Hi, everyone, for this assignment, Carter and I have decided to choose option 3 and print a medical related object.
Specifically, we are trying to fabricate a set of clamps that is capable to grab onto a medical device that our lab is trying to implant into human body. Since this medical device is rod-shaped and has a smooth surface, it’s very easy for it slide down by accident. Therefore, it’s important for us to have a clamp to secure the device at the deaired height or location during surgery.
To better accommodate to the slippery surface of the device, we decided to have TPU as the material between clamps and the device, since it’s a softer material and supposedly should be able to attach to rigid PETG, which is the choice of material for our clamps themselves. Another design consideration is that since the surgical environment is already filled with many tools and medical apparatus, it would be better if we can keep all the tiny screws embedded inside the clamps so that we can ensure that they won’t cause the risk of falling into the patient.
The plan to print with embedded screws is basically to set a pause in the slicer software so that the printer will stop at the point where several pockets for screws are open and we can insert the screws in there. After that, we can continue the printing process manually and cover up portions of the opening where the screws were inserted so that the screws would be secured inside the slots. More specifically, the final diameter of the opening (where a screwdriver can be inserted to operate on the screws) is smaller than the diameter of the screw heads so that the screws won’t be able to fall out, but the surgeon can still have access to the screw heads. A simplified illustration in Figure 1 can better explain this design.
Figure 1.
As shown in Figure 2 and 3 are the CAD designs of the clamp set. Marked in yellow are the two pieces that would compose the clamp and marked in green is the portion we want to print with TPU or any other equivalent soft materials. Those three clamped rods represent the medical devices we are trying to secure. Next, I will talk about how we actually executed our design.
Figure 2.
Figure 3.
As stated above, the original idea was to print TPU and PETG together. However, according to suggestions from technician in the printing section, it seems that the printer we chose (Prusa XL) has an inconsistent record of printing TPU. Therefore, we decided to print two extra batches on Prusa MK3 with one material (TPU) and Bambu with one material (ABS) in two different colors (red and white). In this way, we can ensure success of some results and demonstrate the concepts of embedding the screws and printing from two extruders.
As shown in Video 1, here is the printing with Prusa MK3, where only one material is used. You may wonder why the print looks much thicker than the design in CAD. The thing is that actually we had to raise the half clamp with screws up so that we can later insert the screws (much longer than the thickness of that half clamp). One interesting thing is that the material (TPU) is very soft and normal tree-branch-shaped supports can easily induce much vibrations during printing and cause misalignment of different layers, causing delamination and failure of prints. This is something we learned from printing in our lab printers. Therefore, we had to make our own supports manually and the big chunk of wall-shaped structures are those more rigid and reliable supports. One may question: how could you take those supports off easily if they are so thick and rigid? Well, we actually added a thin cleavage design between the half clamp and its supports so that we can easily break the supports off along the cleavage after the print is finished. Marked by red arrows in Figure 4 is where the cleavage is located.
Video 1.
Figure 4.
As shown in Video 2, this is how the printing looks like after the screws are put in. More specifically, this is the process I mentioned earlier, where portions of the screw heads are cover by prints so that they can be secured inside.
Video 2.
As shown in Figure 5 is then the final result with printing on Prusa MK3, where the upper two are the two half clamps with nuts and the lower two are the two half clamps with screws inside (alleviated because of the thicker supports to house the screws).
Figure 5.
For the duplicate print with two different colors (two extruders), we chose Bambu. As shown in Figure 6 is how the end result look like. Very interestingly, there’s a big chunk of cuboid sitting at the upper portion. This is actually the location where the printer would dump/finish all the rest of material in the first extruder when it’s trying to switch to the second extruder. In this way, the printing with the next “material” in the second extruder would actually be cleaner (without any residuals from the previous extruder).
Figure 6.
Next, we took these prints out and broke all the tree-branch-shaped supports very easily (Figure 7). The reason why we can use these automatically generated supprts this time is because we chose ABS for this printer and it is known for being more rigid/brittle.
Figure 7.
As shown in Figure 8 is how the nuts (heat set inserts) are assembled in the other half clamps so that this whole clamp set is finished.
Figure 8.
As shown in Figure 9 is how these two identical prints are assembled onto the rod with similar diameter as the medical devices. The print finish was quite sturdy and the surface was not too rough after sanding. One thing to note here is that we also tried to use acetone to have a better finish on the ABS one, but the result was not much better. As for TPU, thermal smoothing was utilized to remove any small cracks on the surface (very essential for sterilization of components used in OR). However, though the surface does seem smoother, the warping of material was also very obvious.
Figure 9.
Next, let’s talk about the more experimental trial with the combination of TPU and PETG so that we can have a tighter grasp on the rod. It seems that just as the technician suggested, Prusa XL does seem to have a inconsistent result with printing TPU. Though the PETG seems to be printed very sturdy and clear, all the sections with TPU seems to have too many residuals (Figure 10). My speculation of the cause contains two parts. Firstly, it’s very possible that the temperature we choose for TPU might not be optimal when working with another material (PETG in this case). When I’m talking about the optimal temperature, it’s not just the temperature suggested by the manufacturer, but the working temperature when two materials with different Young’s modulus are blended together. More specifically, I believe what happened might be that the newly extruded TPU layers on top of the PETG layers might be shrinking (caused by cooling down once left the extruder) at a different rate than that of the cooled PETG layers below. Such mechanical mismatch might be the cause of this delamination between two different materials. Another possible cause of this failure might be the problem with this printer, since when we tried to load the fabrication file, it seems to only recognize the extruder size for PETG (0.6 mm, which is too large and might be another possible cause of failure, since our designs have much finer features) and not the extruder size for TPU.
Figure 10.
Cost analysis table is shown here:
| Cost Type | Cost | Price | Source | Quantity | Total |
| Materials | ABS Filament (all iterations) | $15.99 / kg | Amazon | 3.8 g | $0.0608 |
| TPU Filament (all iterations) | $12.67 / kg | AliExpress | 27.6 g | $0.3497 | |
| M2.5 Screws | $0.065 / unit | McMaster-Carr | 8 units | $0.52 | |
| Heat Set Inserts | $0.33 / unit | McMaster-Carr | 8 units | $2.64 | |
| Labor | Prototyping Engineer (X2) | $25.91 / hour | indeed | 10 hr | $259.1 |
| 3D Printing Technician | $31.62 / hour | salary | 0.5 hr | $15.81 | |
| Overhead | Utility Cost (considering both the cost of facility and electricity) | ~$5.13/ hour | TexAgs | 8 hr | $41.04 |
| Depreciation of Assets (3D printers and tools versus their typical life spans) | ~$0.76/hour | ATO | 12 hr | $9.12 | |
| Design | Engineering and Development | $44.34 / hour | ZipRecruiter | 6 hr | $266.04 |
| Iterations (a continuation of the previous category) | $44.34 / hour | ZipRecruiter | 2 hr | $88.68 | |
| Misc. | Waste and Scrap | ~$49.5/L | UltiMaker | 0.085L | $4.2075 |
Total=$687.568
According to the cost analysis, it seems that the most costly part is actually engineering and development. I would say this is within my expectation, as normally the numerous iterations of prototyping is why outsourcing fabrication of custom components is so expensive. However, one thing I want to note is that with a technology like 3D printer, the cost of materials is even more negligible and most importantly the efficiency in prototyping is much improved and thus the cost for labor would still be much cheaper than it used to be.
We also cleaned after prototyping:
