Improving surface finish of hollowed SLA 3D prints: one aspect of blooming.

Share via

Today, I want to talk about an interesting phenomenon I noticed when printing hollow objects. A simple procedure can drastically improve the surface finish of your prints:

An improvement from left to right that can be achieved

What is “blooming”?

When you print on your SLA machine, you might have noticed the effect people often call “blooming”. It is a quite-well known phenomenon, however, it is not well studied and understood. The guys from Photonster started to recently investigate this phenomenon more in-depth. Read their great Facebook post about blooming.

One possible explanation of blooming is that when you print large crossection of the model, the curing reaction catalyzies itself and runs more aggressively. Therefore, it cures some of the resin on over the exposed edge of the model. The other explanation is that when you cure resin, it slightly expands and, therefore, squishes some partially-cured resins to the sides. When printing small areas, the effect is not noticable, however, when you have a large cross-section it becomes noticable.

What is the true cause (at least for me) is unknown. The guys from Photonsters have successfully applied pixel dimming via UVTools to mitigate the effect. Follow their FB page for further details.

Blooming on hollowed models?

However, recently, when I was testing Resione M58 (surprisingly tough, flexible, and overall durable resin) my first prints were total failures:

This was strange. It looked like an enormous blooming, however, the walls of the models are thin – around 5 mm. What was even more strange, that the insides of the models were perfect!

The models were printed flat on the bed. Therefore, both of them form a suction cup when printing as there are no ventilation holes. Therefore, I propose a hypothesis. Resione M58 is a really thick resin. From my (still unpublished) experiment I know that Elegoo Saturn’s Z-column is really flexible and it bends extremely when printing. Often, it does not manage to squish all the resin when printing before it starts to cure the resin. Therefore, I think what I observe on the outside of the models is a partially cured resin that was squished by the build plate during the exposure. I have confirmed with my measurement setup (see photos below, it allows me to measure build plate position and forces acting on it), that the build plate settles in the position for about 10–20 seconds. That means, that during a normal exposure (2–4 seconds), the build plate never settles! During the exposure, the build plate settles down for roughly 0.1–0.3 mm (that is 2–6× the layer height). Why is the inside perfect? Well, I think it is because of the air/liquid cavity inside the model. It counteracts the squishing to the inside and allows it only on the outside.

Therefore, there are two fixes:

  • make Saturn more rigid (just like I made my Mars more rigid). This is complicated and laborous. Also, quite invasive.
  • just use a long enought ligt-off delay in your slicer that allows you the build plate to fully settle before starting the exposure.

This time, I decided to go with the easy fix, that is, introducing a light-off time. One detour: ChiTuBox’es “light-off” parameter has unintuitive semantics. It defines the minimal time, that the light will be off between two exposures. That includes the movement. For our purposes, we need to set it to “duration of the build plate movement” + “large enough constant to settle the printer).

As you can see from the pictures in the intro, the light-off helps a lot. I used 20 seconds in this particular case. To further verify my hypothesis about a suction cup, I designed the following model and printed it with two resins: Siraya Tech Fast and Resione M58. Fast is quite thin and flows easily. M58 is thick and flows badly.

I printed the models twice; one without light-off and the second with a light-off time. To also make sure the results are not random, I repeated each print twice. The results are the following:

  • Siraya Tech Fast does not have a tremendous difference between the models with light-off and without it. There is a small difference and if you look closely, you can tell which one is which.
  • M58 has terrible outside walls without light-off and nice results on the inside. As expected.
  • The inside pattern gets copied on the outside. You can clearly see where the inside pins are. They (partially) copy the spike pattern. I have no clear explanation or hypothesis on this. Possibly there is “just more partially cured resin” to squeeze out.
  • The models that do not form a suction cup (the straight walls) came out perfect every time.
  • With light-off, the elephant foot is smaller.
  • The models printed with light-off did not have holes in them caused by improper ventilation.

What to take from this experiment?

When you print a hollow model from a thick resin, add some ventilation holes to it, or simply introduce (quite) a large light-off delay. It helps a

PS: If you have any theories about blooming, I will be happy if you propose and experiment that we can try to validate it! It would be really good for the community to fully understand blooming and have ways to avoid it!


Recent news: My open letter to the 3D-printing community

I love the 3D-printing community, but I think there is room for improvement. Let's get better in 2023! Read the full letter.

Support my work!

If you like my work (these blog posts, my software and CAD models) and you would like to see more posts on various topics coming, consider supporting me in various ways:

If you are interested in knowing what I am up to and recent sneak-peaks, consider following me on social media (Twitter, Instagram, Facebook).

My store offers

I launched new tank cleaning kits for Elegoo Saturn, Saturn S, Mars 1, and Mars 3. You can find them in my store.

Share via
Back To Top