Build & Test Prep
Table of Contents
Team Vision for Build & Test Prep Phase
The team has surpassed the goals set for MSD I, and is well into the construction of our initial prototype. In addition to completing the first set of test prints for the designed experiment, further analysis was done to establish an early set of targets for the two temperature factors. An enclosure has been designed and constructed for the printer, and testing of its effectiveness is underway. A pair of cameras have been installed and integrated with the Raspberry Pi, and will be mounted to view the print area within the next few weeks. An early sensor array prototype has been constructed, and a significant amount of progress has been made with our new printing software.
Test Plan Summary
Our test plan for our engineering requirements that need them. Each test plan includes a description of the requirement that needs testing and a flow chart describing how the test will be performed.
Initial DOE Results
In order to determine the initial print parameters for a quality Benchy print, a two-factor, three-level experiment was run with three replications, for a total of nine combinations and twenty-seven individual prints. The two factors selected were bed temperature and extruder temperature, with levels set at 10 degree (C) intervals beginning at 60 and 200 degrees, respectively. Run order was randomized for this experiment.
After the first three prints at the low bed temperature setting all failed (one print at each extruder temperature level), the remaining six prints at the low bed temperature setting were removed from the experiment, as it had been determined that the print would not succeed at that temperature.
The above results show clearly that the high bed temperature setting of 80 degrees results in the best quality in completed prints. While the overall quality does not improve significantly from 70 to 80, bed adhesion and warping improve greatly, while only stringing notably suffers, and not by a significant margin. The effects of the extruder temperature factor are less clear, with no significant change in quality. Four linear regression models were created, and the results shown below:
Each of the four model summaries show a clear relationship between bed temperature and the respective quality score, but none of the four shows a distinct relationship between quality and the temperature of the extruder.
Data Collection FeasibilityWith the recent addition of Octoprint, we considered moving several printer features that were initially planned to be implemented in the printer’s firmware to the Octoprint engine. This was being considered for the ease of access to Octoprint (it’s a web app, available from almost anywhere) and the large amount of native data available within Octoprint. Octoprint provides a real-time temperature graph for a given print for the native elements within the printer (extruders, print bed).
The first issue that was discovered within Octoprint was the lack of concert data output. While Octoprint does provide a graph of real-time data, the actual data values are not outputted anywhere and are quickly discarded after a few more polls. The initial solution to the problem was to develop a plugin or modification to Octoprint that would log the output data to a csv file. While this would have been an acceptable solution to the presented problem, two scope issues where identified while attempting to find an optimal way to implement the solution. The first scope issue was since the actual code of interest was written in an unfamiliar coding language, this would make the actual implementation time significantly longer. This issue could have been worked around if not for the other issue.
The other issue proved to be much more critical. Currently Octoprint has no ability to integrate external sensor within the printer due to how Octoprint collects the data it is interested in. While this issue could possibly be solved by improving and extending Octoprint with a new implementation with these capabilities, this is very much out of the scope of our project. Since Octoprint has no ability to provide us with a superior method of logging that we need for our project, going down this avenue for csv logging was abandoned, however we will continue to use Octoprint for the remote access it allows us to the printer and the ability to change parameters on the fly.
Since Octoprint is no longer an available logging option, all ideas of extending and improving it have been abandoned. The optimal solution to our logging issue still appears to be implementing a method of logging within the printer firmware. Since this is where Octoprint receives data from anyway, we should see no difference between our Octoprint data and our printer’s log data. However, the data log that is outputted will only be available off the printer’s native SD card. In order to provide remote access to data, we plan to transfer the logs from the printer’s SD card to the raspberry pi’s SD card. This should allow us to access the file via Octoprint, allowing us to remotely access the data from anywhere Octoprint is available.
Upkeep and Future Updates
- The link to the full document can be found here.
Risk and Problem Tracking
- The full risk management document can be found here.
- The full problem tracking document can be found here.
Bill Of Materials
Plans for next phase
Week 5 Shared VisionAt the next phase, the team will present our next set of data analysis, including plans for further testing, along with an initial prototype with integrated cameras (including thermal history), working sensor array, and new test results to confirm improvements made by the enclosure.
- The parts for the LED Matrix will have been ordered and the building process will have been started. Progress will be shown at the Week 5 review. It will not be complete until the Week 8 review.
- The HD camera will be mounted inside the enclosure and it will be used to monitor the printer/artifact. The camera was recently mounted inside the printer and will be displayed at the Week 5 review. The optimal location for artifact observation will be finalized by the Week 8 review.
- The thermal camera will be mounted inside the enclosure by the Week 5 review. We will be able to record thermal video by the Week 8 review.
- Feasibility of thermal imaging through computer vision will be presented at the Week 5 review.
- The initial tensile testing data will be collected and analyzed by the Week 5 review.
- The initial impact testing data will be collected and analyzed by the Week 5 review.
- The second round of DOE results will be complete and
analyzed by the Week 5 review.
- These results will validate the use of the enclosure and include information on error variance as requested by the customer. We will be testing 200C, 210C, and 220C extruder temperatures as well as 70C and 80C bed temperatures. We will start collecting sensor data once they are mounted, but no conclusions will be made from this DOE test.
- Areas of interest for further experimentation will be determined and presented to the customer at the Week 5 review.
- The printer's firmware will be modified to allow CSV logging of internal and external data by the Week 5 review.
- The printer will collect valuable data from the internal sensors and will be capable of writing this data to a CSV log by the Week 5 review.
- The printer will collect valuable data from the external sensors and will be capable of writing this data to a CSV log by the Week 8 review. Progress will be shown at the Week 5 review.
- External sensors will be mounted with preliminary data findings recorded to the external SD card by the Week 5 review.
- Sensor network progress will be shown at the Week 5 review. Software functionality updates will be continuously worked on and presented at the Week 8 review.
- The thermocouples will be reliably held in place in their respective amplifiers by the Week 5 review.
- Feasibility of PCB design for sensor network will be presented at the Week 5 review.
Three Week Plans
- Mike's three week plan.
- Ryan's three week plan.
- Maseo's three week plan.
- Justin's three week plan.
- Jacob's three week plan.
- Jon's three week plan.