Overview of

Magnetic Measurement
Techniques
Animesh Jain
Brookhaven National Laboratory
Upton, New York 11973-5000, USA
US Particle Accelerator School on Superconducting Accelerator Magnets
Santa Barbara, California, June 23-27, 2003
 Outline
• Nuclear Magnetic Resonance (NMR)/
Electron Paramagnetic Resonance (EPR)
• Hall Probes
• Magnetoresistors
• Fluxgate Magnetometers
• Flux Measurements with Pick Up Coils
• Magnetic Alignment – center and direction
• Summary
USPAS, Santa Barbara, June 23-27, 2003 2 Animesh Jain, BNL
 NMR/EPR Principle
• A particle with a spin and a magnetic
moment precesses around an applied field.
• The quantum energy levels are split into
several discrete levels, depending on the
spin of the particle.
• The energy gap between these levels is
proportional to the applied field.
• A resonant absorption of RF energy occurs
at a frequency corresponding to energy gap.
USPAS, Santa Barbara, June 23-27, 2003 3 Animesh Jain, BNL
 NMR/EPR Principle
I = Spin
Precession γ = Gyromagnetic ratio
Applied M = Magnetic Moment
Field = γ.h.I
B
Energy = B.M
M
Spin component along the
field direction can take
Magnetic integral values from –I to
Moment +I. ⇒ Energy gap = γ.h.B
Frequency = γ.B
USPAS, Santa Barbara, June 23-27, 2003 4 Animesh Jain, BNL
 Gyromagnetic Ratio
Particle γ (MHz/T) Application
–
e 28026.5 0.5 to 3.2 mT
1
H 42.576396 0.04 to 2 T
2
H 6.53569 2 T to 14 T
3
He 32.4336 Cryogenic
27
Al 11.0942 Cryogenic
USPAS, Santa Barbara, June 23-27, 2003 5 Animesh Jain, BNL
 NMR Magnetometer

VCO
RF
Sample
&
Probe Hold

∆B Trigger

B Modulation
Oscillator
∆B/B ~ 10–4 to 10–3; 30-70 Hz

USPAS, Santa Barbara, June 23-27, 2003 6 Animesh Jain, BNL

 Locking RF to NMR Resonance
NMR Signal Resonance occurs
with at non-zero value
f and B0 of modulating
mismatched signal.
Error NMR signals
B0 arrive at uneven
B0+∆B(t) intervals.

NMR Signal
Resonance occurs
with
at Zero value of
f and B0
matched modulating signal.
NMR signals
B0 arrive at even
B0+∆B(t) intervals.

USPAS, Santa Barbara, June 23-27, 2003 7 Animesh Jain, BNL

 Requirements for NMR
NMR can provide measurement of magnetic
field with absolute accuracy of 0.1 ppm.
However, certain requirements must be met:
• Field must be stable (< 1% per second).
• Field must be homogeneous (< 0.1% per cm):
– The signal deteriorates; difficult to lock
– Probe positioning accuracy becomes critical.
One may locally compensate for the gradient using
small gradient coils, to make measurements in
inhomogeneous fields.
USPAS, Santa Barbara, June 23-27, 2003 8 Animesh Jain, BNL
 The Hall Effect
Charge carriers
experience a Lorentz
θ B⊥ force in the presence
B I of a magnetic field.

– This produces a steady

F – + state voltage in a
– direction perpendicular
v to the current and
–

field.
–
VHall = G ⋅ RH ⋅ I ⋅ B cos θ
VHall
G = Geometric factor
RH = Hall Coefficient

USPAS, Santa Barbara, June 23-27, 2003 9 Animesh Jain, BNL

 The Planar Hall Effect
If the field has a
I component in the plane
defined by current flow
B || + and voltage contacts,
B then a signal is
ψ rre
nt
produced given by:
Cu

VPlanar ∝ I ⋅ B||2 sin( 2ψ)

VPlanar Important for
mapping of 3-D fields.

The Planar Hall Effect can be minimized by a suitable choice of

geometry ⇒ sin(2ψ) = 0.
In practice, the response of a Hall probe to the field direction is
considerably more complex, requiring elaborate calibration.
USPAS, Santa Barbara, June 23-27, 2003 10 Animesh Jain, BNL
 Compensating Planar Hall Effect
- 2 Matched Hall probes
- I directions as shown
- Major component = By
is in the plane of the
Hall probes.
Sum of Planar Hall Voltages
is proportional to:

[sin(2ψ1 ) + sin(2ψ 2 )]
= 0; [ψ 2 = 90° + ψ1 ]

Based on:
R. Prigl, IMMW-11, BNL.
USPAS, Santa Barbara, June 23-27, 2003 11 Animesh Jain, BNL
 Hall Measurement Specifications

• Typical Range: < 1 mT to 30 T

• Typical Accuracy ~ 0.01% to 0.1%
• Typical dimensions ~ mm
• Frequency response: DC to ~ 20 kHz
(~ a few Hz for fully compensated signal)
• Time Stability: ±0.1% per year

USPAS, Santa Barbara, June 23-27, 2003 12 Animesh Jain, BNL

 Hall Measurement Advantages
• Simple, inexpensive devices, commercially
available.
• Small probe size makes it suitable for a
large variety of applications.
• Can measure all components of field.
• Particularly suited for complex geometries,
such as detector magnets.
• Can be used for fast measurements.
• Can be used at low temperatures.
USPAS, Santa Barbara, June 23-27, 2003 13 Animesh Jain, BNL
 Hall Measurement Disadvantages
• Non-linear device, requires elaborate
calibration of sensitivity for each probe.
• Sensitive to temperature: Calibrate as a
function of temperature; Keep temperature
stable; Design compensated probes.
• Long term calibration drift.
• Planar Hall effect can pose a problem for
mapping 3-D fields. Special geometries are
needed for measuring minor components.
USPAS, Santa Barbara, June 23-27, 2003 14 Animesh Jain, BNL
 Magneto-Resistors
Field bends
B the current
path, thus
NiSb Precipitates altering the
V resistance.
Hall voltage
InSb Slab I tends to
reduce this
effect.
NiSb precipitates “arrest” the build-up of charge on the
sides; Non-linear device; Insensitive to polarity; Large
temperature dependence; Modest sensitivity.
Based on: L. Bottura, Field Measurement Methods, CERN School on
Superconductivity, Erice, May 8-17, 2002.
USPAS, Santa Barbara, June 23-27, 2003 15 Animesh Jain, BNL
 Fluxgate Magnetometers
B Excitation Coil:
AC current drives
excitation bias detection a pair of
coil coil coil ferromagnetic
needles to
+Bexcitation saturation.
Detection Coil:
Detects Zero
-Bexcitation field condition.
Bias Coil:
Maintains a zero
Courtesy: L. Bottura, CERN. field condition.

USPAS, Santa Barbara, June 23-27, 2003 16 Animesh Jain, BNL

 Fluxgate Principle: Zero Field
M M M1
M1+M2

H t
M2
t
V
Zero output in
Excitation detection coil
Profile (H )
is symmetric t
Based on: L. Bottura, CAS on
H Superconductivity, Erice, 2002
USPAS, Santa Barbara, June 23-27, 2003 17 Animesh Jain, BNL
 Fluxgate Principle: Non-Zero Field
M M M1

M1+M2

H t
M2

t
V
Output of
detection coil

Excitation
t
Profile (H ) is
asymmetric Based on: L. Bottura, CAS on
H0 H Superconductivity, Erice, 2002
USPAS, Santa Barbara, June 23-27, 2003 18 Animesh Jain, BNL
 Fluxgate Characteristics
• Highly sensitive, linear, directional device.
• Typical field range ~ a few mT.
(Limited by capability of the bias coils)
• Bandwidth: DC to ~ 1 kHz.
• Sensitivity: ~ 20 pT (~1 nT commercial).
• Accuracy: ~ 0.1%
(depends on calibration and stability)
• Used in navigation, geology, mapping of
fringe fields, etc.
USPAS, Santa Barbara, June 23-27, 2003 19 Animesh Jain, BNL
 DCCT: A Special Fluxgate
detection
DC Current
coil Transformer
B Senses magnetic
bias field produced by a
coil current carrying
conductor passing
through a toroidal
core.
I Used for accurate
measurement of high
currents
Current to be excitation (~10-100 ppm typical)
measured
coil
Courtesy:
L. Bottura, CERN.
USPAS, Santa Barbara, June 23-27, 2003 20 Animesh Jain, BNL
 Flux Measurements: Induction Law
Flux through a coil defined
by the surface S is:
B Φ = ∫ B⋅d S
S

If the flux linked varies

with time, a loop voltage is
Coil induced, given by:
S
V(t) dΦ d  
V (t ) = − = −  ∫ B ⋅ d S
dt dt  S 
The time dependence may be caused
by either a varying field or a varying
surface area vector, or both.
USPAS, Santa Barbara, June 23-27, 2003 21 Animesh Jain, BNL
 Flux Measurements
Time dependence of flux gives:
dΦ d  
B V (t ) = − = −  ∫ B ⋅ d S
dt dt  S 
The change in flux is given by:
Coil S
t end

V(t) Φ end − Φ start = − ⌠

 V (t )⋅dt
⌡
t start

and can be measured by

integrating the voltage signal.
To know the flux at a given instant, one needs to know Φstart
⇒ (1) Use Φstart = 0; (2) Flip Coil/Rotating coil: Φ end = mΦ start
USPAS, Santa Barbara, June 23-27, 2003 22 Animesh Jain, BNL
 Common Coil Geometries
b
a Point Coil
Y
θ'=θ+π/(2m)
Insensitive up to
4th order spatial Rc
h harmonic with π/m
proper choice of θ

height and radii.

X

Flat Coil (Line or Area Coil) Multipole Coil

–Fixed coil; Varying field Sensitive to only odd
–Flip Coil/Moving Coil; Static field multiples of a specified
harmonic (Morgan Coils)
–Rotating Tangential/Radial
USPAS, Santa Barbara, June 23-27, 2003 23 Animesh Jain, BNL
 Flux Measurements: Hardware
±5V Digital Integrator:
Vin(t) G 0-10 V VFC
10 V = fmax
N Directly gives
change in flux.
Reference
Vref frequency Nref 10-100 ppm
(+5 V) fref = α fmax accuracy.
Counters

Digital Votmeter:
Gives rate of change
of flux. Numerical
Integration and/or
well controlled coil
movement is needed.

USPAS, Santa Barbara, June 23-27, 2003 24 Animesh Jain, BNL

 Measurements with Pick up Coils
• Simple, passive, linear, drift-free devices.
• Require change in flux ⇒ ramp field with
static coil, or move coil in a static field. Pay
attention to ramping/moving details.
• Measure flux, not field. ⇒ Calibration of
geometry very important; limits accuracy.
• Field variations across the coil area must be
accounted for ⇒ harmonic analysis.
• Field harmonics can be measured at ppm level.
• Field direction can be measured to ~ 50 µrad.
USPAS, Santa Barbara, June 23-27, 2003 25 Animesh Jain, BNL
 Determination of Magnetic Center
To Integrator

X-Y Stage Wire

MA G
NET

Colloidal Cell
Stretched Wire Measurements
• Place ferromagnetic
• Move a stretched wire in a magnet fluid in the field
• Measure change in flux for various • Illuminate with
types of motion. polarized light
• Use expected field symmetry to • Observe with
locate the magnetic center. crossed analyzer
USPAS, Santa Barbara, June 23-27, 2003 26 Animesh Jain, BNL
 Determination of Field Direction
Encoder Coil
Rotating Coils
• Angular Encoder and
Gravity Sensors
Gravity • Accuracy 50-100 µrad
Sensors • Frequent re-calibrations

BEAM
SPLITTER Mirror
& Needle
LASER
MAGNETIC OPTICAL
• For solenoids
NEEDLE FILTER
• Resolution
MIRROR
POSITION SENSITIVE DETECTOR ~ 10 µrad

Based on: C. Crawford et al., FNAL and BINP, Proc. PAC'99, p. 3321-3

USPAS, Santa Barbara, June 23-27, 2003 27 Animesh Jain, BNL

 Summary
• Numerous methods exist for measurement of
magnetic fields. Only some of them are in common
use for measuring accelerator magnets.
• NMR technique is the standard for absolute accuracy,
but can not be used in all situations.
• Hall probes are very popular for point measurements,
such as for field mapping of detector magnets.
• A variety of pick up coils are the most often used
tools for characterizing field quality in accelerator
magnets.
• Innovative techniques have been developed for
alignment measurements to suit various applications.
USPAS, Santa Barbara, June 23-27, 2003 28 Animesh Jain, BNL
 For More Information
• Knud Henrichsen’s bibliography:
[Link]
• CERN Accelerator Schools on Magnetic Measurements:
– March 16-20, 1992, Montreux (CERN 92-05, 15 Sep. 1992)
– April 11-17, 1997, Anacapri (CERN 98-05, 4 Aug. 1998)
• Proceedings of Magnet Measurement Workshops:
– IMMW-1 (1977) to IMMW12 (2001); IMMW13 (May 2003)
• Proceedings of Particle Accelerator Conferences:
– PAC (1965-2001); EPAC (1988-2002)
• Proceedings of Magnet Technology Conferences:
– MT-1 (1965) to MT-17 (2001).
USPAS, Santa Barbara, June 23-27, 2003 29 Animesh Jain, BNL