TO:                  Deanna Young

FROM:            KoES Engineering

SBJECT:          Seismometer Testing Plan

DATE:             4/3/03


Goals of Project:

Many K-12 schools have been looking for a cheap and easy way to show students how a seismometer works.  Currently this is not available because a seismometer that would be sensitive enough for useful educational use is far too expensive. The AS-1 seismometer [1], designed by Dr. Leland Timothy Long, is currently the most fit for these criteria but its cost of $500 makes it too expensive for most schools.  It has been the goal of Incorporated Research Institutions for Seismology (IRIS) to develop a cheap and precise seismometer that would be both affordable and useful for K-12 schools [2].

KoES Engineering has been employed by Dr. Tom Boyd and Dr. John Lahr to build a seismometer that will fulfill the following conditions:

·        Inexpensive, $150 or less.

·        Sensitive, will record seismic wave of magnitude 6 or higher from anywhere in the world.

·        PC compatible, records data onto a computer.

·        Designed for and assembled by individuals with minimal knowledge of seismometers and earthquakes.

·        The period needs to be at least 6 s to 12 s.


Subsystem Overviews

The previous team that originally designed the seismometer in the past semester fulfilled the majority of these requirements. A few of the requirements, however, were not completely satisfied. Our team must improve the sensitivity of the seismometer (the ability to detect an earthquake of magnitude 6 or greater on the Richter scale), the magnet coil system, the integrated circuitry (specifically the amplifier/filter circuit), the analog-to-digital converter, and the overall structural design.

     The Folded Pendulum

In order to make a working seismometer, we need a structure that can handle periods as slow as 6 seconds.  A folded pendulum design will best fit our needs.  In addition to being able to detect low periods, the folded pendulum has a small structure with visible movements.

     The Magnet-Coil System

We need some way to record the displacement, velocity, or acceleration of the system. After exploring the different available options we decided to use a magnet coil system which measures the velocity of the ground motion through magnetic induction. A magnet coil system is the cheapest velocity sensor. It does not require the precise calibration that a displacement instrument needs, and is less expensive than an acceleration sensor. As well as being sensitive, the magnet coil system will serve as a dampening system which is necessary for any long-period seismic sensor.

     The Amplifier Filter Circuit

An amplifier/filter circuit is needed for the seismometer to correctly record data. The seismometer needs to detect motion as small as 5 μm.  Because of this minimal motion, the voltage induced by the magnet coil system is extremely small and needs to be amplified in order for a computer to record the data.  Electrical and mechanical interference also need to be filtered to record a clean signal. The amplifier filter circuit will amplify the signal and filter out almost all unwanted noise.

     The Analog to Digital Converter

Finally, we will need to send the signal from the amplifier/filter circuit to the computer to be interpreted. The signal from the amplifier/filter circuit is analog. However, a digital signal is required for the computer to record data. An analog-to-digital converter is needed to convert the analog signal to a digital signal. Once the digital signal is sent to the computer, Alan Jones's software, AmaSeis, will collect and analyze the data [3].


After all of the subsystems are assembled, we must test the final product to make sure the seismometer is working properly. Four tests will be conducted.  The first test is a simple displacement test. This test will measure the period and damping of the system. The second test will compare our seismometer's response to that of an existing seismometer. The third test will be conducted by our clients and will involve the detonation of dynamite near the seismometer. The final test will consist of leaving the system running for extended periods of time waiting to record data from an actual seismic event.


Research and Development


Folded Pendulum




A folded pendulum best meets the specifications presented by the client.  The folded pendulum is small and inexpensive to build, extremely sensitive, and is reasonably easy to construct.  It is also easy for students to visibly see how the folded pendulum works enhancing their learning experience. 


The folded pendulum design appeared to be the best choice since the main goal was for visual learning. With this design, students can discover the motion which creates earthquakes.  Some folded pendulums have reached periods upwards of 40 [4].  Overall the folded pendulum is an affordable seismometer compared to other designs, since one constraint was to keep the seismometer cost under $150.  Please refer to Figure 1.1 to visualize our folded pendulum in completion and refer to Figure 1.2 to see the design of the folded pendulum.


Figure 1.1 Folded Pendulum






Figure 1.2 Folded Pendulum Design


The folded pendulum is constructed from two pendulums, consisting of a gravitational/vertical pendulum on one side of the seismometer and an inverted pendulum on the other side.  These two pendulums suspend a horizontal mass which moves when there is seismic activity.  The horizontal mass is suspended by brass shims which are all in tension.  When the horizontal mass oscillates during an earthquake, the two pendulums counteract each other’s forces. This prevents the seismometer from collapsing [5].  Please refer to Figure 1.3 for a schematic view of the folded pendulum.


Figure 1.3  Diagram of Folded Pendulum



A moveable mass is located on the suspended horizontal beam of the seismometer.  If the mass is moved too close to the inverted pendulum, the seismometer will collapse.  If the mass is moved too close to the vertical pendulum, however, it will shorten the period of the folded pendulum [5].  It is possible for the folded pendulum to collapse if the moveable mass is moved to close to one of the pendulums.  If the mass is placed to close to one of the pendulums, that pendulum would control the entire system which can possible cause it to collapse.






A major problem with last year’s pendulum design is the joints.  The previous folded pendulum used nylon string as joints to suspend the horizontal beam.  There are many problems in using nylon string, however.  The nylon string can hold water and has a high thermal coefficient [6].  This can make is possible for weather changes to affect the accuracy of the seismometer.  Several ideas were considered to solve this problem, including using high strength alloys, flex pivots at all the joints, and finally using  thin brass shims.


High strength alloys are one possible way to build a seismometer’s joints.  One of the many benefits of titanium alloys and maraging steels is their stiffness.  When the dimensions and the stiffness are known, it is easy to calculate the frequency of the folded pendulum [5].  However, as a team we decided not to use the steels because of their high price.  Also, it is difficult to obtain the correct dimensions of the alloy to make the frequency calculations [7].  


Another way to construct the joints of the folded pendulum is with flex pivots.  At first, these seemed like the best way to build the seismometer's joints.  The flex pivots come in 10 different sizes, depending on the load they need to support.  Flex pivots were not used, however, because four flex pivots costing between $10 and $20 each would be needed [8]. 


Thin brass shims stocks are the third possibility for the joints of the folded pendulum.  Shim stock is inexpensive, easy to assemble, and provides a long period of accurate data [7].  Since brass shim stock is very thin, the restoring forces it puts on the horizontal beam are negligible.  Please refer to Figure 1.4 to see the brass shim stock joints.



Figure 1.4 Brass Shim Stock Joint





The pendulums are constructed of aluminum, which is nonmagnetic and easily machined.  Other benefits of aluminum are its availability at local hardware stores, its rigidity, and its affordability [8].   The pendulum could have been built partially out of steel, which would have made the pendulum rigid, but steel’s magnetic characteristics would affect the magnetic field of the magnet-coil system [6].  Aluminum also allows our design to work within the constraints of cost and strength.


          Moveable Mass


One of the most important parts of the folded pendulum is the moveable mass, which is located on top of the horizontal floating beam.  It adjusts the period of the folded pendulum and maximizes it [4].  The beam where the mass is located is constructed from a  nonmagnetic, threaded brass rod, allowing for precise adjustments to lengthen or shorten the period.  To visualize the moveable mass, refer to Figure 1.5.



Figure 1.5 Moveable Mass      


Magnetic Induction




The magnetic induction subsystem is a vital component for detecting an earthquake.  When an earthquake occurs, the P waves (compression waves) and S waves (shear waves) travel perpendicularly through the ground as seismic waves [3].  The ground motion created by the earthquake is translated into a horizontal oscillation by the folded pendulum.  A voltage is then generated through magnetic induction using a magnet-coil system.  This voltage is the signal that is eventually fed in to the computer for data recording.                                                




A magnet-coil system is used to convert seismic movements into an electronic signal.  The resistance of the wire, the size of the coil, the strength of magnet, the number of turns in the coil, and the speed of the coil moving through the magnetic field are all variables that need consideration for the magnet-coil system.  Although many variables are unknown, simple physics equations help solve for these uncertainties.  The first equation represents the induced voltage:   [10]


V=vBl,                                                                         (1)


where V is voltage, v is the velocity at which the coil moves through the magnetic field, B is the strength of the magnetic field, and l is the length of the wire.  The strength of the magnet, B, and velocity, v must be maximized in order to maximize the output voltage, V [11].  The size of the coil will play a key part in the induction process according to: [12]


            ξ = N (dB/dT) A,                                                         (2)


where ξ is the induced voltage, N is the number of turns in the coil, and A is the area containing the changing flux represented by dB/dT.  The flux change is extremely small since the coil is moving slowly.  Therefore to maximize induction, the most turns of coil possible are needed.  By maximizing the value of A, the greatest number of “field lines” is collected.  However, if A is too large or there are too many turns, the resistance of the wire increases due to additional wire length.  This additional resistance will minimize the gain of the amplifier/filter circuit, which will be discussed later.  Therefore the size of the coil and the magnet arrangements are determined purely through experimentation.




A 30 AWG magnet was chosen for the coil due to its availability, size, and properties of the wire.  It has a low resistance, approximately 2.22 mΩ/cm.  The thinness of the wire (0.05 mm2) allows for maximum windings while still keeping the coil thin.


Bar magnets from the Colorado School of Mines Physics department are used for the magnet.  We used these magnets because they were both strong and relatively small.  The bar magnets can support between 15 and 20 lb and are only a few square inches in area [13]. 


Wire coils of radii 3.3 cm, 4.25 cm, and 2.3 cm were selected for experimentation.  Each coil has 50 windings.  The wire is wound 50 times because of a limited amount of wire. 


The physics department gave Team KoES a rigid pendulum of length 40 cm for experimentation.  Since the experiment was comparing voltages, the length of the pendulum was not important.  The four arrangements of magnets used for each coil is seen in Figure 2.1.  We mounted the coils onto the pendulum and give a small initial displacement of approximately 2 cm.  As the coil passed through the magnetic field the voltage range was read from the voltmeter.  There was a lot of error in reading the voltage because the readings fluctuated from the lack of precision of the measuring device.  The experiment was repeated several times until a consistent range was determined for each coil.  The range of values represented one period of the pendulum.


This experiment determined the best magnet arrangement and size of the coil.  The period of the coil through the magnets in experimentation was approximately three seconds.  In conjunction with the seismometer, we will obtain periods of up to 20 seconds.   All other variables were held constant for all experiments so that these two desired unknowns can be determined.

Magnet A – 2 bar magnets separated 1.6 cm with the north and south poles facing

        one another. 

            Magnet B – A horseshoe magnet.  The field was tested using a compass to

         determine where the field was the strongest.  The coil passed through  

         this region.  It turned out that the coil was centered over the magnet

            Magnet C -  Two bar magnets with south and north facing one another.  Offset in

         such a way that the individual field directly affects one side of the


            Magnet D -  One bar magnet offset over the coil


Figure 2.1 - Arrangement of magnets


The magnets were slightly offset so that the center of the coil was not over the center of the magnet.  This offset allows for greater magnetic fluctuation due to the flux changing more rapidly through the area of the coil [14].  The exact positioning of the magnet was found by experimentation.  By moving the magnet in correlation with the coil, peak values were determined.  Where these peak values were recorded was where the magnets were placed.  These offsets are seen in the photographs of Figure 2.1.  Please refer to Table 2.1 for coil resistances and voltages.

Table 2.1  -  Values of resistances and voltages for each trial of experimentation


Resistance (Ω)

Voltage Range (mV) Magnet A

Voltage Range (mV) Magnet B

Voltage Range (mV) Magnet C

Voltage Range (mV) Magnet D

Coil A

(2.3 cm radius)


3.6 to 4.8

1.7 to 2.4

0.7 to 1.0

0.8 to 1.2

Coil B

(3.3 cm radius)


2.3 to 3.4

0.9 to 2.1

0.6 to 1.2

1.0 to 1.9

Coil C

(4.25 cm radius)


1.8 to 2.7

1.4 to 2.4

1.2 to 1.9

1.6 to 2.7


(Table 2.1 note:  The range of voltage is due to error in the voltmeter.  These voltages are the values that are the average minimum and average maximum value as the coil completed 1 period.  These are not the voltages that will be outputted by the seismometer since the period would be much longer and damped.  These values were only used to determine the best size of coil and arrangement of the magnets.)




The best results came from the 2 bar magnets facing each other at a separation of about 1.6 cm and a coil radius of 2.3 cm.  This arrangement yielded the highest voltage of all 12 trials.  This is the size of the coil and the arrangement of the magnets that is used in our final project. 


The number of turns, N, must be maximized according to equation (2).  We constructed a coil consisting of 1000 wraps to satisfy this equation and to produce a larger radius due to additional wraps (please see Figure 2.3).  The 1000 wrap coil was too thick for the 1.6 cm separation, however, so 1.8 cm is used instead.  In addition, the coil was accumulating a relatively high mass, which if becomes too massive, can over damp the system.  The coil had a total resistance of  44.9 Ω.


Figure 2.3 – Arrangement of the magnets and the size of coil for the seismometer


Amplifier/Filter Circuit




The amplifier/filter circuit is a key component of the seismometer.  It is used to amplify the signal from the magnet-coil system and to filter out unwanted or electronic noise detected by the seismometer.  Without the amplifier/filter circuit, the data recorded by the software will contain unwanted noise from footsteps, highway traffic, electronic interference, etc.  The signal from the magnet-coil system will also be too weak for the computer to read.  This is where the amplifier is used.  The amplifier will greatly increase the signal produced by the magnet-coil system, making the signal readable by the computer software.  Without the amplifier/filter circuit, the whole seismometer would be almost completely useless for data recording.



     Design Possibilities


Several different designs for the amplifier/filter circuit were considered.  John Lahr designed the first amplifier/filter circuit that was considered being used (See Fig. 3.1).  Due to some technical difficulties and cost issues, however, this design was not used.  The technical difficulties included unknown electrical problems that could not be fixed before the due date.  Also, the estimated cost of this design was approximately $85.00 for one unit, making it slightly over budget [15].


Figure 3.1 – John Lahr’s Amplifier/Filter Circuit [15]


Andy Loomis designed the second amplifier/filter circuit [16], which is the design that we will use in the final project.  Loomis’s amplifier/filter circuit is very similar to John Lahr’s, but cost less and has a slightly higher degree of accuracy (See Figure 3.2). 



Figure 3.2 – Andy Loomis’s Amplifier/Filter Circuit [16]


The total cost for this amplifier/filter circuit is roughly $35, about $50 less expensive than John Lahr’s design.  The details of this design will be discussed later in the specifications section of this report. 




Structurally, the amplifier/filter circuit consists of two major parts.  They are the amplifier and the filter.  The amplifying unit is where the majority of the signal amplification takes place, as well as a small degree of filtration.  The filtering unit filters out the rest of the unwanted noise.  These two parts are needed to successfully clean and amplify the signal for the computer to use. 




The first section of the amplifier/filter circuit is the amplifier (See Figure 3.3).  The AC signal coming from the magnet-coil system is connected to a MAXIM 420CPA CAZ inverting operational amplifier.  The CAZ, or commuting auto-zero, operational amplifier is unique in its ability to sense its own internal offset voltage and automatically correct for it.  These types of operational amplifiers are commonly used in low-pass filters.  A 10 MW feedback resistor creates a gain of approximately 1000 depending on the resistance of the pickup coil (See equation 3).  The overall gain of the amplifying unit is defined by the equation,


g = 10 MW / (10 kW + R),                             (3)


where the R is the DC resistance of the coil [17].  In parallel with the 10 MW feedback

resistor, a 0.01-µF capacitor is incorporated to filter-out any 60 Hz power line noise.  The 60 Hz power line noise needs to be filtered out because it is a common type of electronic noise associated with AC current from wall outlets.  



Figure 3.3 – Andy Loomis’ Amplifier Unit (without filter) [17]



One drawback in using a CAZ operational amplifier is its extreme sensitivity to the noise produced by the electronic parts it follows.  To eliminate this problem, a load impedance of 10 kW is used prior to the CAZ operational amplifier.  Also, one section of an additional dual operational amplifier (labeled A in Figure 3.3) is used to isolate the CAZ operational amplifier from unwanted electronic noise.  Any typical dual operational amplifier can be used since there is no need for additional amplification. 


With the addition of the dual operational amplifier, a second-order filter is created.  This filter will eliminate any noise with a frequency of 0.31 Hz, or noise with a period less than 3.2 seconds [17].  The CAZ operational amplifiers along with the secondary dual operational amplifier create the main amplifying unit of the amplifier/filter circuit.




The final piece of the amplifier/filter circuit is the filter (see Figure 3.2).  While the amplifying unit may eliminate 60 Hz power line noise, movements with a period less than 3.2 seconds, and some other electronic noise, the cutoff is not sharp enough to remove other faster-moving interferences.  These interferences include things like footsteps, nearby traffic, and other ground disturbances.  The filtering unit’s purpose is to remove any signal that has a period less than 1.0 second.  For the filter, the second section of the dual operational amplifier is used, creating a fourth-order low pass filter.  Because of its much sharper cutoff characteristic, a fourth-order filter has a higher cutoff.  Therefore, the filtering unit along with the other parts of the amplifier/filter circuit will remove almost any disturbance with a period of less than 1.0 second [17]. 




With both sections of the amplifier/filter circuit in place, an amplifier and fourth order, 1.0-Hz, low pass filter is created.  This unit will sufficiently amplify the signal produced by the magnet-coil system, as well as filter out unwanted noise.  This unit is both cheap and accurate, making it an excellent amplifier/filter circuit. 

Analog-to-Digital Converter


The purpose of the AD converter is to take the signal received from the magnet coil, after the signal passes through the amplifier filter circuit, and digitize it.  Analog signals are unreadable as a computer input, whereas digital signals are not.  The team must output the signal from the seismometer to a serial port on a computer.  The data is then read into the computer by a software program, recorded, and displayed on-screen.  For the Fall 2002 design, the Dataq DI-194RS Low Cost Starter Kit was used [4].  This kit costs $25 and includes the following:




Figure 4.1 – DATAQ DI-194RS Project kit from DataQ Instruments



This AD converter works well; however we decided that it could be streamlined.  Logging software, two digital input, and ActiveX Controls can all be eliminated to minimize supplies, manufacturing time, and cost.  An AD converter can be assembled from individual pieces and be tailored to fit the constraints of the project.  The constraints for the project that apply to the AD converter are all provided in Table 4.1.


Several alternatives to the DATAQ Instruments were researched, but few fulfilled the constraints shown in Table 4.1.  For alternatives, mostly homemade kits were considered.  Several electrical component suppliers were considered including, which has the best prices for this part.


Table 4.1 - Subsystem Specifications



$150 (whole system), $25 AD converter

Power Usage

Either powered by serial port/USB BUS or ~110 V wall outlet


Compatible with software client, and Windows or Macintosh operating system

Ease of Use

Easy to assemble and operate; high school teacher with little or no mechanical and fabrication skill can construct




     Design Possibilities


          Design Alternative 1 – 8-Channel 12-Bit AD converter (See Figure 4.2)


This unit would need to be constructed by hand by somebody who has knowledge of both integrated circuitry and soldering [20].  This unit is reported by the web page to cost $25, making it a poor candidate because of the labor and technical knowledge needed to construct it.  All the circuitry for the seismometer will need to be self-contained, whereas this option’s sensitive components are exposed.  We would need to add about $5-10 for a storage/secure box unit to enclose the circuitry.



Fig. 4.2 – 8 Channel 12 Bit ADC Schematic from RadioSky



          Design Alternative 2 - Use an AD and PC Interface Chip, as shown in Figure 4.3.

We could obtain a PIC14000-04/SP Programmable Mixed Signal Controller and program it using firmware that would need to be designed using a programming language such as Pascal. Programming the chip is beyond an average science teacher’s technical knowledge.  This chip also requires more power than can be provided by the serial port on the PC, so an alternative power source would have to be set up and configured, which is also beyond an average science teacher’s technical knowledge.



Figure 4.3 – Circuit Board for Short Period Seismometer.  AD Converter is the black,   horizontal chip on the right, as indicated by the picture text.  Photo from Infiltec.


          Design Alternative 3Public Seismic Network AD converter to serial 16-Bit serial 

                                                  output AD converter board (See Figure 4.4). 


This board is one of the most advanced, efficient boards for a seismometer.  The Public Seismic Network is a group of scientists who help and educate individuals interested in seismic studies. 


An AD converter board that is endorsed by this expert group of avid seismologists suggests that the converter is reliable and of high quality.  However, this board is more than what is needed for the KoES Engineering seismometer.  It has inputs for a regular twenty-five pin connector, and a J4 communication port connector which are difficult to set up with our current seismometer design.  It also has a GPS connector which is used to communicate to a GPS receiver.  To maintain the constraint of price, a GPS receiver we cannot consider.  Therefore, it would be pointless to have the extra inputs if the goal of the project is to streamline the existing AD converter.  This AD converter is also not appropriate considering the price tag of $99 is over one of the constraints of the project.



Fig 4.4 – Integrated PSN-ADC-Serial Board from Seismicnet.


              Project Constraints





Power Usage


Ease of Use

Overall Score

Design Alternatives

Option #1 -                       8-Channel 12-Bit A-D Converter






Option #2 -                   A-D & PC Interface Chip






Option #3 -                  PSN A-D Converter - serial 16-Bit output board






Original Option - DATAQ DI-194RS Project Kit






Figure 4.5 – Decision Matrix.  Scale 1 through 5; 5 highest.


Specification Comparison: A-D Converters


Number of Channels



Software Included?

Technical Knowledge Required?


Option #1 -         8-Channel 12-Bit A-D Converter

8 Input Channels

Serial Port Output


A2DMAX Data Collection Program, not included

Extensive, for circuity construction


Option #2 -         A-D & PC Interface Chip

2 Input Channels

Serial Port Output


Amaseis seismic data acquisition program, not included

Extensive, for circuity construction


Option #3 -         PSN A-D Converter - serial 16-Bit output board

8 Analog Input Channels

Serial Port Output


Compatible with WinSDR datalogging software, not included



Original Option - DATAQ           DI-194RS Project Kit

Four ±10V Analog Inputs

Serial Port Output


Yes - WinDaq/Lite and WinDaq Waveform Browser




Figure 4.6 – Specification Comparison Chart


Considering all of the above options, as presented in the decision matrix(see figure 4.5), and the specification sheet (see figure 4.6), it appears that the total AD converter package is best exemplified by the cheap and efficient DI-194RS Starter Kit produced by Dataq Instruments.  It is the only AD converter that can be used right out of the box for under $30.  It requires almost no technical or circuitry knowledge.   It is by far the most user friendly; however, the search is continuing for other feasible opportunities, because the KoES engineering team wants to design the best and cheapest seismometer. 


Seismometer Testing


     The Swing Test

The purpose of this test will be to maximize the period and to optimize the damping of the folded pendulum. The standard pendulum support creates a maximum leftward displacement for the system, as shown in Figure 5.1. For the swing test we will pull the system to this leftmost position and let it go. Allowing it to swing back to its point of equilibrium will let us test the pendulums period and damping. By repeating this experiment and by adjusting the movable mass we will maximize our systems period and optimize the damping.

The AmaSeis software is used to interpret the data from the seismometer.  The data gathered accomplishes several things. The period is determined by calculating that amount of time it takes for the system to complete one full oscillation. There are two things allow for period adjustment. Adjusting the mass to the right will increase the period. The problem with this is if the mass is shifted too far right, the system becomes unstable. The second thing we can adjust to increase the period is the damping.


Figure 5.1 - Swing Test for a Folded Pendulum

Damping is a way of slowing down the motion of a system. Proper damping will prevent the system from oscillating continuously. In this case proper damping is achieved by allowing a current to pass through the coil.  To optimize the system's damping we will compare the amplitude of the first oscillation to the amplitude of the second oscillation. Research shows that the ideal damping for any seismometer system is between .7 and .8 [21]. This means that with each complete oscillation the amplitude of the system will be between 30% and 20% of the previous amplitude.

     The Step Test

The Step Test is designed to compare the data gathered by our seismometer to that of a known working seismometer. For this experiment we will use the AS-1 seismometer, which is already available to us through the EPICS department. The AS-1 seismometer and our Folded Pendulum Seismometer are placed next to each other on a concrete platform. Each seismometer will be collecting its own set of data.

To conduct the test, a person must stand three to four feet away from the seismometers. The person’s weight is concentrated on a point that is approximately four inches away from each sensor. This experiment is done on concrete because the amount of compression caused by a person standing on concrete is comparable to the amount of motion during an earthquake [22]. Stepping on the concrete will cause both seismometers to shift out of equilibrium. When stepping away from the seismometers, it will allow the pendulums to swing back to equilibrium. The following table (Table 5.1) is the data table we will use to compare our data in this experiment.


Table 5.1 - Step Test Data Table



















Folded Pendulum












The data collected by our seismometer is then compared to the data collected the AS-1 seismometer.  This allows us to see where our seismometer has inaccuracies and inconsistencies.

     Client Testing and the Field Test

Once the folded pendulum is adjusted to collect the best data, two additional tests are conducted. One test will be to set up our system and then explode dynamite nearby. Our clients will conduct this test, which allows them to see if our seismometer is capable of accurately collecting seismic activity. The final test is to leave the seismometer running until an earthquake of magnitude 6 or higher occurs anywhere in the world. We will then compare our data to the data from seismometers at the USGS. 

     Testing Conclusions

No actual testing could be done for several reasons.  The first error occurred when the amplifier/filter circuit was destroyed.  The positive and negative terminals of the operational amplifier were switched, causing the operational amplifier to short out.  Due to time constraints and budget costs, we were unable to reconstruct the amplifier/filter circuit.  Secondly, the potentiometer’s offset voltage used in the amplifier/filter circuit was incorrect.  This caused the dual operation amplifier to be unbalanced, which created inaccurate data.


Final Product


The overall construction and assembly of the folded pendulum took 10 hours (See Figure 6.1).  Team KoES feels that our seismometer successfully fulfills our clients’ needs in every way.  Our seismometer is both inexpensive and accurate, making it ideal for use in K-12 learning conditions.  If there is any additional information that is requested or concerns about the seismometer please use the following contact information to get in touch with the appropriate individual.




Figure 6.1 Inverted Pendulum  





[1]  Long, Dr. Leland T. “Calibration of the AS-1 Seismometer”. [Online] 2001, (Accessed 15 March 2003).


[2]  Aster, Barker, Braile, et al. “IRIS Education and Outreach Program Plan”. [Online] 2002, pg. 15-16, (Accessed: 15 March 2003).


[3]  Seismic Waves, “Seismic Waves”, [Online] 2003


            10  February 2003).


[4]  “Digi-Key Corporation Home Page,” Digi-Key Corporation, [Online] 2003,         (Accessed: April 27, 2003).

[5] A. Bertolini, “High Sensitivity Accelerometers For Gravity Experiments,” Ph.D.        dissertation, Universita Degli Studi Di Pisa.


[6]  L. Woo, “Seismographic Instrumentation,” (Wave Measurements Using A Folded Pendulum), [Online] 2003,  10 February 2003).


[7]  J. Winterflood, “High Performance Vibrartion Isolation For Gravitational Wave         

             Detection,” Ph.D. dissertation, University of Western Australia, Australia, 2001.


[8]  J. Lahr, “Folded Pendulum,” (Folded Pendulum), [Online] 2002, (Accessed:  10 February 2003).


[9]  D. Youden, (, “Re: Folded Pendulum,” Email to D. Hurelle 

            (, 28 February 2003.


[10]  Paul A Tipler, Physics for Scientists and Engineers, 4th ed. New York, NY; WH   

               Freeman and Company, 1999.


[11]  M Young, CSM Physicist, personal interview, March 2003.


[12]   Magnetic Design, Eight Magnetic Axioms, [Online] 2003, 

      (Accessed: 21 February 2003).


[13]   A. Corn, CSM Physics Laboratory Professor, personal interview, March 2003.


[14]   Physics Push-Ups, Answers, [Online] 2003,`jordanrg/push-

               ups_30.htm  (Accessed: 3 February 2003).


[15]   J. Lahr, “Amplifier-Filter Circuit,” [Online] 2003, (Accessed: 24 March, 2003).


[16]   A. Loomis, “Seismic Filters,” [Online] 2001, 

       (Accesses: 18 March, 2003).


[17]   A. Loomis, “The Preamplifier and Filter,” [Online] 1999,

       (Accesses: 18 March, 2003).


[18]  A.R. Hambley, Electrical Engineering, 2nd ed., New Jersey: Prentice Hall, 2002.


[19]  “Microchip Graphic Explorer – Parent Tab: PIC14000 Device,” [Online] 2003,           000/index.htm (Accessed: April 27, 2003).

[20]  “A2dmax,” Radio-SkyPipe [Online] 2003,    (Accessed: April 9, 2003).

[21]  Keller, Fritz. “The Seismometer Demo Applet,” [Online] 2003, http://www.ifg.tu-

       (Accessed: April 9, 2003).


[22]  Lahr, John C. KoES Engineering Project Plan Meeting. Interview, March 2003.


[23]  Young, Matt. KoES Engineering Project Plan Meeting. Interview, March 2003.


[24]  Wire Magnet. [Online]                    2C+Magnet  (Accessed: 12 March 2003).




“a2dmax,” Radio-SkyPipe. (April 9, 2003)

A. Bertolini, “High Sensitivity Accelerometers For Gravity Experiments,” Ph.D. dissertation,      Universita Degli Studi Di Pisa.


A. Corn, CSM Physics Laboratory Professor, personal interview, March 2003.


A. Loomis, “Seismic Filters,” [Online] 2001, (Accesses: 18 March, 2003).


A. Loomis, “The Preamplifier and Filter,” [Online] 1999, (Accesses: 18 March, 2003).


A.R. Hambley, Electrical Engineering, 2nd ed., New Jersey: Prentice Hall, 2002.


Aster, Barker, Braile, and more. “IRIS Education and Outreach Program Plan”. 2002, pg. 15-16.


Chase, Steven B. High-tech Inspection. Civil Engineering. Vol. 71. pg. 62. 9-01-2001.


“Digi-Key Corporation Home Page,” Digi-Key Corporation, (April 27, 2003)


D. Saum, “How to build an Inexpensive Seismometer,” [Online] 2001, (Accesses: 18 March, 2003).


D. Schneider, “Journey to the Center of the Earth,” American Scientist. Vol. 88, p. 400, 9-01-            2000.


D. Youden, (, “Re: Folded Pendulum,” Email to D. Hurelle             (, 28 February 2003.


J. F. Shackelford, Materials Science for Engineers, 5th  ed., New Jersey:  Prentice Hall, 2000.


J. Lahr, “Amplifier-Filter Circuit,” [Online] 2003,           (Accessed: 24 March, 2003).


J. Lahr, “Folded Pendulum,” (Folded Pendulum), [Online] 2002,  (Accessed:  10 February 2003).


J. Lahr C.  “Local Magnitude”. 2002,


J Lahr C. KoES Engineering Project Plan Meeting. Interview, March 2003.


Jones, Alan L. “AmaSeis”.


Jordan. “Physics Push-ups”.


“INFILTEC: How to Build an Inexpensive Seismometer – make seismograph kit,” Infiltec.

   (April 9, 2003)

Jones, Alan L. “AmaSeis”.


J. Winterflood, “High Performance Vibrartion Isolation For Gravitational Wave Detection,”       Ph.D. dissertation, University of Western Australia, Australia, 2001.



Keller, Fritz. “The Seismometer Demo Applet”.  4/8/2002, http://www.ifg.tu-



Long, Dr. Leland T. “Calibration of the AS-1 Seismometer”. 2001,


L. Woo, “Seismographic Instrumentation,” (Wave Measurements Using A Folded Pendulum),   [Online], (Accessed:  10   February 2003).


Magnetic Design, Eight Magnetic Axioms, [online] 2003, (Accessed 21 February 2003).


“Microchip Graphic Explorer – Parent Tab: PIC14000 Device,” Microchip, Inc.

     000/index.htm (April 27, 2003)


M. Mariotti, “Seismic Signal Conditioning,” [Online] 1999, 

   (Accesses: 18 March, 2003).


M Young, CSM Physicist, personal interview, March 2003.



P. A. Tipler, Physics for Scientists and Engineers, 4th ed., New York:  Freeman and Company,         1999. 



Physics Push-Ups, Answers, [online],`jordanrg/push-ups_30.htm    (Accessed 3 February 2003).

Piezo Systems Inc. “History of Piezoelectricity”.

 “PSN-ADC-SERIAL Board,” Seismicnet,

   (April 9, 2003)

 “Real-time PC-based Data Acquisition DI-194RS Low Cost Starter Kit,” Dataq Instruments

   (April 9, 2003)

R. Peters, “Creep and Mechanical Oscillator Damping,” [Online] 2001,    (Accessed:  24 February 2003).


R. King, MEL I Experiments, CSM: Dec. 2002.


S. Chase, “High-tech Inspection,” Civil Engineering. Vol. 71, p. 62, 9-01-2001.


Schneider, David. Journey to the Center of the Earth. American Scientist. Vol. 88. pg. 400. Published in 2000.


Wire Magnet. [Online] Accessed (12 March 2003)   

Young, Matt. KoES Engineering Project Plan Meeting. Interview, March 2003.