P17341: Harris Near-Zero CTE Test Apparatus
/public/

Preliminary Detailed Design

Table of Contents

Team Vision for Preliminary Detailed Design Phase

Plans for this phase involved a refinement of the system design generated during the previous phase. This involved more detailed design of physical components, including updated CAD models of relevant subsystems and parts. Additionally, simulation was performed in order to understand the effects of temperature change on different potential coupon samples and design materials. Finally, preliminary prototyping was carried out with purchased components. The intention is to demonstrate a working prototype of the maglev concept for the upcoming design review.
Preliminary Detailed Design phase goals

Preliminary Detailed Design phase goals

Feasibility & Design Prototyping

Prototype Structure Design

A prototype maglev system was constructed using rudimentary sensors (not sensitive enough to measure CTE but enough to measure gross position), an off-the-shelf electromagnet, and a 3D printed frame. Since this is simply a demonstration of the maglev system, 3D printing was used to quickly generate the support structure with no concern as to its thermal/phsyical properties. The demo frame was designed to mimic the adjustable concept selected in the previous phase for the final design.
CAD model of prototype frame

CAD model of prototype frame

Frame after printing

Frame after printing

The polygonal hole for the magnet was then sanded to accommodate the purchased magnet in a press-fit configuration. Other components shown below:
DC Electromagnet

DC Electromagnet

Circuit components from left to right: Hall Effect sensor, diode, MOSFET transistor

Circuit components from left to right: Hall Effect sensor, diode, MOSFET transistor

Inductive proximity sensor

Inductive proximity sensor

The assembled protoype is shown below.
Prototype maglev system. Components: 3D printed frame, DC Electromagnet, IR sensor, control circuit (shown below)

Prototype maglev system. Components: 3D printed frame, DC Electromagnet, IR sensor, control circuit (shown below)

Control circuit with Arduino board

Control circuit with Arduino board

Electrical Design

The maglev control circuitry involves a Hall-effect sensor for position sensing, an electromagnet to provide lifting force, a transistor to control current through the magnet, and an Arduino board for programming. The control circuit is shown in schematic form below.
Control circuit

Control circuit

The circuit functions by sensing position via the Hall sensor, which transmits a signal to the Arduino controller. Based on this signal, the Arduino emits a pulse-width-modulated signal to the transistor. When the signal is active, the transistor channel opens, allowing current to flow from the positive supply to ground through the magnet, activating it. When the transistor is off, a diode connected parallel to the magnet protects it from reverse current flow. A test of the Arduino's output capabilities is shown below.

Arduino Output test

Arduino Output test

This oscilloscope output shows the input signal on CH1 (yellow) and the output signal on CH2 (blue). The input is a digital signal from an inductive position sensor (unsuitable for the electromagnet controller due to lack of ratiometric output, but easy to use for testing). The output is the pulse-width-modulated signal from the Arduino, specified via a Simulink program.

Magnet operation controlled by sensor input

Magnet operation controlled by sensor input

Using Simulink/Arduino integration we are able to run control systems real time. Below shows the top and sub levels of the current control design. Figure 1 shows the top level architecture where we have our analog input from the IR sensor leading into a conversion box, which can be seen in Figure 2. The conversion box converts the raw sensor data to useful data. From the conversion box the system leads into the controller and following the controller we get to the PWM output. This controller consists of a threshold voltage, if the output from the conversion box is above the threshold the PWM has 100% duty cycle, if it is below the threshold the PWM has 0% duty cycle. The PWM output of the control system then leads to the gate of a MOSFET which turns the electromagnet on or off.

Fig1&2

Fig1&2

Using the Simulink model seen in Figure 1, tests were ran to assure the IR sensor works properly using a 20 Hz sample rate. Reading data post conversion box we were able to plot the following voltage vs distance curve seen in Figure 2.a. This curve can be compared to the curve given in the device's datasheet seen in Figure 2.b. The same voltage data can be extracted from both the oscilloscope and Matlab. Instead of a plot of Voltage vs Distances, these plots are Voltage Vs Time, where an object is constantly being moved further from the IR sensor. These plots can be seen in Figure 3.a and 3.b.
Fig2&3

Fig2&3

Figure 3.a shows a rather noisy signal. Using digital filters this noise can be reduced. Figures 4.a and 4.b show the effects of a 24th order Bessel filter.
Fig4

Fig4

Using the control system and the hardware design, the functionality of the current state of the system can be tested. Figure 5.a shows the physical reaction of the system, Figure 5.b shows the PWM input and output.
Figure 5a

Figure 5a

Fig5b

Fig5b

Figure 6 shows a block diagram/schematic combination of the levitation system.
Fig6b

Fig6b

Engineering Analysis & Simulation

Four different engineering materials were selected as possible candidates for constructing the test apparatus: Invar 36, Unidirectional T300/5200 Carbon Fiber composite, 6061-T6 Aluminum, and Steel, with generalized properties. Each material was analyzed using a temperature range of 50-100 degrees Fahrenheit, or a temperature change of 50 degrees. It was assumed that, at any point, the sample was at a uniform temperature. The samples have initial dimensions of 8"x1"x.007". An instantaneous coefficient of thermal expansion was used for analysis. The Carbon Fiber sample is unidirectional, with an analysis done for vertical and horizontal fiber alignment. The material properties utilized, and the simulation results are shown below. Material Properties
ANSYS Static Analysis of 4 different materials

ANSYS Static Analysis of 4 different materials

Carbon Fiber with fibers oriented horizontally

Carbon Fiber with fibers oriented horizontally

Material Deflection at 100F [in*10^-3]
Invar 36 0.291
Carbon Fiber (T300/5200, vertical) 6.400
Carbon Fiber (T300/5200, horizontal) 0.436
Aluminum (6061-T6) 5.281
Steel (general) 2.617

As expected from using an instantaneous CTE analysis, the deformation of the material is linear with respect to distance along the sample, and change in temperature. From the analysis, Invar 36 would be the most desired material to use for the testing apparatus, as it has the lowest thermal expansion and is isotropic. Carbon Fiber with fibers oriented horizontally would be the next best option, however difficulty may arise in properly orienting the material throughout the apparatus, due to its orthotropic nature. If a more common engineering material needs to be used, then some kind of Steel alloy is preffered. However, expansion of the apparatus will increase by an order of magnitude if Invar or Carbon Fiber cannot be used.

Thermal expansion animation of Invar 36 Alloy

Thermal expansion animation of Invar 36 Alloy

Drawings & Schematics

Prototype base plate

Prototype base plate

Prototype top plate

Prototype top plate

Main concept base plate

Main concept base plate

Main concept top plate

Main concept top plate

Main concept slot fixture

Main concept slot fixture

Main concept support rod

Main concept support rod

CAD to 3D Printed Prototype Flowchart

CAD to 3D Printed Prototype Flowchart

Screw Clamp Concept

Screw clamp concept

Screw clamp concept

Screw clamp concept

Screw clamp concept

Pros:

Cons:

Spring Clamp Concept

Spring clamp concept

Spring clamp concept

Spring clamp concept

Spring clamp concept

Pros:

Cons:

Plate Clamp concept

CAD model with plate clamp

CAD model with plate clamp

Pros:

Cons:

Bill of Material (BOM)

Bill of Materials as of 4/11/17

Bill of Materials as of 4/11/17

The total expenditure at the end of the Preliminary Detailed Design phase is $248.95. From a total budget of $4000, the project has $3,751.05 remaining.

Link to the live bill of materials here.

Test Plans

Draft Test Procedures

Draft Test Procedures

Test procedures for now only detail quantities of interest and data collection methods. Target values and error analysis will be provided as a result of feasibility analysis.

Live document detailing test procedures available here.

Risk Assessment

Risk assessment changed only slightly from previous design phase. Budget concerns had severity lowered after considering the minimal cost of all materials purchased so far. Additionally, a chamber purchase is not necessary, and therefore no longer a factor in budgeting. As individual system components are finalized, care will be taken to ensure integration with Harris fixturing (item 10).

Risk severity will be updated as more feasibility data is available from prototyping & analysis.

Risk Assessment Chart - Updated for Phase 3

Risk Assessment Chart - Updated for Phase 3

Risk Explanation

Risk Explanation

Link to the live document here.

Design Review Materials

The directory containing each team member's individual three-week-plan may be found at /public/Detailed Design Documents/Individual Plans

Plans for next phase

public/Photo Gallery/phase3_plan.PNG

public/Photo Gallery/phase3_plan.PNG


Home | Planning & Execution | Imagine RIT

Problem Definition | Systems Design | Preliminary Detailed Design | Detailed Design

Build & Test Prep | Subsystem Build & Test | Integrated System Build & Test | Integrated System Build & Test with Customer Demo | Customer Handoff & Final Project Documentation