Transcript pressure
P- and S-wave velocities in rock as
a function of pressure and
temperature
I. Lassila 1,T. Elbra 2, E. Hæggström1 and L. J. Pesonen 2
V. Kananen1 and M. Perä
J. Haapalainen1 and R. Lehtiniemi 3
P. Heikkinen 4 and I. Kukkonen 5
1
Electronics Research Unit
2
Division of Geophysics
3
Nokia Research Center
4 Institute
of Seismology
5 Geological
Survey of Finland
Motivation - Understanding the structure of the
earth’s crust
FIRE (Finnish Reflection Experiment) - project
Seismic reflection and refraction measurements (longitudinal
and shear wave modes)
Photo: Seismic signal is produced by vibrators. Courtesy Jukka Yliniemi.
7.12.2007 Ilkka Lassila
Location of the FIRE reflection seismic lines.
TOF and depth
Seismic measurements give TOF data
Need to know Vp and Vs to calibrate the depth
Example of FIRE results from the end of line FIRE 3A in western Finland. The reflector amplitudes of a migrated section are presented as gray tone intensities.
7.12.2007 Ilkka Lassila
Samples
Outokumpu Deep Drilling Project (2516 m)
7.12.2007 Ilkka Lassila
Device: requirements
Vp and Vs measurements
preferably
10
simultanously
m/s accuracy
Controlled pressure
0
- 300 MPa (15 ton for OKU samples)
Controlled temperature
20-300ºC
Preferably
automatic
22 mm
Data acquisition
25 mm
7.12.2007 Ilkka Lassila
Possible measurement setups
Uniaxial
Multianvil
Hydrostatic pressure
7.12.2007 Ilkka Lassila
Timetable
Jan
Feb
Mar
Apr
May
Jun
Jul
Material considerations
Mechanical design
Ultrasonic testing and designing
Transducers, pulser / signal generator, amplifiers, switches, oscilloscope
Pressure generating
Heating
Pressure monitoring
Temperature monitoring
Transducer cooling
Ordering parts
Planning the measurement procedure
Assembling the setup
Programming the DAQ software
7.12.2007 Ilkka Lassila
Validation
Device: Vp and Vs
Pitch-catch method
Two similar transducers, both comprising shear (1,1 MHz)
and longitudinal (1 MHz) piezo (Pz-27) ceramics
At first only the shear crystal was in use
Longitudinal
Caused
mode well present
by silver epoxy?
Removable delay lines
Fused
quartz
Brass
Water cooling
No load over the piezo crystal
7.12.2007 Ilkka Lassila
Device: pressure simulation
Pressure simulations by Mr. Haapalainen
Device
Fused
can withstand the required pressure
quartz can be used as a delay line material in case of
no roughness
7.12.2007 Ilkka Lassila
Device: pressure
Generating: 15 ton jack borrowed from Department of Chemistry
Measuring: Sensotec Model 53 (max 23 ton) + Lebow 7528 amplifier
7.12.2007 Ilkka Lassila
Device: pressure
Problem with sample durability
Solved with a brass jacket
Splitting sample holder allows sample removal after
compression
7.12.2007 Ilkka Lassila
Device: Temperature, simulations
Thermal simulations by PhD Lehtiniemi and Mr.
Haapalainen
160
W heater is sufficient for 300ºC in case of fused quartz
delay lines
Transducer
temperature stays below solder melting / epoxy
softening temperature
7.12.2007 Ilkka Lassila
Device: Temperature
Heating: Nozzle heater ACIM T197 (160 W / 240 Vac)
Max
400ºC
Covers
the sample holder
Cooling: Water cooler (Lauda WK502)
Measuring: Custom AD595 based thermocouple amplifier
K-type
7.12.2007 Ilkka Lassila
Thermocouple inside the sample holder
Device: Data acquisition
US signals:
5072
PR, LeCroy 9410, GPIB, PC, LabVIEW, Matlab
Thermocouple and load cell:
AD-conversion
7.12.2007 Ilkka Lassila
and transfer to PC with NI PCI-6024E
Device
Transducers
Delay lines
Heating element and sample
Thermocouple
Load cell
Water cooling tubes
Jack
7.12.2007 Ilkka Lassila
Preliminary results
7 samples from Outokumpu Deep Drill Core
T: 300ºC20ºC, Load: 7000 kg 500 kg (resembling the
conditions in the Earth’s crust)
Results comparable with literature values
4000
5500
Vs(m/s)
Vp(m/s)
6000
5000
4500
4000
0
7.12.2007 Ilkka Lassila
3500
3000
0.5
1
1.5
T*load(ºC kg)
2
6
x 10
0
0.5
1
1.5
T*load(ºC kg)
2
6
x 10
Pressure test
Pressure test without sample
49.3
the compression of the
sample?
Compression = 0,1 mm
(Δhsample- Δhno sample)
Error Vp = 24-33 m/s
Error Vs = 15-18 m/s
measured height (mm)
The error if we don’t measure
49.2
49.1
49
1000 kg
13000 kg
48.9
48.8
48.7
48.6
Pressure test with sample
71.3
measured height (mm)
71.2
71.1
71
70.9
70.8
70.7
70.6
70.5
7.12.2007 Ilkka Lassila
1000 kg
13000 kg
TOF (time of flight) through the delay lines
Pulse-echo measurement of the delay
line
Subtraction of the TOF through the
delay lines from the total TOF
Pressure and temperature effects to
the delay lines and transducers are
cancelled
7.12.2007 Ilkka Lassila
Damping the transducers
Ringing of the piezo element makes pulse-echo (PE)
measurements difficult.
Ringing can be reduced with applying attenuating,
material with acoustic impedance close to the piezo to the
back side of the transducer
PE responses to water load
a)
zero backing, b) backing of crown glass, c) backing of
tungsten-epoxy, d) backing of material with Z=Ztransducer
7.12.2007 Ilkka Lassila
Egypt. J. Sol., Vol. (23), No. (2), (2000)
Damping test - ok
Reduced ringing time and increased bandwidth
7.12.2007 Ilkka Lassila
Transducers without backing
7.12.2007 Ilkka Lassila
Outcome of applying the backing
No signal
Resistance between transducer electrodes ca. 5 Ω
Short-circuit
Difference between test
Amount
of tungsten in the mixture was higher
- In the test the resistance between the electrodes was ca.
500 Ω
This type of backing method requires isolation of the
electrodes
Instead of scraping out the backing it was decided to build
new transducers
7.12.2007 Ilkka Lassila
New transducers
Increased sample size:
Height
20-70 mm
Diameter
25-62 mm
Better modal purity required
Mode
conversion in the gap between transducer housing
and delay line
Material:
No
7.12.2007 Ilkka Lassila
stainless steel
separate delay lines
New transducer drawings
7.12.2007 Ilkka Lassila
New transducer
7.12.2007 Ilkka Lassila
New thermal simulations
T=?
T=10ºC
Stainless steel:
thermal conductivity=20 W/(m K)
Specific heat=500J/(kg K)
Sample (rock):
thermal conductivity=2 W/(m K)
Specific heat=790J/(kg K)
7.12.2007 Ilkka Lassila
h = 20-70mm
T(t=0)=350ºC
D = 25-62 mm
Temperature as a function of time in the middle of the
sample and on the transducer inner surface where the
piezos are fixed.
Sample D = 25,5 mm, h = 24 mm
7.12.2007 Ilkka Lassila
Sample D = 62 mm, h = 70 mm
Temperature distribution in the sample and the upper
transducer
Sample D = 25,5 mm, h = 24 mm
t = 200s.
7.12.2007 Ilkka Lassila
Sample D = 62 mm, h = 70 mm,
t = 400s.
Other updates
PC controlled pressure generation
Separate heating of samples to increase the throughput
rate
7.12.2007 Ilkka Lassila
New frame
Compressed air controlled one way
hydraulic cylinder replaced with
electric motor controlled two way
hydraulic cylinder
7.12.2007 Ilkka Lassila
Modification for hydraulic control
Controls of the pump replaced with relay circuit that
is controlled from PC DAQ-card
Two valves that are controlled
Valve
1 open increasing pressure
Valve
2 open decreasing pressure
Valves
7.12.2007 Ilkka Lassila
closed no change
Testing new hydraulics
Pressure increase at 0,1 s intervals
OK for loads over 3000 kg
9000
8000
load (kg)
7000
6000
5000
4000
3000
2000
1000
0
0
0,5
1
1,5
throttle time (s)
7.12.2007 Ilkka Lassila
2
2,5
3
Testing of new hydraulics
Pressure decrease at 0,1 s intervals
No control of outcome when decreasing pressure
14000
12000
load (kg)
10000
8000
6000
4000
2000
0
-2000
0
0,05
0,1
0,15
throttle time (s)
7.12.2007 Ilkka Lassila
0,2
0,25
More control needed
Manual shut off valve, needle type control
Slows down the flow of the hydraulic oil
7.12.2007 Ilkka Lassila
Control achieved
Needle valve can be adjusted to allow precise control
of the load
7,00E+03
8,00E+03
7,00E+03
5,00E+03
4,00E+03
3,00E+03
2,00E+03
target
load
5,00E+03
load (kg)
load (kg)
6,00E+03
6,00E+03
measured
load
4,00E+03
3,00E+03
2,00E+03
1,00E+03
1,00E+03
0,00E+00
0,00E+00
7.12.2007 Ilkka Lassila
measured
load
target
load
Measurement diagram
7.12.2007 Ilkka Lassila
Conclusions
Device is used for measuring Vp and Vs values that are
needed to interpret seismic data
Preliminary results ok
At the moment system is going through some changes
7.12.2007 Ilkka Lassila
Future tasks
Temperature inside the sample vs. on the sample surface
Validation tests
Implement a LVDT/gauge to measure the sample
thickness and thickness change inline
Licentiate thesis
7.12.2007 Ilkka Lassila