Hello Community,
ones again here ist Peter from Germany with another challenge in printing small parts.
I tried it with white (50my) and black (25my) - no chance.
The space between the small rips is always half full with already cured resin.
Momentarily i am buffled 
Thank you all in advance for any ideas
Encoder.form (332.3 KB)
It looks like those slots are about 220 microns wide, is that correct? As you can see in this article, we do print things at that size, but it is pretty small.
Are you sure that the resin is already cured? It may be that the viscosity of the uncured resin is making it stick in the slots. One technique is to fill a syringe with IPA and flush out these negative features as soon as you take the part out of the printer. If you wait, then the resin will start curing as it is exposed to UV.
Hi Magarrity,
thank you for your quick response.
Yes these slots are about 220 microns wide.
What i already tried, was to use an airbrush pistol and ipa nevertheless it didnāt work out.
Maybe i waited to long before flushing the part? I will give it another try. Thank you!
I will report on it.
Best regards
Peter
Hi Peter,
I can see that you would have problems printing that part flat with sparse support structures. I am not surprised that you are experiencing some forms of failures. I suggest rotating the part slightly and making sure the support structures are dense enough to get the right geometry to form.
Additionally, I would recommend using 100Āµm layer thickness for this part. While 25Āµm and 50Āµm result in more layers, their primary advantage is a smoother surface finish. Generally, I have found that our resins will produce better dimensional accuracy in prints at 100Āµm. So I would start with that. White and black resins are good, grey (v3) resin has better dimensional accuracy, and dental model is our most accurate resin (at a higher cost).
Here is my suggestion:
Encoder.form (444.9 KB)
The airbrush with IPA is a clever idea.
Hi Community, Hello Formlabs,
i am still trying to get this little peace printed - no chance so long.
Is really a challenge for the Form2?
Regards from Germany
Peter
Just out of interest, how did you intend to manufacture this part in production?
The laser spot size is quoted at 150umā¦ how wide are the veins, think thatās probably more important than how wide the gaps are for SLA purposes no?
PS. I wish there was a measure function in preform.
Clear resin would perhaps have a more successful print for that objectāthe issue is that the spokes are too close to each other and the laser will affect a little bit around where the laser point hits due to it bouncing off particles in the resin. With clear, it doesnāt have any pigment so that effect should be reduced the most.
To follow up on @Zachary_Brackinās clear resin proposal, you can very well paint it later to keep the intended function of the part.
Also, design-wise, the gaps looks like theyāre composed of lines pointing at the center axis. Sketching a simple rectangle may help while again keeping the intended function.
I also am under the impression that with the orientation shown in your last image, the part is oriented almost at a 90Ā° angle from the build platform. I would thy orienting it so that it is more āflatā but still at a minimum 5-10Ā° angle so as to not create a surface area per layer that is to big.
Another thing related to layer sizeā¦ this is totally a gut feeling to take it i with a pinch of salt : A higher resolution means the laser will spend more time in a given area. It will pass 10 times over a geometry that is 1mm in height @100mĀµ but 40 times over the same area @25mĀµ, which will not help the effect of the laser bleeding over other geometries. however, there might be more to this that what I wrote, formlabs may have taken that into account in some wayā¦
I think the problem with all wheels-with-spokes is that the spokes that are nearly horizontal will not be well supported and will tend to sag unless they have extra supports. I would definitely print this hub side up so that support pips on the disk can be sanded on a flat surface.
I just printed it flat direct on the build plate, in the full knowledge that iād break it trying to get it offā¦ came out better than your photo above (i printed it in standard grey) but ridiculously small detail and not something that Iād bother trying to get out of any RP machine regardless of cost or type. Maybe better of looking into a laser etched foil screen or bigger encoder slots.
sorry, didnāt wash it very wellā¦
Weāve found that Grey V3 and Black have the highest feature resolution of any of our materials. Pigment will prevent the laser from ābleedingā into surrounding areas and improve resolution of small features. Each of the resins do have photoblockers that localize the laser spot for the most part, but pigmentation helps a bit as well.
Printing the part in grey v3 at 100Āµm layer thickness yielded OK results. At their thinnest points, the Form 2 could not resolve the negative features, which are already way below the machineās design specs. Using Dental Model Resin in the orientation I provided at 100Āµm may provide marginally better results.
Good Morning Community,
Good Morning Formlabs,
i am overwhelmed about all of your feedback and your help with my printing issue.
First of all i want to thank you all!!
This part shows clearly the limit of the Form2. I/we wish the Formlabs team could improve the laser in both ways - power an accuracy. More power would give us users free choice for third party resins like Nextdent or Detax. More accuracy would help to produce more detailed (smaller) parts.
For the moment we produce this specific part by micro milling on a Kern 5axis HSC- Machine. Very expensive though but we have to deliver. Our customer is waiting. For the future of small RP parts, we will look in other SLA-/DLP-Printer.
Again, thank you all for your support.
Best regards
Peter
Hi Frida, I have no connection with Formlabs, so donāt feel any compulsion to stand up for them. But every beuro service I have used to date, ARRK, Materialise etc, + some others that have gone by the wayside would not even touch this part and tell you to try something elseā¦ & they have invested millions? into SLA / SLS / FDM technology.
I would suggest laser etching talk to someone like http://www.photofab.co.uk/contact/ Again, I have no affiliation with them but they have done good work for me in the past.
So maybe, & its just a suggestion, print or turn the wheel part in acetal, and bond a thin steel sheet (laser cut / etched) to your wheel.
Ha Haaaā¦
Just looked at their site a little, looks like they already produce encoder discs just as I had imaginedā¦ have a lookā¦ http://www.photofab.co.uk/products/encoder-discs-and-actuators/
Sure, but itās impossible to prevent the laser bleeding, the layer needs to be cured enough or itās going to fall apart when the layer peel action happens. Since you canāt stop the laser from going beyond the layer thickness itās going to bleed to some degree. Otherwise you could turn up the cure time to make stronger layers.
I would imagine that Clear would have the worst effect for Z-axis bleeding. For X/Y axis bleeding Clear would have the least effect since the light isnāt going to bounce as much within the resin as it would with the ones that have pigment particles in it. White is probably the worst for that since the white pigment will reflect the most light within itself.
Weāve done independent feature resolution testing for each of the materials and found Grey V3 to be the best followed by Black Resin. Iām unsure of the chemistry involved, but I would think that the darker pigmentations would have the advantage of absorbing stray light and preventing it from bleeding into surrounding regions. White V2 does perform slightly less well than the Black and Grey Resins but does still exceed Clear Resin. Although white pigment will reflect light, the reflections are occurring in a dense region and the net light bleed is still relatively low as compared to a material with no pigment.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.
