Bad prints out of brand New Form 3


#17

Yes this echoes my findings too.I added more supports with no change in the artifacts and I have printed simple geometric boxes in multiple orientations and I have yet to get a successful print of a simple open box shape.I have a support tickets with formlabs, but they are still helping me with a resin dispensing error issue.We have not even started talking about print problems.I am very frustrated at this point.I use my form2 to pritn and sell sci fi model parts and was hoping to use my form3 for this too, but there is no way I can use my form3 at the moment for that purpose.


#18

Glad you got that mixer arm situation handled, thought it was strange how it sort of just went crazy even wihth no resin. Honestly at this point you have to see yourself as a beta tester or pretend that the machine doesn’t exist for a few months or more. Way too many people have been fed information about orientation issues or some other issue that just doesn’t seem right, especially after printing extensively with the form 2 and now the form 3. With my own machine I have a issue with resin dispensing and the cartridge not seating correctly, ending up just pouring the resin in manually.

This machine is just not tuned anywhere near for the work I do which is miniatures. But I really am frustrated with the fact that it is a good hardware machine with potential. Just wish their marketing department was not as zealous with touting features which just weren’t available.


#19

Yep I came across your posts in another thread. I do miniature work, 1/6 scale mostly, and the Form 2 was the best purchase I ever made for it. I figured the Form 3 would give me more of the same and then some, and unfortunately I can’t even get prints EQUAL to what I got out of the Form 2 on IDENTICAL 3D models.

As you can see in my photos I have printed very rounded/spherical objects, barrel shaped, curved, completely straight items… everything… and the same layer shifting occurs on EVERY print.

I have just about completed a lengthy response to my ongoing support ticket with the new photos, as well as airing my frustrations and asking what can be done about this. It’s either an issue with MY printer or a widespread problem, and the number of other photos I have seen lead me to believe the latter.

So I have a question. Is this printer returnable? I haven’t had mine but 2 or 3 weeks. It’d be a huge hassle but right now my $3500 could be better spent elsewhere until I can have more of a guarantee n print quality.


#20

Parts this small and with such a tiny cross-section should not require that much support.

I don’t care if more support fixes it or not, that’s not normal, nor desired.
Parts of that size print fine with very little support on a Form 2.
Form 3 was advertised as having lower peel forces.

Hence - if it’s getting shifts on Form 3, something’s wrong.

Hey, @DoggieDoc83, any chance you could share the .form of the prints that resulted in shifts? I’m curious if it’s going to print the same on my Form 3 and 2.


#21

Yes that’s been exactly my point this whole time. Not only should my miniature items not require so much support, but they literally HAVEN’T on my Form 2, and they printed just fine. I have posted comparison photos in this thread AND sent them to Form Labs, yet I’m still being told more supports are needed.

And absolutely I can share the files. I will attach the two different files I have put together so go hog wild with what’s on there. I’m interested to see this as well.

https://drive.google.com/open?id=1Dga4y5RSa0LaPgy13KFl4KBG19ooW_4b

https://drive.google.com/open?id=1z4Byq8HL9i1z42PIxEJy7ZAsGANLg8w2


#22

Awesome, i’ll give them a spin today/tomorrow.
For what it’s worth, i never got layer shifts with small hollow objects, but did get some with non-hollow ones. Here’s an example i got very early on (photo #3):

Note that the part of the print with the strong layer shifts had them in the very last layers. Which i found super odd at the time but didn’t comment on. How would the last layers lack support if they’re directly attached to all the previous ones, which didn’t have any shifts? Makes no sense.


#23

Yep exactly, I don’t get it. Many of the prints on that test file are hollow objects, only because they are things I printed on my Form 2 previously in clear due to my needs (as you’ll notice many of the items are clear items in real life, such as the sunglasses lenses, the pint glasses, the light bulb). However I also added the gun and the little cubic box (which is actually a model for the transponder used by Anton Chigur in the film No Country for Old Men) because they were straighter and more geometric, and mostly solid.

As you can see in all my photos it does not matter what I am printing. There is layer shifting on every item. Layer shifting on a model with less supports, with more, oriented 90 degrees to each other, solid items, hollow items, rounded items, flat/squared items. This CANNOT be a “number of supports” problem.

There is no way I can print anything on my Form 3 right now and use it for my needs. Beyond frustrating.


#24

Printed the first file.

Same orientation as you, i just regenerated the supports (because i’m printing in grey) and moved everything to the mixer side so i don’t wate an eternity for it to print (in the file they were on the opposite side, meaning the LPU has to move all the way across for each layer).

GUESS WHAT.

I got the shifts in EXACTLY the same places as you.
They seem a bit fainter, but that might just be grey vs. clear - i’d expect the shifts to be way more noticable with clear than gray (because you see two sides of the shift).

So it does not seem like a variation in tank (e.g. film not being as taut on some tanks) or the rest of the machine.
In fact, it makes no sense in terms of peel forces either - parts of the objects that are less supported don’t have shifts, and parts that are more supported do.

From a cursory examination, it looks like shifts always start/happen on areas that are completely facing the platform - similar to how you’d get bad texture on Form 1 / 1+ back in the day on surfaces facing the platform (facing upwards during printing) from the laser diffusing through layers of resin.

I’ll take some photos tomorrow, i’m knackered, it’s almost 5am and the prints are fresh out of Wash anyways.


#25

Looking at it some more, on the thimble-looking things, the layer shifts appear just below the part where a key support point shows up a few layers later.

I’m thinking this might be two things:

1 - the “low peel force” not being that low for some specific geometries (i.e. not offering any advantage and making things slightly worse for some shapes), and

2 - preform’s automatic support placement not doing such a good job with Form 3

I’ll play around with supports a little tomorrow, see what i get.
If i keep getting the same result with this particular orientation over and over again, like you, then something’s really wonky with the whole LFS concept and Form 3’s peel strategy.


#26

Cause, here’s the thing.
(Sorry, having a pre-crash-to-bed brain dump.)

These layer shifts show up only occasionally, not throughout the whole object.

You know what also varies with layers and shows up every N layers?
Form 3’s whole-tank peel. It does use the rollers and movement of LPU to peel things gently, but if you watch closely during a print, you’ll see that every N layers, it will also rock/tilt the whole tank, and basically “unpeel” the film off the LPU / roller holder completely.

My bet’s that’s what causes the significant shift, and that’s why it only shows up occasionally (once or twice per object in your case, since prints are tiny).


#27


“Oh, but look, the shifts correspond to the supports exactly

“… oh.”

I even got the exact same line in the glasses, just slightly fainter:

Same deal with the bulb:

Here’s a fun one. That transponder box.

Front, no shifts:

Right side, no shifts:

Left side, SHIFTS:

It’s not a hollow object. Left side isn’t angled. How. Why.


#28

Exactly. How.


#29

Thinking about it it can’t be layer shifts as it would effect all sides, it looks more like like FDM printing z stretch/squish.


#30

Z squash would also be visible on all sides.


#31

Sorry yes, it’s behaving more like a FDM printer with sloppy belts, so it must be the movement speed/laser power. As I said in another post, the much higher mass of the new LPU will have acceleration/deceleration factors.

I think the inconsistencies are caused by the laser power being modulated for the relative speed of movement and if this is slightly off you get over or under exposure.


#32

That was one of my thoughts too.

So i’ve printed a new test batch of ”thimble things”, one with beefed up supports in place where the issue occurs, just to double check and confirm it’s not a support issue. And with 3 of them all oriented the same way in X and Y, but rotated 90 degrees in Z.

One had the affected area facing in the direction of LPU travel, one perpendicular to LPU travel in one direction, the other in the other.

Lo and behold - all of them had the same artifact. In the same spot.


#33

I’m running out of ideas (and resin).


#34

Affected area facing back of the printer:

Affected area facing front of the printer:

Affected area facing left side of the printer (mixer):

Facing front + twice the support:

Now printing one lying on the side, which is the only other viable orientation, just to confirm.


#35

Man i appreciate you doing this. I’ll forward these photos along after i get my next support response. This will help make my case that adding supports won’t do jack.


#36

Oriented horizontally:

Other side:

And before someone asks - yes, it had plenty of internal supports: