origin

procedures

While working at Origin, I designed a number of protocols and procedures that streamlined work and were performed dozens or hundreds of times by members of the lab team.

materials validation

In order to evaluate printability and determine print settings of new resins, I devised a series of "materials validation" procedures. These procedures allowed a wide range of materials to be quickly tested and compared with each other.

The materials validations consisted of standard print geometries, methods for measurements, and report templates. The procedures were efficient to perform, and the resulting reports were easy to digest.

XY-scale calibration

In DLP stereolithography, the dimensional accuracy of printed parts in the XY-plane depends strongly on two factors: the projected image size and the exposure time.

A projector that is too close or too far away from the curing plane will result in parts that are under- or oversized by a relative amount; this defect requires a scale factor to compensate.

Exposure times for a layer that are too short or too long with result in parts that are under- or oversized by an absolute amount; this defect requires an offset to compensate.

I created a print procedure that allowed a printer operator to calculate these correction factors efficiently and could be result in repeatable dimensional accuracy of better than 50 um.

sample parts

In order to demonstrate the capabilities of Origin's printer, I designed several kinds of sample parts that could shown off or given away to potential customers.

Ahead of the Formnext 2018 conference, I created a sample part that showcased multiple organic and geometric textures and used interlocking male and female threads to demonstrate print accuracy and repeatability.

I also created a number of other sample parts to showcase various materials and geometries that were possible with Origin's printer.

conference talk

At RadTech's 2020 Ultra Violet and Electron Beam Technology Conference, I gave a talk outlining the fundamentals and capabilities of Origin's 3D printing technology.

Enabling advanced material formulation with programmable photopolymerization (P3)

S.J. Kranz

Rad Tech 2020

9 March 2020

Orlando, Florida

covid-19 swabs

In the spring of 2020, Origin rapidly shifted its normal operations from developing 3D printers, to a round-the-clock factory producing Covid-19 nasopharyngeal swabs.

Jenny Chen, MD at 3D heals interviewed me about Origin's challenges and solutions for mass producing medical devices.

Interview with Steve Kranz, Material Science for 3D Printing

J. Chen

3D Heals

13 May 2020

software tools

Below are a few examples of software tools that I created to help diagnose print issues or to augment the printer's capabilities.

"X-ray" visualization

The x-ray visualization is a diagnostic tool that quickly and reliably detects errors that can occur when an STL is sliced prior to 3D printing.

It works by creating three 2D arrays that represent the XY, XZ and YZ planes of the 3D object. It then loads the stack of 2D slices, and for each slice it loops through each pixel and adds that pixel's value into the respective index of each of the 2D arrays. After all the 2D slices have been processed, the arrays are normalized and each yields a greyscale "x-ray" image, where white values represent the thickest sections and black values represent empty space.

This tool is helpful because it allows an operator to visualize hundreds or thousands of slices in a single image and to identify problematic print files before they are sent to a printer.

voxel-based texture

3D printed parts can exhibit artifacts such as layer lines and aliasing that can make the part look unfinished. Adding texture to the part's surface can hide these artifacts. However, adding texture on the 3D model itself dramatically increases the file size and can make the model unfeasible to work with.

The voxel-based texture tool applies a matte texture that hides print artifacts and gives the part a more even appearance.

The tool loops through each layer and identifies vertical surfaces with an edge detection filter. It then applies a several pixels of random noise at the edges. Likewise, the tool identifies horizontal surfaces by finding the difference from one slice to the next. Random noise is applied to the difference. The textured regions are then added onto the original image stack.