P14421: Next Generation Smart PV Panel
/public/

Detailed Design

Table of Contents

Prototyping, Engineering Analysis, Simulation

Heat and Power Analysis

Excel Sheet for Heat Analysis and Power Calculations

Notes on the tabs in the Heat Analysis and Power Calculations with assumptions

Cell-Ink Configurations that were considered

Chosen Cell-Ink Configuration

Chosen Cell-Ink Configuration

Hourly Energy Comparison January 2

Hourly Energy Comparison January 2

Hourly Energy Comparison February 20

Hourly Energy Comparison February 20

Hourly Energy Comparison March 8

Hourly Energy Comparison March 8

Monthly Net Energy Comparison January

Monthly Net Energy Comparison January

Monthly Net Energy Comparison February

Monthly Net Energy Comparison February

Monthly Net Energy Comparison December

Monthly Net Energy Comparison December

Drawings, Schematics, Flow Charts, Simulations

Drawing for Screen

ANSYS Work

Various models were created in ANSYS that depicted different potential scenarios. Variables included the ambient temperature of air (Tbulk), temperature of ink traces(Tink), and convection heat transfer coefficient (h). The model consists of a .001m layer of silicon beneath a .003m layer of glass. The ink traces were modeled as heated areas on top of glass. A selection of important results are shown below, which prove feasibility.

Assumptions made:

Ink Configurations:

Ink Configurations
Configuration 1 Configuration 2 Configuration 3 Configuration 4
Configuration 1

Configuration 1

Configuration 2

Configuration 2

Configuration 3

Configuration 3

Configuration 4

Configuration 4

Constants:

Name Value Units
Thermal conductivity of glass 1.10 W/(m-K)
Thermal conductivity of cell (silicon) 11.0 W/(m-K)

Scenario 1: Configuration 3

Variable Value Units
Tbulk neg5 deg C
Tink 35 deg C
h (top of panel) 15 W/(m2-K)
h (bottom of panel) 2 W/(m2-K)
Scenario 1: (Top View)

Scenario 1: (Top View)

Scenario 1: (Side View)

Scenario 1: (Side View)

Scenario 2: Configuration 4

This scenario mimics Scenario 1, but removed the extra trace going across the midpoint of the cell. All other conditions remain the same. It is visible that without the additional trace, the midpoint of the cell will not reach a temperature high enough to melt snow that is located there.

Variable Value Units
Tbulk neg5 deg C
Tink 35 deg C
h (top of panel) 15 W/(m2-K)
h (bottom of panel) 2 W/(m2-K)
Scenario 2: (Top View)

Scenario 2: (Top View)

Scenario 2: (Side View)

Scenario 2: (Side View)

Scenario 3: Configuration 3

This scenario inputs much higher 'h' values, representing a much greater amount of convection caused by wind. All areas in-between the traces reach a temperature greater than O deg C, but large portions of the areas in-between the traces only barely cross this threshold, reaching approximately .5 deg C.

Variable Value Units
Tbulk neg5 deg C
Tink 35 deg C
h (top of panel) 25 W/(m2-K)
h (bottom of panel) 4 W/(m2-K)
Scenario 3 (top view)

Scenario 3 (top view)

Scenario 3 (side view

Scenario 3 (side view

Scenario 4: Worst Case Scenario (Configuration 3)

In this scenario, extreme values for conduction as well as bulk temperature are input, in order to create a worst case scenario. If the ink is only heated to 35 deg C, as it is in this model, the majority of the surface area of the panel will remain below freezing, and clearly won't be capable of melting snow.

Variable Value Units
Tbulk neg 15 deg C
Tink 35 deg C
h (top of panel) 25 W/(m2-K)
h (bottom of panel) 4 W/(m2-K)
Worst Case Scenario (top view)

Worst Case Scenario (top view)

Worst Case Scenario (side view)

Worst Case Scenario (side view)

Scenario 5: Configuration 2

This scenario inputs much higher 'h' values, representing a much greater amount of convection caused by wind. All areas in-between the traces reach a temperature greater than O deg C, but large portions of the areas in-between the traces only barely cross this threshold, reaching approximately .5 deg C.

Variable Value Units
Tbulk neg5 deg C
Tink 35 deg C
h (top of panel) 25 W/(m2-K)
h (bottom of panel) 0 W/(m2-K)
Scenario 5 (top view)

Scenario 5 (top view)

Conclusions:

These models further prove that the project is feasible. They support the results of the fin calculations, both by proving feasibility and coming to the same conclusion as far as the most effective trace pattern. It should be noted that both the fin calculations and ANSYS models used consist or similar values for all variables and constants. Emphasis needs to be placed on performance under high convection rather than performance under low temperatures.

Control

Control Electronics Schematic

Control Electronics Schematic

Express Sch Schematic of the micro-controller's interface with system peripherals. Schematic nodes can be linked to other sheets/subsystems to facilitate trace mapping and networking of the main board.

Controller Flow Chart Scheme

Controller Flow Chart Scheme

Pseudo Code for Flow Chart

Pseudo Code for Flow Chart

Power

Power Electronics Schematic

Power Electronics Schematic

Sensor Operation and Conditioning

Because of the nature of this project, it is necessary to develop a design that consumes as little power as possible. With that constraint in mind, the team is implementing a system that, when not in use, will "sleep" in a low-power consumption state. Implementing interrupt service routine capabilities on the Arduino will allow for the sensors to provide both the signal to "wake up" and the data necessary to make a decision to melt the snow.

Each sensor is fitted with customized circuitry to clean and condition the signal that interfaces with the Arduino. From the output of the sensor, the signal is first subjected to a low-pass filter that eliminates frequencies outside of a 10Hz bandwidth. This will prevent any noise from being amplified in the rest of the circuit. In the situation where the sensor outputs a negative voltage, the signal is then buffered using a voltage divider to keep the voltage across the capacitor from changing (due to other components in the circuit). Next the signal is shifted and amplified/attenuated to the 0-5Vdc range that the Arduino analog inputs can handle. At this point, the signal splits into two paths. The first path is to the analog inputs of the Arduino. The second path leads to a comparator, which will develop a high or low (0V or 5V) based on its input. These comparators drive a discrete logic block, which, in turn, drives an interrupt pin on the Arduino.

The comparators provided a design challenge, as a standard comparator will switch (high and low) at a single input point. In the situation that the input voltage hovers around that switching point for an extended period of time, the comparator's output will oscillate back and forth between high and low very quickly. This behavior causes errors in discrete logic circuits, and could effectively render our system useless (as the Arduino would continually try to wake up). To remedy this, the comparators in our system will be implemented with hysteresis, which moves the switch-high and switch-low input points away from the single point. This guarantees that the output of the comparator won't oscillate while the input voltage is stagnant.

Sensor and Filters

Sensor and Filters

Level Shifters

Level Shifters

Comparators

Comparators

Interrupt Logic

Interrupt Logic

Theory of Operation

Below are captures of the ambient temperature sensor conditioning schematic and relevant simulation waveforms. The hysteresis is designed to switch high at -1 degC (when temperature is rising from below) and at -3 degC (when temperature is falling from above). This provides a stable switching around -2 degC, which represents when it would make most sense to melt the snow on the panel.

Ambient Temperature Sensor Conditioning Schematic

Ambient Temperature Sensor Conditioning Schematic

In the schematic, Vsupply and -Vsupply are the +5V and -5V supplies used to power the circuitry. Vsense generates a triangular waveform that oscillates between -550mV and 1500mV (this emulates the behavior of the LM35 temperature sensor from TI). R1 and C1 act as a low-pass filter to eliminate noise outside of a 10Hz range. U1 is a TL081 op-amp configured as a voltage follower. This acts as a buffer between the filter and the rest of the circuit and ensures that the voltage across the capacitor matches the voltage output by the sensor. U2 is the level-shifter. It amplifies and shifts the sensor output so that it changes within the 0-5V range accepted by the analog inputs on the Arduino. Resistors R2-R5 designate the offset around which the output will be centered. In this instance, the output of U2 will be centered at 2.5V. R6 and R7 provide the necessary gain for the signal. At this point in the circuit, the signal branches. One path leads to the Arduino analog input, which is used to make control decisions. The second path leads to a third op-amp. U3 is the comparator, which operates as described above. Resistors R8 and R11 determine the amount of hysteresis the comparator has, while R9 and R10 control the offset around which the hysteresis is centered.

Simulation

The signals shown in the simulations are as follows:

Ambient Temperature Sensor Conditioning Schematic Simulation

Ambient Temperature Sensor Conditioning Schematic Simulation

Ambient Temperature Sensor Conditioning Schematic Simulation, Rising Temperature Focus

Ambient Temperature Sensor Conditioning Schematic Simulation, Rising Temperature Focus

The image above provides a closer look at the rising edge of the comparator output. The comparator is designed to output a logic high when a rising input reaches 1.223V, which corresponds to ~ -1 degC.

Ambient Temperature Sensor Conditioning Schematic Simulation, Falling Temperature Focus

Ambient Temperature Sensor Conditioning Schematic Simulation, Falling Temperature Focus

The image above provides a closer look at the falling edge of the comparator output. The comparator is designed to output a logic low when a falling input reaches 1.178V, which corresponds to ~ -3 degC.

Bill of Material (BOM)

BOM

BOM

Final Bill of Materials for the Smart PV Panel System. The row highlighted in Red indicates that the item is being provided by an external source other than RIT.

Test Plans

Control Subsystem Test Plan

Sensor Array Test Plan

Power Electronics Test Plan

Ink Heat Dispersion Test Plan

MSD II Schedule

MSD II Schedule

MSD II Schedule

Risk Assessment

Risk Assessment

Risk Assessment

Risk assessment of the final detailed design along with the owner (Team member) directly responsible for specific risk and mitigation items.

Design Reviews

Detailed Design Review Powerpoint

Notes From Detailed Design Review


Home | Planning & Execution | Problem Definition | Systems Design | Subsystems Design | Detailed Design | Build, Test, Document | Project Review | Photo Gallery