Improvements made since the break and what's to come
February 3, 2024 - By Hedda Grelz, Andres Calderon, Oselumenosen Ekhuemelo, and Adolfo Diaz
Work Period Dec. 11 - Feb. 3:
In Capstone I, Team Navy finalized all analysis and showed an updated design of the vent mechanism that got preliminary approval with some feedback from the UH Machinist. Over the winter break, the team finalized the vent design according to the comments from the machinist. Two meetings with the UH Machinist were held over the break to discuss the adjustments made to the design and verify the material selection. The main changes were made to the pulley system placed at the center of the vent (see Figure 2) which is used to wrap the SMA wire to give enough vent displacement. The team has changed the material of the pulley system to a high-temperature UHMW Polyethylene instead of steel. It was also decided that the pulley system will be machined separately as shown on Figure 1, and press fitted into the case of the battery and held in place with epoxy for security. Implementing these updates improves the design as the new material is non-conductive (avoids having to isolate the induced current from the wire to the rest of the vent), cheaper than steel, and easier for the Machinist to machine and implement into the vent at the small scale that the team is working with.
Figure 1: Assembly of the updated pulley system and vent case.
It was also decided that a 1018 low-carbon steel rod with a diameter of 3.5 inches and a length of 3 inches would be enough to machine the remaining parts of the vent. Lastly, the team decided to use the high-temperature UHMW Polyethylene rather than steel to fabricate the battery testing case as well. The change in materials helped the team reduce the overall cost of the project by reducing the cost of items such as the battery case and vent parts by switching to more cost-effective materials. The machinist committed to finalizing the fabrication of the vent mechanism and the battery testing case before the end of March.
The materials needed to start the fabrication process have also been ordered. This includes the material for the vent, vent case, testing case, and steel springs. Control system components will be ordered at a later point. The team has also decided to not add an O-ring since it has a long shipping time, adds to the project cost, and will not make a significant difference for the educational purposes and validation testing of the vent. For real-life applications, an O-ring should be implemented.
Since the machinist cannot promise a finished prototype until the end of March, the team had a meeting with the TA who is responsible for the ME 3D printing lab to explore the option of printing some parts of the vent for initial testing. The vent part that moves up and down has to be machined as the weight of this is accounted for in the spring selection and changing the material would therefore invalidate the results from that analysis. The TA Bryan Flores looked at the team’s CAD and said that this can easily be printed using ASA filament. The only thing that needs revision for the printing to be possible is the threads that will be used to screw the vent lid and the vent case together. The team will work on adjusting this over the weekend to then send the CAD files to Bryan for printing.
The team has also spent the first part of the semester exploring options for the control system and learning more about how to develop and test the system. After the literature review, the goal of the control system was defined as follows: “Develop a PWM modulated controller to control the position of the SMA wire (open vs closed state). The control system should be triggered when the temperature sensors in the battery case sense a temperature increase greater than 25 C/min. Once triggered, the SMA should be heated (with x% duty cycle) to open the vent. Once opened, the wire will continue to be heated until the temperature inside the battery has cooled to 50 C. This continuation of heating will happen at y% duty cycle, which will be lower than x. If the positional stability of the SMA is not good in the first tests of the control system, a PD or PID feedback control using data from a displacement sensor will be implemented. Note: the temperature sensors in the battery pack will not be used for feedback control, but just for on/off for the active heating”. Developing the control system is a crucial part of the project and one of the most challenging ones. The team does not have much prior knowledge of the subject and is therefore setting up a meeting with their technical advisor Dr. Gangbing Song to get some guidance on the matter. The team is planning on using Raspberry Pi and a PWM control that will be triggered by temperature sensors and have feedback using displacement sensors for positional stability of the SMA wire. The team needs some help from Dr. Song about what order to do things in, and how to set up experiments to get ballpark values for the duty cycle and frequency needed to heat the SMA.
Figure 2: Updated finalized vent design that has been approved by the UH Machinist
The team’s focus for the next two weeks will be on the active opening of the vent and the design and testing of the control system. The following workflow has been determined for the control system:
1. Open loop testing of SMA wire actuator to study the basic properties.
2. Perform open loop testing to investigate PWM parameters (duty cycle and frequency)
3. Based on the results from the open loop testing, design a PW modulator to modulate the PD/PID controller.
4. Make selections for all components.
- Power source, Temperature Sensor (most likely: DS18B20), wires/connectors, SSR, Displacement sensor
5. Wire all components together.
6. Perform testing on the finished system.
7. If necessary, Develop a continuous PD or PID control for the SMA actuator using Simulink.
8. Perform closed-loop testing on the new finished system with displacement sensors to tune PD/PID gains and check if anything else has to be changed
9. Implement in vent.
10. Install in housing in vent case.
This should be completed by the end of March when the assembly of the entire device will happen. However, the team aims to have the control system close to finished before Spring Break. One thing to note is that Dr. Song has informed the team that a feedback control might not be necessary as the goal is an on/off type control where the vent is either open or closed. The team will start with developing an open loop control using PWM, test it, and evaluate if feedback needs to be implemented. During the next two weeks, Team Navy aims to finalize the first four steps. The components need to be ordered before the end of February but the team aims to get this done earlier. Dr. Song’s lab will be used to conduct the experiments.
During the last meeting with the machinist, the team was informed that the components would not be done until the middle or end of March. This becomes a challenge since the validation process should be started towards the end of march, so during our meeting with our capstone advisor Dr. Argwal, he suggested we look into 3D printing in order to be able to do preliminary testing and validation in case the machinist wasn’t able to get our vent parts done by the specified deadline. Following our advisor’s suggestion, we decided to set up a meeting with the 3D printing lab technicians of the university in order to discuss the feasibility of our device. For the most part the team was informed that the device can easily be printed with the exception of the threads. Initially we had 1-80 threads per inch, which resulted in a really small scale which the 3D printer could handle. In order to fix this the thread size was changed to ¼”-20 (imperial) in order to make it printable, the change is displayed in Figure 3. This modification can also be helpful in the manufacturing process since the new threads are easier to machine than the older ones. The team plans to consult the machinist about the change in order to ensure that it does not affect the process or add expenditure to the cost from having to buy specific tools for machining.
Figure 3: Thread size changed to fit 3D Printing constraints.
Anticipated Challenges Ahead:
One of the main obstacles the team is facing is the overall limited expertise in configuring the control system. However, the team has done extensive research and Simulink training and has now developed a workflow for the control system. The team is still working on designing the open-loop testing to analyze the fundamental characteristics of the SMA wire actuator and explore essential PWM parameters like duty cycle and frequency. To ensure correct testing, the team has reached out to Tico Hannah, who is a student in Dr. Song’s lab and has expertise in control systems. Team Navy has allowed a lot of time for the control system development in their schedule, which means that there will be room to make changes if things don’t go according to plan the first time.
Another challenge lies in determining how to attach/house the control system to the battery case. Given the prototype nature of the project, the team plans to affix the controller outside of the battery case, but it is still being determined exactly how this will work. In real-life applications, the control system will interface with a vehicle’s central processing unit, and together with this the control system would be housed.
Last but not least, attaching the SMA wire beneath the vent has been slightly overlooked, as the 3D design depicts the wire simply touching the vent’s bottom without practical implementation considerations. To overcome this obstacle, the team’s plan involves exploring solutions such as utilizing advanced adhesive materials capable of withstanding elevated temperatures as well as resisting the pulling force exerted by the wire. Another option could be to use small screws or hooks to attach the wire.
Comentarios