Instrumentation & Measurement Magazine 23-5 - 46

Fig. 6. Developed PM sensor. (a) Structure of PM sensor; (b) Semi-finished product of PM sensor; (c) Fabricated PM sensor.

resonant circuit, which was composed of two inductors in the
PM sensor and variable capacitor bank in the analog board,
was designed to amplify the FID signal because the output impedance of the PM sensor is an inductive impedance. The Q
value of the series resonant circuit is about 10. Consequently,
the FID signal output from the PM sensor was amplified about
2,000,000 times, and the amplitude of amplified FID signal
was about 2 Vpp, which can be seen from Fig. 1. The dc Driver
circuit in the analog board was used to provide a current to
the coils in the PM sensor to excite the protons, which was
enabled while the magnetic field measurement comparative
experiment was performed and disabled while the frequency
measurement comparative experiment was performed. The
PM sensor and resistor attenuator were connected to the analog board by an SMA connector; therefore, it was easy to
exchange one with the other according to the different experiment requirement.
The PM sensor which was used for the comparative experiment of the magnetic field measurement is shown in Fig. 6.
The structure of PM sensor is shown in Fig 6a, which includes
a shell, two Coils and Aviation kerosene. The aviation kerosene is the working substance, which is a proton-rich solution.
Two Coils in reverse series were adopted to excite the working
substance and induce the FID signal produced by polarized
proton in solution. The Shell was used to seal the working
substance, fix the Coils and shield the electromagnetic interference. Fig. 6b is a semi-finished product of the PM sensor,
which has not been painted with the shielding material and
sealing material. Fig. 6c is a fabricated PM sensor. The working
principle of the PM sensor is discussed in [9] and [22] in detail.
To generate the stable measured magnetic field for comparative experiments, standard magnetic field generation
46

equipment is needed. In the work, a three-dimensional double layer cube standard magnetic field generation coils was
employed as the standard equipment, which is shown in Fig.
7a. The outer coils were used to counteract the geomagnetic
field (without PLL to track and compensate the geomagnetic
field variations), and the inner coils were used to generate
the stable magnetic field to be measured. The edge length of
outer cube is about 2.4 m, and the inner cube is about 2.2 m.
The shape of uniform magnetic field region in the coils is a
cube which is about 20 cm×20 cm×20 cm. The geomagnetic
field was first counteracted by outer coils, and then the PM
sensor was located in the uniform magnetic field region,
which is shown as Fig, 7b. According to the measurement
principle of the PM sensor [9], the SNR of induced FID signal
is the best one if the axis of coils (shown as Fig. 6) is perpendicular to the magnetic field to be measured. So, only
Z- direction standard magnetic field was generated when the
PM sensor was located as shown in Fig. 7b in the comparison
experiments.

Frequency Measurement Experiment
As can be seen from (21), the precision of the proposed method
is decided by dt, N and Tx. Therefore, two experiments were
designed to study the effectiveness of the proposed method.
The first comparative experiment: the root-mean-square deviation (RMSD) of the different method varies as the length
of the measurement time (NTx) at a fixed SNR and Tx. The second comparative experiment: the RMSD varies as the different
SNR at a fixed measurement time length and Tx. The MPM
method mentioned in [16] and the M-MFM method using only
rising or falling edge were implemented to compare the effectiveness with that of the proposed method. What is more, all

IEEE Instrumentation & Measurement Magazine

August 2020

Instrumentation & Measurement Magazine 23-5

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 23-5