Been running Nylon 12 GF for over a year from the start but trying Nylon 12 this week for the first time.
While these parts were from the first two chambers at 100% fresh powder, I’m seeing major, unfinished looking,“burrs” along edges of parts.
Never seen this before on Nylon 12 GF on the same parts in the same orientation.
We will see when I bump down to a refresh rate if that changes but I also noticed that the printer does an outline around each part before filling it in for each layer. Assuming this is to minimize warping but hopefully that is not the cause. It doesn’t perform this outline pass when using nylon 12 GF.
Again first time trying Nylon 12 and with it taking thousands of dollars and countless hours before finally dialing in Nylon 12 GF… hopefully someone can save me some time n money with a suggestion.
As a starting point you can look at the Nylon 12 GF settings and see what the Perimeter settings are. Then look at the differences between the Fill vs the Upskin and Armor. Take that offset and apply it to the base setting for Nylon 12.
Interesting… However, this being my attempt to move from something I had to tinker with for over two years to get dialed in (Nylon 12 GF)… I assumed that Nylon 12 was the standard and “dialed in”… not sure why I would need to adjust anything in this locked down ecosystem.
Never knew that setting was even a thing…not even needed on Nylon 12 GF… so interested to see what Formlabs says
Yeah, I agree that you shouldn’t have to tinker with the settings when it comes to the default materials. At least they now give you the opportunity to change these settings.
Oh my! But this is where I get suspicious.
The component quality shows a surface that I think is clearly too rough. Here is a comparison photo of our PA12 parts and the Formlabs PA12-GF test print. Our PA12 surfaces are actually quite smooth and no less rough than PA12-GF parts. Prolonged blasting should make the PA12-GF parts slightly smoother due to the glass filling, but not by much.
Such a rough surface would make me check both the optics of the laser and the bed temperature.
This overhanging edge happens when the part has an edge that tapers to 0 and is oriented upwards so that the last laser layer is on top of it. I assume that this was printed with the default settings for PA12, not with the speed settings? The v5.1 settings now make a very good impression on us as far as the top layer is concerned and the “dimpling” is hardly visible.
However, we also had to adjust our bed temperature for this. We have been using an offset of -1.5 °C on our Fuse1+ for some time now and are doing quite well with it.
I would recommend cleaning all the optics once and making sure that no smearing remains anywhere. Then I would reset the temperature offset to 0 and print the calibration print again. Here I would expect to see a better surface on one of the parts.
Then also try the v5.1 settings. I don’t really see a reason to use the default settings on the PA12 at the moment.
Very good info and exactly the questions I’ve been having in my head.
Chamber temp was still at a + .75 from my Nylon 12 GF tune. I immediately thought that may be the cause of the burred edge… however I never saw “outlining” of the parts done before the fill so I was afraid that may have something to do with it. Does not perform that when using nylon 12 GF.
All my tests were at 100% refresh however a print job at 40% refresh was just done and I need to post process. (As per the other thread…I have no idea yet what refresh will be acceptable for small wall strength and for small 2mm springs which I need to do at 70%+ with nylon 12 GF)….
But all these were done in Nylon 12’s “Default”
I almost tried “Legacy” when I thought the outlining of the part may be causing the burr edges… I also started to creep the chamber temp down n this last 40% refresh chamber was done at 0 offset. I will say this time I could barely see the surface armor being done on camera. Was very faint…
But your saying the Speed setting has also been giving you consistent and good results? As per our other thread and my parts what do you suggest? Love to get any opinions.
For the burr edge on a “zero edge”… it made me think of always putting a slight chamfer on all my design edges…but again I never had to do that with Nylon 12 GF as the edges always looked refined. These Nylon 12 parts look like they all got “flashing” from a bad injection mold.
*Also as per other thread… my GF parts look way better than Formlabs sample part and so far better than Nylon 12 with my tests. My nylon 12 GF has a slight sheen and so far smoother almost like it’s been peened or vapor hit.
Hopefully I can get Nylon 12 looking at least comparable.
Now that the Blast has been out a while…like to get any long term feedback on it and the vapor smoothing add on?
By “outlining” is probably meant the optimized print line follower in v5.1? Here, the outer wall is printed first and then the interior. This can also be seen on all expensive industrial machines. I would assume that this reduces thermal distortion for larger components.
With the first versions of the speed settings we had certain problems with the accuracies. Since v5.1, however, these are significantly better, which is actually also stated in the description of the settings.
This protruding edge should be more visible with the default settings anyway, as the laser speed is slower here. This means that the last layer on the edge is lasered longer than with the speed settings. We haven’t had artifacts like this for a while now. In my opinion, the higher speed with v5.1 makes almost everything better or just as good as the default settings. We therefore only use the v5.1 settings.
We also print everything at a rate of 30%. We keep the printer and the sift completely clean after each print. In the printer, we also completely vacuum out the flippers on the left and right and keep the interior completely clean so that no powder residue from the previous print comes into contact with the next print. This results in a constant small amount of powder waste, but with a rate of 30% and a success rate of almost 100%, printing is the most fun.
The Fuse Blast would be another dream purchase, also because of the polishing function now, but as with everything, the budget often says no at first Not if you can just stand at the blast cabinet for two hours when needed and then get the bigger job done.
Vapor smoothing has never been necessary for us. Together with the cost of such equipment, this is not on our roadmap for the time being.
Will give v5.1 a try on the next print. My concern with refresh that low is the point at which it gets “brittle”. Nylon 12 GF starts to for our parts below 60% refresh where we get more spring snaps. With the additional flex of Nylon 12 just will have to find that “push/pull” in refresh to get the same durability which with the additional flex before breaking I assume will at least be below 60%.
I am noticing that all the burr’ed edges are edges that were facing “down” in the chamber
Note: I also do a full clean up and vacuum out the trays to include… running three full Doser rotations while vacuuming out the lower level of powder in the hopper that was closest to the heat from the last job. That has been key in being successful with Nylon 12 GF… that and a 70% refresh… $$$$$$$
We carried out our own tests between the rates of 30%, 50% and also 70%. We didn’t notice a big difference, at least with our spring elements, so we said that we would use the lowest possible rate.
I find it very strange that the bad edge should be the bottom edge, as the bottom edge is actually always the cleanest. The design guidelines also state that the visible side of SLS components should actually always face downwards.
Out of curiosity, do the pictures show what packing density you are working with?
Here is a picture of the packing density we used to print the component shown.
We are close to a packing density of 30% for all our jobs. This chamber was just a test sample of different parts. We have run these chambers a hundred times at least with Nylon 12 GF in the same orientation with perfect results.
Bad edge being down… was why I thought maybe that “outline pass” was too thick on the lower edge. As again Nylon 12 GF does not perform an Outline pass.
So on further tests at a 40% refresh and the chamber temp turned down to zero offset from the original +0.75 tests and on the Default mode… I am still getting ugly fedge flashing and I can confirm it is on the lower edge (facing down) in the chamber.
I think I have already read from several users that they have lowered the bed temperature for PA12. The 0°C standard offset setting was too high for us.
I just looked again and saw that the calibration print is only available for PA12. That means you couldn’t properly calibrate the PA12-GF to the correct offset, could you?
With all the problems described with the molten powder cake, I would have expected a negative offset. Now I read that you had a positive one?
Based on another user report, it actually also says that the lower the refeed rate, the lower the temperature should be. Our first prints on the Fuse at 100% were also without any problems. It was only when we went to 30% that the problems started.
I would therefore expect that PA12-GF should also be able to be printed at the specified 50% rate if the bed temperature is adjusted accordingly.
We used this calibration print from Formlabs to calibrate PA12 at 30%:
Yup, that’s what I thought
Since Formlabs doesn’t provide any information about the many support articles or news about new articles, I occasionally look through all the articles to see if there’s anything new.
You can definitely pick up a thing or two if you read through the articles. Even those that are not obviously of interest to you, such as printing with other materials, various printing defects, etc., can be useful.
A lot of support people at Formlabs do not know what they have published on their own website… and some things on the website are self contradictory to support agent stated practices.
To be honest I’ve derived more value from website info than I ever did from their support agents, at least on SLA side of things.
edit: also from reading the above thread it seems to me that Fuse 1+ settings development has begun to diverge significantly compared to Fuse? It seems that Fuse settings have stalled… Shame as I was considering Fuse and it seems that materials I would actually want to print such as Nylon tough and settings will not be ported over. Fuse 1+ is out of my budget sadly.
Hi @LEADNAV, our services team has been discussing this and recommends using
v5.1 settings, which should give you the best part quality outcome.
If you are open to turning on Print Settings Editor, then you can create a custom setting that uses reduced “Perimeter Upskin Laser Powers” and “Perimeter Downskin Laser Powers” which will help you adjust the exposure at the edges of their parts.
Will run a X/Y scaling calibration now then I will try some of our parts with the Speed Optimized v5.1 settings.
Formlabs… why couldn’t we have this temp calibration print available for other materials like Nylon 12 GF? Would have been very nice… especially when these “test prints” cost hundreds of dollars
I would agree. -1.75 or -1.5. Our test looked very similar, a few cones difference in the -2 test. After we set our Fuse1+ to -1.5°C, we also ran a new XY calibration print.
It is important to print the calibration print with the v5.1 settings as well. I think the Formlabs file did not allow this at the time, so we exported the file from the job and printed it with our own settings.
It says in the settings descriptions that the v5.1 settings were tweaked again in the last version with regard to accuracy. I therefore do not think that it makes sense to print the calibration print like the default settings, where the laser path simply moves from the bottom to the top but, as with the v5.1 settings, also moves along the outer wall first.
Before we print the PA12-GF in the future, I would definitely want to have the calibrating pressure for the bed temperature.
I think only with the right adjusted temperature can we hope to print PA12-GF at 50% rate.
But I can already imagine why this calibration print is not available for PA12-GF: If printing at 50% rate at 0°C offset has always failed for you, the temperature window for PA12-GF seems to be even smaller. A print that goes from +2°C to -2°C would therefore probably produce a print defect in any case.
I would ask Formlabs for the part file and then print one plate at a time with PA12-GF. Then start with 0°C offset and slowly decrease until you have a successful print with the right number of cones with orange peel. Admittedly a cumbersome way of calibrating but I think this is currently the best way to calibrate the PA12-GF at a lower rate.
We tried everything for over a year as far as bed temp etc… nothing Formlabs support had us do fixed it.
Bumping up to a 70% refresh rate fixed EVERYTHING. I would say we were at a pretty high success rate at that level. Obviously that is very expensive…hence us trying Nylon 12 to see if it is “good enough” for our applications.
Not if you can just stand at the blast cabinet for two hours when needed and then get the bigger job done.
