You are, I think, correct with idea (1). That is likely what is happening, however I think you are wrong with your assumption that it is not caused by laser aberrations as you state here:
I think you are missing the actual pattern and the correlation it has to your idea (1). The patter is simple these artifacts are only prevalent on faces oriented in one or two directions, these directions happen to coordinate with flare on the laser spot. So the current running theory is that in a single pass this flare which is weaker than the center of the laser partially cures some of the resin and over a few layers does what @Andrew_Miller describes which is also pretty much what you suggested with idea (1). The main difference here is that we have the knowledge of the reason your idea (1) is occurring and the knowledge that only some printers seem to have this problem while most others print fine. We also know the same resin is being used on the printers that work and those that don’t. So your conclusion that the only solution is reformulation of the resin seems completely incorrect, it seems a better solution is to adjust the laser correctly so that the flare is not so severe it causes extra curing, this would likely make the few printers that don’t work correctly work as well as the many that do. This is not to say that a tighter formulation of the resin would not also be good. I just don’t think reformulating the resin is an appropriate solution to a few machines having significantly more poorly configured lasers.
I printed the cavity part of the mold in v1 grey resin over the weekend to test the FL theory that this issue only appears in some parts printed in clear resin at 0.5mm layer heights. Something they are hoping to fix via their material setting in the software.
My 1st attempt failed due to part of the base lifting from the build plate part way through. The base was very thin & I felt that this was due to me carrying over the fine tune settings on the platform height onto a different resin tray. With the previous clear resin tank I had this set at 0.7mm lower.
For my second attempt I set the Platform fine tuning height to 0.5mm and increase the base thickness by 0.5mm. This worked and I got a completed print.
Constants for both prints
Ambient room temperature 18c – 20c
Resin : v1 grey, filtered before both prints
Layer height : 0.5mm
PreForm : printed from exactly the same file
First Attempt
Base of the print ( the first bit of the part printed, way before the failure)
Porosity affecting the forward (good) face of the print
Poor surface quality on the rearward face of the print. This is the face that is perpendicular to my “laser flare”
Second Attempt
Porosity affecting the left facing surface
Porosity affecting the bottom left hand face.
Poor surface quality on the rearward face of the print. This is the face that is perpendicular to my “laser flare”
Good surface quality on the forward facing surface, the surface opposite my “laser flare”
Very poor base. The first couple of layers were firmly attached to the build plate. Next layers look like they were part cured. The top layers of the base were fine.
Is this a symptom of laser flare or simply a case of needing to lower the build tray a bit more???
I suspect there is more porosity in the print that I cant see, but want to use this part so breaking it open is not an issue.
Try lowering the platform a little more. Here is a simple calibration print. Print it without supports @100um. Measure the thickness of the four squares, now adjust your platform height so the thickest one will be close to 2.0mm thick. Print it again to check that they are about the right thickness. Do this for each tray you use and mark each tray with the appropriate platform height so you can set it correctly whenever you change trays.
Is there a way of adjusting the build plate or resin tram mount if I find that they aren’t parallel ?
I sanded and polished the 2nd grey mold test piece last night and as I suspected there is a lot more porosity in the in the part. A realty good indicator is trapped polish in the porous surface that shows up as white when dried. This is even more evident when you back light the part.
The faults aren’t as bad as what we saw with the clear resin. I would guess that this is due to the grey pigment blocking out some of the laser flare. I strained the resin afterward and there must have been a good 2 table spoons of part cured resin that I binned.
I’m not sure how changing the material setting in the software is going to correct this. I think this is a strange approach as this type of fault only affecting a few of us. That is unless the problem is a lot more wide spread then we realize. Hopefully the new software release will be out this week for us to try.
@Steve_Johnstone there is - but whether build platform and the vat surface is parallel is not important if you have the platform lowered enough, and base set thick enough.
What is important is the whether the vat surface is perfectly perpendicular to the z-axis, if it’s not then then your parts will be slightly skewed (like a parallelogram). Apparently FL have “specialist” equipment for calibrating this - but in the end it all goes out the window the first time you replace the silicone yourself in a vat.
Before the fine-tuning option in Preform I found out how to re-level the vat - see here - https://www.youtube.com/watch?v=CB9OeCxT04U as I say though - platform height tuning should make that method mostly irrelevant.
Regarding porosity - I have come around to realising that a bad laser spot profile can cause this. I have been experimenting a lot the past few weeks with extra lenses and some experiments resulted in hollow parts filled with liquid resin and no bubbles - and I know that in those cases the laser spot profile was not ideal.
It would seem that an off-centre cylindrical lens can cause both flare, and a misshapen laser spot, and that a misshapen laser spot can cause uncured resin and hence porosity. I’ll be writing this all up in detail in a couple of new threads in a few days, I haven’t quite finished with the experiments yet.
Hey @Steve_Johnstone, the porosity could be caused by your resin is not getting enough exposure-power delivered to it. (Maybe because the flare puts the power in the wrong place.)
Have you tried increasing your laser power with that handy laser-exposure table I posted around the forum?
Unfortunately the platform is from my experience quite close to perpendicular to the z-axis, thus if the tank and platform are not parallel It is usually becuase the tank is not perpendicular to the z-axis. Formlabs has told me recently that currently they do calibrate for this with fancy equipment, however I have been told in the past, variously that they intentionally made the tray not parallel, and that they didn’t try to make it parallel but used fancy equipment to correct the resultant dimemsional problems both in the calibration and in the software, I have no idea when their process changed, or how reliable their new methods are. However if the part where they said they used fancy equipment to correct the resultant dimemsional problems both in the calibration and in the software is true, and they did that on your printer then straightening the tray to be perpendicular to the z-axis might introduce skewing.
So to test for skewing print a tall cylinder, at least about 15mm wide, and at least about 100mm tall. i designed one with an arrow on top pointing toward the front of my printer for thIs purpose but I don’t have it uploaded yet. Now set the cylinder on a very flat surface with the top (the end that was not touching the platform when it was printed) facing down. Now set an engineers square next to it and rotate the cylinder slowly while holding the two together. If at any rotation the two diverge you have skewing. If you turn it till they diverge the most the angle of divergence is the angle of the skewing, and you can use the arrow th figure out the direction of the skewing.
The other end of this print will show you the angle and direction of truncation you get from printing to the platform.
There are other enlightening measurements you cam take of the above test as well. Measure the diameter from multiple angles, and if you can get my previously posted leveling print to come out so the middle of you build platform is printing 2mm thick objects correctly check the height of your cylinder rash well. The last I did all of these I was shocked how far off everything was. However when I told Formlabs they told me it was all within expected error tolerances, and they don’t support printing to the platform, even though that wasn’t relevant.
@RocusHalbasch, thanks for all the great info. I think I’m leave this side of things until the laser flare issue is resolved. One thing I’ve learnt over the years when trouble shooting 3D printing is to keep the number of variables you change during tests as small as possible.
@JoshK, thanks for the heads up reference your chart .
Out of interest, how did you arrive at the exposure settings for the various FL’s resins?
You’re the first to ask I printed a layer of a model for with a sheet of paper in the tank. One print for each setting. I had my Canon EOS-60D camera in full manual mode and on a tripod above the printer. (15 second, ISO 100, f22, K3900)
Then I used software to read the average pixel value of the green channel in a region of the image I was interested in. And put the data into Excel. Viola, a very accurate table of laser intensities.
@Steve_Johnstone, As you see in the images I just posted the center of a layer gets less power than the perimeter. This is why I believe your porosity inside is caused by not enough laser power.
Thanks
If they are only changing the clear@0.05 setting then I would imagine they are just scaling it a bit. That is another issue FormLabs told me I was wrong about.
I bought my first clear resin August 2014 and noticed right away only Clear@0.05 was not printing successfully, but support told me I was wrong back then so they didn’t have to look into it. Haha.
And fast forward to the laser flare issues, support said I was wrong again, so in November 2014 I designed a laser choke to prove my point. I first mentioned that in the following post. The day I sold my printer support was still telling me I was wrong. Now they are coming around.
And as Rocus showed everyone, support would rather “handle” you than “handle” the issue until it becomes undeniable. Then they start looking into it.
And someone on these forums said a DSLR was a useless piece of kit… Ok just kidding, but seriously I think that the potential DSLRs have for analysis of the laser is still under explored. I would be interested to see a mid ground of this with my laser spot technique. Basically just use ISO 100, and a 15 second exposure, and print one layer like you did but progressively change the the f-stop in full stops. This would result in a flare maps at different intensities, which could show more clearly the cumulative effect of the flare relative to the desired print for a given geometry. It could be pretty neat stuff.
<img
I printed a layer of a model for with a sheet of paper in the tank. One print for each setting. I had my Canon EOS-60D camera in full manual mode and on a tripod above the printer. (15 second, ISO 100, f22, K3900)
