Face Shield First Samples and Testing

Written by Francis Garing

We've made some significant progress on our alternative face shield design since our last post. The prototype tooling we produced in-house was used by the experts at ThermoPro to produce initial samples and pre-production parts. Through quick and dirty early prototyping, we evaluated our design approach and then further validated it using first-off tooled parts. Much was learned through the process, and we'd like to share that with you in this follow-up post.

The first parts we received to verify fit and function were trimmed by hand. However, to make face shields efficiently, automated trimming and finishing are necessary. Using the vacuum formed sheets as a mold, ThermoPro created a set of fixtures for their 5-axis trimming router. The router trimmed each face shield from its formed sheet by tracing a 3D path in space with a routing bit. While the trimming process is automated, it did take a significant amount of time and was one of the largest contributors to the expected production cost per unit. After discussion with ThermoPro’s manufacturing engineers, it was determined that redesigning the shield for 3-axis trimming would improve speed and reduce cost.

We proposed a simple design change that incorporated a slight flange along all the trimmed edges of the shield. This flange would allow the trimming tool to remain vertical, eliminating the need for a 5-axis motion thereby reducing trim time. An additional benefit is that this would provide increased stiffness across the part.

With samples in hand, we performed simple qualitative comparisons of our new face shield design against face shields currently being distributed to hospitals in the Atlanta area. Two tests were performed to evaluate two major functional advantages of the new design.

The first test measures how effective the face shield is against aerosolized droplets sprayed both directly and from below. The purpose of this test was to see if the shield could prolong the usefulness of a face mask worn behind it by preventing droplets from contaminating them. Dyed water was used to visualize the droplets. The spray test was carried-out from two angles to simulate a healthcare professional facing a patient directly or overhead as if they were in a hospital bed. The primary advantage of the new design is that the bottom edge of the shield has a return that provides coverage from below. This return is made possible since we are molding the face shield using a tooled process.

Our findings confirmed that our design performed as intended. While the two other face shields provided similar protection from a straight on spray, the lower angled spray was not impeded from contaminating the face mask behind the shields. No droplets were visible on the face mask beneath our shield, suggesting enhanced coverage and protection. This could prove to be a significant advantage for healthcare professionals, with many needing to re-use face masks due to ongoing PPE shortages.

The second test performed was a rigidity test of the face shield material. Anecdotally, we've heard accounts of issues related to distorted vision associated with other face shields. We suspected that apart from not being able to strictly control the curvature of the shield material (and therefore, distortion), these thin sheet-based designs would be susceptible to increased or variable distortion based on deformation from different events or activities. A sneezing patient, breeze when crossing corridors and thresholds, or even quickly turning one's head could affect visibility and distortion through the shield. To perform a qualitative assessment between the shields, we simulated the causes of this deformation with compressed air. A few quick bursts were aimed directly at the shield to evaluate the severity of deflection in the shield material.

Since our design uses a thicker sheet formed into a 3D structure, it was unphased by the compressed air bursts. Shape and structural integrity were maintained, and thus the level of distortion was minimized. Additionally, we found that several of the other shields attached their shield material to their head bands using just a few holes. While this simplifies construction, this did not fully control the curvature of the sheet material, causing varying degrees of distortion across the shield. This distortion is further exacerbated by any significant air movement against the shield.

With validation complete and early design changes identified, a larger run of sample parts was produced by ThermoPro to gather additional clinical feedback. In the coming weeks, both Formation and ThermoPro will be working to distribute these samples to those who they might benefit in order to gain more insight into what could help the design better address the needs of those on the front lines. If you'd like to learn more or find out how to get involved with our process and testing, reach out to us at collaborate@formationdesign.com

UPDATE: Face masks are available for purchase here.