Magnetic Ordinance Detection By Christopher Fenton.

Magnetic Ordinance Detection

By Christopher Fenton

Goals

• Analyze feasibility of magnetic ordinance detection methods, specifically with IED detection in Iraq in mind

• If feasible, build working prototype

• Successfully detect something metallic

Different Approaches to Object Detection

• Traditional Metal Detectors

• Ground-Penetrating Radar

• Magnetic Detectors

Magnetic Detection Approaches

• Balanced-Loop– Detects change in B-field over time– Covers large areas

• Magnetometers– Measures absolute B-field– Covers small areas

Balanced Loops• First use of Magnetic “Indicator Loops” for

harbor defense in 1915 by British in WWI, adopted by U.S. in 1942 during WWII

• Can only detect moving magnetic disturbances

• Typically large and immobile (>1.6 km^2)

• Abandoned for harbor defense in favor of SONAR following WWII

Balanced Loops in Action

Old detector station in Nahant, MA

Magnetometers

• First invented in 1833 by Carl Gauss

• Can detect magnitude and direction of magnetic field

• Small and lightweight

• Still used for geological surveying and “Magnetic Anomaly Detectors”

Magnetometers in Action

Magnetometer Array used for UXO detection

MicroMag3 3-axis Magnetometer

Circuit model of sensor used in MicroMag3 (Sensor inductance changes with external B-field)

Approach: Magnetometer Array

• Sensors are small (~1”x1”), cheap ($50) and easy to handle – > Even small loops are several m^2

• Insensitive to scanning speed and tilt– > For loops, tilt and speed need to be precisely

monitored

• Arrays can be scaled to arbitrary width for wide-area scanning– > Magnetometers give point measurements, but can

be expanded to cover wide areas like loops do

The MAGNETube

MAGNETube Setup

• 3 x MicroMag3 3-axis SPI magnetometers– Sensors mounted 15” apart– Calibrated so Earth’s B-field = 1 = 0.48568G

• 2 x Picaxe 18X microcontrollers– Expandable through “daisy-chaining”

• 1 Laptop running “Listener” software and outputting to CSV file for analysis in Microsoft Excel®

Setup

A B C

How is the magnitude computed?

1. X, Y, and Z values for all 3 sensors are sent to laptop

2. Calibration offset is subtracted from each direction

3. Magnitude = √(X^2 + Y^2 + Z^2)

4. Magnitude is scaled from 150-200 range to approximately equal “1” in Earth’s B-field

5. Sensor: 1=.48568 Gauss in Los Angeles

Test 1: 80 lbs of Iron

Location: Erdem’s Apartment

Target: 80 lbs of iron weights in a plastic trashcan

Test 1: 80lbs of IronPeak Magnitude vs. Distance

-0.1

-0.05

0

0.05

0.1

0.15

0.2

14 20 26 36

Distance from ground (inches)

Dif

fere

nce f

rom

backg

rou

nd

A

B

C

Conclusion: Readily detectable if directly above pile, drops off quickly

Possibly due to misalignment of sensor during test

Test 2: 4” Brass Artillery Shell

Test 2: 4” Brass Artillery Shell

Magnitude vs. Distance*

0

0.2

0.4

0.6

0.8

1

1.2

1 3 5 7 9 11 13 15 17 19 21 23 25

Ticks

Sca

led

Mag

nit

ud

e

A

B

C

Magnitude vs. Distance

0

0.2

0.4

0.6

0.8

1

1.2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Magnitude

Dis

tan

ce A

B

C

Test 2: 12” above groundBackground: 12” above ground

Conclusion: Brass has no magnetic signature. Only bolts were detectable, and only then at close range.

Test 3: Neodymium Magnets (high sensitivity simulation)

Large 3”x6” Neodymium magnet

Test 3: N.D. Magnet

Peak Magnitude vs. Distance

-2

0

2

4

6

8

10

12

6 12 18 24 30 36

Distance (inches)

Dif

fere

nce

fro

m

Bac

kgro

un

d

A

B

C

Conclusion: Magnet is easily detectable at a reasonable range

Test 4: Attenuation in Water

Test 4: Submerged N.D. Magnet

Magnitude Vs. Water Depth @ 18"

0

0.5

1

1.5

2

0 5 10 15

Water Depth (inches)

Dif

fere

nce

fro

m

Bac

kgro

un

d

Conclusion: Water has no attenuation effect on magnetic field

Future Improvements

• Use faster microcontroller with on-board FPU (~3X improvement in sampling rate)

• Add wireless serial link for easier calibration and field-use

• Experiment with distortion detection vs. simple magnitude detection

• Use higher-sensitivity magnetometers and higher-density array

• Compare vs. traditional metal detector

Conclusion

• Undocumented hardware failure-modes can be extremely difficult to fix

• Magnetic detection appears to be a valid method (and is apparently in-use)

• A simple array can be constructed for less than $250

• With more time, the current design could be greatly improved