Receiving, Filtering and Amplifying Earthquake Signals

Transcription

1 Receiving, Filtering and Amplifying Earthquake Signals Hosseyn Farazar 1, Parviz Amiri 2 1 Ms.C. student, 2 Assist Professor, Faculty of Electrical and Computer Engineering, Shahid Rajaee Teacher Training University, Tehran, Iran Abstract Humans have always required methods for guessing earthquake before its occurrence and minimizing its potential damage resulting with suitable arrangements. Electromagnetic waves which are propagated before earthquake occurrence inside the ground in epicenter of earthquake have high amplitude and broad frequency range. For this purpose, it is necessary to filter and amplify earthquake signals using electronic circuit after receiving them. One of the common methods is to use earthquake alarms. Therefore, electric circuits relating to amplifying, filtering and optical isolation were presented and studied here. Keywords Earthquake sensor, Earthquake Electronic Circuit, Geophone, Filter, Seismic. I. INTRODUCTION Earthquake means sudden and transient motions and vibrations in the earth that originate from a limited area from which they are propagated in all directions. Earthquake is induced by sudden release of energy accumulated in the crust rocks. This release of energy starts from points in depth of earth called epicenter of earthquake and causes vibration of earth by releasing energy as waves and finally, in case buildings and structures are not constructed according to principles, destroys them. Information on the earth plays a main role in many cases. In geology and earthquake studies, this information is a way of discovering very important subjects. One of the very important areas is analysis of such information exploratory studies about underground tanks and mines. Electromagnetic waves radiate from their special epicenter. These waves are greatly absorbed when passing different layers of the earth crust and increase temperature at the surface of ground and even air. These waves can cause anxiety and stress in some animals. Earthquake precursors are various and have cause-and-effect relations with each other, among which electromagnetic waves have a special position. Electromagnetic waves inside the ground have high amplitude and broad frequency spectrum; but, their amplitude is gradually reduced due to the passage of different layers of earth crust. Also, waves with high frequency are absorbed by the earth crust; but, waves with low to very low frequencies are not absorbed and penetrate into the ground surface. Today, different systems have been developed for detecting surface waves of earthquake or vibrations or motions of the ground surface. Earthquake waves can be sensed and detected by different systems. Earthquake waves are divided into two classes in terms of their motion inside or on the earth surface: body waves and surface waves. A. Body Waves Waves, which propagate through the interior of a body. For the Earth, there are two types of seismic body waves: (1) Compressional or longitudinal (P wave) and (2) Shear or Transverse (S wave). [1] P Wave: the primary body wave; the first seismic wave detected by seismographs; able to move through both liquid and solid rock; compressional waves, like sound waves, which compress and expand matter as they move through it. [2] S Wave: secondary body waves that shear, or cut the rock they travel through sideways at right angles to the direction of motion; cannot travel through liquid; produce vertical and horizontal motion in the ground surface. [2] B. Surface Waves Waves that move close to or on the outside surface of the Earth. These are slower than P or S waves, that propagate along the Earth s surface rather than through the deep interior. there are two types of surface waves that propagate along Earth s surface: Rayleigh waves (polarized in the source direction) and Love waves (polarized at right angles to the source direction). [2, 3 and 4] Rayleigh Waves: surface waves that move in an elliptical motion, producing both a vertical and horizontal component of motion in the direction of wave propagation. [2] Love Waves: surface waves that move parallel to the Earth s surface and perpendicular to the direction of wave propagation. [2] 649

2 II. DETECTING EARTHQUAKE WAVES Earthquake alarm acts based on detection of P nondestructive waves which are propagated from epicenter of earthquake. Motion speed of P waves is approximately 6.4 km/s (speed of P wave varies depending on the material of earth layers) and humans are not able to sense them. In addition, these waves have no destructive effects. As is known, some animals become aware of the earthquake occurrence while it is not possible for humans to sense vibrations and P waves. S wave is very destructive. Epicenter and depth of earthquake are calculated based on the difference between reaching time of P and S waves, which allows alarming or sending instruction in a short term before reaching of destructive earthquake vibration (Figure 1). Recently, some systems which predict S waves using P waves in earthquake-prone countries have been developed, according to which some measures such as stopping trains and gas and power flow are taken. Figure 2: seismograph [7] Electromagnetic detection (geophones): On land, the surface moves as a P-wave or S-wave arrives. Generally reflected signals arrive at steep angles of incidence. Thus P-waves produce surface motion that is dominantly vertical. Geophones measure ground motion by converting motion into electrical signals. Most geophones measure a single component (vertical), but multiple component ones are sometimes used. [6] Figure 1: P and S waves and surface waves on the recorded seismograph [] A. Seismic detectors Mechanical seismometer: Measure lower frequencies than geophones. Use a stationary mass. Measures motion of the Earth relative to the mass. Can measure vertical or horizontal motion. [6] For earthquake studies a more permanent installation is usually required. Three components are usually recorded and the sensor is tuned to detect lower frequencies. Often the seismometer is placed in a shallow vault to minimize wind and other forms of noise. [6] FIGURE 3: TYPICAL CONSTRUCTION OF A GEOPHONE [6] Hydrophones: Only sensitive to pressure changes so only P-waves detected. Used in marine surveys. [6] III. VIBRATION SIGNALS OBSERVED USING OSCILLOSCOPE Signals resulting from small shocks which have been received from Geophone Sensor (Earthquake sensor) are shown in the following figures. 6

3 Figure 4 represents signals resulting from a relatively mild shock at 1 m distance from the sensor on the table. Figure demonstrates a signal in static state. In this figure, no shock or vibration has occurred and signals are caused by noise.. The signals which have vibrated the earth in the world and recorded by the seismograph are given in Figures 6 and 7. Figures 4: Oscilloscopes rang in mv - ms Figures 6: The signal recorded at the Tabas earthquake Figures : Oscilloscopes range in mv - ms Figures 7: The signal recorded at the Bam earthquake After the signal is collected by the earthquake sensor, it goes to the initial amplifier; then, it reaches main amplifier and filters. After passing the main amplifier and filters, it is transmitted to the next section using an isolator, which isolates input and output parts. Finally, the signal is digitalized and sent to a computer. The signal is drawn in the computer by MATLAB software. Any processing action such as drawing signal frequency spectrum can be done in computers. IV. SIMULATED ELECTRONIC CIRCUIT This project had two important analog and digital parts. The analog part included receiving information from sensor, amplifying and filtering it and improving the signal for digital part. The digital part in which computer and microcontroller existed included conversion of analog to digital signals and then submission of the serial number of the produced information to the serial port, which was performed by AVR microcontroller. Finally, the produced signals were sent to the computer and drawn and stored using MATLAB software. Receiving (sensor, amplification, filter and impedance matching) Controller (ADC converter, parallel to series conversion + rapid alarming system) Figure 8: A scheme of this project In this project, PSPICE 9.2 software was used to work with analog part and PROTEUS software was used for simulating its microcontroller part. For receiving signals, scope.exe software was applied. Considering that MATLAB software could control the serial port, this software could be used as well. A. Selecting and Designing Amplifier The first class after sensor, into which the signal entered, was called preamplifier. This class had two differential inputs. Preamplifier class performed the following operations: 1- current to voltage conversion, 2- voltage amplification and 3- noise reduction. Two important indices for amplifier of amplitude and frequency included high input impedance and differential input. Any increase in source impedance or vibration sensors caused loss of signal in its transmission line. Therefore, reduction of source impedance can be useful, which was performed in two ways: the first method was to reduce resistance of sensor using a potentiometer or NTC and the second method was isolation of source impedance from load impedance hat was done by increasing input impedance of the amplifier and reducing its output resistance. Pre-amplification coefficient was considered below 1 and placed before filter class in order not to amplify noise signal and not change the signal from its ordinary form. After filter class, an amplifier was used. AD62 is a precision instrument amplifier with very high CMRR (1 db), the gain of which is adjusted only with an external resistance between bases 1 and 8 (adjustable gain from 1 to 1); it was used as preamplifier in this work. Eq. (1) shows gain relation in terms of resistance [1]: (1) Microcontroller information receiving (MATLAB) In this regard, gain value increased with decrease of R G. The table shows necessary resistances for different gains and also their correspondence with the experimental values. 61

4 4 V- V+ 7 Table 1: Values of resistance for different gains in AD62 [1] For designing this filter, datasheets of ANALOG DEVICE and NATIONAL Company were extracted. B. High Pass Filter For simulating and analyzing its frequency, a source was put as sensor in the input and simulated to fulfill demands of the project. For simulation, pre-amplification class was focused on. Considering the papers which have been studied and will be referred to, measurement was performed as reference (source). In other words, the sensor moved with the earth and there was no turbulence-free and stabilized motion reference. For this reason, it was not possible to directly measure convection. Seismic waves caused transient momentums; according to inertial principle, momentum exists when there is acceleration; this principle demonstrates that seismic waves should be in the form of acceleration with which convection can be approximated. To design filtering, the input values should be known. According to some papers, it can be said that amplitude and frequency of seismic signals are very large. However, it is not possible to measure them using a sensor and circuit. This issue also holds true for frequency. Frequency bands can be enhanced from small value of 1 - to 1 khz. IN fact, these values are very high and should generally cover minimum frequency band of.1 to 1 Hz and motion should also show 1 nm to 1 mm. It is not possible to create a precision instrument for covering these values; thus, filters and sensors with different amplitude and frequency are used. TABLE 2 FREQUENCY CONTENT OF SEISMIC SOURCES [6] 1Vac Vdc C2 1u Figure 9: High pass filter Since there were unwanted DC components in input signals such as Emf potential, a high pass filter had to be used for deleting them. But, seismic signal had nonzero mean or DC nature and lower cut-off frequency; the system was slower and had delay in performing its operation which was DC deletion. Therefore, cut-off frequency was considered as two values of.1 and.1 Hz and a shortcircuit state of resistance was also assumed so that the capacitor could be charged with the desired value and then the system could perform its duty. In a low pass filter RC, ; therefore, capacitor of 1 µf and resistance of 1 kω seemed to be suitable. By selecting standard values, cut-off frequency was calculated as.16 Hz. To have cut-off frequency of.16 Hz, it was enough to change resistance to 1 MΩ by putting a selector. C. Low Pass Filter R4 1K 3 U3 1 + RG1 6 8 OUT R2 2 RG2 - REF 1K AD62/AD Most components of sensor signal were in the range of low frequencies; hence, it was necessary to pass a signal which was recorded and then amplified through low pass filter to delete the disruptive components. For filtering, second-order filter was used and, because the highest attenuation was desirable at urban power frequency of Hz, Q of the circuit was selected to be high. IN fact, it should be noted that Q of low pass filters and high pass filters was below 1 while Q of Band pass filter was larger than 1. The following figure 1 shows the filter with gain of 1: - V V3 R 1k 62

5 4 V- V V+ V- To obtain conversion function of this filter, node equations were written: ( ) (2) (3) Figure 1: Low pass filter (2) and (3) ( ) ( ) ( ) (4) V3 - RG2 R4 RG1 C4 1k 33n U1 1Vac Vdc C2 1u V R2 1K AD62/AD 3 U3 RG1 1 + RG1 6 RG2 8 OUT 2 RG2 - REF R6 68k R7 68k C3 1n OS1 OUT OS2 LM V R3 1k The system characteristic equation was as follows: ( ) ( ) () Where quality coefficient and cut-off frequency are as follows: { Figure 12: Pre-Amplifer, Amplifier and Filter circuits (6) (7) (8) FOR DESIGNING, BUTTERWORTH FILTER IN WHICH Q IS.77 AND CUT-OFF FREQUENCY OF 1 HZ WERE SELECTED. ) ) By selecting standard values, cut-off frequency was calculated as 1.2 Hz. Finally, there was a preamplifier with high pass filter, after which low pass filter to filter frequency of.1 Hz to 1 Hz existed. Figure 13 shows the designed circuit for the analog part. 63

6 7 4 V+ V- 1Vac Vdc R1 68k -.422mV C1 33n R2 68k -1.84mV C2-1.82mV.V Figure 11: Filter with the calculated values 2 3 1n V OS1 OUT OS2 - U1 LM V3 V R3 1k Finally, signal isolation is performed using three major methods including capacitor isolation, magnetic isolation and optical isolation. In capacitor modulation, they modulate the signal by a high-frequency carrier and transmit it through capacitor coupling. 6V V 4V 3V 2V 1V V 1uHz 3uHz 1.mHz 3.mHz 1mHz 3mHz 1mHz 3mHz 1.Hz 3.Hz 1Hz 3Hz 1Hz V(U1:OUT) Frequency D. Isolation Figure 13: Frequency response of the analog part Isolation includes signal isolation, individual isolation and supply isolation. Earthquake sensor signal, on which all kinds of hardware processing are performed in analog board, should be digitalized by microcontroller for software processing and digital information is sent to the computer. To transmit signal from analog to digital boards, their grounds should be isolated so that noise of digital switching does not to affect seismo signal. On the other hand, it seems necessary to use isolator due to issues such as ground loop and common mode voltage. Electromagnetic fields which are available in general media (cell phone, radio, WLAN network) induce voltage in case of passing through a loop. If resistance of loop is low, high current does not pass, which is attenuated in the case of interference with signal. In some cases, it is not inevitable to form such a loop with ordinary amplifiers. At these times, using isolated amplifiers cuts the route. Common mode voltage which has a large value burns ordinary amplifiers and probably burns precious equipments afterward. Isolated amplifiers easily tolerate common voltage up to some kilovolt. Such problems lead to considerable use of isolated amplifiers in precision instruments. isolation signal: Signal isolation is performed such that electric signal is first converted from the first system into nonelectrical signal and then passes electrical insulating signal in the output. 64 Figure 14: A scheme of isolated integrated circuit using capacitor isolation In magnetic isolation, isolated transformers are used for this purpose. Electrical signal is first converted into magnetic flow and passes through transformer core which is made of ferrite and is also an electrical insulator and then is converted into electrical signal in its output. As is known, one of the problems of isolated transformers is their low efficiency. On the other hand, transformers face a problem in terms of frequency function at low frequencies (characterized by Band pass transformers) and important components of electroencephalogram signal are also at low frequencies. To solve this problem, electroencephalogram signal should be modulated and put at working frequency of transformer. Then, a demodulator is used for recovering signal frequency components, which causes high cost in addition to errors that are faced in modulator and demodulator system. Figure 1: Magnetic isolation

7 In optical isolation, electrical signal is first converted into light and then electric signal in output by passing through transparent medium of electrical insulator. Among the evident optical isolators are optocouplers and, in this project, optical isolation was applied. Optocouplers in electronic packages include optical diodes and optical transistors. One of the most important specifications of an optocoupler is CRT specification, which is input to output current transfer coefficient. In most optical isolators, this specification is nonlinear, which is one of the main functional problems of optocouplers. According to the figure of isolation circuit, current of, enters optical diode and this diode emits light in its opposite optical transistor. Since there are equal resistances in two bases of op-amp on the second side and voltage of these two bases is also equal, the current which passes upper transistor also passes lower transistor. This event is such that current of two optic diodes is equal to each other; therefore, voltage to resistance ratio of two optic diodes is equal (Eq. 9): (9) Figure 16: Optocoupler isolation To solve this problem, feedback system could be used as follows: 6mV Figure 19: Isolation circuit 4mV 2mV -mv -2mV -4mV Figure 17: Optocoupler single -6mV s.1s.2s.3s.4s.s.6s.7s.8s.9s 1.s V(R2:2) Time In the following figure, diode emits light based on the passing current and this light collides with base of optical transistor and induces current in it. Input-output relation is nonlinear. But, when 2 numbers of these optocouplers (both of which should be in one electronic package) were used, the input-output relation was linear as follows: 3.V 2.V 1.V -.V -1.V Figure 2: An example of the input signal -2.V -3.V s.1s.2s.3s.4s.s.6s.7s.8s.9s 1.s V(U3:OUT) Time Figure 21: After amplification AD62 Figure 18: Optocoupler double 6

8 3.V 2.V 1.V -.V -1.V -2.V -3.V s.1s.2s.3s.4s.s.6s.7s.8s.9s 1.s V(C4:2) Time 3.2V 2.8V 2.4V 2.V 1.6V 1.2V Figure 22: After Filtering.8V s ms 1ms 1ms 2ms 2ms 3ms 3ms 4ms 4ms ms V(R7:2) Time Figure 23: Output Opto-Isolator V. CONCLUSIONS As shown in figures of seismic signal receiving, the simulated electronic circuit was able to detect low frequencies and discover P and S waves. In this work, input signal was designed between frequencies of.1 Hz to 1 Hz for filtering. In this system, the input was usually between and 2 mv and total amplification was equal to 2. Also, loss of filter effect was considered. For this purpose, the first class was amplified only by 1 times to display only main signal. In this class, noise signal was also amplified, which was done after high pass filtering by 3 times so that the signal resulting from earthquake sensor could reach to volts. In future works, analog to digital signal conversion circuit using microcontroller will be designed by giving information to computer. REFERENCES [1] Glossary in Seismology, Available: [2] Seismic Wave Behavior_effect on Building, Available: 6/SeismicWaveBehavior_Building.pdf [3] Peter M.Shearer, Introduction to Seismology:The wave equation and body waves, Institute of Geophysics and Planetary Physics, Scripps Institution of Oceanography, University of California, San Diego, Notes for CIDER class, june 21. [4] Peter M.Shearer, Introduction to Seismology, Second Edition, Scripps Institution of Oceanography University of California,, San Diego. [] Larry Braile, Identifying the S arrival on AS-1 Seismograms and estimating distance using the S minus P method, Purdue University, February 28, Available: htm [6] Basic principles of seismology, Geophysics 21, University of Alberta, September 28, Available: [7] EARTHQUAKE SEISMOLOGY, Available: hquakes/seismologyhandout/labhandout.pdf [8] Giuseppe Olivadoti, Sensing, Analyzing, and Acting in the First Moments of an Earthquake, Analog Dialogue 3-1 (21). [9] Earthquakes and Seismic Waves, Designed to meet South Carolina, Department of Education, 2 Science Academic Standards, Available: ftp:// /geology/education/pdf/earthquakes.pdf [1] AD62, Low Cost Low Power Instrumentation Amplifier, Analog Devices, Available: [11] Bruce Carter, Thomas R. Brown, Handbook of Operational Amplifier Applications, Texas Instruments, October 21, Available: [12] Ron Mancini, Editor in Cief, Op Amps for Everyone Design Refrence, Texas Instruments, August 22. [13] Bruce Carter, Using the Texas Instruments Filter Design Database, Texas Instruments, July 21. [14] LM741 Operational Amplifier, Texas Instruments, May 1998, Available: [1] Murari Kejariwal, Jerome Johnston, Timothy Hopkins and Prashanth Drakshapalli, A Low-Power High-precision Self-testing Data Acquisition System for a Large Seismic Exploration Grid, Cirrus Logic Inc, 291 Via Fortuna, Austin, TX 78746, USA, 2. 66