Purposeful Over Curing of Stereolithography (SLA) Printed Parts – Final Report

Abstract

The purpose of this report is to examine and better understand the discoloration effect over curing has on 3D printed parts to determine what applications may exist. It is of utmost interest to find suitable materials to be used when over curing and to develop a correlation between the discoloration and degree of cure. It is also critical to determine print layouts and software that consistently produce identical results if this method is to be used.

Introduction

One common method of additive manufacturing is stereolithography (SLA), a process which falls under a family of technologies called vat photopolymerization. During this process, a vat of resin is cured with a low power laser to produce a solid part within the resin. For the Formlabs printers used in this project, the part is cured upside down and lifted out of the vat as each layer is cured by a pass from the laser, seen in Figure 1.

Figure 1. Upside-down (inverted) SLA

Over curing of 3D printed parts occurs in the SLA additive manufacturing process when excess UV light begins to degrade material that has already been cured. If using clear resin, it has been observed this degradation can lead to an amber discoloration within the resin. The discoloration also becomes a progressively darker shade as the UV light continues to pass over the same spot. This effect is most often undesirable since the material degradation compromises material properties, and many applications may require a completely translucent or monochromatic part. However, there do exist applications where specific discoloration may be desirable. In the medical field, it may be relevant to print an MRI of a brain, bone, or organ within a clear cube of resin. This would provide medical personnel with visuals they can easily rotate in their hands and peer into, while being a reliable model for each individual patient. 

Another application could be to suspend delicate models within a cube of resin, such as an insect’s skeletal structure. Printing the design within a cube with SLA would ensure the part would not break and the model would be accurate and precise. Finally, an application could be consumer based, which is the focus of this report. In this analysis, a fishing lure model was manipulated and printed to determine if this method of over curing could be applied to its manufacturability. It was hypothesized that a semi-transparent lure with an internal structure could be more appealing to a fish, and could certainly provide strong evidence to whether over curing is a realistic option that can be scaled in a manufacturing setting. In Figure 2, the initial fishing lure model used is shown, though it should be noted certain features such as the hooks were removed since the complexity was not yet needed. 

Figure 2. Initial fishing lure model

Equipment

In this analysis, two types of SLA printers were used. The first printer was a Formlabs 2 SLA printer, shown in Figure 3, and was accessed in the UW Makerspace at Wendt Commons. This machine was first developed in 2019 and utilizes the upside-down method, which is shown in Figure 1. 

Figure 3. Formlabs 2 SLA printer

The second printer used was a Viper si, which was developed by 3D Systems in approximately 2001. This machine prints by moving the build plate down, contrary to the Formlabs printers. The Viper printer was accessed at the Morgridge Institute for Research and was chosen because it was known to successfully over cure parts. 

Materials

Achieving discoloration due to the over curing process in stereolithographic (SLA) 3D printing requires a resin that is susceptible to color change during material degradation, so research was conducted to find the correct resins to test the over curing process. As the Formlabs Form 2 and 3 were going to be used for the initial prints, the resins that could be tested were limited to Formlabs’ resins. From background research into over curing, non-transparent resins were not found to have any color change during degradation, so the material property sheets of Formlabs transparent resins were examined. 

When examining the material properties of each transparent material, there was not a measure of a material’s degradation when subject to UV light. It was assumed that the heat deflection temperature would be the most similar property, but the Formlabs’ blogs were investigated to obtain additional information before this assumption was acted upon. While browsing the Formlabs’ blogs, a discussion on over curing Formlab’s resin to change its color was found. In this discussion, Formlabs’ employees did not think it would be possible to change the color of the resin with their printers, and it would simply make the overcured areas more brittle and easy to break. Further in the discussion, it was found that polycarbonate-like resins such as Accura and Somos Watershed had been successful in creating different colors when over cured. With this in mind, Formlabs Clear resin was chosen for our Formlabs prints and Accura 60 was used for our prints at Morgridge.

Methodology

For the initial proof of concept, the intention was to overlay several parts into one. In doing so, the team hoped to reproduce the discoloration within the region that is over cured. Figure 4 shows the setup in Formlab’s software, PreForm 3D. Unfortunately, the software automatically signaled an error when parts were overlaid on top of each other. In addition, the generation of supports was difficult since the software wanted to generate individual support structures for each part, resulting in a significant amount of support resin that is difficult to remove. A single support structure for the resulting overlain part would have been more desirable since it would waste less material and make post-processing easier.

Figure 4. Setup of initial print

In comparison to the Formlabs printers, it was much easier to adjust the part orientation and support structure within the Viper printer. In the Viper software, support structure could be added to only the main portions of the part, and overcured areas were mostly allowed to use the same support. It was also much easier to bypass any errors/warnings encountered in the process.

After completing the initial proof of concept to confirm whether over curing would result in discoloration, it was desired to more accurately model the degree of over cure. This would hopefully correlate the degree of over cure to the darkness of the amber/orange color within the part. A part printed prior to the project at Morgridge Institute for Research with varying degrees of cure can be seen in Figure 5. However, it was not known how many parts had been overlaid within the STL file since it had been done about 10 years ago, so experiments were carried out to determine whether a correlation existed.

Figure 5. Previously printed part of unknown cure

Following additional research, an STL file was created that incorporated several varying degrees of cure. By overlaying a different number of parts in specific areas, the discoloration can be better understood. It is desired that an in-depth understanding of this correlation will lead to advances controlling the color within the part. 

Results

The initial proof of concept carried out at the UW-Makerspace with the Formlabs printer did not print correctly. The desired print orientation is shown in Figure 4. The model had 7 cubes overlain within the fishing lure, resulting in a total of 8 passes by the UV laser. However, the print failed and was restarted by the UW Makerspace staff. It is assumed the staff had oriented the parts separately on the build plate since none of the cubes were fused together and none had adhered to the fishing lure. The next iteration was identical, and succeeded in overlaying the parts as intended. The result is shown in Figure 6, below the first failed print. 

Figure 6. First (top) and second (bottom) iteration

The second iteration on the Formlabs printer shows some discoloration in the region where multiple parts were overlaid. As compared to the first iteration, it can be observed the entirety of the part is slightly darker, and the over cured region shows some yellowing. The support material was not easily removed from the second iteration since the Formlabs software, PreForm, did not appear to allow generating only one support structure. Instead, it generated individual support structures for each part, resulting in 8 different structures in the same region. This led to a pool of resin being cured beneath the final part, which was difficult to remove.

Following the Formlabs print, an iteration was completed at the Morgridge Institute for Research on the Viper printer. The intention of this iteration was to better correlate the number of parts overlain to the discoloration within the part as a result of overcuring. The model is shown in Figure 7, which has an increasing number of parts overlain in each square.

Figure 7. Increasing number of overlain parts from left to right

In Table 1, the quantity of parts overlain within the cube is given, as well as the effective number of passes experienced by those areas.

Table 1. Viper print specifications

The resulting print on the Viper machine can be observed in Figure 8. Varying degrees of cure are indicated by the progressively darker regions. On the far left, the effective number of UV passes is 4, whereas on the far right is 11 passes. In each case, the over curing occurred within the larger rectangular prism, and can be better visualized looking from the side, as shown in Figure 9. Here, it is clear how over curing can be accomplished at any location within the resin.

Figure 8. Third iteration completed on Viper printer

Figure 9. Side profile showing embedded over curing

Up until the third iteration, the objective had been to recreate previous results seen in the past to confirm feasibility; and to determine whether a correlation existed between discoloration and degradation. Following this, a fourth and fifth iteration were planned to finalize which printer and resin would yield the best results moving forward. An identical iteration was run on the Viper and Formlabs printer, overlaying a total of 30 skeletal structures into a single lure, as seen in Figure 10. This configuration results in 31 effective UV passes by the laser. 

Figure 10. Model used for fourth and fifth iterations

The Morgridge print heeded great results. The 30 overlaid internal features provided a dark amber color within the lure. The only problem with this print was that, as the internals were overlaid flush with the bottom of the lure, the expansion from over curing caused the internals to stick out of the bottom part. The completed print from Morgridge Viper printer is shown in Figure 11.

Figure 11. Fourth iteration printed on Viper printer

The same print that was conducted with the Viper printer was attempted at the Makerspace with the formlabs printers. To get a better print than the previous formlabs iterations, a single support structure was used for each of the 30 internal features. This would lead to less overcuring of the support structure so the support structure could be removed. Despite optimizing the support structure to satisfy the formlabs software while removing as much unnecessary material, the print failed. The overcuring of the support structure caused the build plate to fuse to the bottom plate of the printer.

Discussion

Over curing to create multicolored prints in SLA printing is an attainable feat. With varying printing softwares and many different styles of printers, some printing setups are capable of achieving over curing while others aren’t. The Viper was suitable for over curing as the build plate lowered, removing the need for a “raft” to fuse the print to it. Also, the software used for the Viper printer allowed for free removal of the support structure. The Formlabs printers lacked these qualities. As a result, a successful print could not be obtained from Formlabs printers.

Future Work

Though the application chosen in this project was to design a fishing lure using over curing, many possibilities exist that could potentially utilize this behavior. In the medical field, a brain or other organs could be printed within a block of resin to help neurosurgeons visualize the path for surgery or help explain problems to patients. A magnetic resonance imaging (MRI) could be converted into a file compatible with 3D printers, and subsequently printed with over curing to allow an individual to peer into a part and see blood vessels, arteries, or lungs. Fragile geometries with thin features could be printed inside a transparent layer to protect the feature. Finally, overcuring could be used in general multicolored SLA 3D printing. Regardless of all the different applications, discoloration due to over curing is still not fully understood. There is not yet an explanation to why certain resins are more susceptible than others and why some resins turn different colors. Further research and testing into this subject could lead to a better understanding of over curing and could possibly lead to use of over curing in industry in the future.

Conclusion

Over curing is a unique process where resin used in SLA 3D printing can be hit multiple times with UV light to degrade the material, sometimes leading to a discoloration. From this project, it was concluded that discoloration from over curing can be achieved with the correct printing software and printer. It was also found that the degree of discoloration is significant enough to be a useful printing practice. With many possible applications and little to no research on this topic, over curing for discoloration has the potential to be a significant development in the 3D printing community and industry.