Here is a close look at the rapid prototyping system selection process used by one group of engineers at Novartis Pharma Technical Research and Development.
Several methods exist that will help you match the right rapid prototyping system with your needs. Because it is useful to hear of others’ experience, we at Novartis Pharma Technical Research and Development will describe a process used for a recent purchase. It is not the intent to recommend one technology vendor over another, but instead to map the strengths of different technologies and the tradeoffs that need to be understood to evaluate them properly. The same techniques, applied toward different needs, will likely result in different choices.
Our device development team specializes in the development of dry powder inhalers. These devices have a unique set of needs:
— A high degree of dimensional and surface-property accuracy to ensure that dry powder medications are aerosolized repeatably and reliably,
— Robust to operate even when the powder coats the internal mechanisms,
— Rugged to survive a range of environments.
In past development projects, we have typically relied on a few select service bureaus for SLA parts built on 3D Systems Viper HD machine. These parts gave us a reasonable look as to how final molded parts would perform, but with a 0.002 in. maximum step height, the equipment was not able to faithfully reproduce some of the more delicate device features.
We were asked to develop a complicated new device on an aggressive timeline, which required a different approach to rapid prototyping. Instead of the two to three day turn around we had worked with in the past, getting into the queue and waiting for FedEx, we would need parts overnight or even in hours in order to cycle through the iterations required to deliver the needed features.
Our process was to survey the market, have vendors produce a challenge part, and then construct a decision matrix with weighted factors to guide the final selection. For the survey, we started first by talking to account managers at 3D printing companies and then to end users; some references were provided by the companies themselves and others solicited independently. Each of these vendors was also asked to fabricate a small part with fine detailed features that we felt would give us a good means of discerning the relative accuracy of each process. Each part was photographed at approximately 50X, focusing on the same three critical feature areas to provide a common point of comparison.
Through the survey, we developed a list of parameters that would ultimately drive our decision and constructed a decision matrix. Though the ideal situation would have been to have the matrix fully laid out ahead of time and to simply populate it with information that we gathered, the reality is that it was an iterative process and some important factors emerged relatively late.
The challenge parts showed greater variety than we expected, and ultimately reduced the number of technology platforms considered in the final round. Challenge parts were made from the following systems:
–Dimension Elite
–Fortus XXXX
–ProJet HD3000
–Objet Alaris30
–Objet Eden260V
–Envisiontec Perfactory Mini Multi Lens
Of these six, the last four were considered in the final evaluation round.
When we gathered all of the data to make our decision, we met as a team to finalize weighting the parameters and then rate each of the technologies against them. The matrix was constructed in Excel, and the final score was hidden until the rating process was complete.
Using a matrix to assess our needs against the information in each system, we ultimately selected the Envisiontec Perfactory Mini Multi Lens, largely because of the demonstrated fine feature control that we saw from the challenge parts.
Given the importance of the decision to commit to one technology over others, we felt that the rigorous process we went through helped to clarify our thinking. We also learned a great deal about the relative strengths of all of these systems, and expect that other groups might come to differentconclusions using the same process, if only because the relative ranking of parameters for any given application will be vary from team to team.
Novartis Pharma Technical Research and Development
www.novartis.com
MPF
