Earlier I wrote about my concerns regarding community development of critical medical parts, and today there is an interesting development.
DIY 3D Printing Concerns
My concerns were about the chaotic nature of the many DIY initiatives, many of which seem to be unaware of the challenging requirements to produce proper and usable medical equipment. Some 3D print vendors have smartly focused on building things that are actually producible and usable, such as Prusa and Stratasys’ work on 3D printable face shields.
Some have attempted more complex projects, including building the valves used in ventilators. I was quite skeptical that common 3D printers could produce usable ventilator valves, but one vendor seems to have figured it out.
I received a package of information from UK-based Photocentric, which is one of the major suppliers of photopolymer resin worldwide and also market a line of capable — and rather large — resin 3D printers.
Photocentric Valve Production
They used a design for a ventilator valve that has been open sourced and published on GrabCAD by Filip Kober. The design includes several versions allowing for varying airflow levels.
On this GrabCAD page the comments indicate there are many people intending to 3D print them. In spite of their enthusiasm, I am concerned that they might not be doing the right thing. And my suspicions were confirmed after reading Photocentric’s analysis, which says:
“All of our hard polymers will pass cytotoxicity and skin sensitisation tests if processed correctly. Our polymers are chemically similar to those used in dental amalgams which are used internally in the body for life. If these have been reacted fully by the free radical process, they will pass Class 1. Medical approval would be part of our declared manufacturing procedure. Covance have agreed to fast track approval to Class 1 for our suggested grade for making respirator valves, RG35 (amber, translucent). We have added an antimicrobial agent to the formulation to make the product safer. Because of the circumstances Covance will complete the tests in 4 days.”
(Covance is a US-based certification agency.)
3D Printed Valve Challenges
What’s more interesting is the technical issues in producing these valves:
“The majority of SLA prototype resins in the world are made from oxetane based cationic resins (3DS, DSM etc) and will not pass those tests because they are highly hydroscopic, whereas our chemistry will. They also would take an impractically long time to manufacture by laser beam.
FDM does not provide high enough resolution (the internal air channel hole has a 1.0 mm internal diameter, making it difficult to manufacture), it cannot be autoclaved (as it is thermoplastic), it leaves a surface that is rough and corrugated and will harbour bacterial growth.”
I would add that the layered surface of FFF-produced 3D prints would add complications to the free airflow in the valve, which is quite complex as you can see in this diagram of the 3D model.
Photocentric Ventilator Valve Production
Photocentric set out to see what they could produce on their huge device, the Photocentric Liquid Crystal Magna 3D printer, which has a massive build area of 510 x 280 mm.
The result is a staggering 104 units per job, taking only eight hours.
At full production, one of these units could produce 1,560 per week, and we’re told that Photocentric has a number of them in their facilities.
Photocentric also has a larger device, the Liquid Crystal Titan, which can produce 1,710 units per week in its 700 x 392 mm build area. Finally, their most gigantic device, the Liquid Crystal Maximus, with a build area of 920 x 510 mm, can produce 2,933 valves per week. You can see an image of the build plate at top, filled with dozens of ventilator valves.
Photocentric says they have already produced more than 600 units overnight using three machines, and say they can produce more than 40,000 valves per week using all their available equipment.
Photocentric is currently reaching out to UK hospitals to see how they can deploy these valves.
This is how 3D printing will save us.