Needs Assessment - EDGE



[pic][pic][pic]

[pic]

Launching Station for Sensor Deployment

|Dr. Jeff Kozak |Dr. Wayne Walter |Dr. Ferat Sahin |Brian M. Molnar |

|Project Supervisor |Robotics Mentor |EE Mentor |PROJECT MANAGER |

|jdkeme@rit.edu |wwweme@rit.edu |feseee@rit.edu |bmm4813@rit.edu |

|Shawn McGrady |Joseph Liquore |Kenneth Schroeder |Michael Shahen |

|Mechanical Engineer |MECHANICAL ENGINEER |MECHANICAL ENGINEER |CONSULTANT |

|sdm7316@rit.edu |jml8369@rit.edu |kts6383@rit.edu |mas8940@rit.edu |

Midterm Design Review

Submitted on:

February 21, 2003

10:15AM

Submitted to:

Dr. Jacqueline R. Mozrall

Professor and Department Head of Industrial and Systems Engineering

Dr. Jeff Kozak

Professor of Mechanical Engineering, Project Supervisor

Dr. Vincent Amuso

Professor of Electrical Engineering

Mr. Mark Smith

Industry Representative

Table of Contents

Review and Preview 4

1 Executive Summary 4

2 Nomenclature 5

3 Figure Number and Location 6

1 Introduction 7

1.1 Motivation 7

1.2 Product Description 8

1.3 Scope Limitations 8

1.4 Stake Holders 9

1.5 Critical Performance Parameters 9

1.6 Previous Research in Sensor Deployment 10

1.7 Budget 11

1.8 Markets 11

1.9 Order Qualifiers 11

1.10 Order Winners 11

1.11 Background Literature Review 12

1.12 Mission Statement 14

2 Theory 15

2.1 Sensor Aerodynamics 15

2.2 Wind Effects 18

2.3 Mass Effects on Sensor Range 19

2.4 Stress in Pressurized Vessels 22

3 Design and Analysis 25

3.1 Brainstorming Session 25

3.2 Compressed Air Concept 27

3.3 Spring Concept 28

3.4 Wheel Drive Concept 29

3.5 Rubber Band Concept 29

3.6 Feasibility Assessment 30

3.7 Preliminary Design 33

4 Results and Discussion 34

4.1 Barrel Wall Thickness 34

4.2 Finite Elements Analysis of Barrel 35

4.3 Wind Sensor Resolution 37

5 Final Design 39

5.1 Pneumatics 39

5.2 Wind Sensor 43

5.3 Bolt, Barrel, and Receiver 46

5.4 Sensor 52

5.5 Sensor Encapsulation 52

5.6 G.P.S. 54

5.7 Angle of Launch Actuation 56

5.8 Initial Budget 58

6 Conclusion 60

6.1 Project Wrap-up 60

Bibliography 63

Appendices

A Modification of Needs Assessment

B Re vs. Cd for a Sphere

C Sensor Range Spreadsheet

D Specifications of G.P.S.

E Hand Sketched Drawings of Concepts

F CAD Drawings

Review and Preview

1 Executive Summary

This Midterm Design Review details the design process for a Launching Station for Sensor Deployment. The launching station will operate remotely via a currently existing robotic base. A Global Positioning System mounted onto the robot will allow for the operator to accurately position the launching station as well as keep track of launched sensors. A wind sensor will be mounted on top of the launching station that will detect wind magnitude and direction. With this information fed into the processing unit on the robot, the launch trajectory can compensate for initial wind conditions, for a more accurate placement of sensors. Deployment of only one sensor would not be an efficient utilization of the area on the robot, therefore a holding device for yet to be launched sensors must also be designed.

Dimensions and weight will be in the forefront of the team’s mind when designing any components for the launching station. There is a limited amount of space in which all components must be placed and must also be light enough as not to damage any of the sensitive electrical components below. The following pages explain the complete design process starting at the product description and ending with the final design.

2 Nomenclature

[pic] Drag Force

[pic] Coefficient of Drag

[pic] Air Density

[pic] Sphere Projected Area

[pic] Velocity

[pic] Reynolds Number

[pic] Reynolds Number Characteristic Length

[pic] Sphere Diameter

[pic] Viscosity

[pic] Force

[pic] Mass

[pic] Acceleration

[pic] Gravitational Acceleration

[pic] Vertical (y-axis) Acceleration

[pic] Horizontal (x-axis) Acceleration

[pic] Horizontal (x-axis) Displacement

[pic] Vertical (y-axis) Displacement

[pic] Time

[pic] Drag Constant Simplifier

[pic] Radial Stress

[pic] Tangential Stress

[pic] Yield Stress

[pic] Internal Pressure

[pic] Outer Pressure

[pic] Inner Radius

[pic] Outer Radius

[pic] Radius

[pic] Total Pressure

[pic] Free Stream Pressure

[pic] Change in Pressure

3 Figure Number and Location

1 Introduction

1.6.1 A Design of a Launching Station for Microrockets 10

1.11.1 Smart Dust Module 13

2 Theory

2.3.1 Flight Trajectory: Velocity 32ft/s, Angle 45 degrees 20

2.3.2 Horizontal Displacement: Velocity 32ft/s, Angle 45 degrees 20

2.3.3 Flight Trajectory Velocity 32ft/s, Angle 25 degrees 21

2.3.4 Horizontal Displacement Velocity 32ft/s, Angle 25 degrees 22

3 Design and Analysis

3.1.1 Initial Brainstorming Results 25

3.1.2 Brainstorming Results After Voting 26

3.1.3 Brainstorming Results with Majority of Support 27

3.1.4 Distribution of Concepts for Initial Sketches 27

3.6.1 Radar Plot of Feasibility Assessment 32

4 Results and Discussion

4.2.1 Barrel With 0.05-inch Mesh 35

4.2.2 Barrel With Boundary Conditions 36

4.2.3 Barrel With Applied Internal Pressure 36

4.2.4 Tangential Stress in the Barrel 37

5 Final Design

5.1.1 Pneumatics Flow Chart 43

5.2.1 Wind Sensor Assembly 44

5.2.2 Omega Pressure Sensors 45

5.3.1 Isometric View of Barrel 47

5.3.2 Isometric View of Bolt 48

5.3.3 Isometric View of Connector 49

5.3.4 Isometric Views of Receiver 49

5.3.5 Isometric View of Magazine 50

5.3.6 Receiver Assembly in Position 1: Reloading 51

5.3.7 Receiver Assembly in Position 2: Ready to Fire 51

5.4.1 Transponder used for Sensor 52

5.5.1 Sensor Assembly 54

5.6.1 G.P.S. Receiver 55

5.6.2 G.P.S. Antenna 55

5.7.1 Spur Gears 57

5.7.2 Worm Gears 57

5.8.1 Initial Budget 58

6 Conclusion

6.1.1 Model of Prototype to be Built 60

6.1.2 Spring Quarter Schedule 61

1 Introduction

1.1 Motivation

As technological and scientific knowledge increases there are new products that are created that tend to be smaller, less expensive, and require less human interaction. Smart Dust is the latest and smallest new development that fits this trend. The overall size of a Smart Dust Module is anticipated to be no larger than 1mm in any direction. With a specific sensor onboard, a Smart Dust Module could measure anything from temperature and humidity to nuclear and chemical residue. This lends the use of these modules to civilian use, but the most widespread use will most likely be by the military.

With today’s concerns over chemical and nuclear warfare, a device that is almost invisible to the eye that is capable of measuring the chemical composition of a remote location and transmitting the data back to a safe location would have a great deal of appeal to the military.

Sensors like mentioned above would be of little use alone. If a Smart Dust Module 1mm in size were launched into the air it would soon settle to the ground. To overcome this wings can be added to the sides of the cube. This allows the modules to float in the air like maple or dandelion seeds. Sensors launched into the air could stay up in the air for days before falling to the earth. In order to get the Smart Dust up into the air a launching station is needed.

The need for a launching station has been debated since the advantages are not well known. Launching sensors in a systematic method over a large area over a battlefield, or area of concern, would allow for virtually invisible surveillance for days. Not all of the applications are limited to military purposes. Sensors deployed over a wildfire or in a weather system would provide valuable information on the intensity, changes and movement of these systems. A more immediate use for the launching station that will be described in this report is to deploy miniature robots, or tethered sensors. A 10mm fully functional robot has been built and is currently undergoing testing at the LACOMS department at R.I.T. These robots could carry small cameras with the ability to travel down pipes or get into small crevices in a debris field.

A launching station does not have to be mounted exclusively on a robotic base. The launching station would be much more robust if it could be installed on a variety of buildings and vehicles such as: military vehicles, police cars, fire trucks, and airplanes.

1.2 Product Description

As research and development of millimeter sensor continues, there is an ever-growing need for a device that can accurately launch sensors. Previous research suggests that microrockets, that were initially designed to deploy sensors, are too erratic in flight to effectively and accurately deploy these sensors. Theory suggests that a sensor that is launched from a base station and contains no thrust capabilities onboard should have a much more predictable flight path.

Since the launching station will be placed on a robot, the size and weight of the components must be kept to a minimum. Future teams will pick up where this project leaves off and continue to reduce the size of projectiles that can be launched. This will continue until the launching station can effectively launch 1-2mm sensors. At this point the launching station can be marketed to organizations such as the Department of Defense.

1.3 Scope Limitations

The design of the launching station must be complete by February 21, 2003. A presentation of the design will be given to a review panel that must include the following:

• Complete CAD Drawing Package

• Bill of Material with initial budget

• Preliminary design documentation (Design Planner Notebook)

• Design Review Document detailing design process

The prototype and initial testing must be completed by May 16, 2003. Another presentation will be given to update the review panel on the progress at that time. The following must be included:

• Complete Prototype

• Documentation stating what prototype testing was conducted

• Final budget

• Complete design documentation (Design Planner Notebook)

• Design Review Document detailing design process

1.4 Stakeholders

The major stakeholders in this project are the team members and the faculty advisors associated with it. Another major stakeholder is the Mechanical Engineering Department whom is funding the project. Future senior design teams and graduate students working on sensor deployment will benefit from what is completed in this project.

1.5 Critical Performance Parameters

The minimum required performance parameters are the features that the launching station must include for the project to be deemed a success. The minimum required performance parameters are listed below.

• Launching station shall rest on a remotely controlled vehicle.

• Launching station shall be able to traverse over smooth ground and relocate based on remote inputs.

• System shall have G.P.S. to translate location to base station.

• Launching station shall be no larger than 12” x 12”.

• System shall have onboard wind sensors that automatically reposition the launch trajectory of sensor deployment based on wind conditions observed.*

The desired performance parameters are features that are not required for successful completion for the project and would be nice to have, time permitting. The desired performance parameters are listed below.

• System should have self-contained power supply.

• System should have load capacity to carry at least 20 additional sensors.

• Launching station should be as small as possible

• Launching station should be able to operate in atmospheric conditions ranging from 0°C to 50°C.

*Note: See Appendix A for an explanation of updated made to the critical performance parameters

1.6 Previous Research in Sensor Deployment

Previous work on sensor deployment at the Rochester Institute of Technology has concentrated on using a microrocket that is powered by an onboard power supply. Figure 1.6.1 shows an example of one design of a launching station for microckets. These microrockets were found to be instable and hard to control. By encasing the sensor in a protective cover and giving it in an initial velocity at a launching station, the flight path should much more predictable.

Figure 1.6.1 A Design of a Launching Station for Microrockets

[pic]

1.7 Budget

The Mechanical Engineering Department has provided a budget of $2,000 to fund this project. The budget must cover the following expenditures:

• Raw Materials

• Off the shelf components

• Electrical engineering/technician support

• Wind Tunnel calibration equipment

1.8 Markets

The Primary markets for the launching station are:

• R.I.T. Mechanical Engineering Department

• LACOMS Laboratory for Autonomous, Cooperative Microsystems

• Military monitoring/surveillance

Secondary Markets include, but are not limited to:

• Weather monitoring

• Wild fire monitoring

• Other monitoring/surveillance agencies

1.9 Order Qualifiers

The team shall design, build, and test a prototype launching station that can launch a 0.5–inch sphere 10 to 30 feet away from it. The launching station must rest on a remotely controlled vehicle, have integrated G.P.S., and include a wind sensor capable of measuring wind direction and velocity.

1.10 Order Winners

Bells and whistles that would be nice but are not critical to the project include:

• Completely wireless operation

• Strong enough to stand impact if dropped from aircraft

• Onboard tracking device for all deployed sensors

• Overall size less than a 6-inch cube

• Ability to operate in atmospheric conditions ranging from 0°C to 50°C.

1.11 Background Literature Review

One of the main motivations for the launching station is Smart Dust deployment. Smart Dust is a very small-computerized sensor that is capable of sending information from air borne particles. The size of Smart Dust has been decreasing rapidly due to increases in MEMS technology. The original contract for Smart Dust was for space applications. A Smart Dust module in space could monitor temperature and other conditions. They could serve as an early warning system for satellites in the earth’s orbit by warning of extreme temperatures or hazardous chemicals.

Researchers at the University of California at Berkley have built a prototype the size of a matchbox that is capable of measuring temperature, humidity and barometric pressure. Smart Dust can be used for military applications, weather monitoring, or to monitor the movements of insects and small animals and any situation that would be too dangerous for humans. The sensors have the ability to transfer information for a period of several days; in the future the sensors could have the capability of sending pictures and data. Figure 1.11.1 shows an illustration of a Smart Dust module.

Figure 1.11.1 Smart Dust module

[pic]

This project is not the first to attempt to build a mechanism to deploy Smart Dust Modules. The University of California at Berkeley and the Rochester Institute of Technology have both conducted research projects in sensor deployment. Both of these efforts focused on microrockets. A microrocket is a 10mm to 20mm long rocket powered by solid propellant. The microrocket has a small nozzle where it generates thrust as well as where it is ignited. The microrockets would be able to carry 5-10 Smart Dust Modules in the nose cone or on external storage bays.

When microrockets were manufactured and tested there arose problems with the stability of the system. The flight pattern was very erratic if they lifted off. Fins were added to the external sides of the microrocket to try and improve stability, but the flight path and accuracy of final landing position were never accurate enough for use as sensor deployment devices.

1.12 Mission Statement

The goal of the Launching Station for Sensor Deployment Team is to design and build a fully functional prototype capable of launching a 0.5-inch diameter sphere. The design and prototype will be a proof of concept for sensor deployment. This concept will be utilized by future research projects at the Rochester Institute of Technology as well as other universities.

2 Theory

2.1 Sensor Aerodynamics

The first section of theory will focus on the aerodynamics of the sensor during flight. A key factor in the deployment of the sensors is their aerodynamic qualities. In order to accurately deploy the sensor to a location up to 30 feet from their initial launch point, any aerodynamic effect on the sensors must be predetermined and thus compensated for prior to its launch. Since the launch platform is designed to deploy various types of sensors, which may or may not carry the same aerodynamic characteristics, the idea was to encapsulate the sensors in one-half inch diameter spheres. This reduces the number of aerodynamic factors that need to be compensated for, as well increases the universality of the launch platform.

The spherical shape was chosen due to its advantages versus other projectile shapes, most notably in the area of drag. Drag in general is the resistance force placed upon the projectile by the medium for which it is passing through, in our case air. Ideally any fired projectile will travel in a parabolic path; however with the presence of drag the projectile will follow a more truncated path thus decreasing its overall range. The drag force is typically expressed by Equation 2.1.1 from Shevell, 1989:

[pic] 2.1.1

The drag coefficient represented by “CD” is a dimensionless parameter, which represents the complex relationships between drag, the projectiles shape and its various flow characteristics. The object shape has the most significant effect in regards to the force of drag exerted upon it. For a sphere the drag coefficient typically varies between 0.07 and 0.5. This large variance is as a result of the sphere’s significant dependency on Reynolds number. Reynolds number is dimensionless number that represents the ratio of inertia forces to viscous forces. For sphere the type of flow can vary significantly based on its velocity. With this in mind, in order to calculate the drag on our sensor we have to determine the Reynolds number for our half-inch sphere by means of the means of Equation 2.1.2 from Shevell, 1989:

[pic] 2.1.2

Once the Reynolds number is determined we can determine the sphere’s drag coefficient by looking up the value on a previously developed graph representing their relationship. This graph can be seen in Appendix B.

Another advantage for a sphere is that its projected area remains constant in flight, unlike an oblique shaped object, which can have the tendency to tumble in flight thereby causing changes in drag. A sphere however can spin in flight causing it the curve in a given direction depending on the type of spin imparted upon it. This comes as a result of the changes in the fluid flow very close to the surface of the sphere creating a force in a given direction. This effect is most commonly known as the “Magnus Effect”. For the purpose of deploying our sensor we can neglect this effect, since the short range and accuracy for which we wish to obtain limits its influence.

Knowing the drag effects on the sensor allows us to calculate the range the sensor will travel based on its initial velocity and launch angle. However the process is complicated since the drag is proportional to the square of the objects velocity. Since the velocity in both the horizontal (x) and vertical (y) directions change over time, as a result of a deceleration from drag and both an acceleration and deceleration from gravity. Given Newton’s second law we can determine the accelerations or decelerations affecting the sensor in flight. Newton’s second law is as follows from Serway, 1996:

[pic] 2.1.3

With gravity and drag acting in the vertical direction we can resolve Newton’s second law for the vertical instantaneous acceleration to the following:

[pic] 2.1.4

As for the instantaneous acceleration in horizontal direction in relation to the drag force the second law can be resolved as:

[pic] 2.1.5

Using the standard physics equations for two-dimension motion and the above modified acceleration equations. Since drag varies with velocity over time our net result would be the following differential equations for acceleration in the horizontal and vertical directions.

[pic] 2.1.6

[pic] 2.1.7

It should be noted that “c” represents the collection of constants from equations 2.1.1, 2.1.4 and 2.1.5. For general reference this constant is defined as:

[pic] 2.1.8

In order to solve the aforementioned differential equations Euler’s method was used to approximate the solution in Microsoft Excel. With the solution known the range or x-displacement can be easily found at an instantaneous point in time with the application of the following equation with Euler’s method:

[pic] 2.1.9

An example and description of the Excel spreadsheet used to determine the range of sensor based on initial launch conditions is found in Appendix C. This will allow for calculation of how far the sensor will travel for various sensor materials and other initial conditions possible.

2.2 Wind Effects

When a projectile is launched outdoors, any wind will alter the trajectory and lead to inaccuracies in placement of the projectile. The wind deflection on a projectile must be understood so it can be accurately predicted, and eventually compensated. Wind deflection or sometimes more loosely known as wind drift is the change in a projectiles path as a result of wind. It should be noted that wind deflection is proportional to delay in the time of flight as a result of air resistant, rather than the total time in flight. Another way to see this relationship is that a projectile fired in a vacuum will travel a set distance faster than if it was fired in the presence of air. Since the wind deflection is related to air resistance or drag, the deflection of the sensors can be determined.

However there are some physical and technical difficulties in determining the wind deflection of the sensors. First the sensor launching platform requires some time before it can determine the speed and direction of the wind. Since wind direction and speed are not always constant with time, the actual wind speed upon sensor deployment may differ slightly from the measure value. Also wind gusts can vary during the flight of the sensor. Since these variations in wind gusts can not be easily determined we are forced to approximate the wind deflection under the assumption that wind speed and direction remain constant during sensor deployment.

Knowing the direction of the wind gusts, the components of wind velocity can be determined. For the cases of a headwind or a tailwind component of wind the velocity can be subtracted or added respectively to the forward velocity of sensor, since the drag has already been found in the determination of air resistance effects on sensor range. For case of the acceleration in the lateral (z-direction) resulting from the crosswind and with the inclusion of drag over time can be seen in Equation 2.2.1.

[pic] 2.2.1

The above equation was then included into the existing range spreadsheet so the approximate displacement of the sensor, as a result, could be found at a given time during flight. An example and description of the Excel spreadsheet used to determine the range and wind deflection of the sensor based on initial launch conditions can be found in Appendix C.

For reference it should be noted that there is also an alternative method that can be used to approximation the wind deflection of a projectile using the “Delay Lag Theory” or “Didion Approximation”. The deflection is given by Equation 2.2.2.

[pic] 2.2.2

In short the “Didion Approximation” relates the deflection to the difference in total time between a projectile traveling through air as opposed to a vacuum. However this method is only valid in cases where the initial velocity of the projectile is much greater than the wind speed. Knowing the displacement effects on the sensor based on the wind will allow for the system to be programmed to compensate for initial wind conditions.

2.3 Mass Effects on Sensor Range

In order to pick the best material for the sensor encapsulation material, the ranges of various materials in consideration must be considered. In calculating the drag force on the sensors, there total mass becomes a factor in determining the final location of the sensor. Although the effect of the sensor’s mass has already been compensated for it should be made clear as to what affect the mass plays in the determination of the trajectory. This difference in range can be seen in Figures 2.3.1 and 2.3.2, which show projectiles of different densities launched at a velocity of 32ft/s and at 45 degrees.

Figure 2.3.1 Projectiles: Velocity 32ft/s, Angle 45 degrees

[pic]

Figure 2.3.2 Projectiles: Velocity 32ft/s, Angle 45 degrees

[pic]

If the projectile is shot at a lower angle but at the same velocity the difference in range between the three projectiles (polycarbonate, white delrin and steel) becomes less apparent. We can see this by comparing Figures 2.3.1 and 2.3.2 to 2.3.3 and 2.3.4, which show the same three projectiles launched at the same velocities but this time at an angle of 25 degrees.

Figure 2.3.3 Projectiles: Velocity 32ft/s, Angle 25 degrees

[pic]

Figure 2.3.4 Projectiles: Velocity 32ft/s, Angle 25 degrees

[pic]

In general a projectile of higher mass will fly farther through air than and equal sized object with a lower mass when launched at the same speed and angle. The higher density of the object allows it to have a higher momentum and since the cross-sectional area of the object has not increased, the opposing drag force remains the same. Although it should be noted that in order to launch a projectile of higher mass requires more energy than that of a projectile of lower mass in order to reach the same velocity.

2.4 Stress in Pressurized Vessels

With the use of a pneumatic system, there is a large pressure that is created in the volume between the bolt and the projectile being launched. This will induce stresses in the barrel that must be calculated to ensure that there is no structural damage during the launch. Both the radial and tangential stresses can be calculated using conventional strength of materials theory.

Equation 2.4.1, obtained from Shigley and Mischke 2001, gives the radial stress for a pressure vessel. Equations 2.4.2 through 2.4.5 are modifications of Equation 2.4.1, based on the conditions stated below.

[pic] 2.4.1

There is no external pressure, [pic], so Equation 2.4.1 reduces to Equation 2.4.2.

[pic] 2.4.2

The stress will be maximum at [pic] = [pic], because that is the only location where pressure is applied. This leads to Equation 2.4.3.

[pic] 2.4.3

Rearranging Equation 2.4.3 gives Equation 2.4.4.

[pic] 2.4.4

Canceling the difference in radius terms gives Equation 2.4.5. The radial stress is just the negative of the applied pressure.

[pic] 2.4.5

Equation 2.4.6 gives the tangential stress for a pressure vessel, obtained from Shigley and Mischke 2001. Equations 2.4.7 and 2.4.8 are modifications of Equation 2.4.6, based on the conditions stated below.

[pic] 2.4.6

There is no external pressure, [pic], so Equation 2.4.6 reduces to Equation 2.4.7.

[pic] 2.4.7

Again, the stress will be maximum at [pic] = [pic], because that is the only location where pressure is applied, giving Equation 2.4.8.

[pic] 2.4.8

Equations 2.4.5 and 2.4.8 will be used in Section 4.1 to calculate the stress that will be found in the barrel during launch.

3 Design and Analysis

3.1 Brainstorming Session

A brainstorming session was held Friday December 20, 2002 with all team members present. The goal of the brainstorming session was to take 20 minutes and come up with as many ideas to meet the design objectives. No ideas were disregarded even if they seemed to be unfeasible, complex, or exotic. For our design there are four distinct design areas that were brainstormed individually. These four are: method of propulsion, type of loading and reloading, method of changing angle of launch trajectory, and method of rotating the launch platform. Figure 3.1.1 shows the initial brainstorming results.

Figure 3.1.1 Initial Brainstorming Results

[pic]

This list of ideas was further discussed and many of these ideas seemed to have a lot of promise. We then went through the list and combined any ideas that were identical or similar. From this condensed list, each team member was given votes and voted for concepts at will. The number of votes was determined by taking 20% of the number of concepts from each design area. Figure 3.1.2 shows the consolidated list after the voting was completed.

Figure 3.1.2 Brainstorming Results After Voting

[pic]

With four votes a piece, the pneumatic and spring launch systems were the best-supported methods of propulsion. Many of the other designs were much more complex and the feasibility is uncertain. The gravity feed and the spring-loaded magazine gathered all but one vote for type of loading. Other types of loading would require much more elaborate magazines. In order to change the distance the sensor will be deployed, the propulsion power can be varied or and actuator can move the barrel. The most robust solution would be to have both methods available. Two ways to rotate the launch platform would be to either turn the entire robot or have a turret that could move relative to the robot. Again having both options at this point in the design is beneficial. Figure 3.1.3 shows the brainstorming results with the majority of support. These results will be used to perform a feasibility assessment and to create initial sketches of our project.

Figure 3.1.3 Brainstorming Results with Majority of Support

[pic]

Initial sketches were made by dividing each of the four propulsion methods between each team member, then dividing loading types evenly, and by assuming there is a gear driven turret in all designs. Each team member was given 5 minutes to sketch their concept, and then the sketches were exchanged so everybody was able to contribute to every concept. Figure 3.1.4 shows the distribution of concepts for the sketches. The cleaned up copies of our concepts, along with preliminary bill of materials, can be seen in Appendix E.

Figure 3.1.4 Distribution of Concepts for Initial Sketches

[pic]

3.2 Compressed Air Concept

A pneumatic propulsion system would be a very simple and repeatable launcher. In order for it to work, it would need a barrel, something to hold sensors waiting to be deployed, a reloading mechanism, an air tank, and a valve.

A barrel, which would guide the sensor initially similar to a gun, could be made out of metal and look similar to a tube.

Many things could be used to hold the projectiles waiting to be deployed. One possibility could be to have a tube mounted at a 90-degree angle to the barrel so that gravity could feed the sensors into the barrel to be fired. In order to do this, a hole must be drilled into the side of the barrel for the sensors to fall through.

Since the sensors will be launched via compressed air it would be wise to take advantage of the preexisting air supply and make the reloading system pneumatic. To do this a metal cylinder would slide inside the barrel. The cylinder has a hole drilled through it that allows air to be fired through it and acts as a bolt would in a pneumatic gun. When it slides back it allows a sensor to fall into the barrel from the magazine and into the chamber. It would then need to be pushed forward into the chamber of the barrel and past the hole that the sensors fall though. This movement could be accomplished by using a pneumatic cylinder attached to the bolt.

To control the cylinder and to fire the air through the bolt to launch the sensor a valve or a series of valves that are controlled electronically could be utilized. Additionally, to power the system an air storage tank is needed.

3.3 Spring Concept

The spring loaded and spring launched concept was one of the ideas considered for the deployment of the sensors. It involves using a spring-loaded bolt inside a barrel to store the energy necessary to propel the sensor to its destination. The supply of sensors would be contained in a rectangular shaped, spring-loaded magazine.

Operation would involve using a small motor to wind up and hold a cable attached to the bolt, thus compressing the spring. The retraction of the bolt would also allow for a sensor to be loaded into the barrel. When the bolt is retracted one or two springs inside the magazine would force a new sensor into firing position. Upon firing the motor would release its hold on the cable and allow the previously compressed spring to propel the bolt forward thereby launching the sensor out of the barrel. Angle of the barrel was to be controlled by a motor or some form of actuator. The rotation of the barrel was to be controlled using a gear driven turret design.

Initial problems that were to be expected were centered mainly on the use of springs. Fatigue on the spring used for propelling the bolt forward could result in variation in the initial launch force. Also the spring loaded magazine would be always be applying a force on sensors, which could possibly lead to a jam in the barrel or cause enough resistance to prevent the proper amount of force to be imparted on the sensor awaiting deployment in the barrel. The ability of the motor to accurately wind the cable, hold it, and then freely release it was also called into question.

3.4 Wheel Drive Concept

The third brainstorming idea would consist of a gravity fed hopper and a direct drive wheel mechanism. Multiple sensors could be loaded into the hopper that would roll to the wheel drive mechanism. The wheel drive mechanism would consist of 2 wheels directly connected to motors. The trajectory could be altered by varying the angle of the motors. The sensors would have to be spherical in shape or have a spherical encapsulation for this method to work. The team thought that the motors would require too much power and the sensor trajectory would be inconsistent.

3.5 Rubber Band Concept

The rubber band concept would require a large rubber band that could be attached in one of two ways. It could be mounted to two vertical posts and the middle would be pulled back to provide the launching power, much like a slingshot. The sensor would be placed in a barrel that would guide the initial angle of launch. Alternatively the rubber band could be fixed to a bolt and barrel assembly. The rubber band would be fixed to both the barrel and the bolt. The bolt would be able to slide freely inside the barrel. When the bolt is pulled back, the tension increases in the rubber band. The sensor would be located on the front side of the barrel. When the bolt is let go, the rubber band pulls the bolt forward in the barrel until it reaches a stop. The sensor continues out through the barrel and into the air.

An actuator would be required to pull the rubber band back to the firing position. This could be a linear motor, a pneumatic cylinder, or anything capable of providing linear motion.

A hopper or magazine would be necessary to automatically reload sensors. A hole drilled though the top of the barrel would allow for the hopper to drop sensors into. The hopper could be anything from a straight tube to a large container that funnels the sensors into the barrel.

To vary the angle of launch trajectory, the barrel must be able to rotate. This can be accomplished with a reduction gear set and a motor.

3.6 Feasibility Assessment

A feasibility assessment was conducted Friday January 10, 2003 with all team members. The goal of the feasibility assessment was to have a systematic approach to narrow down our current four concepts to the design that will be the focus from here on out. Our four concepts for launching sensors at this point are: springs, rubber bands, wheel friction, and pneumatics.

Questions were asked in technical, economic, market, schedule, and performance issues for all concepts. Regarding technical feasibility, this project would be classified as Alpha Feasible. There are no new technologies being developed. The focus is integrating systems that have not previously been used together. There are no parts that cannot be either purchased or easily machined in house using simple machining techniques for any of the concepts. The budget of $2,000 for the project must be kept in mind when considering the economic feasibility. All materials of parts to be made in-house must be purchased, as well as of the shelf parts and hardware. Additional electrical engineering support may be required for circuit design/layout, systems integration, as well as programming. When considering the market feasibility, the future of sensor deployment must be kept in mind. Although this project is solely a technology demonstration and not to be commercially available, future edition launching stations will be marketed to military and civilian applications. The main concern regarding the schedule feasibility is everything has to be complete in two quarters. The design must be completed by the end of February, and a fully functional working prototype must be constructed by the end of May. Performance feasibility needs to focus on the safety issues involved with launching projectiles. All of these constraints will influence which concept is pursued further.

The answers to these questions were quantified by assigning a point value from 0 to 3 for each question relative to a baseline concept. The baseline concept needs to be similar sensor deployment device that has already been developed. The microrocket that was developed by a senior design team last year fits this criterion. The pros and cons of the microrocket were previously described in the Introduction (Chapter 1). A base score of 2 is given as an answer for each question applied to the microrocket concept. An answer of 0 would be given if the feasibility had no realistic viability. An answer of 1 is given if the concept is worse than the concept, 2 if the same as the baseline concept, and 3 if better than the baseline concept. There are a total of 12 questions for each of the four concepts, and they were divided among all team members for ranking. Figure 3.6.1 shows the results plotted as a radar plot.

Figure 3.6.1 Radar Plot of Feasibility Assessment

[pic]

Technical Question 1 stated, “Does the team have the skills needed to implement all aspects of the technologies for this concept”, and it received consistently low marks because mechanical engineering is the only discipline represented on the current team. Integrating the launching station onto a robot will require electrical engineering skills. Programming the robot, GPS and other electrical components will also require extensive electrical engineering experience. Finding an electrical engineering co-op or a graduate student may be necessary to increase the technical ability.

The spring-loaded launch and the rubber band launch concepts were most similar in design and had relatively similar average scores of 2.0 and 2.08 respectively. The main concern with both of these concepts was the accuracy of the launch. Springs and rubber bands have the tendency to degrade in effectiveness over time. The accuracy of how far the spring or rubber band is pulled back, crucial for the distance the sensor will travel, would be difficult to keep consistent.

The wheel drive concept had an average score of 2.17 in the feasibility assessment, and thus was slightly better then the baseline micorocket concept. The advantage of two drive motors would be the ability to easily vary the velocity of the sensors launched. The main drawback would be designing a system to vary the angle of launch by tilting botth motors to the desired angle.

The pneumatic launch concept ranked consistently higher than the other concepts, including the baseline microrocket concept with an average score of 2.5. It was strong in both the economic and performance feasibility questions. Most of the purchased components needed are readily available. Using a compressed air system will allow for multiple launches and long distance deployment. The pneumatic system will be the concept concentrated on, but no concept seemed unattainable and all will be kept for reference if needed.

3.7 Preliminary Design

The pneumatic system for launching the sensor that was first developed in the brainstorming session, and found to be the most viable concept in the feasibility assessment, will be used for the initial design. To vary the distance that the sensor will travel, the launch angle will have to be able to vary. The power given to launch the sensor will remain constant to reduce the complexity of the pneumatic system. The robot base that the team will be using to mount the launching station has a separate motor to drive both sides. These motors are DC motors with actuators inside them. This gives the robot the ability to rotate very accurately in place, eliminating the need for a rotating turret on the launching station. A G.P.S. sensor will be integrated into the launching station, but with the lack of electrical engineering support on the team, an external source will be sought to solve this problem. The final item that needs to be designed is the wind sensor. The design that was suggested, and the one the will be pursued, will be a wedge probe. In order to eliminate interference with other launching station parts, the wind sensor will have to be placed at the highest point of the final assembly. The following chapters describe the engineering principles and the thought process that turned the concept into a final design.

4 Results and Discussion

4.1 Barrel Wall Thickness

Equations 2.4.5 and 2.4.8 from Section 2.4 will be used to calculate the radial and tangential stresses in the barrel. If the stresses are found to be over the yield stress for the aluminum, two possible solutions exist to correct the problem. The material can be changed to one with a higher yield strength, or the wall thickness of the barrel must be increased.

The radial stress given from Section 2.4 is given again in Equation 4.1.5.

[pic] 4.1.5

The maximum pressure that is possible from our pressure regulator is 800p.s.i, so the radial stress on the inside of the cylinder is [pic]

Equation 4.1.8 gives the expression for the tangential stress as previously found in Equation 2.4.8.

[pic] 4.1.8

800p.s.i. will be used again as the maximum internal pressure. The inner radius, [pic] = 0.252in, and the outer radius, [pic] = 0.334in, are from drawing number P801A, which can be seen in Appendix E. The weakest point of the barrel will be at the thread, therefore the outer radius, [pic], was taken as the minor diameter of the 3/4 – 16 thread. Equation 4.1.9 shows Equation 4.1.8 with the numerical values inserted.

[pic] 4.1.9

The resulting tangential stress is [pic]. The material the team will be using for the barrel, 6061-T6 aluminum, has a yield stress of [pic] obtained from Beer and Johnston 1992, which leads to a factor of safety of 12.

2. Finite Element Analysis of Barrel

A finite element analysis was conducted to verify the results in Section 4.1. The barrel was modeled in ANSYS with an internal diameter of 0.504” and external diameter of 0.669”. The external diameter was the minor diameter of the 3/4 – 16 thread, same as used in Section 4.1. A 3-D linear 8-noded brick element with 0.05-inch mesh was applied to the model. Figure 4.2.1 shows the barrel with the applied mesh.

Figure 4.2.1 Barrel With 0.05-inch Mesh

[pic]

In order to effectively represent the model, boundary conditions must be accurately applied. One face of the barrel was restrained from translation in the axial direction. Additionally, one node was fixed from translation in the redial direction to prevent ANSYS from giving an error. The other end of the barrel was left completely free in translation and rotation. Figure 4.2.2 shows the model with the applied boundary conditions.

Figure 4.2.2 Barrel With Boundary Conditions

[pic]

The loads were applied as a pressure on the inner area of the barrel. The magnitude was set at 800p.s.i., the same as in Section 4.1. Figure 4.2.3 shows the barrel with the internal pressure applied.

Figure 4.2.3 Barrel With Applied Internal Pressure

[pic]

The maximum tangential stress was found to be 3,056p.s.i., and the stress contour plot can be seen in Figure 4.2.4. This compares to 2,903p.s.i. using convention strength of materials equations.

Figure 4.2.4 Tangential Stress in the Barrel

[pic]

This is less than 5% error, and thus there is correlation and the results can be assumed to be correct. With a factor of safety of 12 and correlation between the analytical and Finite Elements solution, the barrel dimensions as they are in Drawing P801A, do not need to be modified.

3. Wind Sensor Resolution

In order to be sure that the pressure sensor that was chosen would work well in the system, the team wanted to be able to determine the resolution that the pressure sensors were capable of measuring. The pressure sensors have 1000 distinct increments in their range. Therefore, the smallest increment in the pressure sensors are able to measure is 1/1000 of the total range as seen in Equation 4.3.1.

[pic] 4.3.1

The smallest increment that the sensors can accurately measure is determined by dividing the range by the number of increments that the sensor is capable of measuring. This can be seen in Equation 4.3.2.

[pic] 4.3.2

The pressure will be equated into a wind velocity using the following equation. This equation states that the total pressure is equal to the free stream pressure plus the dynamic pressure due to the wind velocity. This is shown in Equation 4.3.3.

[pic] 4.3.3

Solving Equation 4.3.3 for the velocity gives Equation 4.3.4.

[pic] 4.3.4

Then using the smallest pressure change that the sensor can measure and the density of air, the velocity can be calculated as seen in Equation 4.3.5.

[pic] 4.3.5

The resolution of the wedge probe will be 2.4m.p.h. This will be the smallest increment that can be measured.

5 Final Design

5.1 Pneumatics

The method of propulsion that was chosen to be most feasible was the pneumatic system. This section describes the thought process that went into choosing the components of the pneumatic system. There are many parts that make up an effective and efficient pneumatic system which vary from system to system. In our project we need an air supply, pressure regulator, actuator, valves, and tubing. An actuator is needed to move the bolt back and fourth to reload the projectile. A valve system is needed to control the actuator and fire the projectile. An air supply is needed to deliver air to the entire system. Tubing, pipes, hoses or a combination of them all are needed to plumb all of these components together. Lastly, a pressure regulator is needed, since the system has no way of regenerating a compressed gas. A high system pressure can be used with a pressure regulator to bring down the system pressure to operating pressure and deliver a larger quantity of shots and more accuracy, since the pressure remains constant for a longer period of time.

Although an actuator is normally thought of as a simple device that creates motion, it presented a lot of debate in our system. Initially, we thought a double acting actuator would be best since air pressure is needed to move it in both directions. This was a very feasible idea since it would be able to be mounted either in front of the bolt or behind the bolt leaving further design options open. However, using a double acting cylinder and looking at the pneumatics system as a whole it mean that more complex valves would be needed. Using a double acting cylinder meant that three air lines needed to be controlled; cylinder extend, cylinder retract and firing the projectile. To control this either three separate two position valves with three ports were needed, or one three position valve with 5 ports and a single two position three ports would be necessary to control the entire system. This forced us to see if it was feasible to use a single acting cylinder. If it were feasible, it would mean there would only be two air lines to be controlled.

Three different single acting actuator systems were looked at. The first was mounting the cylinder so that the rod end faces the back of the launching station and the cylinder is mounted to the barrel. The cylinder would be powered by air when it was retracting and then an external spring would push the bolt open to allow another projectile into the barrel to fire again. The problem with this design is that it required considerably more machining in the bolt to hold the spring.

The second design used the same mounting configuration but air would be used to extend the cylinder to open the bolt and then an internal spring would close the bolt once a projectile had fallen into the chamber. The problem with this is that the spring in the cylinder is not very powerful. Gravity is working against the bolt closing and in addition to moving the bolt forward it has to move the projectile in front of the bolt forward into the barrel as well. This could be too much force for the spring to overcome repeatedly.

The third design had the same mounting cylinder mounting configuration but used a cylinder that had a spring built into it to extend the cylinder. To use this design the spring only had to be powerful enough to move the bolt to the open position. However this is not a large force and gravity helps open the bolt since the barrel will always be at an angle. Additionally, the cylinder had to be sized large enough that the air would have enough power to retract the bolt on the retract stroke and with the spring working against it. It was found that a 5/16-inch bore would be sufficient in doing so and this is the system we choose since it would be very cost effective and efficient.

The next issue considered was an air storage unit. Several small, inexpensive, aluminum tanks are available that would fit into our system but can only hold up to 150p.s.i. of compressed air. Having a system pressure of 150p.s.i. would work in this system but in order to get several shots out of the system before the pressure fell below operating pressure, a tank larger than we had anticipated would be necessary. If the system does fall below operating pressure, placement of the projectile would be affected.

The next storage tank that was discussed was a high pressure CO2 cartridge that is commercially available at any sporting good store. There is approximately 800p.s.i. in the small cartridges and many shots can easily be fired out of one cartridge. The disadvantage to the system is that a few additional parts are needed and the parts are more expensive. However, since it proved to be more functional, this is what we choose to use.

In any pneumatic system the valves are one of the most critical components to look at. An undersized valve can cause havoc in a system since it will restrict too much flow hence slowing the system down. Over sizing a valve is also not good. It is usually a waste of money and space depending on the size, configuration and manufacturer.

With so many different configurations to choose from, such as the amount of ports, size of the ports, solenoid or manually operated, inline or manifold mounted and coil voltage, just to name a few, selecting the proper valve for the system was the most difficult part to select.

For our system we have to be able to control two things pneumatically, simultaneously and through electric signals that will not require a significant amount of power. The first is the bolt motion to load the projectile into the barrel to be fired. The second is to fire a burst of air through the bolt to launch the projectile out of the barrel. Flow, size, power draw and the amount of valves are all key issues here.

Since we knew that when any of our air lines were not being pressurized we did not want them to hold air pressure that determined that we needed valves that exhausted any remaining air pressure after the valve was closed. So if we decided to use a two-position valve in the system it would have to have three ways so the air in the lines could exhaust through the third way.

Since it was already decided that a single acting cylinder would be used to power the bolt, when we looked at possible valves, it simplified the valve system dramatically. This narrowed down the valve system possibilities to two basic configurations. The first one was to have one single valve control the whole system. In order to do so, a three position 5 way center exhaust valve is necessary. Many issues came up with this valve though. Most of them had to do with cost, functionality, size, and power draw. A valve sized with enough flow to fire the projectile would cost over $150. The advantage to this was that is was one contained unit but it was also large and gave too much flow to the cylinder so a flow control would be needed as well. Another problem is that both lines cannot be pressurized at the same time. So in order to fire a projectile, the valve would have to pressurize the cylinder to retract the bolt and projectile into the barrel, then quickly fire the projectile before the bolt moved back too far exposing the breech and loosing air pressure in the barrel. The last problem with this valve is it required a significant amount more power than using separate valves controlling each airline. Using this valve would create power draw problems since we need to minimize overall system power consumption.

The other valve system alternative would be to go with two separate two position valves, one to control the bolt motion and the other to fire the projectile. This proved to be a more feasible option since each of the valves could be sized according to the application. Additionally, each valve could be controlled independently so that the bolt can be held shut until after the projectile is fired. This choice was made clear once we found the price of this configuration was less than half of the three-position valve system. Figure 5.1.1 shows a flow chart of the final design of the pneumatic system, starting with the CO2 cartridge and leading to the cylinder and bolt.

Figure 5.1.1 Pneumatics Flow Chart

[pic]

5.2 Wind Sensor

In order to compensate for a headwind, tailwind, or crosswind, the team decided that the robot should be able to determine the maximum wind velocity and direction. The robot will use this information to adjust the trajectory and direction of launch. The ultimate goal is for the sensor to hit the target, and this will require compensation for wind velocity and direction. The team will construct a wedge probe consisting of 3 pressure sensors to determine wind velocity and direction.

To monitor wind velocity and direction, four pressure sensors will be utilized. One pressure sensor will be mounted at a location on the robot to determine the atmospheric pressure. The other 3 sensors will be mounted on a wedge probe that has a leading edge angle of 90 degrees. Three 1/8” holes will be drilled through the spoiler with 1/8” tubing leading to the pressure sensors. One of the three holes will be drilled at the leading edge of the wedge probe. The other two holes will be drilled on the sides of the probe that will also have 1/8” tubing leading to the pressure sensors. The spoiler will be mounted at a location approximately 1” above the barrel. The mounting location will eliminate any turbulence caused by the barrel. Figure 5.2.1 shows the wind sensor mounted to the post that locates it above the barrel.

Figure 5.2.1 Wind Sensor Assembly

[pic]

If the probe were facing directly into the wind, the sensor at the vertex would measure the total pressure. The sensors on the sides of the spoiler measure a combination of the static and total pressure of the wind. The differential pressure between the sensors will be used to determine the wind velocity. As the robot turns, the three pressures at the 3 ports will vary, allowing the wind velocity and direction to be determined. If the wind were directly facing one of sides of the probe, the highest pressure would be measured by the sensor that is directly facing the wind. The sensor at the vertex of the probe would then measure the static and total pressure while the pressure sensor on the opposite side would measure the lowest pressure. The team will calibrate the wind sensor using the wind tunnel. The wind sensor will be placed in the tunnel at various angles of attack with wind speeds ranging from 10-30 m.p.h. Calibration charts can then be made to determine wind magnitude and direction, even if the leading edge of the wind sensor is not facing into the wind direction.

Once the robot reaches the proper launching spot, the robot will be required to measure the wind velocity and direction. This will be accomplished by rotating the entire robot 45 degrees and then stopping to allow the pressure sensors to reach equilibrium. The robot will turn an entire 360 degrees, then the maximum wind velocity and direction can be determined.

The pressure sensors that will be used are manufactured by Omega that are capable of measuring up to 3 inches of water. Figure 5.2.2 shows a picture of Omega sensors.

Figure 5.2.2 Omega Pressure Sensors

[pic]

The sensor that was selected has a 1/8” fitting that is 1” long, this work with the 1/8” ID tubing that the team selected. Also, the pressure sensors manufactured by Omega are very accurate. Combining the data learned in Section 2.2, and the wind sensor assembly, after calibrated will allow for the system to be programmed to compensate for the initial wind conditions at the wind sensor.

5.3 Bolt, Barrel, and Receiver

One of the first things that had to be designed once it was decided that a pneumatic system would power the launching station, was something to hold, reload, and physically launch the projectile. Similarly to the rest of the project, this design had to be small and lightweight so it could be mounted and function on the robot.

The team’s original brainstorming idea was to have a cylindrical piece of metal acting similarly to a bolt in a gun, inside a metal tube (barrel) so it could slide in and out via an actuator. The barrel would have a hole placed in it to allow a projectile to be reloaded when the bolt slid back.

The barrel was sized once a sensor-encapsulating object was found. Once this was found the barrel was sized so that the minimum tolerance between the ball and barrel was 0.003” and the maximum would be 0.007”. This allows the projectile to easily come out of the barrel but there would be a minimal amount of blow-by. Blow-by is wasted air that escapes around the projectile and does no work to propel the sensor. Figure 5.3.1 shows an isometric view of the barrel.

Figure 5.3.1 Isometric View of Barrel

[pic]

Once the barrel was sized, bolt development could begin. As stated, the bolt needs to slide freely inside the barrel. Its purposes are to reload projectiles and to fire a burst of air to propel the projectiles out of the barrel. However, similarly to the projectile, the bolt needs to be sized properly as well. If it is too large it will not slide properly or at all inside of the barrel, or if it is too small it will allow air to escape in between the bolt and barrel upon firing the projectile. In order to function properly, the same tolerances between the projectile and barrel were used between the bolt and barrel. So the bolt has the same size and tolerance as the encapsulating material. Additionally, a port to deliver the air and a way to attach the air to the bolt was needed. The simplest was to connect a hose to the bolt was by using a barbed fitting that gets threaded into the bolt. As for the port, since free flow air is needed to fire the projectile it was decided to make the port the dame size as the drill needed to tap the hole for the fitting in the back of the bolt that feeds the air. This diameter is 0.159.” Figure 3.5.2 shows the isometric view of the bolt.

Figure 5.3.2 Isometric View of Bolt

[pic]

As this idea was further developed some important design problems were encountered. One key one was how to fasten the actuator to the bolt. The first thought on this was to have a plate mounted to the back of the bolt that extends down past the barrel so the cylinder could thread into it. At first this sounded good but nothing was preventing the bolt from rotating inside of the barrel. If this happened it would make the cylinder bind and possible not function.

This problem was overcome by removing the plate from the back of the bolt. Instead of this, a slot is going to be machined into the bottom of the barrel and a shoulder bolt will get threaded into the side of the bolt. This prevents the bolt from spinning and it also allows the actuator to be attached to the bolt by a simple adapter.

The adapter will be a small rectangular piece of metal with a hole tapped into one side to attach to the cylinder. A through hole is also drilled into the adapter oriented 90 degrees from the tapped hole for the shoulder screw to go through. With the cylinder mounted to the barrel, this makes the system complete. Figure 5.3.3 shows a picture of this connector.

Figure 5.3.3 Isometric View of Connector

[pic]

The next innovation we thought of was to have the barrel a separate threaded part that screws into the receiver. This was done to allow for the use of different length barrels for future use. Using different barrel lengths will aid in accuracy for longer distance placement of the projectile if it is needed. Additionally, if other barrels are needed, the receiver, which is the most complex part in the system, does not need to be recreated. Figure 5.3.4 shows two isometric views of the receiver.

Figure 5.3.4 Isometric Views of Receiver

[pic]

With all of this design done, a reloading system was needed to satisfy the design criteria. The original plan was to have a tube fastened at a 90-degree angle from the barrel so it was sticking straight out. This was a very simple way of holding the projectiles waiting to be fired. Even though this would be a functional way of storing the sensors, it would be very cumbersome and added additional height to the overall system. This also brought up the issue of the robot hitting low objects as it passed under them. If this were to happen it could damage the robot and/or flip it over. This caused us to abandon this idea and come up with another.

The second development was to have the magazine attached parallel to the barrel and a hole drilled into the side of the magazine to let the balls fall into the chamber and fired as the bolt cycles. In order to mount the magazine to the receiver, ears will be machined into the bottom of the magazine with through holes in them to that holes can be tapped into the receiver and the two can be bolted together. In order to prevent the projectiles from rolling out of the back of the magazine, a threaded hole will be tapped at the proper location and act as a stop so that the sensors can fall into the chamber to be fired. Figure 5.3.5 shows an isometric view of the magazine.

Figure 5.3.5 Isomeric View of Magazine

[pic]

When it is time to reload a sensor, the actuator will extend. This opens the bolt and allows a sensor to fall into the receiver as shown in Figure 5.3.6. This process is done with the barrel tilted so that sensors cannot roll out of the barrel.

Figure 5.3.6 Receiver Assembly in Position 1: Reloading

[pic]

Once a sensor has fallen into the receiver the projectile is ready to be deployed once the bolt is closed by retracting the actuator as shown in Figure 5.3.7. Once the bolt is closed, firing air through the bolt launches the sensor.

Figure 5.3.7 Receiver Assembly in Position 2: Ready to Fire

[pic]

5.4 Sensor

The sensor used in this project is a placeholder for Smart Dust. The sensor that the team will be using is an animal tag transponder that has an over size comparable to Smart Dust, as it exists today. This is an implantable device that has a unique 10-digit code that is programmed into it during the manufacturing process. This leads to over 550 billion unique numbers that are available for use in the transponders. These tags are currently most widely used for pet and agricultural animal identification. Figure 5.4.1 shows a picture of the transponder that will be used as the sensor.

Figure 5.4.1 Transponder used for Sensor

[pic]

The team chose this transponder because it is a very close representation for where the Smart Dust Technology is today. Today Smart Dust technology is a 8-10 mm cube, and the transponder to be utilized in this project is 2.1 mm x 11.5 mm. Also, since each transponder has a unique identification, the robot can be programmed so it can recognize what sensors have been deployed and where they are located.

5.5 Sensor Encapsulation

The sensor encapsulation will protect the sensor from the high forces it would normally see during launch, flight, and impact. The customer has specified a diameter of 0.5-inch sphere that will allow the station to launch a variety of sensors. The team decided that the material should have the following properties:

• Should be easy to machine

• Should not absorb water

• Should have a high impact toughness

• Should have a high sphericity tolerance

• Should not conduct electricity

The selected material to encapsulate the sensor should be easy to machine, this will allow sensors of multiple shapes and sizes to be launched. The selected material should not readily absorb water. Water absorption is the percentage increase in weight due to water absorption. If the material absorbed water the dimensions of the material would change. The result would be a sensor jammed inside of the barrel. The material should also have high impact toughness. Impact toughness is the area under the stress-strain curve. A material with high impact toughness will not break on impact with the ground, protecting the sensor from damage. The material should also be relatively light, which will allow the station to obtain the desired ranges at low pressures.

The selected material should be available in 0.5-inch diameter spheres that have a high sphericity tolerance. The 0.5-inch diameter will allow for the sensor, described in Section 5.4, to be completely sheltered inside the sphere. When launching the sensor the tolerances will be critical to eliminate jamming or excess blow-by. The material for the sensors should not conduct electricity; a conductive material could possibly short the sensors. A shorted sensor would not operate properly and another sensor would have to be launched to obtain the missing data. Figure 5.5.1 shows the sensor with the plastic encapsulation around it.

Figure 5.5.1 Sensor Assembly

[pic]

The team decided that a plastic encapsulation would be the best choice because of the required design parameters. The two types of plastic chosen are White Derlin and Polycarbonate. The team will order many of each sensor for testing.

5.6 G.P.S.

A G.P.S. receiver is required in order for the system to know where it is and where it has to go. Additionally, G.P.S. will allow for the location of the deployed sensors to be recorded. G.P.S. stands for Global Positioning System, which is a network of satellites that orbit the Earth. Each of 24 satellites transmit a constant signal that contains an accurate time stamp along with other information that is picked up by GPS receivers. The orbit of these 24 satellites is very carefully controlled by the US Department of Defense. GPS receivers receive signals from 3 or more satellites to determine the location of the receiver on the Earth. Figure 5.6.1 shows a G.P.S. receiver and Figure 5.6.2 shows the antenna required to allow for the receiver to communicate with the satellites.

Figure 5.6.1 G.P.S Receiver

[pic]

Figure 5.6.2 G.P.S. Antenna

[pic]

GPS is used for many purposes. The military uses GPS guided “smart” bombs to hit targets, boaters use GPS to navigate to their port and hikers use GPS to find their way home. This technology is now widely available and has even been incorporated into some automobiles to provide mapping capabilities. GPS could be used for sensor deployment by allowing the launching station to determine its current location and where the sensor must be deployed.

Many GPS sensors that are available from are designated as WAAS enabled. WAAS stands for Wide Area Augmentation System, basically this allows for greater positional accuracy by combining GPS signals from satellites as well and GPS signals from ground stations. A WAAS capable receiver can give a positional accuracy of less than 3 meters 95 percent of the time. WAAS corrects for satellite orbit errors, timing and atmospheric disturbances, this system has not yet been approved for aviation and is currently only available in North America.

Selective Availability was used in the original GPS systems and only allowed for positional accuracies of 100m. This was due to government-imposed accuracy degradation to limit GPS capabilities.

OEM stands for Original Equipment Manufacturer; an OEM GPS sensor does not include an antenna or protective casing. It is recommended that these sensors have a protecting cover. This design results in a reduction in the overall weight of the sensor with a more compact design to minimize space requirements. The antenna can then be mounted to receive the best possible signal without any obstructions.

Due to size, weight and power requirements of the launching station any GPS sensor should be as small as possible, be as light as possible, low power requirements and high positional accuracy. The use of an OEM GPS sensor would minimize size and weight. Two GPS sensor that meet the requirements for this project are the: GARMIN GPS15L and the MOTOROLA M12+ ONCORE. The team selected the Garmin GPS15L because it was much more accurate than the Motorola M12+. The specifications of these two sensors can be seen in Appendix D.

5.7 Angle of Launch Actuation

Since the pressure in the pneumatic system is fixed, the angle of the barrel must be altered to vary the distance of the fired projectile. To do this, a way of tilting and holding the barrel stationary was developed.

In order to tilt the barrel, either a linear actuator or a motor driven gear assembly could have been used. A linear actuator proved to be too large to fit into the system, so a gear driven tilting mechanism was chosen that is powered by an electric motor.

Three types of motors were considered while designing the system. The motors that were looked at were stepper motors, servo motors and DC motors with encoders. The proper motor had to be able to angle the barrel to the correct position and hold it stationary while the sensor was launched. This includes any recoil from the projectile being launched. It also had to have enough torque to rotate to barrel and overcome friction in the gears. Additionally, the motor had to draw as little power as possible and be as small to fit into the design scope of the project.

The first motor that was looked at was the servo motor. This motor would have been a good choice since it has excellent precision and is capable of high torque. However, it proved to be too large. The next motor was the DC motor. This motor was the easiest to find and also the smallest available. The drawbacks to this type of motor were it would need an encoder to sense position. Additionally, these motors are the least accurate and have the fewest amount of stopping positions, or steps, per revolution. The best motor for the launcher is a stepper. A stepper motor has a high amount of steps per revolution, is a high torque motor, and has great holding torque.

In terms of gearing, two types of gears were looked at, spur gears, Figure 5.7.1, and worm gears, Figure 5.7.2.

| Figure 5.7.1 Spur Gears | Figure 5.7.2 Worm Gears |

|[pic] |[pic] |

Spur gears would have been an excellent choice because of their low cost, high efficiency, and high availability. However, to keep the amount of parts and size of the launcher down, it was preferable to use only two gears. Additionally, using more than two gears introduces more error in barrel position since there would be backlash between three gears instead of two. Using only two spur gears in this application required that the motor be able to hold the barrel perfectly still during launching since these gears are non-locking (either gear can drive the other). Therefore, if these gears were used, a much larger motor would have to be used just to assure it would be able to hold the barrel still.

To overcome this, worm gears were chosen since they are locking (the worm can drive the gear but the gear can not drive the worm). The drawback to worm gears is that they have much more friction than spur gears. However, the increase in friction can easily be overcome by the low gear ratios available. A ratio of 1:40 was chosen to offset the friction. Another advantage of the worm gears combined with a stepper motor is the precision. For every complete turn of the motor it will move the barrel 9 degrees. The chosen stepper motor is able to stop at increments of 15 degrees. This means that changes in barrel inclination of 3/8 of a degree are possible.

5.8 Initial Budget

In designing the components that will go into the initial design, the team always kept the budget in mind. All parts that need to be purchased are either off the shelf parts, or raw materials that are commonly kept in stock. Figure 5.8.1 shows the initial budget that the team needs to build the prototype. Included in the budget are off the shelf parts and raw materials.

Figure 5.8.1 Initial Budget

[pic]

As can be seen in Figure 5.8.1, the total budget up to this point is $639.21. This accounts for approximately 32% of the team’s allowed budget of $2,000. A portion of the remaining budget will cover unforeseen expenses, such as additional fittings and fasteners. Another expense that will be covered by the budget could be the hiring of an electrical engineering consultant that will help with the integration of the launching station onto the robot.

6 Conclusion

6.1 Project Wrap-up

Figure 6.1.1 shows a complete 3-D CAD model of the team’s launching station laid out on a 12” square. This is what will be built and tested next quarter. There are no parts that require special machining techniques, and thus all parts will be fabricated in the R.I.T. Mechanical Engineering Machine Shop.

Figure 6.1.1 Model of Prototype to be built

[pic]

The team chose all of the off the shelf parts and raw materials carefully so that they are standard stocked items, requiring less than two weeks of lead time to be shipped. This will allow for purchase orders to be completed and shipping to take place during the Spring Break. When the team arrives back on campus from the break, all of the parts should be waiting, and the prototype building and testing can begin.

The major problem that will set back the timely completion of a fully functional prototype will be the interfacing between the mechanical components and the electrical components. The team has no electrical engineer, and therefore must outsource some work to an external consultant. There is over half of the budget still remaining after the purchased parts are bought, leaving room to pay for the electrical engineering services.

After the wind sensor is made and assembled, it will be calibrated in the R.I.T. closed circuit wind tunnel. The G.P.S. sensor will have to be interfaced to a computer and calibrated to assure that the readings are accurate.

As previously stated in Section 1.5, the launching station will be built on a 12” x 12” square plate. After testing has been completed with all three major components, launcher, wind sensor and G.P.S., the focus will be on integrating as many components as possible onto a 10cm robot. Figure 6.1.2 shows an initial schedule that the team will use to complete all tasks required.

Figure 6.1.2 Spring Quarter Schedule

[pic]

In order to successfully complete the prototype, the launching station should be able to travel to a remote location (G.P.S.), detect the wind magnitude and direction (wind sensor), find a target (G.P.S.), correct launch trajectory for wind (robot and actuator), and launch a sensor to the target.

Bibliography

Beer, Ferdinand P., Johnston, E. Russell, Jr., Mechanics of Materials Second Edition, McGraw-Hill Inc. 1992, Pg. 700.

Benson, Tom. “Shape Effects on Drag.” NASA: Glenn Research Center. 8 May 2002.

12 Feb. 2003

.

Fowler, Michael. “Using Spreadsheets for Projectile Motion.” University of Virginia, VA. 12 Feb. 2003

.

Gilbertson, Mark. “Variation of drag coefficient with Reynolds number for circular disk and a sphere.” University of Bristol, England. 12 Feb. 2003 .

Guarente, Chris, et al, “Microrocket System,” Senior Design Study, Rochester Institute of Technology 2002

Mangonon, Pat L., The Principles of Materials Selection of Engineering Design First Edition, Prentice-Hall, Inc. 1999, Pg. 709

Peake, A.D. “Didion Approximation Derivation.” Apr. 2000. 12 Feb. 2003 .

Pister, Kris, “SMART DUST Autonomous sensing and communication in a cubic millimeter” University of California Berekely, Berkeley Sensor & Actuator Center, July 2001. 2/10/2003



Serway, Raymond A. Physics for Scientists and Engineers. Fourth Edition. Philadelphia: Saunders, 1996.

Shevell, Richard S. Fundamentals of Flight. Second Edition. Upper Saddle River, NJ: Prentice, 1989.

Shigley, Joseph E., Mischke, Charles, Mechanical Engineering Design Sixth Edition, McGraw-Hill Inc. 2001, Pg. 133.

Teasdale, Dana, “Solid Propellant Microrockets,” Master Thesis, University of California at Berkeley 2000

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download