Fast printing with a PDMS resin tank

One of the 3D printing community's strong interests is fast printing without waiting hours to print 1 inch tall object. A print speed of 1 inch/hr is considered as standard for DLP based printers. The most time consuming step besides actual exposure is separating the cured layer from container bottom and refilling the gap with liquid resin. To do this, the build platform needs to move vertically or horizontally, which takes 2-7 seconds per each layer depending on mechanism, settings, build area, and resin. 

A key is how to make an inhibition layer. CLIP technology from Carbon3D offers a true continuous printing, however its patented, expensive, and gas cylinder required system is not an option for most 3D printer enthusiasts. Similar but much cheaper and straightforward approach was tested by Autodesk Ember team and they already mentioned that fast printing is possible using just PDMS coated resin container. 

Please see the link for what Ember team tried. 

http://www.instructables.com/id/How-to-Configure-Ember-for-High-Speed-3D-Printing/

http://learn.ember.autodesk.com/announcing-high-speed-printing-on-ember

In brief, they achieved 440mm/hr speed at 250 um layer thickness. They claimed the continuous printing is 24x faster than normal printing, but this is actually inaccurate since they compared 25 um layer thickness at normal printing vs. 250 um layer thickness at fast printing. And the resin used in the normal vs. fast printings were different to obtain deeper light penetration. They might simplify the equation just to show the fast printing is fast enough to be surprised. 

So I was curious if I can configure Mono1 to work like that. However, the resin tank with a Teflon film cannot be used since the film is flexible and it will prevent the resin refilling between the gap without platform movement. So I prepared a PDMS coated resin tank (PDMS thickness ~2.5mm). Then the platform up&down movement after each layer was turned off (lift distance = 0) and I just kept the shutter open (disable shutter). 

The picture below is a test print with Fun to do Industrial Resin with black pigment. It usually requires 0.6 sec per 50 um thick layer, so average printing speed is around 300 mm/hr!! If I print with a layer thickness of 200 um, which might require around 1.5 sec, the printing speed would be 480 mm/hr! This value is close to Ember's test. The actual print time is around 20 min for 40 mm tall mesh structure because I used a longer exposure time and platform up&down for layers under h=1mm. I would be able to print a 200 mm tall mesh in an hour if I continue to print.  

The limitations of this approach were also well stated on the first link above. It's not a general method that can't be applied to all sort of objects. This is effective only to objects whose cross-section area is small and local area is also small, for example, mesh structures or Eiffel tower. I will try to print the Eiffel tower next time. 

After I found out the method works with Mono1 and the PDMS resin tank, I tried another print. But I encountered a problem that I read somewhere regarding PDMS incompatibility with FTD resins. After the second print is completed, there was a big PDMS detached area at the center of the resin tank. It looks like the resin was reacted and aggregated together with PDMS. It happened at the end of the second printing, i.e. on around 1500th layer. 

I'm not sure if using other resin will solve this problem? I will try other resins like Makerjuice G+, Monocure, or B9C resins if they have similar issues or not. Does it work with SLA printers like Form2? Probably, but the benefit is not huge since the scanning itself is slow.

In conclusion, I confirm that the fast printing is possible with some design and material limitation. Resin to PDMS issue will be tested again and posted here later. 

Thanks for reading. 

Yun


Leave a comment