RIT will permanently shut down the RIT Wiki (Confluence wiki) on February 2, 2024. Review the Wiki Shut Down Overview for information on alternative tools, exporting content from the wiki, and other FAQs.
Team Vision for Final Demo and Handoff
During this phase, our team planned to tie up all the loose ends on our project as well as put the finishing touches on. We also wanted to leave detailed documentation for the next ConcRIT team so that they can, as best as possible, pick up from where we are leaving off. We also had a few scheduled demos for the local news as well as for our customer. Between all these demos we printed a lot of concrete, so collecting data and information during all these printing sessions was another goal of ours. Over the course of this phase we got a lot of information on how a hand-fed operation of our printer would actually work, as well as the shortcomings and strengths of our project. We did actually accomplish what we planned to do this phase, with respect to the testing and final pieces to the project. We got through all the test plans we made at the beginning of MSD II. The Duet was mounted onto the side of the printer and all the wires were routed through the channels of the 80/20 rails. The hopper was made and tested as well, which was a whole design process that included welding done by the Machine Shop.
Test Results Summary
During this phase all system tests were conducted, these tests resulted in a large number of conclusions being drawn and a most of our engineering requirements being met. The conclusions of each test are summarized below and the live Test Results document can be found here.
Table 1: Summary table of Test Conclusions
| Test Plan ID: | Test Name: | Description: |
| S1 | Inter-Print Repeatability Test | Testing the repeatability between multiple separate prints. Use standardized samples. 10 separate prints, 10 cubes |
| Result Conclusions | The inter-print repeatability of the project is not exact down to the millimeter but it is accurate enough to run the same GCode in two different prints and get two objects that are extremely similar. For set time, we found that if you try and print more than 5 layers the bottom layer will experience crushing due to the weight of the additional layers. This will cause the print to flatten out and not retain proper dimensions. Therefore, it is advisable to print a large structure in parts, where the bottom 1-4 layers have time to set before additional layers are printed on top of it. Those layers should set sufficiently after 24 hours. This can be reduced but we have not collected enough data due to the time-frame being so large—it makes it hard to check in on the print at midnight. | |
| S2 | Teardown test | Testing the disassembly procedure of the printer. Needs to be timed, number of tools needed should be tracked, and size should be measured. |
| Result Conclusions | Through this test plan the team learned a few things. Previously the tear down size was going to be much smaller than 5x6x8 ft. It was learned this would be extremely demanding for users to fully disassemble. This would be too much to ask of students who only may use the printer for a few days. Instead, with customer approval, we decided that as long as it can roll through the double doors in Gleason it fits its transport/storage requirement. | |
| S3 | Cleaning Time Test | Time how long it takes to full clean all concrete from the printer. |
| Result Conclusions | Overall these each were successful with some contingencies. First, ER3 was passed since it is possible to exterchange nozzles in about a minutes time. The duet board allows the auger to stop allowing the flow to stop making this possible. However the printer itself does not consistently work with the printer. It frequently causes jams and is better off with no nozzle. Second, ER20 is passed because two allen keys are used to take eoff the extruder from its mount. More would be need to fully deconstruct the extruder but only two are needed to get the parts to a state for adequate cleaning. Lastly, ER21 was passed since through various trials it always under a 30 minute mark. We tried this with a compination of working fast and slow and full time or part team. Each combination was under the required time. The hopper was marked as it takes the longest to clean due to its size and need to be pat tried. Something else to note, these times are with the expectation you are throwing away the corrugated tube. The price is not worth the time to clean or the risk of it not being cleaned properly. | |
| S4 | Extruder Bead and Flow Test | Subsystem test of the extruder. Test the bead production and flow rate with various standard nozzles. |
| Result Conclusions | Therefore in conclusion we have failed ER13 and ER27 both because our extruder cannot extrude mortar with nozzles attached. We pass ER17, ER28, and ER30 for the reasons listed above. | |
| S5 | Intra-Print Repeatability Test | Testing the repeatability of geometries within one print. 1 print, 10 cubes. |
| Result Conclusions | As the extruder was hand fed the amount of mix in the extruder varied and caused varying back pressure. This variation in back pressure made the bead width vary as well. This variation caused the first line to be disregarded as it was visibly failing. The Second line was a full pass and was within .5" of the expected length of 27.56" (700mm). The third line was short because we ran out of mix which can also be noted in the decreased bead widths. For the vertical width there was slight variation with layer heights. This is partly due to the hand feeding of the extruder, causing "floods" and "droughts" in mix flow. I believe with the auger constantly being fed concrete via hopper/tube system this variation could be tuned out. We do pass the variation for the test. We would start to fail it at over 6 layers with the current setup. | |
| S6 | Strength Test of Print | Test the compressive strength of a printed concrete sample |
| Result Conclusions | All our tests showed a compressive strength of at 909 psi. This was the maximum pressure that was possible to measure due to methodology used for this test and the limitation of measurable weight due to the scale that was used. To get a better measurement of maximum force a more rigid push rod and scale capable of measuring up to a higher weight could be used. These results are also highly dependent on the exact type of mortar mix used and level of hydration the mortar is mixed to for this sample we used Secrete type S mortar mix at a 18% hydration level. We expect that the compressive strength should always meet the minimum value, but results have risk of varying if a different mortar mix is used or a different hydration level is used. For a most accurate method of measuring compressive strength with the ASTM C873/C873M-15 Standard Test Method for Compressive Strength of Concrete Cylinders Cast in Place in Cylindrical Molds test method. This testing based off our preliminary results should still pass our minimum compressive strength requirement. For the tensile strength of this sample a test method was unable to be determined and manufacturer does not prove and expected values for the materials that were tested with. | |
| S7 | Student Usability Testing | Walk through the entire printing process and determine areas of improvement in the procedures etc. |
| Result Conclusions | The slicing software (Cura) needs more work and testing that we are not able to conduct due to time constraints. The GCode used in any test or print must be manually created. While this is a massive setback in terms of student usability, once GCode is generated the system is highly usable. The generation of GCode is the crux of the issue and so for that reason, we conclude that the system does not meet our expectation for student usability. | |
| S8 | Printer Measurements | Get the final dimensions of the printer |
| Result Conclusions | The printer passes all tests that the current team had a hand in designing. It only failed the x and y axis envelope due to a probable oversight by the previous team. As the current team is lacking funds and time to rectify this problem it is considered a failure, however it should be easily amendable as there is approximately 2" extra in each dimension. This should be verified by the purchasers if they want to fix this problem to make sure the axis still fit after replacing them. | |
| S9 | Hopper Capacity Test | Test the capacity of the reservoir |
| Result Conclusions | This test revealed a significant amount of information about the hopper that had been previously unknown such as how the mass of the tube would interact with the motion of the printer, how successfully the hopper could supply material to the extruder, how low the hopper could be lowered to fill with mortar, and how accessible the hopper would be during a print. It was found that when the hopper was attached to the extruder the mass of the tube caused significant amounts of binding in the y-axis of the motion of the extruder, thus making it impossible to conduct any full system operation tests that included motion. To resolve this, it is recommended to find a to provide the tube support such that the motion axis does not bear all the stresses that are a part of the tube, another alternative would be modifying the motion system to better accommodate the stresses applied to it by increasing the length of the bearings in the ball screw and linear rails this could help counteract the torques that are applied by the tube. This test also found that the hopper could feed the extruder with consistent amounts of mortar, but it needed 'coaxing' down the tube and out the funnel of the hopper by inducing vibrations. Finding a method to vibrate the tube and hopper would considerably increase the reliability of the flow of mortar to the extruder. Additionally, it was found that the tube was both too short to be hung off the side so that the hopper could be easily accessible during the print but too long to have the hopper hanging in the middle of the printer. It is our opinion that the safety benefits that having the hopper hung in the middle of the printer outweigh the need to have the hopper be easily accessible during a print but with this the tube length needs to be optimized to prevent the kinking seen in image 1. It was found that with the tubing used in testing that the hopper if lowered with the printer is out of the way is very easy and comfortable to fill. Additionally, the test showed that when assembled with hose clamps the hopper to tube connection or the tube to extruder connection leaked mortar. When filled with a full 80lb bag of mortar that had been mixed to an 18 percent water to mortar mix ratio there was still plenty of room in the hopper to accommodate more mortar. However, it is our recommendation that this fill level is not exceeded by more than an additional 10lbs of mixed mortar to ensure that there is enough room at the top of the hopper to prevent any accidental spillages during lifting. Overall, the hopper is in a very good place but needs optimization with tube length, supporting the mass of the tube, and introducing a vibrating element to induce the flow of mortar. | |
| S10 | Complex Geometry Test | Print portion of arborloo |
| Result Conclusions | Overall the ERs pass but there is work to be done. The team extruded the arborloo. The team attempted to smooth the edges like frosting a cake. This was with the assumption future students may attempt this if they did not want the harsh seams the printer creates. The picture above was the first arborloo print at the time of this report which explains its rough state. Overall the mold version was slightly better visually but recall the molding took several attempts due to cracking and demolding breakage. The internal diameter was a mistake in the coding by user error not a print error and was not included in this test. | |
| S11 | Duet & Extruder Integration | Control the extruder DC motor from the Duet |
| Result Conclusions | ER5 The minimum achievable mass flow rate was 2.15 grams per second. This is 0.0047 lbs per second. This surpasses our ER because we can achieve a mass flow rate of less than the identified value. | |
These conclusions resulted in the following table that summarizes our engineering requirements and whether or not our test results met desire values.
Final Risk Management
The final Risk Management file is linked here.
Final Project Documentation
- P21652 Technical Paper
- P21652 Final Poster
- Links to final design documentation:
- CAD/Schematics
Functional Demo Materials
Demonstration was performed in-person on Friday April 7, 2021.
Plans for Wrap-up
As a team, all that is needed to wrap up this project is complete documentation and the customer handoff. The customer handoff also includes what to do with all the materials we have around, which will be an undertaking of reasonable effort. We have several bags of concrete, tons of nuts and bolts around, as well as wires and connectors that we have and would like to give to the next team working on this project. Currently most of our materials are stored in the Active Learning Lab so these might have to be moved back up to the MSD Cubicle upstairs that is assigned to us. Once all the documentation is complete, our work on this project is finished!