Development of Rapid TDR Measurement Capability for Mobile ... · Development of Rapid TDR...
Transcript of Development of Rapid TDR Measurement Capability for Mobile ... · Development of Rapid TDR...
WSSS Annual Meeting, Park City, UT June 19-21, 20061
Development of Rapid TDR Measurement Capability for Mobile Water Content Mapping
Scott Jones, Utah State UniversityDavid Robinson, Stanford University
Shmulik P. Friedman, Volcani Institute, ARO, Israel
WSSS Annual Meeting, Park City, UT June 19-21, 20062
On-The-Go Mapping
• Precision Agriculture – Variable Rate Technology– It is especially important to develop and implement
sensor technology for those parameters, such as soil nitrate and soil moisture, which can change rapidly (both spatially and temporally) and must be measured in real-time or near real-time to be useful for input control.
• K.A. Sudduth, http://mpac.missouri.edu/pubs/eng_tech.pdf
s
nv
wb ECECEC +⎟⎟⎠
⎞⎜⎜⎝
⎛=
φθ
Mapped
Unknown Texture
WSSS Annual Meeting, Park City, UT June 19-21, 20063
Project Rationale• Soil water content measurements are needed for on-the-go
agricultural, engineering and environmental activities.– Plowing, planting, reclamation, management, irrigation– Road base materials, dam materials– Hydrological studies, ecological work
• We propose an invasive TDR-based water content probe– provide robust direct permittivity measurements– Capacitance probes previously suggested– Short TDR probes have been proposed– An open-ended probe has advantages
Blade w/ open-ended TDR probes/ and temperature sensors
Depth Sensor
Tractor/Implement
Hand-heldPC
EMI
GPS
Datalogger
TDR
Blade w/ open-ended TDR probes/ and temperature sensors
Depth Sensor
Tractor/Implement
Hand-heldPC
EMI
GPS
Datalogger
TDR
WSSS Annual Meeting, Park City, UT June 19-21, 20064
TDR Travel-Time Analysis
Travel time [ns]0 10 20 30 40 160 170
Ref
lect
ion
Coe
ffici
ent
-1.0
-0.5
0.0
0.5
1.0
ECw = 0 S m-1
0.1 S m-1
0.6 S m-1
t1 t2
V0 Vf
212
2)(⎟⎠⎞
⎜⎝⎛ −
=L
ttcbε
⎟⎟⎠
⎞⎜⎜⎝
⎛−= 12 0
fc VV
ZKEC
PermittivityPermittivity
WSSS Annual Meeting, Park City, UT June 19-21, 20065
Open-Ended Dielectric Probe
• HP – Network Analyzer– Dielectric Probe Kit
• Frequency-dependent permittivity– 300 kHz – 6 GHz
• Frequency domain advantages– Relaxation phenomena– Fixed frequency
Blade (-) (+) (-)
Soil
Blade (-) (+) (-)
Soil
WSSS Annual Meeting, Park City, UT June 19-21, 20066
Open-ended Probe Waveform Analysis
• Bilinear function (Berberian, 1993; Cole et al, 1989)– Laplace transform– Compute admittance values
• Sample, cell, coaxial line, etc.
– Reflection coefficient related to• Complex permittivity• Bilinear coefficients
– f (reference dielectric, TL admittance, connectors)
– Bilinear coefficients are determined from:• Short, open and 50 ohm termination • Or three known references (water, alcohol, air)
Castiglione and WraithMontana State Univ.
WSSS Annual Meeting, Park City, UT June 19-21, 20067
Probe 1- Near Surface Probe
• UHF to BNC adapter• UHF end machined flush• One-sided measurements
WSSS Annual Meeting, Park City, UT June 19-21, 20068
Testing Effects of Probe Travel Velocity
• Porous media– Dry, moist, saturated sand
• Travel Speeds – 0 to 5 mph
WSSS Annual Meeting, Park City, UT June 19-21, 20069
Open-Ended Probe Fringing Field
Blade (-) (+) (-)
Soil
Soil
Blade (-) (+) (-)
Soil
Soil
• Modeled using– ATLC (Kirkby, 1996)– Plots energy density– Darker regions
• Higher field density• Greater influence on permittivity
WSSS Annual Meeting, Park City, UT June 19-21, 200610
Prototype Probe 2 – Single-depth Probe
• Custom-made– Dual sided
• RG-178 cable• 0.66 cm dia. Rod
– Inner conductor
• 1 cm wide blade– Shield
1.55 cm
0.66 cm
1.55 cm
0.66 cm
WSSS Annual Meeting, Park City, UT June 19-21, 200611
Field Testing of Probe Travel Velocity
Open-endedprobe in blade
a) Open-endedprobe in plate
Open-endedprobe in blade
a) Open-endedprobe in plate
-0.10
0.10.20.30.40.50.60.70.80.9
1 51 101 151 201Waveform points
Ref
lect
ion
coef
ficie
nt
0m12 m24 m36 m48 m
Open-ended probe12 m spacing5 km/h
Wettedregion
0.15
0.2
0.25
0.3
0.35
0.4
0 20 40 60 80 100
Transect distance (m)
Wat
er c
onte
nt e
stim
ate
b)
c)
4 km/h
-0.10
0.10.20.30.40.50.60.70.80.9
1 51 101 151 201Waveform points
Ref
lect
ion
coef
ficie
nt
0m12 m24 m36 m48 m
Open-ended probe12 m spacing5 km/h
Wettedregion
0.15
0.2
0.25
0.3
0.35
0.4
0 20 40 60 80 100
Transect distance (m)
Wat
er c
onte
nt e
stim
ate
b)
c)
4 km/h
• Waveform consistency• Wet and dry
– Travel-time analysis– Relative water content
WSSS Annual Meeting, Park City, UT June 19-21, 200612
On-The-Go Probe in Action
• 12 kph• Single blade• Probe at 15 cm
WSSS Annual Meeting, Park City, UT June 19-21, 200613
Prototype Probe 3 – Multi-depth Probe
• 3 – BNC Tee’s– Dual sided
• RG-58 cable• Epoxied in place then machined flush• 1 cm wide steel blade
Probe 3
10 cm
0.78 cm
0.23 cm
Probe 3
10 cm
0.78 cm
0.23 cm
WSSS Annual Meeting, Park City, UT June 19-21, 200614
Multi-depth Probe Results
• 9 reps
• Water– static
• Soil– θv ~ F.C.– 3.2 km/h
Water Soil Blade
Waveform pixels
Ref
lect
ion
coef
ficie
nt5 cm
15
25
RMSE = 0.012
0.010
0.019
0.035
0.019
0.045
WSSS Annual Meeting, Park City, UT June 19-21, 200615
Probe/Cable Attenuation
• 1 UHF• 2 Custom• 3 BNC T• 4 UHF T
Probe 2Probe 1
Probe 3b
1 MHz 10 100 1000 1 MHz 10 100 1000
1 MHz 10 100 1000 1 MHz 10 100 1000
UHF Tee
High attenuation
Low attenuation
NA Probe - Reference
WSSS Annual Meeting, Park City, UT June 19-21, 200616
Summary
• On-the-go water content measurements would add valuable GIS information to many mobile activities.
• An open-ended design is the most feasible probe configuration but requires an advanced analysis approach that is being developed for TDR.
• Hardware testing indicate high quality cable and probe can reduce losses to levels observed in a network analyzer.
• Ongoing work is looking to provide real-time water content determination within the next few years.