Integrated System Build & Test
Table of Contents
In Phase 3, Team Spring Breaker received the cylinder, machined more parts, continued with documentation, and finished the MATLAB script with GUI. We also had some setbacks with the piston and budget. The piston came in with a rating of 1800 psi and we needed one that was rated to 3000 psi. The budget that the MSD office had was different than the original amount we thought it was supposed to be. We then had to confirm it with our customer so we could continue ordering the rest of the materials. This phase content of our Edge page includes detailed information and outcomes of our journey through our Integrated System Build & Test Phase. It also includes links to important documentation completed during this design phase.
Team Vision for Integrated System Build & Test PhaseThe plans coming into this phase were to complete the following:
- Finalize calculations with the baseplate, key, and retaining block
- Purchase the final items needed
- Finish the MATLAB GUI
- Start looking into the sensor and microcontroller integration
- Obtain cylinder and confirm specifications and dimensions
- Machine the rest of the spring disc packs
- Select hydraulic fittings needed
- Finalize documentation for all machined parts
- Update machining plans to reflect what was done and how long it took
- Maintain an up to date schedule for when the cylinder comes in
- Finish outline of documentation for Meggitt
What was actually completed was:
- Microcontroller was ordered and received
- MATLAB GUI is almost finalized
- Baseplate and key calculations were made simpler once the cylinder arrived
- The spring disc packs are fully machined
- Documentation control was implemented for the machine drawings
MATLABThe purpose of the MATLAB GUI is to provide an easy way for Meggitt engineers to find the spring discs needed to add a new plane to the AABS. With the GUI, all the engineers have to do is input the desired pressure vs displacement curve they want, and the GUI will give them the best spring discs to mimic that curve. The GUI opens with the window shown below. This window contains instructions on how to run the program, several areas for inputs, a “calculate” button, and a console for printing messages. The user will first input the name of the spring disc catalog, wait for the console to say the catalog was parsed, and then will enter the values of the desired pressure vs displacement curve. The diameter of the piston can be changed as well, in case Meggitt changes to a different hydraulic piston in the future. Once all the inputs have been entered, the user can press the “calculate” button to start the calculations. The console will inform the user when the calculations are complete.
Once the calculations have finished, the user can select the “Graphs” tab at the top to view the different pressure vs displacement graphs. There are four options for displaying graphs: display all spring disc options, display the best disc options, display the second best spring disc options, and display the third best spring disc options. Each graph also shows the desired pressure vs displacement graph for comparison.
Next, the user can switch to the Data tab. This contains all the spring disc information for each curve. The information is also written to a spreadsheet so the user can look at it later. The table gives the names of the spring discs, their dimensions, the series/parallel combination, the height of the spring disc stack, the K constant they create, and the error compared to the desired K value (calculated from the desired pressure vs displacement curve). At the bottom of the table is a legend to explain what all of the values mean.
Finally, there is a “Something went wrong” tab that will contain information on common errors and how to fix them.
Microcontroller & SensorsThe AABS has two sensors: a pressure sensor and a linear position sensor. To read these values, we are going to use a Beaglebone Black microcontroller. The microcontroller will contain a program that will collect the pressure and displacement data at regular intervals and will store those values on a USB. The USB can then be plugged into a computer so the Meggitt engineers can access the data. We have received the microcontroller and have tested some of the initial programs it comes with. We have an outline for our program and will start coding and testing it in the following weeks. The pressure sensor requires a 5V source and the linear position sensor needs a 24V source. The outputs of these sensors are mV and V accordingly.
MachiningThis phase has seen the construction and completion of both Pack A and Pack B for the small and large jet programs. Base plate, stiffening parallels and block material has been sourced and will be completed at the beginning of the next phase. Due to a budgeting error, the Klein order will be received and machining completed the following week.
Most parts in this phase were comprised of nylon and low-carbon steel. As such the time required to manufacture these components was significantly longer than that required to machine the aluminum components of Pack A. Moving forward weldments will be performed on the base plate to secure the stiffening parallels. As shown below, the remaining items, standing as the main nylon housing, the base plate and block, and final integration.
Bill of Materials
The BOM shown in the figure above displays the final list of items that need to be purchased or manufactured. Items highlighted in green are delivered/completed, items highlighted in blue are on order, and items highlighted in red are under review pending calculations, quotes, or specifications. Items that are not highlighted are assemblies that need to be built, and will be marked green when completed and tested. At this point in time, all essential items have either been delivered, or are on order. Non-essential, but useful items include things such as a case for the microcontroller to mount on the cart assembly.
Risk and Problem Tracking
We added three risks to our risk tracker this phase. The first two have to do with the base plate we’re using to attach the piston to the cart. Depending on the forces involved during operation, the base plate could fail, which would damage the system and render it unusable. The team has been working on calculations to prove that this failure won’t happen, and we have shown those calculations to mechanical engineering professors and our client.
The third risk is outsourced orders not meeting our specifications. To speed up our machining time, the team has asked workers in the machine shop to manufacture some of our components. While this is a great help in terms of schedule, we have to make sure that we communicate the exact dimensions needed so that the parts come out the way we want them to.
For this phase we added a new category of risks – missed. These are risks that we didn’t consider, but came up and delayed our progress. The first one is Parker delivering an incorrect cylinder. We didn’t anticipate this being a risk, as Parker is an established company that should be able to manufacture parts correctly. There is not much we could have done to mitigate this risk, as we communicated our specs in writing and Parker didn’t follow them.
The second missed risk is MSD not allocating our budget correctly. This is due to miscommunications between MSD and Meggitt. The problem is that we were only given a budget of $4000 when our budget was supposed to be $10,000. When the team asked MSD about this earlier in the semester, we were told we could order what we needed regardless of the budget discrepancy. We reached a point where we went over that $4000 limit, however, and were unable to purchase more items. We are working to resolve this with MSD and Meggitt.
Finally, we updated several risks. The first one is CE compliance. All of our electrical components (pressure sensor, potentiometer, and microcontroller) meet CE compliance, so this is no longer a risk.
The second is the lead time for the piston. Since the piston has arrived, this is no longer a risk.
The third risk concerned the MATLAB GUI for our spring disc calculator. The team had no prior experience in coding GUIs, so it was unclear how difficult this task would be. The GUI is now functional and nearly complete (pending more verification), so this risk can be removed.
The final risk is not documenting any changes we make to the design during manufacturing. Evan and Mike implemented a revision control system to store our machining documents, and they make sure to update it after every change, so this is no longer a risk.
Design Review Materials
Plans for next phase
- Evan's P4 Plan
- Lily's P4 Plan
- Elias's P4 Plan
- Tony's P4 Plan
- Sabrina's P4 Plan
- Mike's P4 Plan
- Lori's P4 Plan
- Garrett's P4 Plan