Table of Contents
Team Vision for Detailed Design PhaseSummarize: The team planned to make decisions on major components, while expanding on the issues brought up by our major risks of cost, power dissipation, and safety. As a reminder of our system-level design, we've attached our systems level representation from the previous review.
Software DesignThe first revision of the software was completed in Simulink. This model doubles as the documentation for this design.
Test Stand Design
DesignThe frame was modeled using CAD Software and is detailed below. The dimensions are subject to change depending on the housing needs of the sub-assemblies.
A 3D render of the bench with a simple resistor block model is shown below.
As mentioned in previous reviews, the frame is composed of T-slot aluminum.A fiberglass pane will cover the area on the bottom right containing the motor and belt drive assembly.
Motor Mounting Bracket
DesignUsing the PERM 132 motor as our candidate load generator and unit under test, we created a mounting system that would be compatible with a belt drive system. The bracket is made out of 6" x 6" extruded aluminum angle bracket, and is mounted on a block spec'd to rest some of the weight of the motor on the frame plate, though we've calculated that the bracket could easily hold the motor itself. The drawing used for the PERM 132 motor is below:
The mounting dimensions and threads used in the mounting bracket are based on the above dimensions. Converting units to inches and drawing this up on the 1/4" plate yields the following:
In order to ensure the height was proper (greater than 6", the maximum size of the angle bracket), we designed the following block out of aluminum bar stock:
AnalysisThe biggest area of concern off the bat is that the bracket is able to withstand the high torque loading from the test. 1/4" aluminum was selected to reinforce the bracket, but we need to simulate the results of our loading to ensure we are in the right area. We started by simulating the worst-case torque loading of 37500 ozf-in.
The results of the simulation were very good, with 14.01 Ksi Von Mises stress resulting at the maximum torque. The simulation showed convergence to this value with successively finer meshes.
After we got these results, we chose to throw all of the loading conditions on the bracket (including full motor weight and belt tension). The simulation is shown below:
This test showed convergence to 17.1 Ksi maximum Von Mises stress resulting from maximum loading conditions.
For both tests we showed quite a distance from the yield strength of 35 Ksi. Our factors of safety are 2.50 and 2.05 respectively.
AssemblyOur design goal for the mounting bracket is the following:
Load GenerationWe've worked with the customer to constrain the requirement for overall torque load generation to 30% of the stall torque on the PERM PMG-132 motor. This equates to about 8kW when converted. Since then, the team has aggressively pursued sponsors for this element from all across the world. The full specifications provided to potential sponsors were the following:
- 4 ohms (+/- 5%)
- 8 kW continuous dissipation
- Maximum Voltage 72 Volts
- Any additional proprietary cooling options appreciated
The chosen sponsor for this component is Powerohm. They provided the resistor completely free of charge, including shipping. In exchange, Powerohm will be brought on as a kilowatt sponsor for EVT, giving them the requisite sponsor privileges.
Due to the large size of this resistor, it has changed some of the dimensions on our bench structure itself. Below is a diagram showing the resistor dimensions.
Sensing/Electronics SubsystemsThe array of sensors that will be used to provide users with live test data have been researched, and are fully specified to interact with the simulink models discussed above via the dSpace hardware. The following pinout diagram shows where each sensor will connect to the dSpace dS1103 controller during test. In addition, we have designed a set of control and safety circuits to make sure that loads are applied safely and as specified to the UUT motor.
Each of these sensors, with the exception of the IGBT device will be feeding back raw voltage outputs to the ADC ports on the dSpace controller. Then those inputs will be calibrated and interpreted using the models shown above in the models and simulation section.
The IGBT will be configured to receive a generated PWM signal from the dSpace controller at its gate. This device will then quickly switch on/off the loop coupling the resistive cell to the generator - modulating the effective resistance provided to the UUT motor's belt drive. All the details of PWM parameters will be handled at the model level, and simply are passed through the ADCH5 pin to the transistor.
Run/Control ModuleEngineering a safe high power system dictates that hardware and software should both be involved in safety shutoff control, as a matter of redundancy. Thus, we've designed a run/control module to facilitate both hardware and software inputs to the shutoff relays on the UUT motor and the generator loop.
Braking ChopperProviding a dynamic load on the input motor is achieved by using PWM on the generator loop, which requires fast high power DC switching capabilities. We've put together a Braking Chopper component that combines 2 IGBT's (Insulated Gate Bipolar Transistors) with a driver circuit to safely switch the load on/off, while protecting the switching IGBT during high duty cycle operation.
The braking chopper and drawing can be seen below:
Thermal ConsiderationsThe Braking Chopper and IGBT will be generating considerable heat due to the power throughput. The maximum temperature that the module can reach is 150 degrees C, so knowing we'd have margin we started with 125 degrees C.
A heat sink able to offer this low of a resistance will be larger than the module itself, and will cost around $40.
Our mounting solution would look similar to this:
High Power Electrical RiskOne major area of risk occurs in the actual electrical wiring of our components come phase II. Since we're working with such high power connections, we need to be extremely careful about choice of wiring, especially in light of thermal hot-spots and high current conditions. We plan on having an electrician look over the design in the spring and assist with the assembly to ensure the safety of all involved.
Budgetary RiskCurrently a major risk factor on the EVT Test Fixture project is the budget. This undertaking has proven to be much more capital intensive than perhaps anyone involved anticipated at the outset. Currently, the project is sitting at a ~$700 deficit, including some corporate sponsorships that the team has pursued. The following spreadsheet demonstrates the current overview of our budget by subsystem. A detailed breakdown of each component and associated cost is available as well.
- Risk Evaluation
The two major risks remaining for this project are budgetary concern, and electrical safety. Regarding the budget concerns, we are actively seeking additional sponsorship money going into the second semester, and hope to eliminate the remaining $678 deficit by the end of the first phase of MSD2. Electrical safety is now the main concern we are considering as we work into the build portion of this project. We are finalizing the electrical designs so that we can quickly seek the assistance of the RIT electricians, and get their assistance wiring and confirming our designs.
- Design Finalization
The majority of the design work for the Test Bench is complete. We anticipate a few minor design changes as the sponsored components are finalized. The long lead time components are mostly being included as part of the sponsorships, and are all scheduled to arrive during the winter intersession.
The three different teams will be building subsystems independently early in the semester, with integration following quickly. The majority of the semester will be used for full system integration and test.
- Phase 6 Plans