The main focus of this blog post will pertain to the discussion of optimizations made for our designs, and go into detail on some problems we have stumbled upon, and how we have managed to overcome them. Before I begin discussing our optimizations I will first touch on the some noteworthy advances made in the project. Before the last post on November 23, only the beer tap had printed correctly, and both the pin wheel and beer-flipper had failed during printing. In the time since the last print, all three models have been successfully printed with multiple times with FDM, and the beer tap, and beer-flipper had been successfully printed with SLS.
Beer Tap
I will first discuss the optimization of the beer tap that Brad and Josh are working on. After printing the first beer tap in FDM, one problem that arose was water tightness. Due to the nature of this project, water tightness is an extremely important design constraint that must be meet. In an attempt to solve this problem acetone was brushed on the surface. Unfortunately, after testing this did not appear to solve the leaks that were in the tap. Another tap was printed in FDM; however, this time the wall thickness was increased from 1 mm to 2 mm. This seemed to do the trick, and the tap no longer leaked. Another problem that occurred during the printing of the first two taps, was that they required internal support structures. A third FDM print was done in an orientation that eliminated the need for internal support structures, which can be seen in the picture below (the tap on the right requires internal support structures, tap on the left does not).
Unfortunately, when printing in this orientation, the holes turned into elliptical shapes. This was not acceptable, seeing as there are critical dimensions in the assembly. So, the original orientation for FDM printing will be preferred.
Pin Wheel
Next I will talk about the pin wheel which Jon, Ryan, and Tom are working on. After a successful FDM print, the main problem which arose was how the pin wheel attached to the tap. The print had been made with an inner diameter of 0.75 inches (the tap itself has an outer diameter of 0.75 inches where they will connect), so the first design was to be a very tight fit. Unfortunately due to thermal expansion, the pin wheel did not properly fit onto the tap. To alleviate this problem, for the second print the inner diameter of the pinwheel was changed to 0.78 inches, which then successfully fit the tap. Also, in the second print the wall thickness was increased to 1.4 mm and the infill percentage was increased as well. Looking forward the pin wheel will soon be printed in SLS as one part.
Beer-Flipper/Dual Filler
Finally, I will discuss the beer-flipper/dual filler that Alex and Matt are working on. The beer-flipper/dual filler has gone some fairly large changes since the first FDM print. As was mentioned in the previous post, the first print took approximately 18 hours to print, and the printer actually ran out of material during the final few layers resulting in an unusable part. In order to try and reduce the print time as well as save material, the model was greatly reduced in size for the second FDM print. This included shortening the loop, decreasing the outer diameter, and shortening the distance between the two ends that split. Also, issues arose with accuracy of the mating surfaces, so the design was split into three components for the FDM prints, the connector where the it will attach to the tap, the loop, and the split (Seen in the picture below). This will then allow for simply the connector to be reprinted if issues arise in attaching the flipper to the tap, which will drastically reduce the amount of time needed for a reprint, and also save a lot of material. Both prints had been done with 1.5 mm walls, and there was no apparent leaking through the walls. There was leakage at the connection between the connector and loop; however, this can be solved through simply modifying the tolerance and reprinting the connector.
For the first SLS print, there threads were incorrectly made on the model, and this resulted in issues with assembling the flipper to the nozzle. Since only the threads were incorrect, this still gave us the ability to see how the parts would fit together and if there was any significant warpage, which there was not. Another issue that arose during the first SLS print was that an error occurred in our STL file, which printed a layer in the area where the connector and the loop were mated. This was not good, because beer would not be able to flow. So a second SLS part is currently being printed, with fixed threads, and an updated STL file that does not have the error. Below is a picture of the SLS part and the first FDM part to get a comparison of the change of size.
