At the time of prototyping, ventilator splitters had not been FDA-approved under emergency use authorization as they currently are for hospital use. Splitter designs must overcome the challenge of providing consistent pressure and airflow to two or more patients with different lung capacities and respiratory needs. So as they continued prototyping, the design continued to evolve around building effective airflow controls.
Dresher and his team attempted to overcome this issue by designing a ventilator splitter that could regulate the amount of air flowing to each patient. The splitter design would have detents and an indicator tab on each arm of the splitter to allow 30%, 50%, 70%, 90%, or 100% airflow. Dresher and O’Halloran repeatedly adjusted the valve geometry based on O’Halloran’s computational fluid dynamics (CFD) analysis of the valve flow characteristics. This design also aimed to prevent cross-contamination by having as few as 3 pieces in each assembly—one part to split the airway and one part on each tube arm’s end that could adjust airflow.
After 9 prototypes, the latter 5 of which successfully fit into existing ventilator tubing with attachments for regulated airflow, it was time to do physical testing. Prototypes were sent to Scott and the respiratory team at Memorial Hospital for testing with ventilators. Ultimately testing revealed the splitter was able to modify airflow but not enough to match the initial CFD modeling and targeted flow percentages. This meant more development would be necessary. At the same time, the US’s circumstances changed in April as hospital ventilator capacity improved with the completed production of more ventilator units.
After comprehensive prototyping, testing, and the change in the market, the team decided to close the project and focus their efforts elsewhere. Xometry’s rapid prototyping service empowered a cross-functional team of engineers, medical professionals, and manufacturing experts to test a hypothesis, build trial designs, and experiment without wasting any time. They successfully used the “fail fast” method to make their product development agile by prototyping and getting engineering feedback early on. This allowed them to make the decision to move on to other projects before sinking additional weeks or months of time into the project, or prematurely investing in an injection mold tool.
What to Do When Your Prototype Design Fails
Failure in prototyping is crucial, as it is a natural part of realizing a successful version of a hardware component. Sometimes it can lead to product modifications or improvements, and other times, like in this instance of prototyping a ventilator valve splitter, it can lead to the hard decision to move on to different projects. In the time when Dresher’s team was prototyping this device, there were still many unanswered questions for what was needed for a ventilator valve splitter for emergency use medical care. In the end, the Dresher and his team’s project was not a success, but was another important data point in the life-saving efforts of millions of engineers and scientists to prevent COVID-19 deaths.
With the ongoing development of 3D printing technologies, rapid prototyping is a viable solution for fast innovation across all industries. So when part designs do not produce the intended result, they can be used as feedback to make better decisions on the product’s path. Rapid prototyping is both for validation and learning. At worst, failures can be cherished as lessons learned in the prototyping journey. At best, they may save millions of lives.
