Sensors

ION MagNav Workshop 2023, Monterey, CA

Aaron Nielsen

The views expressed in this article are those of the author and do not necessarily reflect the official policy or position of the United States Government, Department of Defense, United States Air Force or Air University.

Distribution A: Authorized for public release. Distribution is unlimited. Case No. 2023-0427.

Introduction

  • Conventional MagNav sensor configuration
    • A little bit of history
    • Scalar magnetometers
    • Vector magnetometers
  • NV Diamond Sensors (Linh Pham, MIT-LL)
  • Magneto-inductive sensors (PNI)

Magnetometer history

  • Compass use documented in China 4th Century BCE (Yan 475-221 BCE)
  • Gauss invented magnetometer in 1833 (Gauss 1833)
  • Aschenbrenner and Goubau invented flux-gate magnetometer in 1936 (Aschenbrenner 1936)
    • Vector sensor, measures one component of \(\vec{B}\)
    • 3 sensors required to measure all components
  • Vacquier invented airborne version of fluxgate (Vacquier 1945)
    • Used gimbal to align sensor axis with total field
  • Alkali vapor magnetometers (Bloom 1962) (Bell and Bloom 1957)
    • Measure vector magnitude or total field \(|\vec{B}|\)
    • Review article in (Tierney et al. 2019)

Scalar measurements

Sensors for scalar measurements

Scalar sensors used to avoid rotation issues

Vector sensors as attitude sensors

\[ \begin{align*} DC_x = \arccos{a} &= \frac{B_x}{\sqrt{B_x^2 + B_y^2 + B_z^2}}\\ DC_y = \arccos{b} &= \frac{B_y}{\sqrt{B_x^2 + B_y^2 + B_z^2}}\\ DC_z = \arccos{c} &= \frac{B_z}{\sqrt{B_x^2 + B_y^2 + B_z^2}} \end{align*} \]

Direction cosines

  • Non-linear transform that hides some potential artifacts
    • For example an overall scaling of vector components divides out.
  • DC derived from vector components of total field
    • Computed DC will include effects of dipoles in platform reference frame

Why not just use a vector?

  • Sensor requires calibration
    • Mechanical - orthogonalization
    • Electronic artifacts - gain and bias
    • Magnetic artifacts - hard and soft bias
  • Temperature dependence

Gain and bias: \[ \vec{B}_m = {\mathbf{A}_\mathrm{soft}} \vec{B}_r + \vec{b}_\mathrm{hard} \]

Hard and soft magnetic moments: \[ \vec{B}_m = {\mathbf{A}_\mathrm{soft}} \vec{B}_r + \vec{b}_\mathrm{hard} \]

Vector orthogonalization calibration techniques

  • Calibration with scalar value - good review is Merayo 2000 Merayo et al. (2000)
    • Fix the sensor in place and spin around
    • Magnitude of \(|\vec{B}| = \mathrm{constant}\)
    • Uses external reference for \(|\vec{B}|\)
  • Orthogonalization correction
    • \(\vec{B}_m = A \left[\vec{B}_r + \vec{b} \right]\)
    • \(A\) is a triangular matrix
    • \(b\) is a bias vector

Hard and soft iron bias calibration

  • Commonly advertised technique for inexpensive magnetometers
    • Hard - permanent moments
    • Soft - induced moments
  • Typically a rotational test \(|\vec{B}| = \mathrm{constant}\)
  • \(\vec{B}_m = A \left[\vec{B}_r + \vec{b} \right]\)
  • \(A\) is symmetric matrix
  • \(b\) is the permanent moment bias

Vector calibration example

  • Before calibration, the vector magnetometer lies on a shifted ellipse.
  • After calibration, the vector magnetomter values will lie on the surface of a sphere.

  • blue - before calibration
  • green -after calibration
  • red - fixed magnitude sphere

Temperature dependence

Newton’s law of cooling \[T(t) = T_s +(T_0 - T_s)\exp(-kt)\] which is the solution to \[\frac{dT}{dt} = -k(T_0 - T_s)\] if \(H \propto T\) then we can write by analogy \[H(t) = H_f + (H_0 - H_f)\exp(-kt)\]

Temperature compensation

Newton’s law of cooling \[T(t) = T_s +(T_0 - T_s)\exp(-kt)\] which is the solution to \[\frac{dT}{dt} = -k(T_0 - T_s)\] if \(H \propto T\) then we can write by analogy \[H(t) = H_f + (H_0 - H_f)\exp(-kt)\]

Scalar sensors

Scalar sensor has advantage that it’s output independent of magnetic field orientation.

Generally we refer to Atomic vapor sensors.

Geometrics MFAM (MFAM Module Specifications, Laser Pumped Cesium Magnetometer 2020)

Interpretation of scalar value in geomagnetic context

\[ \begin{align*} |\vec{B}_\mathrm{total}| &= |\vec{B}_\mathrm{Earth} + \vec{B}_\mathrm{anomaly}|\\ &= \sqrt{ |B_\mathrm{Earth}|^2 + |B_\mathrm{anomaly}|^2 + 2 |B_\mathrm{Earth}||B_\mathrm{anomaly}|\cos\theta}\\ |B_\mathrm{total}| &= |B_\mathrm{Earth}| \sqrt{ 1 + \frac{|B_\mathrm{anomaly}|^2}{|B_\mathrm{Earth}|^2} + 2\frac{|B_\mathrm{anomaly}|}{|B_\mathrm{Earth}|}\cos\theta } \\ &\approx |B_\mathrm{Earth}| + |B_\mathrm{anomaly}|\cos\theta + \cdots \end{align*} \]

\(|B_\mathrm{anomaly}|\) is the projection of \(\vec{B}_\mathrm{anomaly}\) onto \(\vec{B}_\mathrm{Earth}\)

Atomic sensors - Larmor precession

Atomic vapor sensors operate on the principle of Larmor precession

These are gases that have an unpaired electron that has a specific magnetic moment.

Cesium \(\gamma = 3.5\ \text{nT/Hz}\)

In Earth field of \(50000\ \text{nT} \rightarrow 175 \text{kHz}\)

Geometrics

  • \(\vec{A}\) is angular momentum
  • \(\vec{T}\) is torque
  • \(\vec{M}\) is magnetic moment
  • \(\vec{H}\) is magnetic field
  • \(\gamma\) is gyromagnetic ratio

Atomic sensor operation

Inside the sensor head is the Ce gas. Without doing anything the atoms in the gas have a random orientation.

A pump laser with a specific polarization orients the Ce spins and excites them to a specific state.

A probe laser reads out the precesion rate and the magnetic field is inferred.

The probe laser is polarized so that it is absorbed by the Ce gas when the spin aligns with the beam polarization.

  • Geometrics
  • Red - probe laser beam
  • Blue - magnetic field
  • Yellow - magnetic moment

Atomic sensor issues

  • How to interpret signal in geomagnetic context
  • Primary sensor issues
    • Deadzones
    • Heading errors

Deadzone

Related to the pump laser optical axis.

Pump laser is circular polarized and transfers angular momentum to the atoms so that they spin align to the optical axis \((\vec{M}\parallel\text{optical\ axis})\)

Torque \(\vec{T}\) on \(\vec{M}\) from \(\vec{H}\) is: \[ \vec{T} = \vec{M} \times \vec{H} \] If \(\vec{M} \parallel \vec{H}\), then \(\vec{T} = \vec{M} \times \vec{H} = 0\)

No torque \(\rightarrow\) no precession and no signal to monitor.

Details of laser-gas interaction define deadzone size.

Deadzone avoidance

  • Orient optical axis based on knowledge of external field
    • OK if external field does not move - not practical for vehicles
  • Use multiple sensors
    • arrange so that one is always out of deadzone, how many do you need?
    • travel from Northern to Southern hemisphere

Geometrics MFAM (MFAM Module Specifications, Laser Pumped Cesium Magnetometer 2020)

Heading error

Heading error is due to small amounts of magnetization near the sensor and leads to an orientation dependent bias.

\(\cos\theta\) is the angle between \(\vec{B}_\mathrm{external}\) and \(\vec{B}_\mathrm{sensor}\).

\[ \begin{align*} |\vec{B}_\mathrm{measure}| &= |\vec{B}_\mathrm{external} + \vec{B}_\mathrm{sensor}|\\ &= \sqrt{ |B_\mathrm{external}|^2 + |B_\mathrm{sensor}|^2 + 2 |B_\mathrm{external}||B_\mathrm{sensor}|\cos\theta}\\ |B_\mathrm{measure}| &= |B_\mathrm{external}| \sqrt{ 1 + \frac{|B_\mathrm{sensor}|^2}{|B_\mathrm{external}|^2} + 2\frac{|B_\mathrm{sensor}|}{|B_\mathrm{external}|}\cos\theta} \\ &\approx |B_\mathrm{external}| + |B_\mathrm{sensor}|\cos\theta + \cdots \end{align*} \]

Multiple vector sensors

The full magnetic vector gradient has 9 components \[ \nabla \vec{B} = \begin{bmatrix} \frac{\partial}{\partial x}\\ \frac{\partial}{\partial y}\\ \frac{\partial}{\partial z} \end{bmatrix} \begin{bmatrix} B_x & B_y & B_z \end{bmatrix} = \begin{bmatrix} \frac{\partial B_x}{\partial x} & \frac{\partial B_x}{\partial y} & \frac{\partial B_x}{\partial z} \\ \frac{\partial B_y}{\partial x} & \frac{\partial B_y}{\partial y} & \frac{\partial B_y}{\partial z} \\ \frac{\partial B_z}{\partial x} & \frac{\partial B_z}{\partial y} & \frac{\partial B_z}{\partial z} \end{bmatrix} \]

Maxwell’s equations provide constraints, trace must be zero \[ \nabla \cdot \vec{B} = 0 \] In free space \[ \nabla \times \vec{B} = 0 \] Which means that \(\nabla \vec{B}\) must be symmetric

Only 5 independent elements (Bracken and Brown 2006).

Vector Gradiometer aka Tensiometer

Because of the contraints, a vector gradiometer can be made with \(\ge 4\) vector sensors in a cross (plane) or in a tetrahedron or similar shape.

Magnetic field gradients

\[ \vec{B}_\mathrm{dipole} = \frac{\mu_0 }{4\pi} { \frac{1}{r^3} }\left[ 3 (\vec{m}\cdot \hat{r})\hat{r} - \vec{m} \right] \propto \frac{1}{r^3} \]

\[ | (\nabla \vec{B}_\mathrm{dipole})_{ij}| \propto \frac{1}{r^4} \]

Differencing two measurements will will remove a constant and leave behind what’s different.

The closer a dipole, the steeper the gradient.

Gradient measurements will be dominated by local dipoles.

References

Aschenbrenner, Hans. 1936. “Eine Anordnung Zur Regisrierung Rauscher Magnetischer Storungen.” Hochfrequenztechnik Und Elektoakustik 47 (6): 177–81.
Bell, William E., and Arnold L. Bloom. 1957. “Optical Detection of Magnetic Resonance in Alkali Metal Vapor.” Physical Review 107 (6): 1559–65. https://doi.org/10.1103/PhysRev.107.1559.
Bloom, Arnold L. 1962. “Principles of Operation of the Rubidium Vapor Magnetometer.” Applied Optics 1 (1): 61. https://doi.org/10.1364/ao.1.000061.
Bracken, Robert E., and Philip J. Brown. 2006. “Concepts and Procedures Required for Successful Reduction of Tensor Magnetic Gradiometer Data Obtained from an Unexploded Ordnance DetectionDemonstration at Yuma Proving Grounds,arizona.” 2006-1027. U.S. DEPARTMENT OF THE INTERIOR, U.S. GEOLOGICAL SURVEY.
Gauss, C. F. 1833. Intensitas Vis Magneticae Terrestris Ad Mensuram Absolutam Revocata. Sumtibus Dieterichianis. https://books.google.com/books?id=dJA\_AAAAcAAJ.
Leliak, Paul. 1961. “Identification and Evaluation of Magnetic-Field Sources of Magnetic Airborne Detector Equipped Aircraft.” IRE Transactions on Aeronautical and Navigational Electronics ANE-8 (3): 95–105. https://doi.org/10.1109/TANE3.1961.4201799.
Merayo, J M G, P Brauer, F Primdahl, J R Petersen, and O V Nielsen. 2000. “Scalar Calibration of Vector Magnetometers.” Measurement Science and Technology 11 (2): 120–32. https://doi.org/10.1088/0957-0233/11/2/304.
MFAM Module Specifications, Laser Pumped Cesium Magnetometer. 2020. Geometrics, Inc. http://mfam.geometrics.com/.
Smithsonian Institution. n.d. “Developing Inertial Navigation.” On-line. https://timeandnavigation.si.edu/satellite-navigation/reliable-global-navigation/inertial-navigation/developing-inertial-navigation.
Tierney, Tim M., Niall Holmes, Stephanie Mellor, José David López, Gillian Roberts, Ryan M. Hill, Elena Boto, et al. 2019. “Optically Pumped Magnetometers: From Quantum Origins to Multi-Channel Magnetoencephalography.” NeuroImage 199 (October): 598–608. https://doi.org/10.1016/j.neuroimage.2019.05.063.
Tolles, W E, and J D Lawson. 1950. Magnetic compensation of MAD equipped aircraft.” Airborne Instruments Lab. Inc., Mineola, NY, Rept, 201–1.
Tolles, W. E. 1954. Compensation of aircraft magnetic fields. 2692970, issued 1954.
———. 1955. Magnetic field compensation system. 2706801, issued 1955.
Vacquier, Victor V. 1945. Apparatus for responding to magnetic fields. 2407202, issued 1945. https://patents.google.com/patent/US2407202A/en.
Yan, Zou. 475-221 BCE. Zou Yan Lu (Book of the Devil Valley Master). China: Unknown.