clays. The range of N values for clays
from the Table is 0 to >30. Most frequent N values for normally
consolidated clays at depths less than 100 ft would range from 0 to
15 Blows.
In cohesionless soils, the SPT has been correlated to the
relative density and friction angle of the soil These correlations
were developed with large scale laboratory chamber tests by several
researchers (Gibbs and Holtz, 1957, Schultze and Menzenbach, 1961,
and Bieganousky and Marcuson, 1977). Table 1.2 summarizes the range
of N value and relative density in sands. Of primary use to
geotechnical engineers are design charts developed by Peck, Hanson
and Thornburn,(1974), for settlement of footings on sands. This
empirical series of design charts provides footing design governed by
1-inch of settlement. The charts have a built in safety factor due
to their empirical nature.
An important application for SPT testing in sands is the
evaluation of earthquake liquefaction resistance evaluations. The
correlation of ground liquefaction occurrence and liquefaction began
in the 1960's in Japan and additional data was collected (Tatsuoka,
8


et al, 1980, Seed, 1979, and Seed et al, 1983). Attempts to obtain
undisturbed samples for dynamic testing of sands were unsuccessful
and the SPT remains the primary method for evaluating liquefaction.
These evaluations require a high level of accuracy since the
potential damages from ground liquefaction are costly and may include
loss of human lives. The liquefaction studies were a driving force
in collecting and gathering energy measurements in the 1980's.
The range of useful N values for sands are 0 to 50 blows/ft.
For most normally consolidated sands at depths of less than 100 ft.
values ranging from 10 to 30 blows/ft. are anticipated.
1.3 Variability in SPT
The SPT has been subject to much debate regarding the
variability of the test throughout its development. Even with the
advent of a standard test method in ASTM, variability is still a
problem. Although the procedure specifies hammer weight and drop
height many different hammer systems and drop methods are employed in
practice (Ireland, 1966). Numerous discussions regarding the
difficulty in application of the SPT test for interpretation of
engineering parameters can be found in the Fourth Pan-American
Conference on Soil Mechanics and Foundation Engineering (deMello,
Adam, and DeGodoy, 1971).
The sources of variability in the SPT test can be grouped into
two classes; equipment effects and procedural/operator effects. An
9


Table 1.1. Traditional N Value Su Correlation
(Terzaghi and Peck, 1967)
Approximate Undrained Shear S trength, s u
N Value Consistency kN/m^ tsf
0 2 very soft < 12 < 1/8
2 4 soft 12 25 1/8 1/4
00 1 medium 25 50 1/4 1/2
8 15 stiff 50 100 1/2 1
15 30 very stiff 100 200 1 2
> 30 hard > 200 > 2
Source: Terzaghi and Peck (1), p. 347.
Table 1.2. Traditional N Value Correlations For Sand
(Terzaghi and Peck, 1967)
Approximate Angle of
N Value Relative Density Friction, * (degrees)
0 4 very loose < 28
4 10 loose 28 - 30
10 30 medium 30 - 36
30 50 dense 36 - 41
> 50 very dense > 41
Source: Terzaghi and Peck (1), p. 310.
10


excellent detailed review of all factors affecting SPT variability
can be found in a recent EPRI report by Orchant, Kulhawy and
Trautman, (1988). In 1965, one of the developers of the SPT,
Fletcher described the abuses of this test. He Stated "The use of
improper methods and equipment subjects the SPT to abuses that can be
corrected only by better specifications for borings and competent
field control." Fletcher identified thirteen important factors which
affect SPT results.
1. Inadequate cleaning of the borehole
2. Failure to maintain sufficient hydrostatic head in the
borehole
3. Variations from the exact 30-inch drop of the fall weight
4. Use of drill rods heavier than one inch extra heavy pipe
or A rods
5. Extreme length of drill rods (over 175 ft.)
6. Interference of free fall of drive weight from any cause
7. Use of 140 lb weight without hardwood cushion, block, or
guide rod
8. Use of sliding weight that can strike the drive cap
eccentrically
9. Use of deformed tip on the sampler spoon
10. Excessive driving of sample spoon before blow count
11. Failure to seat sample spoon on undisturbed material
12. Driving of sample spoon above bottom of casing
13. Carelessness in counting blows and measuring penetration
With the identification of these factors by Fletcher, a framework for
future study and research was laid out for others to understand the
11


SPT variability. Over the past 20 years much additional information
was gained. The SPT has today been investigated more extensively in
terms of variability than any other insitu test.
All of equipment used in performing the SPT can cause
variability in the N value. Equipment such as the hammer, hammer
drop system, drill rods, and sampler can all cause variations to
differing extent. Table 1.3 summarizes the relative importance of
equipment variables.
Prior to standardization of the SPT many different
configurations of samplers were used for testing. In 1948, Burmister
recognized the effects of different samplers and recommended
corrections for different configurations. Even after U. S.
standardization, different samplers were used in some countries
(deMello, 1971). In the U.S. barrel liners were first used in early
SPT testing since in some cases lab tests were made on the recovered
soil. With time the liners were omitted and today most all U.S.
practice uses a sampler barrel with no liner and resulting larger
interior diameter of 1-1/2-inch (Kovacs, 1981). Meanwhile other
countries such have Japan have adhered to the original ASTM inside
diameter requirement of 1-3/8-inch inside diameter. Since the
component of soil side friction to penetration is estimated to range
from 50 percent in sands and up to 80 percent in clays, differences
in inside barrel diameter can cause differences of 10 to 30 percent
in N value (Schmertman, 1979, Kovacs and Salamone, 1984). The
12


Table 1.3. SPT Equipment Variables
(Orchant, J. C., Kulhawy, F. H., and C. H. Trautmann, 1988)
Variable Relative Effect or. Results
Non-standard, sampler moderate
Deformed or damaged sampler moderate
Bod diameter/weight minor
Bod length minor
Deformed drill rods minor
Type of hammer moderate to significant
Hammer drop system significant
Hammer weig.it minor
Size of anvil moderate to significant
Type of drill rig minor
Table 1.3. SPT Procedure / Operator Variables
(Orchant, J. C., Kulhawy, F. H. , and C. H. Trautmann, 1988)
Variable Relative Effect on SPT Results
Size of borehole moderate
Method of maintaining hole minor to significant
Cleaning of borehole moderate to significant
Failure to maintain sufficient hydrostatic head moderate to significant
Seating of sampler moderate to significant
Hammer drop method moderate to significant
Error in counting blows minor
13


current ASTM procedure allows for both samplers with and without
inner liners.
The influence in rod diameter in penetration testing has been
studied by many researchers. The existing ASTM procedure allows the
use of A through N rods. Some studies initially performed in
laboratory controlled conditions indicated that there was not a
significant difference in A through N rods (Gibbs and Holtz, 1957).
Many field studies performed in actual field conditions indicate no
significant effect of drill rod diameter (Fletcher, 1965, Clark,
1969, DeGodoy, 1971 and Brown, 1977). Wave equation studies provide
useful insight into the possible effect of drill rod size. These
studies which will be reviewed later indicate only slight effects due
to drill rod size.
Important procedural/operator variables are size of borehole,
method for advancing and cleaning the borehole, maintenance of
hydrostatic pressures, seating of the sampler, counting of blows and
hammer drop method. A summary of the significance of these variables
is summarized in Table 1.4. Even though the ASTM procedure has some
guidance on procedures there are still many possibilities for human
errors. The procedural/operator variables are by very significant
and can only be overcome with careful attention to details.
The primary difficulty with advancement of the borehole is to
assure undisturbed soil is tested. Early problems were identified by
Parsons (1966) in borings under the water table where SPT borings
using water or drill mud differed by a factor of 2.5. ASTM
14


procedures now precaution to maintain hydrostatic pressures to
overcome these problems. However, ASTM does allow any drilling
method which can assure the soil is undisturbed. Some drilling
methods are just not suitable for specific formations to be tested.
A good example of poor drilling techniques causing variation in SPT
is presented by Sanglerat and Sanglerat (1982). Many procedural
errors cannot be adequately quantified since the drilling method
combines with geologic variability. The most reliable method for
reducing procedural error is through adequate training of drill staff
and field inspection.
The most significant equipment variable in SPT is the hammer and
drop system employed. Some hammers operated with the cathead rope
method are subject to procedural variability depending on the
performance of the operator. Considerable progress has been made in
understanding effects of hammer system variability. The studies of
the hammer systems will be reviewed in detail later in this paper.
The most significant advancement was made by Palacios in 1977 when a
system was developed to measure the stress wave energy in the drill
rods immediately below the hammer. By measuring at the drill rod,
energy losses in the drop system and the hammer-anvil contact could
be measured. Kovacs and others continued measurements of hammer
systems and found that the energy delivered in SPT varied from 40 to
90 percent. In these studies it was found that the SPT N value was
inversely proportional to the energy delivered. If an average energy
of 60 percent of free fall energy is assumed, the following Table
15


illustrates a possible range in N values that could be obtained with
different hammer systems;
Table 1.5 Illustration of Variation of N at Differing Energies
N Value at 60 % Energy N Value at 40 % Energy N Value at 90 % Energy
5 8 3
10 15 7
20 30 13
30 45 20
40 60 26
As illustrated on Table 1.5 the effect of the hammer system is
very significant. The possible differences in N value are more
pronounced for larger values. Therefore, the variation is more
significant for sands than for clays of low blow count. The question
as to whether the N value is truly inversely proportional to energy
will also be addressed in this report.
Thesis Approach
From the preceding introduction the importance of energy
measurements in SPT are clearly evident. The purpose of this report
is to add important additional data regarding measurements of new and
existing hammer systems. Hopefully this data can be of benefit for
16


future SPT studies requiring a greater degree of accuracy and will
help to advance methods of SPT measurements.
To gain insight into how energy measurements are obtained and
how the reliability of the data can be evaluated, it will be
necessary to first review theory of energy transfer in SPT.
Principles of one dimensional wave equation applied to SPT will be
reviewed. This review will follow the precedent setting study by
Palacios. Methods of measuring and computing efficiency and energy
content will be reviewed. A uniform presentation of energy
distribution in SPT will be presented.
In order to compare data in this report with others it will be
necessary to review the data presented by others. Early measurements
began with Hammer velocity determinations. Development and operation
of the Binary Instruments device will be reviewed. Stress wave
energy data presented by Kovacs and others will be summarized.
Unfortunately a uniform test procedure for stress wave energy
measurements was not developed until 1985. This procedure will be
presented since it was followed in this study. Important
considerations in obtaining stress wave energy will be reviewed.
Field data obtained in this study will be presented along with a
discussion of the purposes of testing at each site and difficulties
encountered during testing. Test procedures will be presented and
raw data will be included in the appendix.
Analysis of the data will include statistical analyses and
comparison of results with previous data collected by others.
17


Recommendations for further study an revisions to existing testing
will be given.
18


CHAPTER 2
THEORY OF ENERGY TRANSFER IN THE SPT
One Dimensional Wave Equation Theory
The complex problem of impact of a hammer onto drill rods
requires some simplifying assumptions regarding the nature of energy
transfer and propagation in the system. Attempts to model the
problem using Newtonian coefficient of restitution are not
productive. The one dimensional wave equation theory is used to
explain the propagation of energy along a double ended rod. It is
assumed that the energy is delivered to the rods in the form of
longitudinal vibrations. It is further assumed that the plane cross
sections of the rod remain plane during passage of the strain pulse
and that the stress over the section is uniform. The one-dimensional
wave equation relates the displacement of the rod material with time
as follows;
2 d u 1 > ro 2 d u
2 d t 2 d x
(2.1)
Where;
u is the displacement of the bar cross-section a distance x
along the bar from its undisturbed position
t is the time of the pulse
and,
19


c is the velocity of the strain wave propagation, which is
found to be related to the modulus of elasticity of the
bar material, E, and density p in the following way;
c = (E/p)m (2.2)
Detailed derivations and discussion of the general solution have
been presented by Palacios, 1977, in his detailed review of wave
mechanics applied to SPT testing. The one-dimensional wave equation
relationship assumes that the stress pulse travels with constant
velocity without change in shape. This assumption is not completely
valid since the velocity of wave propagation varies with the
frequency of the wave. Kolsky, 1951, has shown that for engineering
purposes the error is negligible for wave lengths greater than about
ten times the bar radius which is typical in SPT.
Assuming that the elementary wave equation relationship is valid
for SPT testing the following properties of the elastic pulse may be
readily demonstrated;
1. The velocity of propagation c is independent of du/dt
or the local velocity of the elements transmitting the wave, c
depends only on the elastic properties of the propagating
medium, c for carbon steel typically ranges from 16,800 to
17,000 ft/sec.
2. A simple relationship between stress amplitude of the
pulse (or total force F across a section of the bar with area
a) and particle velocity, V, exists;
F = (E a/c) *V = z V (2.3)
20


where;
z = E a/c is known as the characteristic impedance of the rod
In terms of stress;
a = E/c V
(2.4)
3. Particle motion or velocity is in the same direction
as wave propagation for a compressive pulse and in the opposite
direction to a tensile pulse.
4. A pulse reflected from a "free end" has the same shape
but opposite in sign as the incident pulse. The reflected
pulse travels back the rod at the same velocity c. For
instance, a compressive incident pulse reflects as a tensile
pulse from the free end and vice versa.
5. A pulse reflected from a "fixed end" has the same
shape and sign as the incident pulse. Thus, a compressive
incident pulse reflects as a compressive pulse from a fixed end
and vice versa.
6. For a pulse, the components of which are travelling in
the same direction, the total energy acquired by the rod over
the duration of the pulse is made up of;
From assumption number 2. above, it was shown that Force and
particle velocity are uniquely related such that under one-
a. Kinetic Energy.
b. Stored Strain Ene
(2.5)
a2 dt (2.6)
21


dimensional wave equation assumptions both strain and kinetic energy
are equal and individually comprise 1/2 of the total energy of the
pulse.
Of prime concern is how energy is transferred from a hammer
impacting on drill rods. To solve the impact problem two assumptions
are made across the plane of impact of the hammer and rod (Fairhurst,
1961).
1. The force in the hammer must equal the force in the rod.
2. The absolute spatial velocities of the striking end of the
hammer and the struck end of the rod must be equal at all times
that the two surfaces are in contact.
Fairhurst performed a study of percussion drilling and applied
one-dimensional wave equation to the impact conditions. During
initial impact a compressive wave is sent down the drill rods and a
compressive wave is also reflected back up the hammer. From the
above assumptions, the initial particle velocities in a cylindrical
bar v^ and hammer vho can be predicted as follows;
where;
V is the initial velocity of the hammer
r = Zh/zh = ab pb Cj/aj, ph ch is the characteristic
impedance ratio.
Hammer-Rod Impact Conditions
vho = {r/(1+r)} V
vb0 = {1/(1+r)} V
(2.7)
(2.8)
22


For bar and hammer of the same material the value of r is simply the
ratio of areas.
In terms of stress the initial stress in the bar from impact is
the maximum stress of the pulse;
omax = p c tl/(l+r)} Vh, (2.9)
Ideally, energy transfer to the drill rods continues to occur in
the form of a stepped pulse transfer as long as the hammer and rod
remain in contact. The tensile wave reflected from the free end of
the hammer cancels the outgoing compression wave progressively
releasing the hammer from strain. By the time the tensile wave in
the hammer has arrived at the impact surface no strain exists in the
hammer and the hammer is travelling at a reduced uniform velocity Vr
A second pulse of particle velocity vb1 is then delivered. This
cyclic transfer is illustrated on Figure 2.1 after Fairhurst. Cycles
of energy transfer continue until impact is terminated when the
velocity of the bar exceeds that of the hammer. In practice transfer
is terminated when the head of the reflected tensile wave in the rod
reaches the top of the rod. This occurs at time 21/c where 1 is the
drill rod length. If the velocities are known the stresses can be
computed from Equation 2.3.
The result of these studies show that, theoretically, the energy
pulse is in the form of a stepped wave. The number of cycles and
therefore the duration of the pulse in an infinitely long drill rod
are only dependent on the length of the hammer. Also the magnitude
23


of peak stress and subsequent stress pulses are a function of
impedance ratio and initial velocity of the hammer. These
relationships were proven experimentally by Fisher, 1959. The step
down nature of energy pulses for SPT 140 lb. hammers of different
aspect ratios and lengths is illustrated on Figure 2.2.
For a hammer-rod system with impedance ratio of 1 (area of
hammer and rod equal) there is only a single step pulse of stress;
*bo = = 1/2 fi c vhi (2.10)
For a hammer rod system with impedance ratio of 0 (area of rod
infinitely small with respect to hammer) Fairhurst, 1961 derived the
expression for a continuously decaying wave of infinite steps as;
a = p c Vh} exp{-apct/m) (2.11)
where;
m = mass of hammer and t = time from impact
Some researchers have studied the SPT wave equation for r = 0 to
make conclusions regarding effect of drill rods (Yokel, 1982). These
analyses provide insight but, are not exactly correct because many
SPT hammer rod systems have impedance ratios, r, ranging from 0.2 to
0.4.
24


V

l
(a)
Figure 2.1. Mechanism of impact of a cylindrical hammer on a finite
cylindrical rod. (Fairhurst, 1961)
25


Fraction of Theoretical Max. Streas
~1 .0625
i 1 L i r .25 1 1 1
L_ i 1 l^r .44 1
i+. L fh '].!
Pn I 1 r i
y
O'-------------------------------------------------------------3
0 0.4 0.8 1.2 1.6 2.0
Tlae, 2L/c
r .0625
Figure 2.2. Effect of impedance ratio on the theoretical wave shape
for same hammer mass and drop (Palacios, 1977)
26


Wave Transmission at Area Changes
Sudden changes in cross sectional area of the bar will cause an
oncoming wave partially transmitted and partially reflected. The
magnitude and sign of the transmitted and reflected waves can be
determined using the same assumptions for hammer impact, namely
forces and particle velocities at the area of juncture are always
equal. For a rod were material properties E and p are constant the
following equations describe the relationship between incident
stress, o- and reflected ar and transmitted stresses, where A1 and A2 represent the changes in area in the rod. The sign
conventions indicate that the incident and reflective waves have the
same or opposite signs according to the increase or decrease in area.
The transmitted stress has the same sign as the incident stress but
its magnitude increases or decreases depending on the area change. A
graphical presentation of the complexity of wave transmission at a
change in area is shown on Figure 2.3.
Equations 2.7 through 2.13 are only approximate because the
assumption that the spatial particle velocities are equal on both
surfaces are not completely satisfied. This condition is only
satisfied inside the material and not on the free surfaces at the
area change. As a result, complicated stress wave interaction occurs
at the boundaries.
and
ot = {2A1/(A1+A2)} a,
or = {(A2-A1)/(A2+A1)} a,
(2.12)
(2.13)
27


In the SPT test, an impact anvil is always present between the
rods and the hammer. Typical shapes of anvils were shown on Figures
1.2 and 1.3. The anvil causes and increase in pulse amplitude as
illustrated in Figure 2.3b. The tensile wave reflected back up the
anvil will return to add a smaller compressive pulse upon return to
the connection. The net energy transmission, however is reduced
because some energy is lost in wave excursions in the anvil. The
most efficient energy transfer occurs for smaller anvils of short
length.
Energy Transmission Along the Drill Rods
The importance of the behavior of wave transmission at area
changes becomes even more apparent when one considers the design of
drill rods used in SPT testing. The Diamond Drill Core Manufacturers
(DCDMA) have standardized drill rod design in the U.S. Drill rods
designated as A through N series are allowed for penetration testing.
A summary of DCDMA drill rod sizes are tabulated on Table 2.1.
The "W" series drill rods find common use today. The drill rods have
coupling pins at every joint which represent changes in
crosssectional area. In the past five years the use of new tapered
thread drill rods has increased. The inclined surfaces in these
couplings present complex surfaces for wave transmission.
For most SPT, sections of ten foot drill rod are used. To further
complicate matters, drill rods have either upset wall or parallel
28


X
t < 0
I
o
IConctision
*
I
|b
j Tension
Figure 2.3. Wave transmission and reflection due to an abrupt area
change (Palacios, 1977)
29


Table 2.1. Summary of Typical SPT Drill Rod Information from Manufacturers Catalogs
(Kovacs, W. D., Salomone, L. A., and Yokel, F. Y., 1983)
Drill Rod Size Upset Wall ID ID Weight Per Foot of Rod8
Wall Area OD Parallel Upset r 2' 5' 10
in^ In in in lbs lbs lbs lbs
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)
E X 0.834 1 5/16 13/16 3 6 14.5 29
A X 0.847 1 5/8 1 1/8 4.5 8.5 20 38
B X 1.301 1 29/32 1 13/32 6C 10.5C 25 47
N X 1.289 2 3/8 2 10c 13.5C 27 49
EW X 0.795 1 3/8 15/16 3.5 6.5 15 31
AW X X 1.178 1 3/4 1 1/4 (1 7/16b) 5 8 21 18b 42 33b
BW X 1.141 2 1/8 1 3/8c 1 3/4 9C llc 24 43
NW X 1.212 (2.270)c 2 5/8 2C 2 5/16 llc 15C 31 55
a Without coupling
^ Upset wall available
c Usually furnished with parallel wall for 1 and 2 ft lengths.
30


wall construction as shown on Figure 2.4. The complex nature of
drill rods present the major problem in application of pure 1-D wave
equation modeling of the system.
The most important factors which influence energy transmission
along the drill rod is attenuation or reduction in pulse amplitude.
The primary sources of attenuation include internal friction and rod
joints. Internal friction is a term to account for damping
properties of the rod. For steel drill rod the attenuation constant
is frequency dependent. Using a typical attenuation constant for
steel rod and impact wave frequency of SPT it is estimated that
losses due to internal friction of only 1 percent per 10 foot of rod
would occur. Takaota, Hayamizu and Misawa (1958) investigated energy
losses in a single rod joint in a 7/8-inch hollow hexagonal drill
rod. Three types of joints including a tapered joint were
investigated. The results showed 3.5 to 5.5 percent losses due to
friction and 3 to 4 percent losses due to reflection. Goble et
al.,1967, investigated rod connections in 2-1/2-inch steel pipe under
impact loading. With the joint configuration tested it was found
that there was no change in the elastic wave. In 1973 Uto et al.
performed a study of 40.5 mm drill rod under SPT impact conditions.
Up to 206 ft. of rod were instrumented with strain gauges varying
intervals allowing for a determination of strain energy loss due to
rod length. Their results indicated a strain energy loss of 1.1
Percent/ft at constant rate with length. These studies seem to
confirm estimated losses due to internal friction. In the study of
31


ROD
COUPLING
ROD
V///////Z//ZZZ I If
vzmmA *L\\\\\\\\>

LENGTH
DRILL ROD WITH COUPLING, PARALLEL WALL TYPE
Figure 2.4. Illustration of drill rod wall type (a) Parallel wall (b) Upset wall
(Kovacs, W. D., Salomone, L. A., and Yokel, F. Y., 1983)
32


SPT energy transmission by Palacios, 1977, both top and bottom of the
drill string were instrumented but the energy losses in the drill
string were not quantified. Since the study only considered depths
of up to 75 ft. energy losses were considered to be minor. Finally,
Matsumoto and Matsubara, 1982, performed field test of SPT with drill
rods of 40.5, 50, and 60 mm instrumented with strain gauges at
intervals of 9 to 12 m. Drill rods of up to 39m were tested.
Results showed that energy losses of 1.7 %/ft. These results exceed
those estimated by internal friction alone and may reflect effect of
rod joints These losses are significant and could especially effect
deeper borings. Undoubtably if loose joints are present, energy
losses could be higher but these losses are difficult to quantify.
Considerable additional study is required to determine if energy
losses are significant, especially with the new tapered joints being
used today.
There is some data regarding the effect of drill rod diameter on
energy transfer. McLean et al., 1975, modeled the SPT using wave
equation analysis. The results indicated that the effect of drill
rod diameter was dependent on the driving system and soil resistance.
For the donut hammer modeled, the N rod resulted in slightly higher
blow count than A rod. The difference was generally less than 2
blows in 6-inches and depended on soil resistance. Matsumoto and
Matsubara, 1982, recorded stress wave data under SPT conditions with
rod diameters ranging from 40 to 60 mm and concluded that there was
no effect on energy transfer but only on the shape of the wave. The
33


larger rods contained a higher peak stress and the energy content of
the wave diminished more rapidly in the tail of the wave.
34


CHAPTER 3
DEVELOPMENT OF SPT ENERGY MEASUREMENT METHODS
Possible Methods to Measure Energy or Efficiency
Because of the significance of the N value dependency on
transferred energy and the preponderance of different forms of
equipment used for SPT it is important to develop a reliable and
reproducible method to measure efficiency in the test. Many
approaches may be taken. A schematic illustration of energy transfer
in the SPT is shown on Figure 3.1. The SPT hammer has potential
energy of 4200 in-lbf if dropped from exactly 30-inches. In
practice, operators using the rope and cathead method may not deliver
the actual 30-inch drop. To measure the efficiency of the hammer
drop mechanism the velocity (kinetic energy) just prior to impact
could be measured. The velocity can be determined by displacement
transducer, radar, or optical methods. Both Goble and Ruchti, 1981
and Kovacs et al., 1975, developed optical measuring systems to
measure hammer velocity at impact. These investigations gave
important insight into factors affecting the SPT variability. The
investigations showed that for the safety hammer operated by the rope
and cathead method, the number of wraps around the cathead greatly
influenced the energy. The energy was reduced by approximately 20
percent when the number of wraps was increased from 2 to 3.
35


STANDARD
SPT ENERGY
FOR 30 in. (76 cmj FALL
ksKD
-(a) potential energy
mg-H = 4200 in-lbf
B) KINETIC ENERGY OF HAMMER
Ev = '/2 m V-,2
LOAD CELL
ENERGY TRANSMITTED
THROUGH THE DRILL ROD
Ej = f {force2 -time)
l
l'
BORE HOLE
DRILL ROD
mg = Hammer weight
H = Hammer fall
Ey = Kinetic energy just
before impact
m = Hammer mass
V j = Hammer velocity
just before impact
Ej = Energy in drill rod
SAMPLER
gure 3.1. Schematic illustration of energy transfer in the SPT
and configuration for stress wave testing (Kovacs, W.
D., Salomone, L. A., and Yokel, F. Y., 1983)
36


Additional factors such as drill rig type and rope age, and cathead
speed became apparent. It quickly became apparent that the friction
involved with the rope and cathead method greatly influenced the
test. Kovacs, 1979 also measured hammer velocity of several
mechanical trip hammers used overseas. The hammers have the
advantage of releasing from the hoist rope after 30-inch drop is
obtained thus the rope pulley friction is absent. The studies,
however, indicated that the true trip height was dependent on the
rate of operation.
An important consideration to field instrumentation of the SPT
is the complexity of the measuring system and its ability to
withstand harsh operating conditions. Field conditions during SPT
are very harsh and very adverse to electronic equipment. The optical
systems developed in the 1970's were rather complex electronic
systems. The systems did not allow for direct analysis of the data
in the field. Today simple radar based equipment is available for
measuring hammer velocity. This equipment will play an important
role in future hammer performance studies.
The application of one dimensional wave equation has provided
additional insight into evaluation of the efficiency of SPT. A very
important aspect of SPT energy transfer is losses in the impact
anvil. Much of SPT prior to the advent of the safety hammer used
donut or pin guided hammers with cushioning material. Also, many
donut hammers have massive anvils which act as energy traps further
reducing energy transfer. In order to measure the true performance
37


of the hammer drop system and impact anvil the energy measurements
should be performed on the drill rods. In order to evaluate rod
energy loss it would be desirable to measure energy immediately above
the sampler, however the conditions within the borehole are even more
difficult. In the borehole, tests would have to be performed in
submerged conditions and there is difficulty with lead wire and
damage to equipment.
As shown by the application of wave equation, the energy
transmitted should be comprised equally of both strain and kinetic
energy. If one-dimensional wave equation conditions exist, energy
may be determined by integrating the product of force and velocity or
the square of either force or velocity versus time. If assumptions of
wave equation are not completely satisfied it would be ideal to
measure both force and velocity because any inequalities in strain
and kinetic energy could be evaluated.
Measurement of the velocity of a drill rod is considerably more
difficult than measuring velocity of the hammer because the
velocities are much higher and contain much higher frequency
components. Optical methods have been applied with limited extent
and they suffer from the complexity problems experienced in the
field. Optical studies were performed by Yamada, Fuyuki and Uto, in
1980, (ISSMFE, 1981) to analyze penetration and reflection behavior
of the SPT. They showed displacement-time behavior of the hammer and
rods during penetration for a wide variety of penetration resistance.
The dynamic bearing capacity behavior could be modeled using
38


reflection theory and wave equation but kinetic energy was not
determined. Another possibility for measuring velocity is by the use
of accelerometers. Earlier attempts to use accelerometers by
Palacios, (1977), were unsuccessful due to the high frequency
vibrations exceeded the capability of the equipment. With
advancements in accelerometer designs today, successful applications
may be attained. Measurements with accelerometers are currently
being studied by Dr. G Goble at the University of Colorado at
Boulder.
Measurement of force is somewhat easier. Strain gauges can be
applied to the drill rods but one must consider such problems as
localized bending, non-uniformity of cross sectional area, and
dynamic response of the gauges.
The first studies of stress wave energy transfer in the SPT were
performed by Uto and Fuyuki, (1973). They instrumented drill rods in
SPT with strain gauges below the hammer and along the drill string.
The details of the instrumentation systems are not known since the
research reports are in Japanese and only summaries are given in
English (ISSMFE, 1981). They successfully determined energy transfer
from the hammer drop system and along the drill rods. Energy content
of the stress wave was determined by integration as follows;
Development of the Stress Wave Energy Measurements
a2 dt
(3.1)
39


Where:
Ej = Drill rod stress wave energy.
The energy is double of the strain energy given by Equation 2.5
using the assumption that strain and kinetic energy are equal.
Measurements were obtained with the Japanese Industrial Standard SPT
hammer both in rope-pulley operation and in a free fall mode (ISSMFE,
1977). The percentage of nominal energy transfer to the rods
immediately below the hammer was 53 to 72% for rope-pulley and 80 to
90% for free drop.
In 1977, Palacios developed a method for determining the stress
wave energy in SPT by the use of load cells placed within the drill
string. These studies were a significant contribution to
understanding of the variability of SPT in the U.S. and they showed
significant variation in the test. The object of the study was to
analyze the transfer of energy in SPT testing. The study consisted
of field measurements of SPT with two different hammers and two
different drill rods with lengths of up to 75 ft. Hollow strain
gauge load cells were placed in the drill string immediately below
the hammer and above the sampler. The 40,000 lb capacity strain
gauge load cells manufactured by Sensotec were 3-1/2-inch diameter by
2-1/2-inch in height. A threaded center hole of 1-1/4-inch diameter
was provided to allow water to pass through.
The energy was calculated according to wave equation principles
by integrating the square of the force as given in Equation 3.1. It
was recognized that for short drill rod lengths the total available
40


energy would not be transmitted to the rods because the reflecting
wave in the drill rod would terminate the stepped pulse input of the
hammer (Section 2.2). A correction factor, K2, was derived to
account for incomplete energy transfer. The correction was derived
based on the theoretical hammer impact relationships as discussed in
Section 2.2 which assume a theoretical stepdown pulse of the stress
wave.
K2 =1/{1- exp [-4 a p (1-a1)/Mh]) (3.2)
Where:
Mh = Mass of hammer
a! = Distance from impact surface to load cell (Figure 3.1)
1 = Length of drill rods (Figure 3.1)
Table 3.1 illustrates the effect of the correction factor K2.
When drill rod length exceeds 45 ft the correction factor approaches
one. For short rod lengths the correction is significant.
Table 3.1. K2 Factors (ASTM, Method D-4633, 1986)
L X2 Factor
ft m
10 (3.0) 1.45
15 (4.6) 1.22
20 (6.1) 1.1t
25 (7.6) 1.06
30 (9.1) 1.03
35 (10.7) 1.02
40 (12.2) 1.01
45 (13.7) 1.00
41


Typical results of the study by Palacios are shown on Figure 3.2
for AW drill rods and safety hammer combination. The efficiency of
the energy transmission is expressed as follows;
ER, E, / E* (3.3)
Where:
ER,. = Measured stress wave energy ratio
E* = Nominal kinetic energy of SPT of 140 lb hammer dropped
from 30-inch height = 4200 in-lbf
The results shown on Figure 3.2 show a general efficiency of 60
percent for the AW rod-safety hammer combination. The theoretical
maximum efficiency of the system based on manual computations using
hammer impact theory are shown with a dotted line. Note that at
shallow depths from 10 to 30 ft, efficiency is slightly less
indicating that the correction, K2, may not be completely accurate.
The study by Palacios considered such issues as the energy used
in sampling, the proportionality of blow count with delivered energy,
the displacement of the sampler with cycles of energy transfer, and
equipment variables such as drill rods and hammer types. Findings
from this study were summarized in a paper by Schmertmann and
Palacios, 1979. Significant findings are summarized below;
Efficiency ranged from 30% for donut hammers to up to 70%
for safety hammers.
42


Figure 3.2. Variation of E-/E* with rod length for AW rod-safety
hammer combination (Palacios, 1977)
43
Length of roda t (in)


- Energy transfer at depths shallower than 40 ft is not
complete due to return of the tensile wave causing termination
of stepped pulse transfer at impact. This effect is more
pronounced for short lengths and the deficit in energy can be
estimated by the K2 factor given above. There was no
significant reduction in efficiency with depth exceeding 40 ft
confirming that there were minor losses in drill rods.
- The compression wave energy entering the drill rods depends
only on the hammer-rod system and not on the soil resistance.
- The speed of wave travel was in close agreement with the
theoretical stress wave speed of 16,700 ft/sec.
- The sampler set depended on the resistance of the material
tested which are in agreement with wave equation analysis of
SPT.
Since there was minor energy loss in the rods to estimated
depths of 100 ft., the measurement of stress wave energy in the rod
below the hammer is the best method for evaluating energy transfer of
SPT hammer drop systems. The wide variation of energy measured had a
significant impact on the N value and aroused concern in the
engineering community (Schmertmann, 1978). The inefficient energy
transfer in short drill string caused Seed, (1979, 1983) to recommend
that N values obtained at depths of less than 10 ft be reduced by a
factor of 0.75 in liquefaction studies. A more exact correction
could be applied by using the inverse of the K2 factor. Several
44


groups of researchers began an extensive series of field measurements
to further evaluate energy variation in the U.S.
Application of the Binary Instruments Device
The Binary Instruments device was developed by Hall, (1982), to
perform stress wave energy measurements using a load cell and
electronic analog device to determine the drill rod energy, Ep in
real time after each blow of the SPT hammer. A block diagram of the
key elements of this device is shown on Figure 3.3. Either drill rod
energy, Ei, or peak force can be displayed on the LCD display. The
device uses force detection system to sense the on coming pulse and
begin electronic integration of the wave until the tensile wave
return is sensed. The device was housed in a single package which
was easy to set up in the field. The use of a load cell immediately
below the hammer did not present any significant problems in field
monitoring. Only ten Binary Instruments Calibrators were
manufactured and used for testing in the U.S. Some of the devices
were used by the following organizations;
National Bureau of Standards
Schmertmann and Crapps
Law Engineering
Central Mine Equipment Company
Mobile Drilling Company
Earth Technology Corporation
45


Figure 3.3. Schematic showing key elements of the Binary Instruments SPT energy measurments device
(Kovacs, W. D., Salomone, L. A., and Yokel, F. Y., 1983)
46


The devices were used extensively from a period of 1979 to 1983
to collect the bulk of data regarding energy transfer in SPT for U.S.
practice.
The principle of integration of stress wave energy is the same
as proposed by Palacios with an additional correction factor, Kr It
is not clear in the literature when this correction was derived but
it was applied to all data collected with the Binary Instruments
device. The factor was not apparent in the thesis by Palacios. The
correction is first found in Hall's report on the Binary Instruments
device. The factor is a correction to account for the fact that the
load cell is not located at the impact surface. The load cell should
be located a minimum of 10 rod diameters from the impact surface to
avoid effects from discontinuities in the drill rod. According to
Hall (1982), "The first compressive pulse energy at this point will
be less than if the measurement were made at the impact surface."
The correction factor can be expressed as;
K1 = {1 exp [-4apl/Mh]}/{l exp [-4apAl/Mh]} (3.4)
Where all of the variables have been presented previously. The K1
factor has a significant effect in short drill strings combined with
increased distance from the impact surface. The range of values of
K., are illustrated on Table 3.2. In practice the load cell is
frequently separated by a 2 ft section of drill rod below the impact
surface. In addition most of the energy measurements are attempted
at depths greater than 30 to 40 ft. to minimize effects of both
47


correction factors.
Table 3.2. K., Factors (ASTM, Method D-4633, 1986)
uisiance Distance from Impact to Load Cell, ft
from Imoaa _______________________________________________
Samoler. ft 2 3 4 5 6 7 8 9
10 1.14 1.24 1.37 1.57 1.86 2.35 3.35 6.33
12 1.10 1.16 1.25 1.36 1.51 1.72 2.04 2.58
14 1.07 1.12 1.17 1.24 1.33 1.45 1.61 1.84
16 1.05 1.09 1.13 1.17 1.23 1.31 1.40 1.53
18 1.04 1.06 1.09 1.13 1.17 1.22 1.28 1.36
20 1.03 1.05 1.07 1.10 1.13 1.16 1.21 1.26
22 1.02 1.04 1.05 1.07 1.10 1.12 1.15 1.19
24 1.02 1.03 1.04 1.06 1.07 1.09 1.12 1.14
26 1.01 1.02 1.03 1.04 1.06 1.07 1.09 1.11
28 1.01 1.02 1.03 1.03 1.04 1.06 1.07 1.08
30 1.01 1.01 1.02 1.03 1.03 1.04 1.05 1.07
32 1.01 1.01 1.02 1.02 1.03 1.03 1.04 1.05
34 . 1.01 1.01 1.01 1.02 1.02 1.03 1.03 1.04
36 1.00 1.01 1.01 1.01 1.02 1.02 1.03 1.03
38 1.00 1.01 1.01 1.01 1.01 1.02 1.02 1.02
40 1.00 1.00 1.01 1.01 1.01 1.01 1.02 1.02
factors is illustrated below;
a2 dt (3.5)
The operations manual for the Binary Instruments device was
detailed and was continuously updated with modifications (Binary
Instruments, 1980). The calibrator system has outputs for the force
signal and trigger pulse which delineates the integration time. The
outputs allow for viewing of the force time trace through an
oscilloscope. The instrument has necessary zeroing circuits for load
cells and internal scaling circuits which adjust the for a variety of
drill rod sizes.
Application of both correction
E, = K, K2 a c / E I
48


The Binary Instruments (Serial No. 202) device used in this
study was the apparatus on loan from the National Bureau of
Standards. Its operation was documented by Kovacs, et al., 1983, but
will be reviewed here for the reader. The equipment was modified
over the standard device as listed below;
a. The internal circuitry was altered to allow two different
strain gauge load cells to be used interchangeably with the
flip of a switch instead of going through an elaborate internal
instrument calibration;
b. A new circuit was added to allow for the use of new high
capacity piezoelectric load cells. The new circuit allowed for
zeroing of the load cell and different load cells with
differing calibration factors could be used by adjusting the
Ecal and Fcal values.
c. A new circuit was added to allow for the use of a tape
recorded force-time signal as input to the calibrator to check
actual field readings.
d. Added a DC 12 volt battery power supply.
e. Added a timer circuit to provide readout of integration
time.
Of these modifications, items a, c, and d were not important to
this study. Kovacs, et al., 1983, performed an extensive comparison
of the integration performed by the Binary Instruments device and
integration of the force time history recorded on tape from the
device output which was integrated on a digital processing
49


oscilloscope. The comparison indicated excellent agreement of the
two methods validating the Binary Instruments device. The NBS
calibrator was used to collect the bulk of SPT energy data collected
both in the U.S. and in Japan.
In this field study two piezoelectric load cells (sn #157 and
#233) were used for testing. The 60,000 lb load cells were
manufactured by PCB Piezotronics model number 227M04. Piezoelectric
load cells work on a static discharge system and as such only respond
to sudden changes in force. Calibration of these cells requires
rapid changes of controlled loading. During this study the cells
were sent to the manufacturer frequently for calibration checks.
Results of the calibrations showed extreme stability with maximum
variation in calibration factor of .001 mv/lb. The internal
circuitry of the Binary Instruments device was calibrated by the
manufacturer prior to testing and once during testing with only minor
trim adjustments necessary.
The Binary Instruments device uses two internal calibration
circuits with variable resistors to adjust for load cell calibration
factor and rod size (area used scale energy for the appropriate area.
These two calibration factors are adjusted prior to each series of
testing. The calibrator is matched to piezoelectric load cell # 157
which has a calibration factor of 0.108 mv/lb. An internal
calibration voltage is generated equal to the output of the load cell
at force level of 20 kips. For load cell #157 this voltage is
equivalent to 2.16 volts. If substitute the substitute load cell
50


#233 with calibration factor 0.116 mv/lb is used, the force readout
in the force calibration mode, Fcal, should be adjusted using the
front panel potentiometer as follows;
Fcal = 2.16/0.116 = 18.62
When the matched load cell #157 is used, Fcal is set equal to 20.0.
The rod size factor, Ecal, is provided with a switch on the
front panel for rocj sizes A, AW, N and NW. The rod size factor is
used for scaling the final energy output from the apparatus as
illustrated in the block diagram. The default values from the switch
setting, Efp, are to be used with the matched load cell. They are
calculated by internal integration of a 20 kip square wave with
duration of 8.19 msec (Kovacs, 1983). The front panel default values
are dependent on cross sectional area of the drill rods used. The
default values are shown on Table 3.3.
The Ecal values can be calculated given the area of drill rods
from the following relationship;
Ecal = 531.9/a (3.6)
where: a = drill rod crossectional area in2
If the unmatched load cell is used with the Binary Instruments
device, the rod area factor, Ecal, must be adjusted according to the
change in Fcal. This adjustment is performed by potentiometer
located on the front panel. For example if standard AW rods are used
the default value is adjusted by the following factor;
Ecal = (Fcal/20.0)2 Efp (3.7)
51


Table 3.3. Tabulation of Calibrator Display Numbers for Drill Rod
Areas (Kovacs, W. D., Salomone, L. A., and Yokel, F. Y.,
1983)
Calibrator Display No. Rod Size Area (in2)
(1) (2) (3)
493 A 1.079
452 AH 1.177
413 N 1.288
[466] [BW] [1.141]
235 NWa 2.264
[439] [NWb] [1.212]
a Parallel wall rod
b Upset wall rod
Numbers In brackets do not appear on Calibrator selection switch area.
For load cell #233 Fcal is equal to 18.62 and Ecal for standard AW
rods would be 391.
The area of drill rods used by different manufacturers varies
slightly and many measurements of energy lack accurate determination
of rod area. Such was the case in this study. It is very difficult
to determine inside area of upset wall drill rods as apparent in
Figure 2.4. The DCDMA specifications set the outside diameter of
classes of drill rod constant, however, the inside diameter varies
slightly with manufacturer. In any class of rod these variations
52


should be minor. In this study a significant number of measurements
were obtained for NW class of drill rods. The default rod size
factor, Efp, for the Binary Instruments device is based on cross
sectional area of parallel rod, a = 2.264-in2. Most all of the NW
rods in practice are of upset wall configuration to minimize weight.
All of the drill rods measured in this study were of upset wall
construction with a = 1.212-in2. If an incorrect Ecal factor is used
during testing, adjustments to the Ei data can be determined after
testing based on the ratio of the actual Ecal factors to the faulty
Ecal factor used in the field and data is not lost. This adjustment
was required on some of the data in this report.
In 1982, a timer apparatus was added to the Binary Instruments
device to display the integration time. This addition was added
based on measurements with Kovacs in which the author took part. In
hard driving conditions, simulating fixed end conditions in wave
equation analysis, the reflected wave is a compression wave and the
Binary Instruments device does not sense tensile wave cutoff and
continues integration giving erroneous energy values. This
occurrence is illustrated in Figure 3.4. The Ei value from the
calibrator will almost be double of the actual energy content in the
first pulse. Prior to use of the timer, measurements performed with
the Binary Instruments device but without visual confirmation of the
wave integration with oscilloscope could be subject to errors. The
timer can also assist with evaluating other electrical problems such
as shorting of load cells as illustrated on Figure 3.5.
53


FORCE kips
Figure 3.
er; = 54%
20
10
0
ER | = 53% integration, At = 8.3 ms, DPO; = 88%, integration, At = 13.2 ms, DPO
ERj = 90% calibrator, At = 13.5 ms ^
10 11 12 13 14
End trigger v
123456769
TIME ms |
Tape recorder DpQ
amplifier overload^ first crossover V
Trigger f
Calib.
integration
time = 13.5 ms
Illustration of compressive wave return, force time history, and trigger duration time
(Kovacs, W. D., Salomone, L. A., and Yokel, F. Y., 1983)
54


Series 59, blow 9, safety hammer, l' 33.4 ft.
ERj = 30 calibrator, t = 1.2 ms
Figure 3.5. Example of electronic shorting of load cells during impact
(Kovacs, W. D., Salomone, L. A., and Yokel, F. Y., 1983)
55


Electrical shorting was a problem with the early versions of strain
gauge load cells but has not been evident with the new piezoelectric
cells. Piezoelectric cells are more durable and when they fail they
fail completely.
The disadvantages to the Binary Instruments device is its lack
of hard copy output. During testing the Ei values must be read and
recorded for each blow. After each blow the instrument must be
reset. The instrument can be reset either manually by switch or
automatically. The automatic timed reset is of little used since SPT
is not performed at a constant rate unless an automatic hammer is
used. SPT is normally performed at blow count rates of 50 to 15
blows per minute leaving only 1 to 4 seconds for instrument reset.
It is also recommended to record the integration time from a separate
LCD display. This leaves no time to switch and display peak force.
The most frequent method used in the field to obtain the maximum
amount of data is to use a handheld tape recorder to monitor a series
of blows.
Although the binary instruments device seems outdated, it is
still highly functional. With the addition of a timer, there is no
need for an oscilloscope or other complex electronic instrumentation
in the adverse SPT environment. Even today the most expensive
digital oscilloscope cannot record and save blows at the rate
generated in SPT. In order to save a waveform of duration of
milliseconds an extremely fast sampling rate is required for an
oscilloscope. The memory requirements therefore do not allow for
56


rapid storage of multiple waves which occur at a rate of 1 to 4
seconds. The Binary Instruments device solves this problem since the
waveform is scaled ^and integrated in a matter of seconds.
Standardization of Stress Wave Energy Measurement
In order to assure uniformity in energy measurements, there must
be stringent guidelines for methods of recording, correcting and
reporting data. In 1978, Schmertmann, Smith and Ho, reported an
example calibration of SPT on a drill rig. They used a strain gauge
load cell, oscilloscope, and Polaroid camera to obtain a copy of the
stress wave. The Polaroid photo was then later digitized and
integrated. Their estimated accuracy in integration of stress wave
energy was within 5 percent. Their report included information on
the drill rig and examples of the waveforms obtained. The
presentation of typical waveforms from an oscilloscope gives
considerable confidence in testing, yet does not allow recording of
multiple SPT N values during testing.
Efforts to standardize SPT energy measurement did not begin
until the late 1980's when Schmertmann began efforts to ballot a
procedure in the American Society for Testing and Materials (ASTM).
This is unfortunate since the bulk of SPT energy measurements used to
make engineering decisions today, were obtained prior to
standardization. The experienced gained was beneficial in developing
a standard which could precaution against potential pitfalls.
57


Numerous exchanges in information on stress wave energy measurements
were disseminated in journals prior to standardization.
Standardization was completed in 1986 with approval of an ASTM
standard D-4633, "Standard Test Method for Stress Wave Energy
Measurement for Dynamic Penetrometer Systems". This procedure was
adhered to in this study and is attached in Appendix A for reference.
The procedure provides for determination of the stress wave energy
ratio by means of load cell and processing instrument which
calculates the energy in accordance with Equation 3.4. The type of
load cell is not specified and the use of strain gauges is not
discussed. Although reporting of an oscilloscope waveform trace is
not required it was recommended as a useful adjunct. As a follow up
to the ASTM procedure an international procedure was first published
at the proceedings of the First International Symposium on
Penetration Testing on March, 1988 by the ISSMFE technical committee
on penetration testing. The procedure is virtually identical to the
ASTM method with exception that it is written specifically for SPT.
The method requires periodic recording of the duration of the
pulse until cutoff, At. The theoretical travel time is calculated as
follows;
At = 21 7c (3.8)
where: 1' = Distance from load cell to bottom of rods
(Figure 3.1)
c = recommended value for steel = 16,800 ft/sec
If the wave travel time recorded exceeds the theoretical value by 1.2
58


a reflected compressive wave is included in the integration and Ei is
not reliable. If the recorded wave travel time is less than 0.9 of
theoretical there is a premature cutoff which needs to be determined.
This premature cutoff can be caused by electrical shorting or
unreliable zeroing of the cell. The premature cutoff problem was
significant in this study. The author believes that if an
oscilloscope is not used the travel time should be recorded for every
data point for reliability evaluations.
Another controversial issue arose in this standard. An
additional correction factor was added to Equation 3.4. The
correction factor Kc was defined as;
Kc = eye, (3.9)
where: ca = actual compression wave velocity measured
cf = theoretical value of compression wave velocity used
in integration or programmed in processing instrument
The factor is intended to account for differing speed of compressive
wave velocity measured. Kovacs, 1983, found that the travel times in
his studies normally were slower than theoretical. Data from other
researchers was corrected by a factor of 16,000/16848=0.95 in his
report to account for this correction in previous studies, yet
statistical basis and data for the value of 16,000 ft/sec were not
given. One cannot argue that if c is truly different, then Equation
3.4 requires the appropriate value of c. However the author contends
that this correction is not justified due to the following reasons;
59


- The correction assumes that the measurement system has
sufficient accuracy to measure c.
- The true value of c near the measurement point may be indeed
close to theoretical, yet travel times measured are slower due
to discontinuities in the drill rod causing attenuation. The
modulus of steel does not vary appreciably.
-The occurrence of loose rod joints may also cause premature
cutoff, but in this case a new drill string length should be
used for computations of travel time and other correction
factors K1 and K2.
If Kc is applied, correction factors can range from 1.1 to 0.83 given
the allowable range in travel times provided for in the procedure.
In practice Kovacs estimated this correction normally results in
corrections of 3 to 5 percent. In this study the travel time is
recorded for every blow. The correction Kc will be evaluated later
in this report.
60


CHAPTER 4
REVIEW OF EXISTING STRESS WAVE DATA IN SPT
Initial Study bv Palacios & Schmertmann
The initial study by Palacios under direction of Schmertmann
were instrumental in pointing out the variability of SPT testing.
The study by Palacios included analysis of energy effects in the SPT
along with complete analysis of statics and dynamics of both SPT and
cone penetration testing for possible correlation of the two tests.
Results of the study were documented in a doctors thesis at the
University of Florida (Palacios, 1977), and a series of articles
written by Schmertmann and Palacios (Schmertmann, 1978, Schmertmann
and Palacios, 1979). For comparison purposes of this study only the
drill rod energy data is reviewed.
Energy measurements were performed in nine borings with three
different drill rigs. The scope of the work included three hammers;
a Mobile safety hammer (S), a Sprauge and Henwood donut hammer (B),
and a modified donut hammer referred to as the Florida Testing hammer
(F). The Florida Testing hammer had a large anvil which resulted in
low energy transfer. The hammers were tested in combination with
three type of drill rod, A, AW, and N. Results of the study are best
summarized on Figure 4.1 which is in similar format to Figure 3.2.
The comparison shown on Figure 4.1 also includes effects from
61


Figure 4.1. Summary of variation of E./E* with rod length for
different rod-hammer conbinations (Palacios, 1977)
62
Length ot rode t (a)


different drill rigs and operators which can have significant effects
when using the rope and cathead method of hammer drop. The effects
of drill rig and operator were not systematically investigated in
this study. Since these variables are not constant conclusions
regarding the effect of drill rods cannot be made. The test results
show that the safety hammer is more efficient that donut hammers,
especially the Florida hammer which had efficiency of only 40
percent. The donut hammers had efficiencies ranging from 40 to 53
percent while the safety hammer ranged from 50 to 55 percent.
Summary Study by the National Bureau of Standards
Kovacs, while working at the National Bureau of Standards,
performed a comprehensive study and compilation of drill rod energy
measurements. He performed 1087 energy measurements at seven sites
comprising 5 drill rig types and 11 operators. Kovacs attempted to
quantify the following variable effects on SPT;
- Establish the efficiency of the hammer delivery system
through measurement of hammer impact velocity
- Compare results between drill rod energy from the Binary
Instruments device and integration by digital processing
oscilloscope
- Study the efficiency of energy transfer through the anvil by
comparison of hammer impact velocity and drill rod energy
- Study the effects of operators, equipment and procedures on
the energy transmitted to the drill rods
In addition to the significant quantity of field measurements
Kovacs, included a compilation of drill rod energy measurements
63


performed by others. The results of the study were presented in
detail in a Nuclear Regulatory Agency report in 1983 (Kovacs et al.,
1983). Prior to the 1983 study, Kovacs had reported stress wave
energy data on and additional four drill rigs and two hammer types
(Kovacs and Salomone, 1982). These reports serve as an excellent
summary of drill rod stress wave energy measurement data collected up
until 1983.
Summary Table 4.1 is a compilation of all average drill rod
stress wave energy measurements summarized in the report by Kovacs.
The purpose of this table was to compile data regarding national
average energy data for typical SPT testing conditions. The
measurements by Schmertmann and Smith were performed by integration
of Polaroid image from oscilloscope. The remainder of testing is
assumed to be collected with the Binary Instruments Device with
strain gauge load cells. Data collected by researchers other than
Kovacs may include errors since wave travel times might not have been
verified. For example, the data by Brown, 1980 for a donut hammer
with 3 cathead wraps could easily be almost double that of actual if
reflected compressive waves occurred. Since the data by Kovacs was
checked by digital integration reliable travel time data is assured.
The data were further summarized on Figures 4.2 and 4.3 where the ERj
data are shown for different drill rigs. Any correlation to drill
rig type was unsuccessful, but the report showed the average
arithmatic energy for safety and donut hammers to be 59 and 47
percent respectively.
64


Table 4.1 Tabulation of Average ERi (Kovacs et al., 1983)
[Cotpatid iitalai CMptmlvt Hm Velocity In lodl 16818 ft/e (SIM /)]
(Z) Haaear Type Drill Rig Model tf0e of Tune Rod Type Remarks Reference
U) (11 (h (5> (63 in
71 s as is 3 AH Scbmerrmesn end Smith (1977)*
74 s QOS 45 3 AW Schmartmean end Smith (1977)*
53 s Qfd 45B 4 AW Schmertmmsm end Smith (1977)*
59 s Falling 1500 3-4 MV Scbmertmenn end Smith (1977)*
47 s Kayhsw 1000 3 AW Sctamertmmna end Smith (1977)*
57 s Falling 1500 3 Ntf Scfamertmean end Smith (1977)*
55 s GKE 45 3 AW Sobmmrtme&a end Smith (1977)*
56 s CMS IS 3 AW Sefamercmena end Smith (1977)*
54 s OS 65 3 AW Sehmmrcmmaa end Smith (1977)*
71 s OS 55 4 AW Sehmeraana end Smith (1977)*
57 s Falling 1S00 N/A HU Wire-Drum Sefamertmmnm end Smith (1977)*
56 s Aeker K-2 2 Brown, R.E. (1980) Frlvntn Cnaanaleatlon
53 s Acker M-2 2 Brown. R.E* (1980) Private fnmlurIon
55 s Mobile B-31 N/A Safe-T-Drlver Brown, R.E* (1980) Private Communication
76 D CMS ss 3 Brown, R.E* (1980) Pvlvetm fnunl ration
44 s OS 55 2 Brown, R.E. (1980) Frlvntn Cnnannlration
64 s Mobile B-33 ATV 2 Brown, R.E. (I960) Private Comaamicetloa
50 s Mobile B-80 N/A Safe-T-Drlve Schmertnana, J.H. (1980) Private rnunlretlon
41 s Mobile B-80 2 Scheartaann, J.B. (1980) Frlvntn Ciill ration
60 s CHE 45 2 Mud Bug Scbmertmenn. J.H. (1980) Prlvete CoKnlcetlon
51 D CMS 45 1 Swaap Buggy Sttlabarg, S.B. (1980) Prlvaet CoMaalcatlon
36* 0 CMS IS 2 Swaap Buggy Steinberg, S.B. (1980) Prlvete Coammniceclon
45 D CMS IS 2 Stalnbarg, S.S. (1980) Frlvaca CriMinl cation
71 D CHE 55 2 Oper. D Steinberg, S.B. (I960) Prlvete Comamaieetion
60* D OK 55 2 Opar. F Steinberg, S.B* (1980) Prlvete Commo&leatloa
42 D Joy B-12 2 Oper. A Stalnbarg, S.B. (1980) Frlvata firm!cation
55* D Joy B-12 1 Oper. o Stalnbarg, S.B. (1980) Frlvata CiTnlration
16 S Mobile B-31 N/A Safe-T-Drlver, J Stalnbarg, S.B. (1980) Frlvata Ceeaalcatlon
32 D Mobile B-61 3 Oper. A Steinberg, S.B* (1980) Prlvete Coaameetloa
40* D Mobile B-61 2 Oper* B Steinberg, S.B* (1960) Prlvete Coamnalcetlon
37* D Mobile B-61 2 Oper* C Steinberg, S.B. (1980) Private r Tint cation
66 S Qffi 55 2.2 N Old Rope Kovacs, et el., 1981, Series lb
69 s CMS 55 2.2 N Old Rope Xoveee, et el., 1981, Series 3
71 s CMS 55 2.2 H Old Rope Kovacs, et al., 1961, Series 6
73 s CMS 55 2.2 H Old Rope Kovaca, et el., 1981, Series 8
81 s CHE 55 2.2 N Nav Rope toveca, et al., 1981, Sarlaa 9C
77 s CHE 5S 2.2 N Hew Rope Kovscs, et sle, 1981, Series 11
73 s Qffi 55 2.2 N Hev Rope Kovscs, et el*, 1981, Series 17
62 s CHE 7S0 2.73 AW Kovscs, et el*, 1981, Series 19^
63 s CMS 750 2.75 AW Kovscs, et el*, 1981, Series 20
65 s CHE 7S0 2.75 AW Kovscs, et el*, 1981, Series 22
70 s CHE 750 2.75 AW Kovscs, et el*, 1981, Series 25
57 D CME 55 2.2 AW Kovacs, sc al*, 1981, Series 28*
Notes: The caablaatlon of the ueuel operator, hie drill rig with cathead and rope, and tha ha^er are considered
as a separata data point. Whan a second operator uaaa a rig that la not hla own, that data point la
denoted by an asterisk, a, and la not plotted on figure 1.4.
a. Tha values shown In colon (1) haws bean revised to reflect a theoretical coapresalvs wars velocity of
16,818 ft/aac. (5111 a/a). Tha original data were computed using 016000 ft/a (1877 n/a). The text
discusses how all tha data in table 1.1 ara used la figure 1.1.
b. Tha weighted avarags of Serlea, 1, 3, 6, and 8 taken aa 72Z
c. Tha weighted average of Series 9, 11, and 17 taken aa 77Z
d. Tha weighted avarga of Sarlaa 19, 20, 22, and 25 takan aa 63Z
e. Tha weighted average of Serlea 28 and 31 takan aa 57Z
f. Tha weighted avaraga for Series \? s 33 takan aa 35Z
g. Tha weighted avaraga for operator A takan a 66Z
h. These data reduced using Be correction froa actual wave velocity naaaureaents. The corresponding corrected
values are 60, 70 and 61 percent, respectively.
65


Table 4.1
(contd.) Tabulation of Average ER,-
(Kovacs et al
1983)
Tabla 4*1 Continued
t (I) Blair Type Drill Rig Modal No* of Turna Rod Type Remarks Reference
-m U) (3i (4) <> a)
55 D CMS 55 2.2 AH Eovace, et al.v 1981, Serlea 31
31 D CHE 45 2.2 AH Eovaes, et al. 1981, Serlea 32*
46 D CHE 45 2.2 AH Kovaca, et al.9 1981, Serlea 33
78 S CHE 55 2 AH Brovn, R.E. (1981) Private Comnicatlon
40 s Falling 1500 2 AH Schaartaann, J.H. (1962) Private Coaaunlcatlon
51 s Falling 1500 2 AH Schaartaann, J.H. (1982) Private Coaaunlcation
63 s Falling 1500 2 AH Schaercaann, J.H. (1982) Private Cowunlcatlon
68 s CHE 45 1 AH S~25 Schaartaann, J.H. (1982) Private Comualcaclon
31 D Longyear 34 3 - N~14, Lg. Anvil - Schaercaann, J.H. (1982) Private Conaunleaclo
57 D Falling 1500 2 N V. Snail Anvil Schaercaann,'J.H. (1982) Private Coaaunlcatlon
59 s Falling 250 2.75 N Opar. Af N*i5 Schaartaann, J.H. (1982) Private CoaaunleatlonS
69 s Falling 250 2.75 N Oper. A, N-62 Schaercaann, J.H. (1982) Private CoMumicatloo*
59* s Falling 2S0 2.75 H Oper. 3, N-23 Schaercaann, J.H. (1982) Prlvata Coaaunlcatlon
71* s Falling 250 . 2.75 N Opar. C, N22 Schaartaann, J.H. (1982) Private Communication
55 s Mobil. B-S0 2 AH N-10 Schaartaann, J.H. (1982) Prlvaea Cunnl cation
62 s Falling 1500 2.75 N Opar. A, N41 Schaartaann, J.H. (1982) Private Cosaunlcation
59* s Falling 1500 2.75 H Oper. B, N17 Schaartaann, J.H. (1982) Private Coaaunlcatlon
6** s Falling 1500 2.75 N Oper. C, K*28 Schaartaann, J.H. (1982) Private Coaannlcation
50* s Falling 1500 2.75 N Oper. 0, N-19 Schaartaann, J.H. (1982) Private CoHxmicatlon
62 s Falling 1500 2.75 N Oper. Ct N"6 Schaartaann, J.H. (1982) Private Covunlcatlon
63 s Falling 1500 2.75 N Opar. Cv H-14 Schaercaann, J.H. (1962) Private Coamnleatlon
49* s Falling 1500 2.75 H Opar. D, N>100 Schaertmann, J.H. (1982) Private Coaaunlcatlon
60* s Falling 1500 2.75 N Oper. A, N-15 Schaartaann, J.H. (1982) Private Comnlcatlon
64 s CHE 55 1.75 jiH Oper. A, N"12 Schaercaann, J.H. (1982) Private Coananlcaclonh
75* s CHE 55 1.75 AH Oper. B Schaartaann, J.H. (1982) Private Coaaunlcatlon
68 s CHE 45C 1.75 AH Oper. A. (Ser. H) Schaartaann, J.H. (1982) Private Coaunleaclon
43 D Longyear 34 2 Caapanella, R.G. and Robertaon, R.E. (1982),
Private Coaaunlcatlon
62 s Longyear 34 2 Caapanella, R.G. and Robertaon, R.E. (1982),
Prlvata Coaaunlcatlon
39 D Longyear 38 2 Caapanella, R.G. and Robertaon, R.E. (1982),
Private Coaaunlcatlon
47 S Mobile B61 2 NV Oper. A Table 3.3, Series 79
48 S Mobile B61 2 Ntf Oper. B Tabla 3.3, Serlea 95
60 s Mobile B61 2 BW Oper. A T.bl. 3.3, Serls. 10Z
49* s Mobile B61 2 Btf Oper. B Table 3.3, Series 103
54* s Hoblla B61 2 BH Oper. C Tabla 3.3, Series 104
70 s Longyear HC150 2 8V Oper. C Table 3.3, Serlea 111
72 s CHE 75 2 BV Oper. B Table 3.3, Serlea 113 and 119
60 s Falling 1500 2 Btf Oper. D Table 3.3, Serlea 128B and 130
33 D CHE 45 2 AH Tabla 3.4, Series 52
40 s CHE 45 2 AH Table 3.4, Serlea 61
66


The data include a range of cathead rope wraps from 1 to 4 turns
around the cathead. Earlier studies by Kovacs (1975, 1979, 1981, and
1982) have illustrated significant effect of number of wraps. As a
result of studies of number of cathead wraps, the ASTM SPT procedure
was revised to standardize testing at two wraps of the cathead. In
this study all rope and cathead data were performed with two wraps.
Measurements of Automatic Hammers
The only Automatic hammer system finding increased use in the
1980's was the hammer developed by Central Mine Equipment (CME)
company. Recently an automatic hammer has been introduced by Mobile
drilling company yet, as of 1990, the only measurements have been
performed by the manufacturer and have not been independently
verified. The CME hammer is illustrated in Figure 4.4. The hammer
works on a principle originally developed by the Corps of Engineers.
Outside of the enclosed sleeve, where the hammer moves, there is a
hydraulically driven chain which has a finger-cam which picks up the
hammer at the bottom of the sleeve and carries it to the top. At the
top, the hammer has upward velocity and the cam disengages. As a
result of the upward momentum, the hammer travels a distance higher
than where the hammer disengages the cam. The distance and therefore
drop height of the hammer depends on the velocity of the chain. The
speed of the chain is dependent on flow control adjustment of the
hydraulic motor used to drive the chain. The complete assembly is
67


ENERGY RATIO FOR F(t) = Ej/E* = ENTHRU = ERj
DRILL RIG MODEL q 3 q.4 0.5 0.6 0.7 0.8 0.9
Acker M-2 1 1 1 1 0 0 I 1 T
CME 45 0 8 0 0 0
CME 45B 0
CME 45C 0
CME 55 0 0 000 CO
CME 65 0
CME 75 0
CME 750 0
Failing 250 0
Failing 1500 0 0 00 @3
Joy B-12
Longyear 34 0
Longyear 38
Longyear HC-150 0
Mayhew 1000 0 SAFETY HAMMERS
Mobile B33 0
Mobile B34 0 0 Data corrected to represent maximum
Mobile B50 0 theoretical energy, i.e.
Mobile B61 00 0 for a 40 ft. (13 m)
Mobile B80 1 0 0 1 1 1 drill rod length 1 1
0.3 0.4 0.5 0.6 0.7 0.8 0.9
CORRECTED ENTHRU, ERj FOR Kc = 0948
COMPRESSIVE WAVE VELOCITY = 16,000 ft/s (4877 m/s)
Figure 4.2.
Summary of ER, data for safety hammers.
(Kovacs et al., 1983)
68


ENERGY RATIO FOR F(t) = Ej/E* = ENTHRU = ER|
DfllL RI6 MODEL 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Acker M-2 I I 1 1 1 1 1
CME45 o o o o
CME 45B
CME 45C
CME 55 o 0 o
CME 65
CME 75
CME 750
Failing 250
Failing 1500 o
Joy B~12 o
Longyear 34 o o
Longyear 38 o
Longyear HC-150 -
Mayhew 1000 DONUT HAMMERS
Mobile B33 Mobile B34 Data corrected to
represent maximum
Mobile B50 theoretical energy, i.e.
Mobile B61 o for a 40 ft. (13 mj
Mobile B80 drill rod length
0.3 0.4 0.5 0.6 0.7 0.8 0.9
CORRECTED ENTHRU, ERj FOR Kc = 0.948
COMPRESSIVE WAVE VELOCITY = 16,000 ft/s (4877 m/s)
Figure 4.3.
Summary of ER? data for donut hammers.
(Kovacs et al., 1983)
69


mounted on a hydraulic ram which is designed to apply 250 lb downward
force on the rods and follow the rods as the test progresses. Both
the CME automatic hammer and Vicksburg hydraulic hammer of the Corps
of Engineers were measured in this study and their operation will be
discussed in more detail in Chapters 5 and 6.
Dr. Riggs of CME performed an extensive series of energy
measurements comparing the automatic hammer to safety hammers. The
reports were published in the ASTM Geotechnical Testing Journal
(Riggs et al., 1983 and 1984). The energy measurements were
performed with the Binary Instruments device, strain gauge load cells
and travel times were not measured. His interests were of course
purely commercial. The automatic hammer has an impedance ratio
closer to 1.0 and the fact that it drops freely with no trailing
hoist rope results in higher energy transfer. CME showed through
both drop height and velocity data (Goble and Ruchti, 1981) that
safety hammers were normally dropped from heights greater than 30-
inches. This is possible since the hammer is designed to allow for 3
to 4-inch overstroke. The argument for higher energy in the safety
hammer was to result in only minor differences between the safety and
automatic hammer.
The first report by Riggs et al., 1983 is generally considered
unreliable by experts in stress wave energy testing. Riggs performed
comparisons between two Binary Instruments devices and showed
differences between the two. The comparison was invalidated since it
was discovered on of the devices was out of calibration. In this
70


Figure 4.
The CME automatic SPT hammer during performance of the
SPT. (Riggs et at., 1983)
71


report there was significant amount of ER1- data exceeding 100 percent
with values as high as 118 percent. In the testing the bottom 3 ft
of 47 ft deep boreholes were filled with gravels. The gravel with
high penetration resistance, may have reflected compressive waves
resulting in almost doubling of energy from the Binary Instruments
device. Since wave travel times were not monitored the results are
questionable.
The second report by Riggs et al., 1984 reported results of
comparisons between automatic and safety hammers in 14 drill holes.
The SPT was performed in uniform loess deposits with N value ranging
from 11 to 5. There was little difference between N values from the
safety and automatic hammer, generally less than 1 blow per foot.
Three series of energy measurements were performed on the automatic
hammer and four series on the safety hammer. Results were as follows;
Hammer Mean ER; % Standard Deviation ER;%
CME Automatic 88 1.1 to 1.3
CME Safety 80 5.6 to 8.8
The higher energy of the automatic hammer was also measured in a
series of measurements performed by Schmertmann for Florida
Department of Transportation (Schmertmann and Smith, 1977).
Schmertmann found an average of ER,. of 90 percent for the automatic
hammer. The higher energy of the CME safety hammer is felt to be a
result of higher drop height and solid steel guide rod. Other

72


manufacturer safety hammers may provide lower energies as reported by
Kovacs.
Energy Data from Other Countries
In addition to data collected in the U.S. attempts were made to
quantify differences in ER,- for hammer systems used in different
countries. Measurements were obtained for the Pilcon trip hammer
which is used extensively in the United Kingdom. Measurements of the
Pilcon hammer will be reported in this study. Due to the concern
with earthquake liquefaction data collected in Japan a series of
energy measurements were obtained there.
In 1983 a report was issued by the University of British
Columbia at Vancouver reporting results of drill rod energy
measurements obtained with donut, safety and Pilcon trip hammers
(Liang, 1983). The measurements were performed with the Binary
Instruments Device. The load cell type was not reported but it is
assumed that it was a strain gauge type. Measurements were performed
at two sites and hammer types were alternated. BW parallel drill
rods were used during testing. These drill rods are not typically
used in SPT testing and have more cross sectional area than NW upset
wall rods. The integration rate of the Binary Instruments device was
evaluated. It was concluded that the sampling rate of the Binary
Instruments device was sampling at a coarse rate of 0.23 msec and
there was some error between ER,. from the device and digital
73


integration. These conclusions are in direct conflict with those of
Kovacs, 1983.
The results of Liang's study summarized the range of ERf for
hammers as follows;
Pi Icon Trip Hammer = 57%
Safety Hammer = 59 to 62%
Donut Hammer = 43 to 48%
The energy measurements of the Pi Icon trip hammer were lower than
those reported by Douglas, et al., 1981. ER{ values from Douglas
ranged from 65 to 80 %. The values were also obtained with the
Binary Instruments device but details of testing were not provided.
The National Bureau of Standards in cooperation with several
U.S. government agencies and the government of Japan performed an
extensive series of energy measurements in Japan (Kovacs and
Salomone, 1984). The author participated in collection of data in
this study. The purpose of the study was to document ERf in Japanese
practice and to obtain measurements at actual soil liquefaction sites
at Akita and Niigata. Over 2,200 drill rod energy measurements were
taken at 15 sites. A total of 78 SPT tests were performed using 19
different testing conditions (equipment and operator conditions).
Measurements included hammer velocity and drill rod energy using the
Binary Instruments device. Waveforms were recorded on tape for later
processing yet this data was never analyzed. Piezoelectric load
cells were used for 37 SPT series in Akita until breakdown. The
remainder of testing was performed with strain gauge load cells.
74


Japan performs SPT in accordance with Japanese Industrial
Standards (JIS). The JIS standards require use of a donut hammer
whose dimensions are fixed. The impact anvil is rather small at 3 to
3.5-inches in outside diameter. Therefore, these donut hammers were
more efficient in energy transfer. Figure 4.5 shows a frequency
diagram of all data collected. The data include variation in cathead
wraps and drop method. Some data with free fall conditions using the
Tombi hook-drop method are included, yet the majority of data is by
rope and cathead method. The average energy for rope and cathead
method was 67 percent.
The author was impressed with an apparent change in data when
load cell type was switched. Energy data using the Piezoelectric
load cell was averaging approximately 70 percent in Akita. In
Niigata when strain gauge load cells were used, energy data was
averaging approximately 60 percent. Inconsistencies in data are
apparent on Figure 4.6 where the number of rope turns is evaluated.
As the number of rope turns increases the energy should drop due to
increased rope friction on the cathead. The opposite trend is
apparent on Figure 4.6. The data for 2-3/4 turns was collected in
Akita with piezoelectric load cell while the data for 3/4 turn was
collected in Niigata with strain gauge load cell. Since the hammer
and anvil dimensions are restricted by JIS standard the apparent
difference in load cell can be significant. The differences energy
may be as much as 10 percent. This possible difference should be
evaluated by systematic comparisons. The difference may be more
75


FREQUENCY OF OCCURRENCE
58
77.8

20
18
16 Ail Data Avg 68.3%
14 a 9.6%
12 CV 13.8%
10
8 -
8 -
4 -
2 -
0
30 40 50
8 88.3
* a
-1________I________I________L.
60 70 80 90
ERi <%)
Figure 4.5. Frequency diagram of ER- results Japan study.
(Kovacs, et al., 1984)
76


40
30
Tombi 3/4 1-3/4 2-3/4 3-3/4
NUMBER OF ROPE TURNS
AROUND CATHEAD
Figure 4.6. ERT- versus number of rope turns around the cathead
Japan study. (Kovacs et al., 1984)
77


significant for small rods such as the 40.5 mm rods used in Japan.
Recommendations for Energy Corrections
With collection of a significant amount of drill rod energy data
showing wide variation delivered in SPT recommendations were made to
correct N values. N values were shown to be inversely proportional
to drill rod stress wave energy in the study by Palacios, 1977. N
values could then be corrected according to the following
relationship;
Nc Nm ERim/ ERjc (4-1)
Where:
Nc = corrected blow count for the selected energy ratio ERic
Nm = measured blow count
ERim = measured drill rod stress wave energy ratio
ER^ = selected reference drill rod stress wave energy ratio
Using this correction, data using different hammer systems could be
compared on an equal basis.
Kovacs in his 1983 study made recommendations for a National
Average Energy value for the U.S. A value for the reference drill
rod stress wave energy ratio of 55 percent was recommended by
combining the data in Table 4.1 with a weighting system based on
drill rig model usage. This value of reference energy is in
agreement with recommendations by Schmertmann and Smith, 1977.
78


After the evaluation of energy data in Japan, Seed et al., 1984
reevaluated the field penetration resistance data base on earthquake
liquefaction. SPT energy data from various countries was summarized
as shown on Table 4.2. The data for China's trip hammer was
estimated based on comparison with the Pilcon hammer (Douglas and
Strutznksy, 1984) and not by direct measurement. It should be
pointed out that there is a wide variety of trip hammers in use in
China. A report by Shi-Ming, 1982 shows several hammers with extreme
difference in anvils which significantly alter energy transfer. The
values for Argentina were estimated. Seed argued to set ERic at 60
percent to avoid necessity for correcting U.S. safety hammers which
are in predominant use in the U.S. The use of 60 percent also
requires no correction for the United Kingdom. The liquefaction
resistance chart boundary lines were adjusted based on reevaluation
of the data base corrected to 60 percent reference energy using
Equation 4.1. Since that time it has been accepted to use ERic=60%.
79


Table 4. Summary of Energy Ratios for SPT Procedures
(Seed et al., 1984)
Country Hammer type Hammer release Estimated rod energy (%) Correction factor for 60% rod energy
Japan' Donut Free-fall 78 78/60 = 1.30
Donut2 Rope and pulley with special throw release 67 67/60= 1.12
United States Safety2 Rope and pulley 60 60/60 = 1.00
Donut Rope and pulley 45 45/60 = 0.75
Argentina Donut2 Rope and pulley 45 45/60 = 0.75
China Donut2 Free-fall3 60 60/60 = 1.00
Donut Rope and pulley 50 50/60 = 0.83
1 Japanese SPT results have additional correction for borehole diameter and frequency
effects.
2 Prevalent method in this country today.
3 Pilcon-type hammers develop an energy ratio of about 60 percent.
80


CHAPTER 5
PRESENTATION OF FIELD DATA
5.1 Summary of Testing Program
The field data was collected over a time period from January 1986 to
August 1987 while the Binary Instruments device was being used by the
Bureau of Reclamation. Reclamation was using the instrument to
determine energy transmission of varying drilling rigs during
investigations of dams subject to soil liquefaction. The data was
required to correct site penetration resistance values (Seed et al.,
1984). During this time period the author performed or directed the
testing program. In cases where the instruments remained at a
project for periodic monitoring other personnel were trained in
operations and continued to perform tests.
The testing program was not specifically designed to answer
questions on energy testing but was actually driven by project need.
On some projects the data was required for immediate data
interpretation which put added pressure on the measurement
reliability. For example the data collected at Jackson Lake Dam was
required to monitor the effectiveness of a multimillion dollar ground
improvement program. The following projects and general goals of the
program are summarized below;
81


Project
Testing Goals
Corp of Engineers Study
Initial Tests Pilcon Hammer
Check on Binary Equipment
Confirm ER1- of Special SPT
Hammers
Jackson Lake Safety Hammers ER- of Ground Improvement Stage I
Contractor.
Jackson Lake Auto Hammers
ER,. of Ground Improvement Stage
II Contractor
ER,. of Specific Drill Rig Model
Gus Pech Drill Rig
Since the testing program was not designed specifically for this
thesis, the field data will be presented and discussed in
chronological order. For each major project the goals and results of
the program will be summarized. Major data trends will be summarized
in Chapter 6. Any difficulties with data collection will be reviewed
along with any important lessons learned.
Field data was collected on data sheets which transcribed in to
a computer spreadsheet in this report. In all cases we attempted to
conform with the ASTM standard D-4633 and all exceptions to the
procedure are noted.
Over 60 series of SPT energy measurements were obtained in this
report. With the thousands of data points collected a consistent
5.1 Format of Data Presentation
82


method for reporting data was developed. Each series of tests
consists of collecting a number of blows of data ranging up to 40
blows. In some cases the data is collected during the performance of
SPT over the 1.5 ft. drive interval but in other cases the sampler
was overdriven in excess of 1.5 ft.. Data from the Binary Instruments
device is collected by operator in written form on two data sheets.
An example xerox copy of the field data sheets are shown in appendix
B. The first sheet provides information on the testing arrangements
such as the SPT equipment and Binary Instruments settings. The
second sheet contains a blow by blow record of the ERi value and
travel time. A major deficiency to the Binary Instruments device is
the lack of hard copy record. For that reason a portable hand held
tape recorder is used to check records during testing. After a
series of testing is complete the tape is replayed and data is
checked and verified. Field data sheets remain the permanent record
of the test. The data reported in this study were analyzed
previously to provide design recommendations. Normally the average
ERi value and travel time were determined for multiplication by
correction factors Kt, K2, and Kc (Equations 3.4, 3.2 and 3.9).
For this analysis data was transcribed from the field data
sheets to computer spreadsheet format into a template which
facilitated statistical analysis and summary graphing. The computer
spreadsheet used was Lotus 123 version 2.0 for IBM compatible
personal computers. A standard template was developed which .
simulates the field data sheet but includes automatic calculation of
83


correction factors and statistical evaluation of data. The range of
acceptable wave travel times is automatically determined and data is
screened for validity in accordance with the ASTM procedure. A group
of cells was reserved for frequency distribution analysis with
automatic summary graphing capability. This allowed for easy summary
graphing of important trends in any series.
A complete transcription of all data and analysis is attached in
appendix C for all data collected. An inspection of the data shows
the basic two sheet format used in the field. The first sheet
contains information on the drilling equipment including data on
cathead, rope, hammer and drill rods. Information on the Binary
Instruments load cell, Ecal and Fcal settings, and zero load setting
is also included. Based on the lengths from impact surface to load
cell and total length of drill rods from impact surface to sampler
the wave travel time is computed assuming c=16,800 ft/sec. A
comments area on the first sheet notes any unusual occurrences during
testing.
On the second sheet, the ERi and wave travel time data, if
obtained, are tabulated. If the wave travel time falls within the
acceptable range the calculated speed of wave travel is given and the
Kc factor is computed for each blow of data. Three columns of ERi
data are shown on the right hand portion of the sheet, ERi with
unacceptable travel time, ERi corrected for factors K1 and K2 only,
and ERi corrected for Kv K2, and Kc. The factors K1 and K2 were not
calculated in accordance with Equations 3.2 and 3.4. The values were
84


estimated from the tables in the ASTM procedure. This introduces
slight error since the factors are a function of the mass of hammer
and area of drill rod. Each blow of data is corrected and
statistical analysis of mean, standard deviation and coefficient of
variation for the series are summarized at the bottom of the table.
Data distribution is summarized by a series of graphs which follow
each series. It is important to note that the frequency distribution
graphs show the number of occurrences less than or equal to the value
shown on the graph.
In order to provide a basic summary of the testing and easier
reference for presentation of field testing Table 5.1 was developed.
Table 5.1 provides a summary of test results with basic information
on each test series configuration and statistical summary of ERi test
results. The following discussion will provide a chronological
review of the testing along with additional summary graphs and
figures to stress important findings. In the following discussions,
when stress wave energy ERi is mentioned without reference to
correction factor it will be data with acceptable travel times but
not corrected for Kc.
85


Table 5.1. Summary of Energy Measurments Performed
ERi FOR TIMES LESS THAN 2L'/C ERi GOOD TIMES Kl/<2 CORRECTED
SERIES DATE HAMMER DROP METHOD DRILL RIG DRILL ROD NUMBER DATA PTS ERi AVG. STND. DEV.COEFF. S.D. VARIA.% tRANGE NUMBER msec PTS. ERi AVG. STND. DEV. S.D. COEFF. VARIA.% NUMBER PTS. COMMENTS
1 1/31/86 PILCON TRIP S&H HOLEMASTER AW 25 45 7 16 1.1 6 56 2 4 6
2 1/31/86 PILCON TRIP S&H HOLEMASTER AW 31 51 6 12 1.4 7 70 10 15 22
3 1/31/86 PILCON TRIP S&H HOLEMASTER AW 28 51 4 7 1.4 7 70 5 7 21
4A 2/21/86 PILCON TRIP FAILING 1500S AW 47 67 6 9 4.6 12 65 6 9 17
4B 2/21/86 SAFETY ROPE/CAT FAILING 1500S AW 44 - - - - - 74 9 13 12
5 3/19/88 DAMCO #2 TRIP FAILING 1500 N 14 - - - - - 78 6 7 14
6 3/19/86 DAMCO #2 TRIP FAILING 1500 N 37 75 7 9 4.3 11 83 5 6 26
8 3/19/86 DAMCO #2 TRIP FAILING 1500 N 40 79 8 10 4.3 5 81 5 6 35
g 3/19/86 DAMCO #3 TRIP FAILING 1500 N 6 - - - - - 60 4 7 18 MSMTCH ANVt. DIRTY GUIDE ROO
10 3/19/86 DAMCO #3 TRIP FAILING 1500 N 38 68 6 10 1 TO 5 20 61 4 7 18 DIRTY GUIDE ROD
11 3/19/86 SOILTEST TRIP FAILING 1500 N 36 73 5 7 1 TO 5 34 77 - - 2
12 3/19/86 WES AUTO HYD. TRIP FAILING 1500 N 38 94 2 2 2 TO 6 38 - - - -
13 3/19/86 WES AUTO HYD. TRIP FAILING 1500 N 38 98 4 4 3.6 38 - - - -
14 3*20/86 DAMCO #2 TRIP FAILING 1500 N 26 65 7 11 1 TO 7 22 66 - - 4
15 3/20/86 DAMCO #2 TRIP FAILING 1500 N 38 61 11 18 1 TO 4 11 70 3 5 27
16 3*20/88 SOILTEST TRIP FAILING 1500 N 37 75 5 6 1 TO 6 34 78 - - 3
17 3*20*86 DAMCO #2 TRIP FAILING 1500 N 36 64 5 7 2 TO 6 27 70 2 3 9
18 3/20/86 DAMCO #3 TRIP FAILING 1500 N 36 48 9 19 1 TO 8 13 66 5 7 1 2 REFLECTED COA/FRES -SJVE WAVES
19 3*20*86 SOILTEST TRIP FAILING 1500 N 35 78 5 6 2 TO 6 32 82 3
21 7/18/88 SAFETY ROPE/CAT C ME-550 NW M 37 - - - - - 72 3 5 35
86


Table 5.1. (Contd.) Summary of Energy Measurments Performed
ERiFOR TIMES LESS THAN a_7C ERr GOOD TIMES K1 X2 CORRECTED COMMENTS
HAMMER DROP METHOD DRILL RIG DRILL NUMBER ROD DATA PTS ERI AVG. STND. DEV.COEFF. S.D. VARIA. t RANGE msec NUMBER PTS. ERi AVG. STND. DEV. S.D. COEFF. VARIA. NUMBER PTS.
22 7/1986 SAFETY ROPE/CAT CME-550 NW-M 40 - - - - - 80 3 3 40
23 7/1988 SAFETY ROPE/CAT DIEDRICH D-50 NW M 19 - - - - - 67 2 3 19
29 91986 SAFETY ROPE/CAT MOBILE B-40 NW ML 2-1 83 3 3 3 TO 6 19 85 2 2 5USBR
30 91986 SAFETY ROPE/CAT MOBILE B-40 NW-M. 38 92 3 3 2 TO 6 3i 93 4 4 7USBR
31 918186 SAFETY ROPE/CAT CME-550 NW ML 22 - - - - - 75 5 7 22
32 9/9/86 SAFETY ROPE/CAT CME-550 NW M. 29 44 - - 1.8 1 69 5 7 28
34 92986 SAFETY ROPE/CAT MOBLE B-80 NW-M. 37 70 3 6 1.8 37 - - - USBR
30A 4/22/87 SAFETY ROPE/CAT GUS PECH 22R NW-M 14 65 8 12 1 TO 3 14 - - - -
32A 4/22/87 SAFETY ROPE/CAT GUS PECH 22R NW ML 16 70 3 4 3.1 16 - - - -
33A 4/22/87 SAFETY ROPE/CAT GUS PECH 22R NW-M 28 70 3 4 3.1 28 - - - -
35 91987 CME AUTO HYD. TRIP CME-55 NW-M. 13 104 3 3 2.7 8 108 - - 4
36 5/1987 CME AUTO HYD. TRIP CME55 NW ML 19 88 0 0 2 4 97 3 3 15
37 91987 CME AUTO HYD. TRIP CME-55 NW ML 23 124 10 8 3.2 15 - - - -
39 926/87 CME AUTO HYD. TRIP CME 750 NW-M. 23 104 1 1 2.5 23 - - - -
0 1 CME AUTO HYD. TRIP CME-55 NW ML 27 117 2 1 2.5 27 - - - -
41 92987 CME AUTO HYD. TRIP CME 750 NW ML 18 122 1 1 3.6 6 122 1 1 12
42 92987 CME AUTO HYD. TRIP CME-750 NW ML 19 99 4 4 1.7-2.9 19 - - - -
43 92987 CME AUTO HYD. TRIP CME-750 NW ML 22 112 3 3 3.3 22 - - - -
45 92987 CME AUTO HYD. TRIP CME-55 NW ML 40 109 6 6 2 TO 5 40 - - - -
46 92/87 CME AUTO HYD. TRIP CME-750 NW ML 11 - - - - - 90 1 2 11 USBR. W/O SPACER
87