Minimum Feature Size Test


#1

Here is a stretched 1cm block with a series of simple hole features. There is a round and a square hole for each size. The person testing is allowed to support and orient it any way they feel will give them the best results. When post-processing feel free to squirt nice and hard with the IPA squeeze bottle to clean them out. But no drill bits or other manufacturing please. Please post your results here.

1.5mm (1500 micron)
1.2mm (1200 micron)
.9mm (900 micron)
.6mm (600 micron)
.3mm (300 micron)

Min Feature Test.STL (83.1 KB)

If you are doing another print soon, please throw this tiny model in the bed beside your other one.


#2

Hi @JoshK, what layer heights should we run the test at?


#3

I would say the best, 0.025 (25 micron).


#4

You will find that in most cases the Form 1+ can better resolve negative features using a thicker layer height (e.g. 100 microns over 25). This phenomenon is a consequence of the curing properties of the resin in conjunction with a gaussian light source.


#5

Well we should try them both then, because any results we get at 100 microns would be discounted because of perception.


#6

I’ve successfully printed 700um round holes through a 1mm wall (not 300, I know), and they were dead on. I have also made 400um deep 1.5mm circular indents that were quite crisp as well.

I have your model, and will print it @ 100um, as I cannot actually make any print work at better resolutions.

-C


#7

What’s wrong with your printer? The .7mm holes sound good. The indents would just be texture though.


#8

You mean about the resolution? I think it’s fairly normal that sub-100um is pretty dicey with the Form1+. To be fair though, I have actually printed one thing good once @ 50um, but the diff between the same part @100um was not worth double the time. I’ve never had a 25um print work, but again to be fair, I stopped trying…

They were more than just texture, they were pockets to inset and adhese 1/16"x1/32" neodymium disk magnets into, and they worked well for that. There was an extremely minute radius at the bottom edge of the pocket, but the adhesive in the bottom came up to the tangent point so it was not an issue.

-C


#9

I see.

This wouldn’t be acceptable to most customers.


#10

It isn’t acceptable, it’s kinda false advertising, but I need the machine, and I cannot send it back and wait for another that history shows is unlikely to be better. Yes, it should work @ 50 and 25, but at the end of the day I have work to do, and I can actually get it done @100 - even if it’s not as pretty as it should be.

-C


#11

@ChristopherBarr — contact our support team, and they’ll take a look at what’s going on. Sounds like you know that and are taking a break — but we’ll be there for you if you want to take a deeper dive at any point.

Printing at 25 microns can be more challenging than printing at 50 or 100 microns, and works best with small parts — but if your printer is failing, please contact our support team.


#12

Here is a test print of the Min Feature Test. Printed at 100 micron in black v 2. Auto orient. My laser spot is not very good, flair and “rabbit ears”. Will try higher resolution when I get a little more time.


#13

Wow! I never had a printer that could print like that. I can actually see the 300 micron dimple. It’s not a fully formed hole, but at least it can be located. And the 600 are nice. I would put your min feature size between 300 and 600. Now I wonder what others will look like.


#14

Thank You Sam. When the storm subsides (if ever) I definitely will.


#15

Hole resolution looks good, but the surface near them does not look that good. Wonder how tight the holes would be if the tiny holes were last out of the vat?

Also, while looking at your laser spot something occured to me like a poof! (or epoofiny, for the full technical term), which was triggered by you saying

I guess seeing both ‘flair’ and ‘rabbit ears’ together in a short, sweet sentence clicked for me.

What if what we’re seeing in these ‘spot shots’ is somehow really just a function of persistence of vision? My intuitive understanding of how a galvo operates is that there are two competing and opposing electrical forces on the galvo for it to target a coordinate, and maintain directionalty (e.g. supply and restrict current). Is it possible that the signals to both sides to do that are ocillating, or vibrating such that the actual dot is simply spending most of it’s time on it’s target location, but the frequency is jiggling the beam outside that ideal for a frequency-related timeframe?

If anything close to that were true, it seems like a test would be to add some mass to the gavo mirrors to see if the spot changes. Could it be weird if the frequecy of it all was close to the resonant frequency of the mirrors, like an unintended feedback loop? (or hopefully not quite like this: https://www.youtube.com/watch?v=3mclp9QmCGs) :scream:


#16

I printed two of these with some other parts I was running today.

Printer Info: I used a Form1+ I’ve had the machine for about a month and a half so far, and I’ve performed no calibration on it. I haven’t inspected my laser, so I have no idea if it’s exhibiting flare, bunny ears, ect.

Set Up: Settings were .1mm with black reference. For reference this has been my most successful resin/ resolution combination, though I haven’t actually tried very many so far. The resin tank used has had somewhere in the area of 1.1-1.2 liters worth of use so far. The first test piece was located right in the center of the platform and was oriented flat relative to the tank with supports. The second test piece was located right on the edge of the platform and was oriented using the automatic preform settings.

Other info: Ambient temperature was ~67 Degrees F. I left the parts hanging for ~4.5 Hours after the print completed as the parts were printing while I was at work.

Results: .9mm-1.5mm holes printed properly on both parts. The .6mm holes are clearly visible, but are fully obstructed very close to the surface. There are tiny dimples at the .mm hole location which appear to be the correct shape, There doesn’t appear to be much difference in the holes based on their location and orientation.


#17

@ChristopherBarr, That’s a good thought. But the flare has been seen equally with the laser pointed at the wall.

@Scott_Morgan, Thanks for running that. That is about the same level of detail I was getting back when I tried to print my case with the .635mm holes.


#18

@ChristopherBarr If it was anything caused by vibration of galvos (when they should be standing still) you’d get a lissajous curve, not something looking like our spots. And, like JoshK says, the flare has been very clearly isolated to the laser optics.


#19

BTW, this isn’t really the “minimum feature” test. Printing “pimples” of varying sizes would be a “minimum feature” test, this is the “minimum negative feature” test.

You’ll find you’re typically capable of printing smaller holes than “pimples”, although it depends on the profile of the beam.

Think of it this way - a pencil can draw a dot no bigger than it’s point, but you can make tiny circles that get progressively smaller and smaller than that, encircling an empty dot of paper that can teoretically be infinitely smaller than the pencil point.

The “negative” feature is limited by the pencil’s leakyness (beam profile) and your hand-eye coordination (galvo resolution). Whereas the actual feature size is limited by the pencil sharpness (beam focus).


#20

Have now printed the same .form file at 50 micron, part on the right. No big difference in detail. The various surface imperfections are in the same location on both prints so I guess this is from dust in the optical pathway. Ambient temperature 25° C. Printer is an F1 upgraded to F1+.