We are a group of junior year college students attending Worcester Polytechnic Institute. In an effort to get us involved with the current COVID-19 situation, we were tasked with either improving a current design of a face shield or creating our own.
Currently, the healthcare system is bottlenecked by a lack of PPE availability, making manufacturing speed a high priority. Each individual face shield needed to be printed quickly as well as be adaptable to printing more than one at once. Additionally, the height of the face shield was a consideration as it directly correlates to prolonged print time. A brow guard also directly correlates to print time, but when speaking to medical professionals, they expressed a strong preference to have it. This was due to safety concerns as it provides more protection for the wearer. We also had to consider the weight of the face shield; if it was too heavy there was a possibility for contact sores and more fatigue for the wearer over time. Sizing was also an important consideration as the product should fit many different sized heads. An additional requirement that we needed to consider was the minimization of assembly. As these face shields are to be given to medical personnel whose main concern, besides safety, is to get the mask on with minimal fuss, we needed to come up with a design where the ease of use was the main selling point. An adaptation of the materials at hand was important for the design as well. Acetate is a necessary assembly material as it is the primary element of the shielding. However, we wanted to make something where the elastic was not needed, as elastic is a limited material. Instead, our design uses tape in the back to connect the arms. Some other designs use a strap that goes up and over the head to reduce fatigue on the wearer, however, our design is light enough that this is not necessary.
We were unable to find conclusive information on the sterilizatibility of FDM parts. Our design, being so rapidly printable and low on material usage, can be easily justified as one time use consumable, especially when the time required to sterilize non-disposable equipment is taken into account.
The face shield design we have is a single, 3D printable component upon which an acetate, polycarbonate or similar transparent sheet can be attached. This version has spacing for the standard 4.5” spaced, 6/16” 3 hole punch. The part is printed with no supports and is compact enough such that most printers can manufacture two with the brow guard at once. Print time is exceptionally low, with a 0.32mm layer height and a print speed of 100mm/s producing the part in roughly 13 minutes. While the print speed is high, most geometry follows a curved path, so the motors of most common printers will not be stressed too much. Additionally, the little area is required to be filled in, meaning backflow issues commonly preventing Bowden tube-based extruders from printing quickly is eliminated. Material usage is very low, which also means that the mask is very light, reducing contact fatigue from prolonged use. The mask is worn by taping the two rolled up sections of the band together. We decided tape was readily enough available both at home and in hospitals that this would be more preferable than sacrificing additional print speed to add a fastening method. We tested the band with scotch tape, duct tape, electrical tape, and masking tape, and found that all provided more than enough stiction to secure the mask, even with vigorous shaking of the head. The face shield is secured on the wearer by the spring created by the attached band. Even PLA, which is very stiff, provides more than enough flexibility with the thickness of the band to not break. Assembly is quick, and the protective shield can easily be swapped out if needed. The headband has been sized to securely fit on most adult head sizes. If a tighter fit is needed, the two band halves can simply be taped in a tighter position. See the attached image sidebyside.PNG for a top view of the initial folded state, and with the band linked together.
Our proposed face shield is designed to optimize manufacturing time on an FDM 3D printer while still providing the required safety. During our design process, we developed a design strategy to create parts that are strong and functional, yet print on the order of minutes instead of hours. As we have not yet done extensive testing, we recognize our design is very likely unfinished. However, we think our approach offers many benefits over some of the other designs currently in use.
Design Strategy Explanation:
Standard FDM printers are capable of printing decently fast, however thick walls (shells) and filled in areas are the enemy of speed. Additionally, time spent by the extruder not laying down material should be kept to a minimum or completely avoided (ie. no retractions, travel moves, etc). Most slicers have the capability to optimize many of these to a minimum already, however it is not always perfect. Instead, we have designed the part so it can be printed in one continuous path. The walls of the model geometry are dimensioned to be slightly smaller than twice the extrusion width. In our case, with a 0.4mm nozzle, walls are kept at 0.6mm thick. When a set of walls intersect to form a closed area, a small gap should be left. We found that 0.05mm works well, and gives very strong wall-intersection adhesion. You can see these gaps highlighted in the attached image combinedimages.PNG. Also shown is a preview of the slicing path.
Some thinking is needed on the designer’s part to visualize how the extruder will travel throughout the path. We found observing when SOLIDWORKS no longer identifies enclosed areas in a sketch to be a fairly reliable way to check against intuition. Lastly, filled in areas are not always entirely unavoidable, such as with the brow guard on our design. However, we found that keeping the thickness of these areas to exactly twice the layer height to be ideal. A single layer sometimes can have small gaps, which can be problematic in cases such as for the brow guard that must protect against liquid ingress.
We have provided a printing profile for Ultimaker Cura that has the settings that worked well for us. We used an ender 3, which has a 220x200mm build area, though only ~210x100mm is required for a single part. We printed this in PLA. Settings will need to be adjusted depending on material and printer of course. Note that due to the higher printing speeds, we found that increasing the extruder temperature by 10-20c above what is normally used to produce the best results.
WHO :
Sales and Marketing: Julia Cuendet jmcuendet@wpi.edu
Manufacturing Engineer: Dana Landry dlandry@wpi.edu
Design Engineer: Connor Dietz ctd1513@gmail.com
Project Manager: Lena Sophia Thompson lsthompson@wpi.edu
STL File :
