Objective



Objective

The objective of this senior design project entails the conceptual and detailed design of a device exhibiting the artistic nature of motion and balance. Currently, kinetic sculpture is an exciting new field in the art world. Until now, however, computer programming has never been implemented as a means to control the movement of a sculpture. The chosen sculpture design combines elements of computer engineering, electrical engineering, and mechanical engineering to accomplish a controlled movement inspired by the work of Alexander Calder, a graduate of Stevens Institute of Technology in 1919.

The design has elaborate motions, a carefully designed-in sense of balance, and conveys a central theme or message. The final design was selected based on:

• The theme and its expression.

• Elegance and artistic aspect.

• The quality of engineering design, modeling, control, and the ability to construct a prototype.

The sculptures movement will be controlled via a package of components, including a programmable logic controller (PLC), multiple motors, and several types of valves.

The objective that was to be completed by the end of this spring semester was to build a maquette of the sculpture design and submit a proposal to commission a full sized version to be built in front of the Palmer dormitory. At this location, the design will function to beautify the area and serve to create a fun place for students to study or relax outdoors. The maquette was assembled and ready to display on Senior Design Poster Day, April 28th, 2004. We have high hopes that the school might be willing to hear our proposal and pursue the actual construction of a full-sized version of our maquette as a result of the positive feedback we received on Poster Day.

Acknowledgement

Our group would like to thank the following people for their various contributions to our project:

Dr. Arthur Shapiro is our Mechanical Engineering advisor. He is an avid sculptor, Provost Emeritus and Dean of Faculty Emeritus. Professor Shapiro has donated innumerable hours of technical expertise, artistic experience and overall guidance.

George Wohlrab, Joe and their student workers (especially Dave and Ronny) helped us fabricate our maquette in the machine shop. They have been a tremendous help in all aspects of the assembly of the model fountain and the maquette would not exist without them. They were great resources as well as letting us use their machine shop for all of our work and testing.

Dr. Sumit Ghosh is our Electrical and Computer Engineering advisor. He is a Hattrick Chair Professor of Electrical and Computer Engineering at Stevens Institute of Technology. Professor Ghosh has provided helpful advice on the technical aspects of programmable, specifically programmable logic controllers (PLCs).

Barton Rubenstein, a thriving kinetic sculptor, allowed us to visit his studio in Maryland, showed us his works involving kinetic water sculptures. Mr. Rubenstein also gave us useful advice for building a full-size outdoor kinetic sculpture.

Alexander Calder (1898-1976), a graduate of Stevens Institute of Technology, is the most acclaimed and influential sculptor of our time. Calder is renowned in the world of modern art as the founding father of kinetic sculpture. Famous for his mobiles, whose abstract elements are suspended harmoniously through balanced movement, Calder has been an enormous source of inspiration for our project.

Introduction

The backgrounds of each group member varies through a range of fields and interests, including computer engineering, electrical engineering, mechanical engineering, carpentry, and art. After initially brainstorming, several ideas were presented, including a programmable autostereogram, a 3D moving toy, and a water sculpture (See Appendix A for more information). Based on the criteria for technical aspects, beauty, cost, feasibility and labor, the group chose to further develop the water sculpture concept.

Originally, this design included a fountain mechanism to create a uniform sheet of fluid to work with as a medium. It would consist of many paths with computerized valves that will change the direction of water flow to form a pleasant image for the observer. Possibly, colored lights and/or musical notes would be added to correspond to each vane of water to enhance the sculpture. The entire structure could, then, be programmed according to light shows or musical compositions. This concept was intended for a relatively small-scale sculpture with the potential to be mass-produced.

Considering, however, the potential artistic significance of this project creating a new type of kinetic sculpture, the group changed direction of the project to create a design for a single large-scale sculpture to potentially install on the Stevens Institute of Technology campus. After more brainstorming, several new conceptual ideas for a large-scale installation piece / water fountain were presented, including a rolling sphere fountain, a Calder stabile fountain, and a rotating sphere fountain (See Appendix B for more information). In this design, the group combined a series of components to control the intended movement of the sculpture. A PLC driven controller/driver was used to run a computer program with instructions for the predetermined motion of the piece and send instructions to the motors. The motors, in turn, control the rotation of the spheres, which creates the desired changes in the sculpture’s orientation.

Final Design

The design the group proposes to construct is made up of three spheres that are connected with the largest sphere on the bottom and the globes get progressively smaller as they move up. CAD drawings showing this design from alternate angles:

Two additional top views of basic construction are given below to illustrate more detail of the proposed design. For more background information, see Appendix C.

[pic]

The movement of the central piece would first be created by the rotation of the base sphere. Added to this movement, each of the two other spheres move on tracks located in a straight line on top of the previous sphere. Each of the spheres would have valves similar to the sphere nozzle, creating the appearance of three moving balls of water, and four jumping jet valves would be installed in the fountain base. The jets will shoot water at given periods of time to coordinate with the movement of the main piece.

A full-size prototype of the programmable kinetic sculpture will not be feasible due to its large size and cost. Consequently during this semester, the group researched the technical requirements, built a maquette of the design, and displayed a modified version on Poster Day.

The maquette is altered from the sculpture design; in that, each sphere is clear to allow the viewer to see the internal structure and movement, incorporating the engineering aspects into the appeal of the piece, as well as reducing the cost. Additionally, the water pattern created by the valves is simplified, and the jumping jet valves in the base of the fountain are omitted.

The final sculpture design contains three spheres that rotate on a 180˚, 2D track in the sphere below it and a base that is capable of rotating 360˚ to give the effect of unlimited movement. The maquette, however, had limitations for what the group would be able to build in one semester with a limited budget; therefore, the implemented design did not allow the base to rotate and was made mainly out of clear plastic and PVC pipe instead of a sturdier material that would be appropriate for an outdoor installation.

The components necessary to build the maquette include: for the body - clear plastic spheres and PVC pipe, wooden base, and pool; to move the spheres - stepper motors, controllers, and drivers; and to produce the water flow – clear plastic tubing, PVC pipe, pumps, and valves.

Parts List

Since weight was an important issue for the construction of the maquette and life-size sculpture, the group decided to use PVC and plastic materials wherever possible to minimize weight. However, we found that there were some parts that simply didn’t come in plastics or PVC. Also, we substituted metal for certain parts to create more support, for example in the brass swivel joint.

Later during the assembly period, we learned that the choice to use PVC presented certain problems because PVC cement was necessary to bond two pieces. Here is a table summarizing parts used to construct the maquette:

[pic]

For a better idea of these parts, here are some diagrams of some of the parts used for the construction of the model:

[pic]

PVC Hemispheres

• 60 gauge clear PVC

• 1/16th of an inch thick

• Sizes

o 12” Diameter

o 18” Diameter

o 24” Diameter

Water Ace   1/2 HP 115 Volt

Submersible Sump Pump

• For use in 12" to 18" diameter sumps

• 1 1/4" discharge

• 53 gallons per minute at 10' lift

Stepper Motors

#ORM 2913E-4 Size 34

• 1200 oz-in

• 8-Wire

• 5.7 Amps

• 32.7V

• 1.8° step size

• 200 steps/rev

#ORM 268E-2 Size 23

• 240 oz-in

• 8-Wire

• 2.9 Amps

• 3.2V

• 1.8° step size

• 200 steps/rev

[pic]

[pic]

Controller/Driver and Power Supply

[pic] [pic]

[pic][pic]

[pic]

[pic]

[pic]

Motor Control

One of the most common devices in industries requiring electronic and machine control is the programmable logic controller (PLC). These devices were created as a relatively inexpensive and less complex way to replace sequential relay circuits. They are readily available and adaptable for virtually any electro-mechanical purpose. The group’s fountain-sculpture design relies on PLC driven devices to store information input by computer and send signals to control the stepper motors.

All PLCs consist of a central processing unit, memory and data storage areas, and circuits that can handle data for input and output. PLCs are able to simulate counters and timers, which makes it very useful for the final design.

The sculpture will need to not only move mechanically, but at pre-programmed intervals of time as well. Although many PLCs have the ability to increment in milliseconds, that level of precision is not necessary for this application. The PLCs that can increment between 1 and 655.35 seconds (roughly 10 minutes) are sufficient. The sculpture will not be in a constant state of movement, for both economic and aesthetic purposes. For example, constantly moving gears wear down more quickly and require more power, and at slower speeds the motors have more torque, which is necessary to lift the weight of the sculpture. It is also desired for viewers to observe each new position of the sculpture, rather than to see a blur of constant movement.

In addition to a PLC, a stepper motor needs an indexer and driver to control each stepper motor. A block diagram of the setup follows:

[pic]

To simplify the list of necessary components, the group opted to purchase PLC driven combination Controller/Drivers from to combine the PLC, Indexer, and Driver in the above diagram into one component. A SMD 483I-232 was used to control the smaller stepper motor (ORM 268E-2), and a SMD 1007IE was used to control the larger stepper motor (ORM 2913E-4).

The SMD 483I-232 and SMD 1007IE are motion controllers that can connect a computer and a stepping motor to allow the user to control all of the motor’s movement via a computer program.  The program can be set to run as soon as power is applied to the device, after a time delay using a specific command, or be transferred to the motor as it is written. This project utilizes HyperTerminal to communicate with the motion controllers, but any programming software with output commands, such as VisualBasic or C++, is compatible. A few basic capabilities of the controllers are:

• Storing programs of up to 2048 bytes

• Setting initial and slew velocities, between 20 and 20,000 full steps per second

• Setting step resolution, between full steps and 1/256 step

• Moving at a constant velocity indefinitely, until the user interrupts the command

• Moving a specific number of steps two directions

• Moving to a specific position relative to the origin

• Waiting for a specific amount of time before executing a new command

• Setting acceleration and deceleration rates

• Loop through any previously written programs

In order to get the necessary torque from each motor, they were both wired in parallel. The 8-wires from each motor were extended out of the sculpture to a controller/driver that was protected from potential splashes of water by a PVC box, made in the machine shop, and connected to power supplies contained in a separate PVC box. A wiring diagram for the smaller motor and controller/driver is as follows:

[pic]

A wiring diagram for the larger stepper motor and controller/driver is as follows:

[pic]

Communications were established with each controller/driver via RS-232 9-pin cable. The SMD 483I-232 came with a compatible connector on the controller; however, the SMD 1007IE is only compatible with RS-422 data transfers. To convert the RS-422 to RS-232, a COMT 285 was purchased, also from , attached the 5 terminal blocks to the RS-422 pin connector on the controller/driver, and connected to the computer via a RS-232 25-pin to 9-pin cable.

To power each system, a power supply was provided for each controller/driver to convert the voltage output of a typical wall outlet to the correct voltage and current values.

Each power source does not exceed the respective controller’s maximum ratings for voltage or current and, therefore, are appropriate. Since the controller’s temperature also must be kept in a given range, each was attached to a heat sink to facilitate dispersing the heat it produced.

The last component of the wiring scheme is the AC line conditioner, which was recommended to prevent voltage spikes that could damage necessary equipment. Instead of purchasing an AC line conditioner specifically for this purpose (approximately $600), a UPS system with automatic voltage regulation (AVR) intended to provide backup for a computer was substituted. The applicable specifications for this component are:

|General |Model |CPS825AVR |

| |Capacity |410Watts |

|Input |Frequency |57Hz to 63Hz |

| |Voltage Range |85V - 150V |

|UPS-Output |Line Conditioning Regulation |AVR Max. Boost 13% of input voltage for output |

| | |regulation while input voltage is from 8% to 30% |

| | |under nominal. |

| | |AVR buck 3% of input voltage for output regulation |

| | |while input voltage is from 8% to 26% over nominal. |

| |On Battery Output Voltage Regulation |120VAC ± 5% (user adjustable) |

| |Transfer Time |4ms typical, including detection time |

| |On Battery Output Wave Form |Simulated Sine Wave Form |

| |Max. Load |410W |

As a result, the output voltage of the AVR is consistently between 110V and 120V, which is a safe range for these components. Calculations to ensure the required outputs do not exceed the maximum rating of the UPS system:

Power Requirement for Small Power Supply - [pic]

Power Requirements for Large Power Supply - [pic]

Total Power Requirement - [pic]

The 320 Watts needed is less than the maximum rating of 410 Watts; therefore, the UPS system can be used as an AC line conditioner.

Data

• Weight

o Small Sphere: 21 oz.

o Medium Sphere: 42 oz.

• Torque

o Small Sphere: 315 oz-in

o Medium Sphere: 1386 oz-in

• Gears

o Small to Large for larger torque and

slow speed

o 2:5 ratio (larger Sphere)

o 1:3 ratio (medium Sphere)

This was done before assembly and all the parts were in. During assemble parts and modifications were added which caused an increase in weight. We will later discus how this may have affected our final out come of the model because it resulted in the wrong calculated torque and gear ratios.

Size Ratio:

Model Full Size

|Large Sphere Diameter |24 inches |4 feet |

|Medium Sphere Diameter |18 inches |3 feet |

|Small Sphere Diameter |12 inches |2 feet |

|Pond Diameter |8 feet |18 feet |

Progress of Design

Joint Movement Progress

[pic]

[pic]

[pic]

Design Improvements

Final Model Design

Assembly

Between our proposal presentation this spring and the progress presentation, we started ordering parts for our model. We chose to use PVC and plastics wherever possible due to weight. How ever in returned this turned out to be a little more trouble do to have to glue/cement pieces together, which was permanent; also it was hard to get exact measurements and lengths until piece was glued in (error of fractions of an inch which sometimes could cause problems). When we started piecing things together we realizes the plastic hemispheres need supporting and also need a way to attach it the PVC structure of the moving systems, so we designed plexi glass cross pieces (as seen in picture). We also had to come up with a new way of mounting motor once we got the design for the joint worked out (after swivel joint and motor ordered). The design for the motor mount can be seen below; the bottom plate was mounted around the T-joint that was directly underneath it. These pieces added unexpected weight to the spheres and were not included in the calculation of the torque because that was done before parts were ordered. We tried to accommodate for the change with the gear ratio. Because we wanted slow speeds and high torque we went with a small to large gear (more revolution of the motor to cause less of a turn of the rotating arm).

Cost

|Barnard LTD. |193.18 |  |  |

| | |112 |24" Spheres |

| | |60 |18" Spheres |

|Barnard Ltd |62.25 |  |  |

| | |10 |6" Spheres |

| | |16.5 |9" Spheres |

| | |26 |12" Spheres |

|Home depot |18.09 |  |  |

| | |7.08 |6 @ 1.18 |

| | |2.17 |  |

| | |1.84 |2 @ 0.92 |

| | |5.98 |  |

| | |  |  |

|Home Depot |13.78 |  |  |

| | |0.73 |Fittings |

| | |3.48 |1.5" PVC |

| | |2.97 |  |

| | |3.98 |Goop |

| | |1.04 |  |

| | |0.8 |  |

|Home Depot |38.02 |  |  |

| | |10.33 |Luan |

| | |24.99 |Corner molding |

|Home Depot |134.67 |  |  |

| | |119 |1/2 HP Sump Pump |

| | |0.6 |4 fittings @ 0.15 |

| | |1.19 |Fitting |

| | |0.95 |5 fittings @ 0.19 |

| | |3.47 |Fitting |

| | |1.84 |Fitting |

|Lowes |134.91 |  |  |

| | |5.54 |Cleaner |

| | |2.56 |2 @ 1.28 |

| | |2.21 |13 @ .17 |

| | |1.52 |1/2" Tee 10 Pk |

| | |1.28 |Copper coil |

| | |37.96 |  |

| | |59.94 |  |

| | |19.97 |  |

|Lowes |0.88 |1/2" PVC conduit |

|McMaster-Carr |128.95 |  |  |

| | |93.1 |2 @ 46.55 |

| | |12.9 |10 @ 1.29 |

| | |6 |PVC tubing |

|McMaster-Carr |12.76 |PVC tubing |

Water Testing

Water and Motor Testing

Test Run Problems

During our test run with water and motors, we ran into a serious problem. Our gears would slip out of place, causing the motors to be unable to hold the structures above it. While looking at the assembly, we realized the arm holding the gear would lose its placing due to the weight of the sphere above it. There needed to be something keeping the gears together in order for them to mesh correctly and keep the elbow joint in place. If you look closely to the picture on the bottom right, you can see the gears are not in contact. Also, our large motor, which was located in the large sphere, was not responding to our programming. As a result of the fact that we ran this water and motor test the day before Poster day, we decided as a group that the important thing was to have something presentable for the next day. After exploring several possible solutions, we decided to separate the sculpture so that we could show the movement of the middle sphere with its working motor and came up with a temporary solution to hold the gears in place. Then we made the large and small sphere a rigid body. This way people were able to get the idea and motion of the movement that would have occurred if working correctly, while also being able to see the effect of the water.

Senior Design Poster Day

[pic]

The above picture shows our final product that we presented on poster day. This model is not what the final design would look like; this was what we were able to get together with the time and money we had. We moved the medium sized middle sphere to a table and pump of its own and put a nozzle at the end of its rotating arm. We tried leaving the small sphere on top of the medium sphere so that the medium sphere arm would rotate the small sphere but because of either our gear problem which we tried to temporarily fix by bracing the gear are or because of the motor not having the torque needed because of added weight during assembly or a combination of the two, the small sphere was not rotating as it should. So we replaced it with the nozzle and put the small sphere on top of the large sphere, but were not able to get it to move because we lost connection with it. We also had the existing problem of the gears.

Summary of problems we ran into just days before poster day:

• Large motor not communicating with computer due to faulty wiring.

• Gears not connecting due to arm/joint displacement

• Possibly not enough torque in small motor (was not able to test the large motor)

We also ran into some problems the day of the poster day. First we had a bit of trouble getting water, due to a bad washer in our hose. However, the help of the physical plant workers on duty solved this problem that day. Once we had the pool filled we ran the pumps, but because we had both pumps as well as the power for the small motor all coming from one outlet, we overloaded it causing to trip the circuit breaker.

After some plug rearranging, we finally got water in the pool and the pumps and small motor working. The only other slight problem we ran into was water pressure or pool size depends on how you look at it. When we had the valves between the pump and the spheres fully open, then the lower streams of water would shoot outside the diameter of the pool. When we would partially close the valve to where the lower streams made it inside the pool, then the streams at the top of the sphere would not be a strong. In short, we would have liked to use a bigger pool except kiddie pools only come so big!

We also ran into the problem of the ground outside the gym not being level. We had this problem when we tested it so we glue 18” pieces of 1x3 wood every 2’ around the inner edge of the pool. However when we filled it with water on poster day, some of the wood pieces that we were using as a support system became unglued. We had tested this structure before and it worked great, but with the long-term water exposure, the glue must have lost its bond. When the side of the pool started to give, we came up with the solution of moving one of the large flowerpots that sits outside the gym right up against the poolside to support it.

Prototype Performance and Evaluation

The difficulty of this project was that we were trying to make a model of a sculpture that would be permanently installed; the whole model had to be a mobile as possible. Such challenging examples were the pool as well as the pump and it housing, normally in permanent installation the pump would be in a built in hole lower than the pool surface.

While evaluating the design, the first task assigned for the Spring semester, the group realized that a total separation of the design and assembly phases would not be possible because certain specifications for components could not be obtained before it was received delaying further calculations. Therefore, the project's progress was initially delayed while the design was determined partially by trial and error of the ordered parts.

-----------------------

Final Design

Top View

Front View

Construction Views

Top View Two

Top View One

PVC Elbow PVC T- Joint PVC Male Adaptor Brass Swivel Joint

PVC List Nylon Single-Barbed Tube Fittings

Fabricated Manifolds made out of solid square gray PVC. Picture shows barbed fittings teffloned and screwed into manifolds

This image is of the swivel joint attached to the gear, which was welded on the brass fitting

SMD 483I-232 48V, 3A Microstepping Driver with Controller, RS-232

MUS 40-02 40VDC, 2A Unregulated DC Power Supply, 104-132VAC Input

SMD 1007IE

80V, 7A Microstepping Driver, with Controller and Encoder Input

MUS 40-06

40VDC, 6A Unregulated DC Power Supply, 104-132VAC Input

Both Microstepping Driver, with Controller for both motors wired up and in their gray PVC boxes, which we made. Under neigh this box sits another box just like it, which holds the two Power supplies for both motors.

Typical Stepper Motor System

These are front and side views of our early basic design ideas for the whole body. These diagrams are helpful to be on the same page as to see how the actual sculpture was going to work and where structures would go. The basic idea was established. Water would be pumped from underneath through the pipes or members that were the structure holding and moving the spheres. The plastic spheres were basically a shell cover to hide the interior motors and gears and manifolds. The reason being is so they don’t add more unnecessary weight that the arms had to rotate. After developing these draws and basic concept, we needed to focus on the interior joint and how the mechanics were going to work.

Beginning ideas of how the arm connecting the spheres should rotate. However we couldn’t come up with a way of having the rotating joint move while having water run through it and not loose the water pressure or have it leak.

We then found the use of a swivel join would help to solve our problem. We then came up with the design. However when we got the swivel joint, it didn’t swivel quite in the way we thought it would and also had concerns of welding the gear directly to the swivel joint in concern that that would melt an O-ring in the swivel joint.

We then decided to rearrange things and came up with this new design and moved to weld the gear to a brass fitting that treated into the swivel joint and connected the swivel joint to the PVC elbow. At this point we could work on the motor mount design better since we knew where we wanted it.

Original CAD drawings before we changed the manifold

design and the PVC lengths to fit gears and motor in

Most updated and resent design before full assembly. We will talk more about adjustments we would make to this design if this project were to go farther in production past Poster Day.

Small Top Sphere

D=12”

Medium Middle Sphere

D=18”

Large Bottom Sphere

D=24”

This is the full sculpture with box platform and pool. The only thing not included in this CAD drawing that was in our model is the barbed tube fittings that were in the hemisphere shells and the plastic tubing running from the barbed fittings on the manifolds to those that would be on the hemisphere.

After turning on water

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

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

Google Online Preview   Download