The low cost Proton Precession Magnetometer developed at the Indian Institute of Geomagnetism

Transcription

1 Journal of Instrumentation The low cost Proton Precession Magnetometer developed at the Indian Institute of Geomagnetism To cite this article: P. Mahavarkar et al View the article online for updates and enhancements. Related content - Construction of an Overhauser magnetic gradiometer and the applications in geomagnetic observation and ferromagnetic target localization H. Liu, H. Dong, Z. Liu et al. - Noise characterization for the FID signal from proton precession magnetometer H. Liu, H. Dong, Z. Liu et al. - Earth's field magnetometry W F Stuart This content was downloaded from IP address on 18/06/2018 at 11:30

2 Published by IOP Publishing for Sissa Medialab Received: March 20, 2017 Accepted: May 4, 2017 Published: May 15, 2017 TECHNICAL REPORT The low cost Proton Precession Magnetometer developed at the Indian Institute of Geomagnetism P. Mahavarkar, 1,2 S. Singh, S. Labde, V. Dongre and A. Patil Instrumentation Division, Indian Institute of Geomagnetism, Dept. of Science and Technology, Govt. of India, Plot No. 05, Sector 18, New Panvel, Navi Mumbai , India Abstract: Proton magnetometers are the oldest scalar magnetometers. The first commercial units were produced in early 1960s as portable instruments. In continuation airborne instruments appeared with optimized speed of readings and sensitivity, large sensors etc. Later development of Overhauser and optically pumped magnetometers has eliminated Proton magnetometers from airborne surveys. However they remain very popular in various ground surveys and observatories. With this primary purpose of generating the ground based magnetic data, the Indian Institute of Geomagnetism (IIG) for the last 3 decades have been developing low cost Proton Precession Magnetometers (PPM). Beginning with the 1 nt PPM which has undergone several changes in design, the successor PM7 the advanced version has been successfully developed by the institute and is installed at various observatories of the institute. PM7 records the total field F with accuracy of 0.1 nt and a sampling rate of 10 seconds/sample. This article briefly discusses the design and development of this IIG make PM7 and compares the data recorded by this instrument with one of the commercially available Overhauser magnetometer in the world market. The quality of data recorded by PM7 is in excellent agreement with the Overhauser. With the available quality of data generated by this instrument, PM7 is an affordable PPM for scientific institutions, schools and colleges intending to carry out geomagnetic studies. The commercial cost of PM7 is 20% of the cost of Overhauser available in market. Keywords: Manufacturing; Detector design and construction technologies and materials 1Corresponding author. 2Mobile phone: c 2017 IOP Publishing Ltd and Sissa Medialab srl doi: / /12/05/t05002

3 Contents 1 The PM7 operating principle 2 2 PM7 design & development 2 3 The sensor 3 4 Sensor installation 5 5 The signal sensing and the signal processing unit 6 6 Data recorded by PM7 and Overhauser 7 7 Future plans 8 A Proton Precession Magnetometer is an instrument that measures the scalar intensity of the local magnetic field and relies upon the proton-precession measurement technique. This is a well established technique and is successfully implemented in classical proton precession magnetometers and the Overhauser magnetometers. By the virtue of the Overhauser principles the Overhauser magnetometer always has advantages over the classical proton precession magnetometers which do not work on the principle of Overhauser effect. The sampling rate, accuracy and power consumption is always better in case of these magnetometers. But it also has disadvantages. The polarization of radical solution in the probe of an Overhauser magnetometer always needs a radio frequency to be excited to get the free induction decay (FID) signal. In some exploration projects, such as gradient measurement at close range, instruments will interfere with each other, which leads to the error of magnetic field measurement; when the Overhauser magnetometer works with other systems, it will also affect the entire system, and it is not easy to completely shield the radio frequency. In addition, since the Overhauser magnetometer is expensive, in a multipoint weak magnetic monitoring application, the cost is high. On the other hand, due to the different principles, the proton magnetometer has no mutual interference problems and is cheaper than the Overhauser magnetometer [1]. Utilizing the proton precession principles, PM7 has been designed to have an accuracy of 0.1 nt, a minimum 5 sec sampling rate and 15W power consumption. However the data recorded by PM7 is in excellent agreement with the Overhauser. This quality of data generated by PM7 is achieved with a low cost sensor design and a less complicated electronics. 1

4 1 The PM7 operating principle Figure 1. PM7 unit with the sensor. PM7 utilizes the precession of spinning protons or nuclei of the hydrogen atoms in a sample of Hexane to measure the total magnetic intensity. The spinning protons in this sample are temporarily aligned or polarized by application of a uniform magnetic field (hundreds of Gauss) generated by passing a current of 1-2 A through a coil of inductance 30 mh [2]. After 7-8 seconds, maximum number of protons are aligned in the direction of the applied field. When the current is stopped (the external field is switched off suddenly), the spin of the proton causes them to precess about the direction of the ambient or Earth s magnetic field. The precessing protons then generate a small signal (an induced emf) in the same coil which is used to polarize them, whose frequency is precisely proportional to the total magnetic field intensity and independent of the orientation of the coil. This emf is of the order of a few micro volts. It is amplified to get an output of 5-6 V. After passing through the PLL and processing by the micro-controller, frequency is determined from this signal and then displayed. Also the value of total field is available on serial port, to be stored by using third party Data loggers. 2 PM7 design & development PM7 comprises of the following 2 units: The sensor The signal sensing and the signal processing unit (electronics) The sensor is a separate assembly which is placed in the field where measurements are to be made and the signal sensing and signal processing unit (the electronics) is an assembly of 2 PCB cards housed inside a parallelepipedic metal box referred to as the control unit. Figure 1 shows the PM7 set up. The electronics included mainly consists of the stages like the excitation of protons, tuning of signal, the amplifier, the phase locked loop (PLL) and the micro-controller with associated 2

5 Figure 2. PM7 block diagram. peripherals. One PCB card is designed to accommodate the excitation, amplifier and the phase locked loop (PLL). The other card accommodates the micro-controller and the peripherals. Figure 2 shows the layout of the PM7 in the form of a block diagram. First the sensor design is briefly discussed which is followed by the signal sensing and signal processing unit. 3 The sensor The proton signal is very weak; hence a specific designed sensor (figure 3) is needed to pickup such low amplitude signal. The technical specifications of the sensor design are not mentioned here. The sensor used for this purpose is of passive type and essentially works on the principle of proton precession phenomenon. It is a well-sealed cylindrical housing of about 1000 cc filled with a proton rich fluid [5] such as Hexane (C 6 H 14 ). A pair of coils connected in series is placed inside the sensor. The sensor coil is the most critical component of the system. It s inductance and resistance determine how fast it can be switched on and off, how much current it can carry, and how sensitively it can detect the weak oscillating magnetic field produced by the sample [3]. According to the theory of operation of the proton magnetometer, the total intensity, measured as the frequency of precession, is independent of the orientation of the sensor. The amplitude of the signal, however, does vary (as sin 2 θ) with the angle between the direction of the applied field within the sensor and the Earth s field direction. Variation of signal amplitude does not normally affect the 3

6 Figure 3. The PM7 sensor. Figure 4. A pair of sensor coils. readings unless there is simply insufficient signal to be measured accurately, i.e., a minimum signal amplitude is required above which a variation in amplitude does not affect the readings. Ideally, the applied field in the sensor should be at right angles to the Earth s field direction. The direction of the applied field is governed by the configuration of the polarizing coils in the sensor which are commonly either solenoids (cylindrical) or toroids (ring or doughnut-shaped). The solenoid produces an applied field parallel to its axis, whereas the toroid produces a field which is ring-shaped about the axis of the toroid. Solenoids are used because they produce somewhat higher signal than a toroid [4]. In the ideal case, the solenoid axis should be held perpendicular to the field direction for maximum signal amplitude. Also the solenoid configuration has one great advantage. The advantage is that it is very easy to wind a solenoid. 4

7 Figure 5. The noise cancellation model in sensor coils. A major problem that needs to be considered in designing a sensor coil is the environmental noise as this noise is picked up by the sensor coil. This problem is overcome by using two identical sensor coils which are wound in opposite directions with respect to each other and connected in series. They are mounted parallel to each other so that external noise common to both coils is canceled out (figure 4). As shown in figure 5a, the two loops in sensor are connected back to back and polarizing current is made to pass through both coils. The loops are wound in opposite direction which serves a special advantage in noise cancellation. As the loops are connected back to back, the precession signal gets added up as shown in figure 5b. The noise voltage induced due to stray magnetic fields (magnetic lines) induce voltages of same polarity in both the coils and cancel each other as shown in figure 5c. If the coil inductance and dimensions are well matched, noise cancellation takes place but the induced signal gets added in the coils. Hence stray noise pickups are eliminated in such a sensor design while the signal is boosted. Using this noise cancellation technique, two anti wound Cu-coil loops are kept immersed in the working liquid (figure 6). The solenoid geometry allows easy access to and quick replacement of the sample. The sensor coil acquires the proton precession signal from the sample which is further amplified and digitized by electronic console. The sensor is separated by a cable from electronics (console unit) to avoid its stray magnetic fields. 4 Sensor installation While measuring the geomagnetic field, the sensor is always mounted 5-6 feet high above ground level with help of any nonmagnetic staff and pedestal (wooden stool, brass or aluminum rods are preferred) (figure 7). This is so because the magnetic anomalies present in Earth s crust and its interior create magnetic gradients in geomagnetic field to be measured thus distorting it. Due to these magnetic gradients the signal decays faster than usual. In such a case the precession time reduces. This ultimately leads to added difficulties in measuring the field in such a small span of time. To avoid this problem the sensor is always elevated at some height above ground level. This gives optimum signal precession performance. Thus the overall performance of sensor is collectively enhanced. 5

8 PM7 Sensor Overhauser Sensor Figure 7. The sensor is mounted 5-6 feet above the ground level to improve the sensing action of the coil. 5 The signal sensing and the signal processing unit Signal sensing is done in two parts (figure 8); the first of these is the polarization of the sample in which the working liquid is subjected to a strong field in order to magnetize (i.e., line up) the protons. The second part is the actual measurement of the precession frequency in order to determine the external magnetic field. During polarization, a large current of 1 A is passed through the solenoid to generate a magnetic field which is of the order of hundreds of Gauss which will line up the protons along coil axis. A polarization current of about 10 sec duration is fed through the relay to the sensor coil. The function 6 Figure 6. A pair of coils connected in series is placed inside the sensor filled with the working fluid.

9 of relay here is to switch between polarization and sensing electronics. The actual geomagnetic signal information picked up by the sensor in form of sinusoidal variations of proton precession is extremely weak in amplitude. It typically ranges in few micro volts and lasts for about for a few seconds. Therefore, the signal is tuned and then amplified by a special low noise amplifier. The output of the amplifier is required to be converted to a square wave to be used as an input to the PLL. For this a zero crossing method is implemented by a phase comparator, designed using another OPAMP. The PLL is a high Q filter [8]. It is a negative feedback system whose function is to force a voltage controlled oscillator (VCO), to be coherent with the input reference signal in both phase and frequency. By virtue of its inherent design, PLL is able to lock firmly on to any desired signal and then it can easily hold on to that particular frequency. Therefore the frequency of desired signal can be easily recovered from a noisy signal using phase locked loop circuit and can be multiplied by a suitable factor. Although multiplication factor in the PLL reduces the measurement time and increases the measurement accuracy it cannot be increased indefinitely to achieve an increase in measurement accuracy. Thus to achieve higher accuracy it is necessary to use different measurement techniques. One way is to measure the period of the precession signal by recording the zero crossings of the sinusoidal precession signal. Because of the noise riding over the signal there is an error at zero crossing which introduces an error T in the measurement of period T. This error is given by T = 1 ωr where ω is the angular frequency and R is the peak to peak signal to noise ratio. A least square estimation technique is used to treat this error at zero crossings. Time for N number of periods can be written as t 0 + pt c = t p where p = 0, 1, 2,..., N and t 0 is the uncertainty in the timing of the first crossing. The techniques used derive T c in such a way that its variance is as small as possible. The standard technique used attempts to minimize the mean square difference of the actual zero crossing times and those computed from the estimated fit. The estimated period can be written as [7] where S A = N p=0 ( 1 σ 2 p ) ; S B = N p=0 ( p σ 2 p T C = S AS E S B S D S A S C SB 2 ) ( ; S C = N p 2 ) ; S D = N p=0 σ 2 p p=0 ( t p σ 2 p ) ; S E = N p=0 ( pt p σ 2 p ). 6 Data recorded by PM7 and Overhauser The plots below compare the quality of data recorded by PM7 (in black line) and Overhauser (in red line) installed at Alibaug Magnetic Observatory (18.64 N, E geographic coordinates) for 7

10 Table 1. IIG specifications for PM7 magnetometer. Specifications PM7 Environmental 0 to 60 Celsius Operating Range (Tuning: manual mode) 30,000 to 80,000 nt Input/Output All control and communication by RS-232 link Power 12 V, 1250 ma peak (during polarization), 200 ma standby Accuracy 0.1 nt Sampling rate Optional (Default 10 seconds) Display Monochrome character display (20 character 4 line) Weight (Console and Sensor) 4.9 kg Console 2.6 kg Sensor 2.3 kg a few typical days 15 February 2015 (figure 9a), 15 March 2015 (figure 9b) and 15 April 2015 (figure 9c). Here the variation of total field in nt versus time in hours is plotted. Similarity can be readily seen from these figures; however the offset in the recorded field is due to the different location of sensors in the same observatory under study. The total F near the equator (figure 10) is in the known range of 42,000 to 43,000 nt [6] which can be observed from the recorded data. All the curves follow a similar trend, however some spikes are seen in the PM7 data. This is attributed to low sampling rate of 1 sample/minute for PM7 compared to the 60 samples/minute for the Overhauser. The quality of data recorded by PM7 is generated with a low cost electronics and sensor design. Table 1 gives the technical specifications of PM7. 7 Future plans We are in a process of popularizing PM7 among the scientific community, schools, colleges and institutes of higher education in India. Although immense efforts have gone into the development 8 Figure 8. The PM7 signal sensing and signal detection PCB s (Controlling electronics).

11 (a) (b) (c) Figure 9. The quality of data recorded by IIG PPM (PM7) is compared with the data recorded by the Overhauser magnetometer. of the PPM, the cost is kept marginally low ( $4000) to ensure that it is affordable to budget conscious users. Further towards the design, efforts are being made in the existing design to improve the accuracy and power efficiency. Also the portability of instrument is being considered for development with battery pack included in the console. 9

12 Figure 10. Isodynamic map showing total intensity or strength of Earth s magnetic field, with higher values indicating greater intensity. References [1] H. Dong, H. Liu, J. Ge, Z. Yuan and Z. Zhao, A high-precision frequency measurement algorithm for FID signal of proton magnetometer, IEEE Trans. Instrum. Measur. 65 (2016) 898. [2] W. Bayot, Practical guidelines for building a magnetometer by hobbyists; part 1: introduction to magnetometer technology, version 1.2, 22 June [3] Hackaday.io, Details PyPPM: a proton precession magnetometer for all, [4] S. Breiner, Applications manual for portable magnetometers, Geometrics 2190 Fortune Drive San Jose CA U.S.A. [5] J. Jankowski and C. Suckdorff, Guide for magnetic measurements and observatory practice, Warsaw Poland, (1996). [6] Avian navigation and orientation webpage, [7] A. Patil and R. Rajaram, Numerical techniques for proton magnetometers, in Proceedings of the X th IAGA Workshop on Geomagnetic Instruments Data Acquisition and Processing, April [8] Analog Devices, Fundamentals of Phase Locked Loops (PLLs), Tutorial MT-086, (2009). 10