Directional Borehole Radar Tests
of an Oil Injection Experiment at
the Colorado School of Mines,
Golden, Colorado
by Jared Abraham 1, Craig Moulton1, and Philip J. Brown II 1
Open-File Report 02-63
2002
This report is preliminary and has not been reviewed for conformity with U.S. Geological Survey editorial standards or with the North American Stratigraphic Code. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.
U.S. DEPARTMENT OF THE INTERIOR
U.S. GEOLOGICAL SURVEY
1
Denver, ColoradoCONTENTS
In October 2001, the U.S. Geological Survey conducted borehole radar surveys of an oil injection experiment at the Colorado School of Mines (CSM), in Golden Colorado using the prototype U.S. Geological Survey (USGS)-developed directional borehole radar system (DBOR). A explanation of the system can be found in Wright and others (2001). The USGS was invited to the CSM to deploy the prototype directional borehole radar system during an oil injection experiment conducted to investigate the applicability of radar to monitoring formation invasion from a horizontal borehole. This work was conducted by a student at the CSM and is summarized in Moita (2001). The purpose of this report is to release the data and to summarize the experiments conducted with the DBOR system. This report contains, (1) a description of the system as deployed in the experiments, (2) a description of the data collected and data parameters used, (3) a simple display of some of the data collected, and (4) a description of the DBOR data files.
USGS DBOR system was deployed at the CSM oil injection experiment. The system used was a modification of the system developed by the USGS for use in boreholes.
The modifications were:
The system used for well logging could not be used as is because the outer fiberglass tube, which housed the inner fiberglass tube holding the antennas, had a diameter too large to fit in the hole of the experimental sand tank. The inner tube was placed in the hole, with the motor drive section on one side and the radar pulser drive on the other. Small tables on either side of the tank were used to support the different sections. The TX and RX antennas were visually centered to be in the middle of the sand tank hole. A block diagram of the system is shown in Figure 1.
Figure 1. Block diagram of the USGS DBOR.
The microprocessor-controlled motor/encoder in the tool positioned the antennas azimuthally. The radar signal was digitized by a 6 GHz bandwidth digital oscilloscope. A LabVIEW™ program written specifically for this system controlled tool operation, data acquisition, and real-time data display. Post-acquisition data visualization and analysis were accomplished using another program written in Research Systems Incorporated (RSI), Interactive Data Language (IDL).
Mechanical
The radar system uses identical directional cavity-backed monopole transmitting and receiving antennas that can be mechanically rotated. A geared stepper motor/encoder system in the tool provides the ability to rotate and position the paired antennas to a commanded angle, accurate to a few degrees. The paired rotating antennas are connected to their stationary pulsing and receiving electronic circuitry through rotary coaxial connectors.
Electrical and Software
The stepper motor used to rotate the antenna is controlled by a microprocessor in the tool, using an encoder for feedback of the antenna’s angular position. The microprocessor receives positioning commands from a software control program operating on a separate PC.
The radar pulses sent to the transmitting antenna are generated from a Power Spectra PGS405 device, which is triggered from a free-running clock oscillator circuit at an 20 kHz rate. The receiving antenna signal passes through a programmable attenuator and RF amplifier before being sent to the Tektronics TDS 820 sampling oscilloscope (6 GHz bandwidth), where it is averaged a user-selectable number of times. The TDS 820 is also triggered by the same signal used to trigger the PGS405. The adjustable delay on the scope is set to allow for the signal propagation delays. The TDS 820’s waveform data is sent over a GPIB interface to a PC. The PC, running a LabVIEW™ program, stores the waveform and displays it as a colorized intensity graph. The program also performs sequencing tasks during each cycle such as positioning the antenna, waiting for move confirmation, and arming the scope to acquire data.
The DBOR tool was placed in the sand tank model with the zero degree reference point pointing directly up at the top of the horizontal borehole as described in Moita (2001) (
Figure 2). All of the Data files collected with the USGS DBOR system at the Colorado School of Mines oil injection experiment of 10-03-01 and 10-04-01 are summarized in Table 1. Note that all data are for a single depth, 0.000 m, because the tool was not moved relative to the interior of the CSM tank. Some of the files are repeat surveys. These repeat surveys can be used for averaging and verifying system stability (see Table 1). The number in the file extension (see Appendix A) denotes which repeat survey-multiple. During data collection the tool rotated unintentionally in the mounting bracket. This rotation was estimated at the beginning of each run and is indicated by the degrees offset in Table 1, and should be removed during data processing. Additionally, all electronic setup information was logged in the headers per the acquisition software version 10-03-01 (see Appendix A) except the presence of an external RF amplifier which supplied 10 dB of gain. The following table summarizes the collected data files.
Table 1. Data file collected during the CSM oil injection experiment.
|
File Name |
Pre- or Post- injection |
Time after injection (hr) |
Length of records (ns) |
Sample interval (ΔT) (ps) |
External Amplifier (dB) |
Angle offset in degrees positive direction (clockwise) |
|
Pre |
NA |
20 ns |
40 ps |
10 |
0.0° |
|
|
Pre |
NA |
40 ns |
40 ps |
10 |
0.0° |
|
|
Post |
1.0 hr |
40 ns |
40 ps |
10 |
15.0° |
|
|
Post |
1.0 hr |
20 ns |
40 ps |
10 |
15.0° |
|
|
Post |
1.0 hr |
10 ns |
20 ps |
10 |
15.0° |
|
|
Post |
3.0 hr |
10 ns |
20 ps |
10 |
30.0° |
|
|
Post |
3.0 hr |
20 ns |
40 ps |
10 |
30.0° |
|
|
Post |
3.0 hr |
40 ns |
40 ps |
10 |
30.0° |
|
|
Post |
6.0 hr |
10 ns |
20 ps |
10 |
15.0° |
|
|
Post |
6.0 hr |
20 ns |
40 ps |
10 |
15.0° |
|
|
Post |
6.0 hr |
40 ns |
40 ps |
10 |
15.0° |
|
|
Post |
20.0 hr |
10 ns |
20 ps |
10 |
5.0° |
|
|
Post |
20.0 hr |
20 ns |
40 ps |
10 |
5.0° |
|
|
Post |
20.0 hr |
40 ns |
40 ps |
10 |
5.0° |
* file extensions(.100-.103), ** file extensions (.100-.102)
The data collected using an 20 ps sample interval and an 10 ns record length( file:
csmpost5, csmposat6, csmpost9, and csmpost12) were selected for plotting. This plot is to verify that the DBOR system had detected the movement of the oil injection. Figure 3 is an azimuthal plot of the data with background subtraction. There appears to be a shift or other system change that occurred as seen in the three hour injection plot (file csmpost6.100). Further analysis needs to be conducted on the data to determine the cause of this change in the data. No further analysis of the data has occurred to determine if this is the only fault in the data set or if all the other data collected at the three-hour point after injection has a similar behavior.
Figure 3. Plots of the post injection data collected at a 20 ps sample interval with a 10 ns record length with background removal.
MicroStrain, 1995-2000, MicroStrain 3DM: MicroStrain, Inc., Burlington, VT, p. 30.
Moita, C., 2001, Formation invasion from a horizontal borehole: A two-phase flow experiment monitored with ground-penetrating radar and modeled with ECLIPSE numerical simulator: MSc. Thesis, Colorado School of Mines, Department of Geophysics, p.177 +CD-ROM.
Wright, D.L., J.D. Abraham, D.V. Smith, and S. R. Hutton, 2001, A high resolution, short range, directional radar: in Proceedings of the Symposium on the Application of Geophysics to Engineering and Environmental Problems (SAGEEP), 2001
, March. 4-7, 2001, Denver, CO.
DIRECTIONAL BOREHOLE RADAR LOGGING TOOL (DBOR) FILE INFORMATION
Each DBOR file is composed of 24 records. Each record contains a 512 byte header and the 32 bit floating point data. The number of data points in the record is recorded in the header (see below). Each record corresponds to a single azimuthal data point (i.e. 15°). Each file begins at 0° and ends at 345°. The angle that the data was collected at can be found in the header under the Theta: entry. All data are in volts, however any changes in the amplitude induced by the programmable attenuator are not compensated for in the data. The programmable attenuator setting can be found in the header under the Attenu: entry.
Time spacing needs to be calculated based on the value in the samp_int: entry. The first entry is at 0.0 seconds and all other data points follow at the sample interval rate.
Header
The header is composed of 512 bytes of binary data the following is a description of the items in the file.
Site: Description of site
Oper: Operator name
Well: Description of Well
Z(0): Initial depth (meters)
Z: Depth increments (meters)
Theta: Angle increments (degrees)
Date: Date (MM/DD/YY)
Time: Time (HH:MM:SS)
#pts: Number of points in scope record
samp_int: Interval between scope sample points (seconds)
vert_sens: Scope vertical sensitivity (volts)
Trig_lev: Scope trigger level (volts)
Ant_load: Antenna relative dielectric permittivity loading factor (1.0 for air)
Comments: User comments
Scp_VOff: Scope vertical offset (volts).
TX/RX_Offset: Offset between the TX and RX antennas (meters)
#avg: Number of scope averages
Attenu: Programmable attenuator setting (dB)
Scp_HPos: Horizontal position offset on the scope (seconds)
Pit: Pitch angle from the 3DM mag/accelerometer (degrees)
Rol: Roll angle from the 3DM mag/accelerometer (degrees)
Yaw: Yaw angle from the 3DM mag/accelerometer (degrees)
Mx: Magnetic x axis info from the 3DM mag/accelerometer (digitizer units1)
My: Magnetic y-axis info from the 3DM mag/accelerometer (digitizer units1)
Mz: Magnetic z axis info from the 3DM mag/accelerometer (digitizer units1)
Ax: Accelerometer x axis info from the 3DM mag/accelerometer (digitizer units1)
Ax: Accelerometer y axis info from the 3DM mag/accelerometer (digitizer units1)
Az: Accelerometer z axis info from the 3DM mag/accelerometer (digitizer units1)
EVal: Value of the motor encoder at end of rotation(degrees)
1
The MicroStrain 3DM unit outputs digitizer units that can be interpreted by using the company supplied software (Microstrain, 1995-2000)Example Header:
Site:CSM Sand Tank Oper:JDA Well:Post-injection Z(0):0.000 Z:0 Theta: 0 Date:10/3/01 Time:12:00:57 #pts:1000 samp_int:4.00E-11 vert_sens:0.100 Trig_lev:0.338 Ant_load:1.0 Comments:Post-injection 1hour Scp_VOff:0.050 TX/RX_Offset:0.000 #avg:16 Attenu:10 Scp_HPos:1.05E-7 Pit:82.5 Rol: 0.3 Yaw:16.8 Mx:1.548 My:2.470 Mz:2.296 Ax:0.637 Ax:1.224 Az:0.846 EVal:0
DATA
Following the header are the 32-bit floating point binary data in MS-DOS® personal computer format. This is a modification of IEEE 32-bit floating point format with the byte order reversed. The number of data points in the file is found in the header next to the #pts: entry. The number of points are variable and to read the data the data reading program needs to look at the header to know how many floating point numbers follow. Thus a single record (one angle at 15° increment) with 1000 data points is 4,512 bytes long and a complete data file (one 360° rotation) is 108,288 bytes long.
