Form1+ laser flare issues illustrated - pics and video

This also seems mostly unlikely as @KevinHolmes blob test to show the flare exhibits the same glow around the dot itself, and no paper is involved at all with his method. It’s possible the resin diffuses the beam about the same way and amount, but that seems unlikely.

1 Like

Fantastic photos!

@KenCitron - it seems likely that most if not all F1+ lasers have some form of flare - as @CraigBroady effectively admits in post42 - and the FL line up to this point has been - “it’s too low intensity to cause issues”. And then - “it’s too low intensity to cause issues beyond special cases”.

I don’t agree with that. My parts from post1 are not special cases - they’re very plain small parts.

It now seems obvious that there is a large degree of variation in flare intensity, and hence in part quality between printers. Compare my tall test pieces from post3 and post26 with @SachaGloor’s from post47.

Perhaps your flare is as benign as his.

I suspect most printers including mine will pass basic tests - but stretch the ones with more intense flare - like printing close the hinge side - and the parts are no good.

I think it is @RocusHalbasch’s contention that pretty much any flare at all causes observable issues (on certain parts) - and that the 25micron rook will print with quality issues on most F1+ printers.

This may well be the case - I don’t know - no-one’s posted perfect 25 micron rook yet and I haven’t managed to print anything at all at 25 microns, I gave up and focused on 50 microns where I could quantify flare effects. Personally I would be happy enough with results like Sacha’s - and to be able to use my entire build platform - which I can’t currently.

2 Likes

Hey @KevinHolmes could you try the rook at 50um? I would like to see how it compares to your test print.

I haven’t had a chance to find the file through all the posts, I also have been using a 3rd party resin that is friendlier for printing flat objects but it seems to be limited on printing tall objects since the blockers tend to settle quickly so I am doubtful it would be a good comparison. Reason I tried another brand is because the resin is much thinner making peeling much easier on the machine. So far this is working out really well. The other resin also hasn’t given me any blowouts on objects with rims such as coins when printing flat.
The parts are also semi flexible so they are not nearly as fragile as the Formlabs resin which did give me some issues when making molds where the parts may crack.

I will be ordering some more Formlabs resin along with a tank soon and will test your rook on that.

Here is the first of my simpler test prints this is a replacement for the Rook. It is smaller, has way less volume and is simpler so you know more clearly what the problem is. Still print it at 0.025 with no supports. I haven’t tried this version yet I printed one with a different rotation and thicker walls and it worked so this one should work even better. Anyway this should probably be used instead of the rook from this point forward,

1 Like

I have taken down my variant of the rook as it is no longer needed and is not my work. If you need my version which was modified for the Form1 let me know and I cand give it to you.

This first pic is of the side where my rabbit ears are. You can see the rough and bumpy texture.

This second pic is of the opposite side where I have a small defined carret. You can see the carret causes the spirals to bulge.

This third pic is of the side in between the two it’s mostly clear and what artifacts it does have are attributable mostly to the bad sides.

Oh and lastly the big line all the way around it near the bottom is from formlabs going back to over curing the bottom of the print. They have varied the amount of the print they do this to across versions. In different versions I’ve seen it be anywhere from 0mm to 2mm of the bottom of the print but now it is 5mm. Unfortunately formlabs doesn’t give us control over this setting so when they have it over 0mm it messes up all prints done with no supports. I’ve requested user control of it but it has been ignored.

Printed the helix out at 50% with 25µ layer thickness. Front and back had failures which is parallel to my flair but I don’t think in this case the flair has anything to do with it because I have waves which is typical of the particular resin I am using in which the blocker settles too quickly during the print. Reason I say that is because the early part of the print is fine. In this case with this resin I am sure the flair doesn’t help matters though.

There might be a mix issue with the FL resin and flairs but not nearly as severe as the resin I happen to be using. I think the blockers in the resins all settle as prints progress, why thicker is better on bigger prints. As blockers settle the laser will refract throughout the resin and part causing blobs, blips, partial cures and eventually layer failure from poor adhesion. I guess until resins can be designed not to settle with the print time it is the nature of the beast with these machines.

When I get more FL resin in I’ll re-run the test. I’m sure it will be much better than this one.

Hi

i switched to clear 02. The Tower was print perfect @0.25mm. The first try in black failed. Will post the pic later. Black is new for me. I think black is not made for printing @0.25mm.

Sacha

It’s possible that the clear resin doesn’t require much power from the laser, vs the black and grey resins, and therefore the flare doesn’t really affect it as it does with the other resins.

Add me to the list.
Less than a liter put through the machine, parts are usable by us as at the moment fine details are not required, but the surface finish has always been poor.
Thought it was something we were doing!
Marks in the pic are cm

Dave

Actually, no, the exact opposite happens, although the end result is as you describe it.
As the pigment settles on the PDMS, it blocks the laser more, so the resin doesn’t cure completely (or at all in extreme cases).

Which is something that might be possible to solve in software, actually, for resins with known settling times - all you’d have to do is gradually increase the exposure time and/or laser power as the print progresses.

Sorry I am both bad with a camera and in a hurry today. This was one of my two plus signs printed using the pattern on this thread. It was done at 0.1mm. I am doing another one now at 0.025mm and expect the experiment to show the flakes growing higher in density.

I like the way you can see them accumulate the farther along the print job ran. Yep we can now add me to the list of people with the laser flare bug.

All right a few comments.

My first comment is I would not print the test print at 50%. I built it because it showed particular problems at it’s given size. This was dependent on the size of the flare when you scale it down you don’t also scale down the size of your flare so results will be pretty drastically different, making comparisons to other prints useless.

My second comment if you base your assumption that the print failure is due to settling on the fact that the early part of the print is fine. This logic does not hold up for a variety of reasons.

The first reason is your part first fails immediately after the over-cured layers. The printer on this release of preform runs two passes on each layer in the first 5mm of the print to over-cure them. This portion will as a result print thicker and stronger. The ring around the print 5mm up is the ridge where the over-cure stops and the walls therefore get thinner. It seems clear from your second pic that the print would have failed earlier if the first part of your print had not been over-cured.

The second reason is because although the part is designed to show flaws on virtually any flare, it is not designed to show them all along the part. The point of the spirals inside the print is so that no matter what direction your flare may be oriented at some range of layers the spirals will line up to be close to parallel with your flare, in this range the problems should be noticeable, elsewhere not necessarily. So not having problems until partway up the print is not only possible but expected for a lot of printers.

The third reason is because the print first starts to fail quite noticeably at 5mm in, knowing roughly how long it takes my printer to go through 38mm at 200% the size you where printing leaves me pretty sure that 5mm point is less than an hour into your print, my guess would be closer to 30 minutes in. I know of no resin that works on the Form1+ that settles so quickly that it will result in print failure under an hour into a print. I recognize this may just be me being naive of the resin your using, but if so that is a ridiculously short time span.

My last comment is that I’ve seen waves in a lot of my prints developing this sample and for me it wasn’t settling. I can print two in a row without mixing my resin and they will come out the same, and both will have about the same amount of waves. If it was settling the second would obviously come out quite notably worse.

Wow @SachaGloor that does look good, although it’s difficult to see in those pics what the surface finish looks like. Between the clear resin and the fact that the part hasn’t been cleaned and dried. In minor cases all I was expecting was for surfaces to be somewhat rough on one side, not for complete failure. So the thing that is really interesting is that the print failed in black but is perfect in clear v2. I have printed the rook in black at 0.025mm at least 6 times without one of them failing, however it hasn’t come out without roughness once, and @Steve_Johnstone tried printing it at 0.025mm in clear v1 twice one of which it failed the other it showed roughness.

So how did the black one fail? Would you be willing to try my new sample in black? It is far less complicated and I would guess far less likely to fail, and it should take less time and a lot less resin. Either way thanks a lot for taking the time you have to give us samples from a machine that hasn’t been exhibiting problems.

Indeed, clear resin requires way less power and much shorter exposures.
I imagine the cumulative effect of the flare over multiple runs across the perimeter / multiple scans is much reduced with clear.

What is the significance, if any, the base surface is very smooth and walls (oriented diagonally) print poorly- would that help rule out tank issues, and point to the laser ?

BTW right click and select ‘view image’ to see that first one.