Laser Flare & the effects on your prints - Part II


#1

Continuing on from -

After upgrading to Perform 1.8.2 I was hoping to see some improvement. Sadly so far this is not the case.

@KevinHolmes (I think) picked up on something that had me thinking. It was his observations that the porosity often occurs on the support side, especially where the supports are closely clumped. I more closely examined the clear cavity mold piece and this seems to be the case.

I think this happens for the exact same reasons why we see it on @KevinHolmes cross test piece. Regardless of orientation, the supports are always printed vertically and therefore the laser flare is part curing the same area of the resin tray whilst the support is been printed.

Clump a bunch of support together, the problem multiplies and there is a greater chance of part cured resin causing porosity in your print. Similarly, the taller the supports are the greater the problem.

I designed a flat plate (100x80x3mm) to test this theory. I orientated at 45 degrees to the build plate and then purposely grouped the support as seen below -

Both tests were printed at 0.5mm layer heights, in V1 FL Grey Resin

Test 1

I strained the v1 grey resin and printed it in the “correct orientation” with the longer supports closest to the hinge.

Test 2

I rotated it 90 degrees and printed it with the long supports facing the back of the printer.

Results

Because my laser flare faces to the rear of my printer, I was expecting the 2nd test to be have more of the porosity. This wasn’t the case, it was the first test piece came out really bad, covered in porosity.

I think was because the longer supports were close to the hinge and there was less chance for the part cured resin, caused but the laser flare, to get washed out of the way during the peel processes.

Even this morning, 2 days after the test, there is still resin seeping out of the porosity.

Test 1 Results

Test 2 Results

What Next

I’m going to print the same tests in FL V2 Clear and then printed is going to have to go back as I don’t see what else I can,do


Retro Form 1+ Guide
#2

@Steve_Johnstone interesting stuff - I missed the conclusion though - do more supports cause worse porosity or not? Test 1 seems to show the worst effects in between the support strips?

Also have you tried printing grey using clear settings?

Clear has longer exposure time than grey, and I’m now thinking that porosity like this where trapped bubbles can pretty much be ruled out - is caused by incompletely cured in-fill resin being stripped out by suction during peel, and co-incidentally, making flaking worse.

So the extra exposure could help - although depending on whether the extra exposure is offset by the reduced free-floating partially cured in-fill or not - flare flaking might be slightly worse.

Not sure really what they did in 1.8.2.1 - print times for clear 50m seem pretty much the same, so it doesn’t seem as though they increased exposure time.


#3

I’m not really sure what my conclusions would be, apart from I think the peel wave motion, helps move the resin abound a bit. If the the part cured resin is moved out of the way then obviously it can’t affect the print.

Just grasping at straws at the moment. :flushed:… very frustrated.


#4

@Steve_Johnstone - so will you try another experiment with grey running as clear? I’ve found that it can help …


#5

@KevinHolmes I’m hoping to print the clear version this evening / tomorrow and I will take it from there.


#6

From my experience, it looks like a support issue. I think the thin wall is deforming when the printer peals a layer. See this post:

I would try your test print again, but manually add a grid of 10mm spaced supports to the entire slopped side.


#7

Did you try printing that shape at an angle instead of 90 degrees to the hinge?
I have had similar problems with the gray resin and worse with a 3rd party resin. Thing is it doesn’t always happen and orientation seems to be a big player.
Reducing the peel force seems to help tremendously. It seems the more square the part the more prone it is to failure. I say this because I had a part that had no flat surfaces, part was huge and had no flaws on any of the surfaces, yet smaller flat pieces give me fits. It it was the flair causing it then at some point the large part would have some surface problems.

Big flat sheet like you have probably have problems stemming from the supports. There seems to be too big a gap between them to hold up from the peel.


#8

When I orient parts, I think about how the shape of the layer against the PDMS will peel, e.g. if I was to manually peel that layer with a putty knife in the direction the machine peels from, how much force would it take? Generally speaking, you’ll want to optimize an average such that the bulk of the width on the PDMS at any given time goes in the direction of peel to hinge, or in your first example /not/ how you’ve arranged it. With the long layer direction going front to back on the platform, you’re forcing the machine to have to peel the entire length at once, for only a short distance. That seems to be what’s causing most of your issues.

Let’s say the x, y, and z dimensions of your “part” are 100, 80, and 3 respectively. Now imagine it getting printed and pulled up from the tank, and you are looking at it from the front. The 80 dimension should be oriented 90 to the hinge, pointing straight at the peel side. The 100 dimension should be pointing down @ 90 degrees from the platform, and the 3 dimension should be going front to back of the printer. 100 and 80 can be reversed, but having it ‘taller’ puts less stress on the part as less surface area is getting built per layer (but it’ll take longer).

Now, to optimize the stresses, the part should be rotated at least 30 degrees (as if pivoting the hinge side of the piece down with the locus being the 3mm edge at the platform and the peel side) such that the first layer of the piece to print is just the 3mm corner edge near the platform and toward the peel side. The last layer to print is 3mm layer at the hinge side. This will place all supports on the 3mm edges, leaving the faces clean, and minimize stresses during the peel process.

The peel happens like a lever, meaning the peel side moves down further per unit time (stepper rotation) than the hinge side. Having the last layers near the hinge means the peel is slower, and thus gentler.

try that orientation, and see how that part comes out.

-C


#9

Hi @Charles_Rogers. Yes I agree, the support placement isn’t ideal. I purposely arranged the supports this way to test my theory that clusters of supports creates localized clumps of part cured resin due to the laser flare i have. It was great to read that you managed to print your part successfully. :smiley:

Hi @KenCitron. The second test was 90 degrees to the hing and printed successfully, with minimal porosity. The forward facing surface quality was good, but the surface facing towards the rear was poor. I believe the poor rearward faces of my prints are a product of my laser flare orientation. I’m 100% certain that if I were to rotate my laser 90 degrees in its mounts, as @KevinHolmes has demonstrated here, the poor surface quality would follow the flare orientation to the side.

@ChristopherBarr, thanks for the detailed advice. I totally agree with everything you’ve suggested. My big problem is that I’m having to compromise with part orientation because of my laser flare. It’s a variable that the majority of Form1+ users don’t need to consider. If this problem was affecting everyone then it wouldn’t be an issue, but judging by some of the awesome results others are having, this isn’t the case.

Formlabs support have been very proactive and I really feel that we have been working together to get to the bottom of this. Unfortunately I can’t see much more I can do my end to resolve what I consider a mechanical issue. I’m unable to use the printer for core business projects so I’m going to ask for it to be returned for a 3rd time.

My worst fear is that its returned with the same laser spot test image…


#10

I had a problem that produced the same results where the slope facing the build platform had porosity and the slope side facing the tank came out fine.

What it turned out to be was poor mixing of the resin and not the laser. Poor mixing cause inconsistent curing causing waves, blobs and voids (partly from inconsistent blocker mix and inconsistent photo sensitivity). Variations were more pronounced near build supports so my first thoughts was there was something wrong with that.

The peel process actually helps mix the resin and attributed to the better surface on the peel side.

With that being said, the laser flare will definitely be a problem where it will cure partially unwanted areas.


#11

Hi all,

Support asked me to run the same test in clear to see if the revised software setting had made any difference. Ideally would have liked to reprint the cavity mold in the same orientation etc. but its a long costly print so decided against it.

In the end I printed the 100x80x30mm test twice as per the previous tests in grey resin.

Both printed @0.05mm LH, Clear v2 resin

Test 1
Orientation

Forward facing surface is good

Rearward facing surface is poor, some flaking but no porosity :smile:

Other observations -

  • There was a large amount of part cured resin after in the tank that required filtering.

  • There was no porosity at all in the print which was great.

  • Compared to the front, the rear face quality was very poor.

Test 2
Orientation

The Print failed

Looks like this is where things started to go wrong - porosity caused by laser flare or the peal process, who knows.

Other observations -

Its difficult to say whether the new software setting for the clear v2 resin @0.05mm LH have made a difference. I’ve some new large prints that I’ve agreed to try and print for a client, so we will see. I’m going to print them in v2 clear at 0.1mm, to give me the biggest chance of success… fingers crossed :wink:


#12

@Steve_Johnstone,

Interesting results for sure! Just to play devil’s advocate here, do you have to print this at 50 microns? Have you printed it at 100 yet? Have you tried printing the plate at the same orientations with supports that are less dense and evenly spaced across the surface of the part?


#13

The Test 2 one seems like a peel issue.

It started right between the supports. That span between the central set of supports and the side ones is too long for such a thin peel profile. The supports in the middle and on the sides keep it stretched, but then the peel rips it in the middle.

Think of it like holding a piece of sticky tape stretched between your fingers, then letting the middle stick to something (your table) and pulling. That’s essentially what happens during a peel with this kind of geometry.

I never had these issues with my DIY printer, as i was solely using MakerJuice resins, which cure to a very hard and rigid (but extremely brittle) state - they’ll sooner damage the PDMS than rip like that. Formlabs resins are way more flexible and “floppy”, which i’ve come to realize while trying to print casings with very thin, straight faces the past few days. I haven’t had any rips (because i’ve been printing much smaller objects with shorter unsupported distances), but i’ve had plenty of bad surfaces and completely bent corners and edges - exactly because of the process i’ve described.


#14

Hi @Aaron_Silidker,

For the mold tools, yes I need to print at as fine as I get away with as it greatly helps with post finishing. This part is just a test to replicate the main cavity part surfaces that I previously had problems printing in both clear and grey. I purposelessly grouped the supports on the test piece to test my theory described at the top of the thread.

The cavity mold part shown below was printed in clear & greay resin, in a very similar orientation to Test 1.

Hi @Aaron_Silidker,

I totally agree with you. This is nearly the worse way you could orientate the model, but printed it anyway, as a comparison to the grey resin test.


#15

Not great news.

The first part printed for the client at 0.1mm in clear v2, was a mess. It looks to have failed at about 80% and to me, it looks like the problem is getting worse.

I spent ages orientating this part to insure minimal peal force is required during the print. The wall thicknesses of the part are between 2.5mm & 3mm .

Orientation of the part.

Approximate failure point

Loads of part cured resin on the rearward face (the face perpendicular to my laser flare).


Some serious flaking running parallel with the my laser flare. This large piece was well before the print failed.

Poor rearward facing surfaces

I’m now seeing flaking on the supports again orientated parallel with my laser flare.


Is this forum going the same way as Makerbots?
#16

@Steve_Johnstone so what’s the word from FL on a replacement machine?

Couldn’t tell from those shots, was there any porosity as well as the flaking?

My machine seems to be behaving very well now with the choke in place; flaking is pretty much gone, and I run prints at one higher exposure setting to make sure I don’t get porosity. That is - I set preform to print in clear when I actually have grey in the vat, black for clear, and so on. See @JoshK’s list of resin exposure times here Resin list in order of exposure/curing?.

Not only do I have almost no flaking, but surface slime also seems much reduced and requires less IPA brush scrubbing to remove. I even suspect small holes are more accurate; I just reprinted my glasses arms and they slotted on first try, whereas previously I was having to carve out gummy surface resin from the fixture slots with a scalpel.


#17

Just wanted to chime in here. Please don’t dismiss the condition of the resin tanks. I just changed my resin tank and got perfect prints. I noticed some clouding on the old tank, so decided to change the tank and try the same test file and it was perfect. I tried it on 50 and 25 microns and got the same perfect results.

I still have the flare on this printer (replacement), however with the new tank, it prints perfectly. Will keep track of how long it last and before I see any issues.


#18

There wasn’t any porosity in the print at all. Your printed glasses arms look awesome. Are they as printed or have they been sanded etc?

I don’t think fitting a choke to my laser is an option. It’s not because I don’t think it would work, it’s because I shouldn’t have to. For the same reason I haven’t tried using different material profiles.

@Monger_Designs, Thanks for the advice and to be honest I nearly broke out a new tank this morning to retry the print. I’ve decided against it as at some point I need to draw the line. This part was printed straight after printing the Test 1 clear, which printed well, apart from the poor rear facing surface. The only thing I did was strain the resin.

Here are some more pictures and observations of the part now that it’s cured -

The first 30mm of the part printed fine. I expect that this was because early on there is little part cured resin build up. Form this point the print gets significantly worse, as the part cured resin builds up over time, until it fails.

The side of the print facing the hing and running parallel with my laser flare printed fine. In the picture below please note the orientation on the flakes in the supports. They all face toward the rear of the printer, in the same direction as my laser flare. This is the same for all the affected supports.

The flakes on the supports are something new. I wonder if the changed the material setting, longer exposure perhaps, for the 0.1mm clear resin as well???

The poor rearward facing is a lot sharper and has an almost sandpapers texture.

Toward the point of failure flakes appear on the good, forward facing surface.

I think there is more then enough evidence here to prove that the printer hasn’t operated as advertised from day 1.

I’ve had 3 months of this and all the printer is doing is adding cost to my business, not value. I’ve worked really hard with FL support but seem to be getting no where. They’ve had the printer back twice for “calibration / repair” so sending it back a third time, for more of the same makes very little scene. They wont acknowledge what we in the community are referring to as laser flare, so can’t see them changing the laser. Therefore, there is something fundamentally wrong with my printer and I have asked for it to be replaced.

By replaced, I expect a new printer. Not a repaired or refurbished one, but a brand new one. I don’t at all feel confident having a second hand printer and having no idea of it’s history. At least this way I will have a good chance of getting a printer with a decent laser.

I apologize for my rant and none of this is directed at any of the support team. I expect they are just following company policy.


#19

@Steve_Johnstone I agree you shouldn’t have to try repairing the printer yourself and that you should get a replacement - which is why I asked how it was going with FL on that front :worried: keep us posted ?

As for my print - no finishing other than a light scrubbing in the ipa bath with a baby bottle brush.


#20

That awesome and where I would like to be.