CATI recently had the opportunity to work with Gillette Children's Specialty Healthcare in St. Paul, MN to develop a unique solution for a traditionally expensive and time-consuming procedure.
Infants are often born with, or develop a condition known as plagiocephaly, or flat head syndrome, where the head becomes misshapen due to regular pressure against a surface such as a crib or stroller during skull growth. To correct this, a specially fitted helmet, or "CranioCap" is custom made for each infant to reshape their skulls.
The previous process for creating these caps involved outsourcing to have a high density foam milled to the shape of the infant's head, then vacuum forming a sheet of plastic over the foam to create the cap. The milling process was expensive and took over a week to arrive. With Gillette's high volume need (1,100 caps a year), and the small window in which the helmets are effective (due to rapid growth in infant skulls) this was a less than ideal process.
3D Printing Method
This is where 3D printing enters the picture. Armed with a Stratasys Fortus 450mc, Gillette Children's Hospital is now able to take a 3D scan of the infant's head, and print out a vacuum mold for the cap in a matter of hours. Their overall turnaround from initial infant visit to cap fitting can be as fast as a day instead of a week. Additionally, they are now able to build these caps at half the cost of their previous method, even with the cost of the machine amortized in!
Throughout Gillette's investigation into 3D printing, CATI was able to provide solutions to some of the roadblocks that other methods were unable to overcome. This includes utilizing two powerful tools, the Insight software that drives the Fortus systems, and netfabb.
Other than cost, the main driver for this project was throughput. Without any optimization, a typical head would take over 7 hours to print. With a throughput requirement of 4-6 heads per day, 7 hours was not acceptable. By the time we had optimized the part, we had cut the time down to 3 hours per head, easily meeting the throughput requirements.
In order to achieve this, a number of methods were used. First we needed to reduce the overall amount of material being printed. We did this by removing a self supporting cone from the inside of the head using the Boolean Subtract feature in netfabb.
First we imported both files into netfabb, then overlaid them so the cone was centered inside the head. Then, using the Boolean operations function, we subtracted the cone from the head.
The new, hollowed out .stl file was then brought into Insight to prepare for printing. Since this part doesn't need to undergo large amounts of stress, we decided to further reduce material by dramatically decreasing the amount of sparse fill. With the custom settings in Insight, we were able to create a 0.5" spacing between each of the internal rasters while still maintaining enough strength to withstand the vacuum forming process.
Lastly, we manually removed the support material tower that is supporting the eyebrow area of the model. This area is close to self supporting, and since the helmet is cut above the eyebrows, it was not necessary to have perfect geometry in this specific location.
Tip: Reducing the number of layers that have support material is one of the greatest build time reducers. This is because the printer no longer has to switch between model and support material during each layer.
Using these three techniques, we were able to cut down over 50% of the build time versus a standard sparse filled part while still meeting dimensional and structural requirements.
Owen Lu | Application Engineer