I just gave it a shot, at 0.05mm, and I'm pretty happy with the result 
This is one of the shoes of my character. because of the long, slightly curved point going upward at a very low angle, it was a good candidate, you could see the layers a lot at 0.1, like on the top of a sphere. In the end, it makes the other artifacts only more visible (especially the backslash echo) I had a raft built in the model.
Some of those tiny frogs are insane!
I recently ran the this at .05mm. Didn't do any fine tuning or anything. I just plugged in .05mm layer height in Cura.. There's some blobs on the really thin parts and the very top part got a bit iffy, but it worked pretty good for the most part.
(I think the print head was too hot and fast at the top which is what created that swirly thing) The two other dots under that are suppose to be there though, it's in the 3d.
Small note it's better to use 0.06mm then 0.05, because of the stepper motor on the Z and how the steps divide on that.
I've been printing quite a bit at 0.06mm layers (which is also the high quality setting in Cura these days) prints start to become smooth at that point, which is really awesome.
Bed leveling is only critical if you don't keep the default 0.3mm initial layer.
Thanks for the tip, that's probably why I have seen more people printing at 0.04 and not 0.05.
That's good to know.. will have to try that.
Small note it's better to use 0.06mm then 0.05, because of the stepper motor on the Z and how the steps divide on that.
I've been printing quite a bit at 0.06mm layers (which is also the high quality setting in Cura these days) prints start to become smooth at that point, which is really awesome.
Bed leveling is only critical if you don't keep the default 0.3mm initial layer.
I always print at 0.08 because that's the default in NetFabb, however when I got my machine (about 2 years ago now!) I printed as low as 0.04 with no issues... Basically freshly built with no tuning whatsoever... And that was on the old extruder and hot-end. The 0.1mm layers advertised for the UM is highly conservative in my opinion and any properly assembled machine should be able to easily achieve better... I'm not convinced that the same is true for other machines like Cube-x, Leapfrog, or MakerBot's replicators... From what I've seen they really seem to be at their limit at 0.1mm..
Regards,
Troy.
I always print at 0.08 because that's the default in NetFabb, however when I got my machine (about 2 years ago now!) I printed as low as 0.04 with no issues... Basically freshly built with no tuning whatsoever... And that was on the old extruder and hot-end. The 0.1mm layers advertised for the UM is highly conservative in my opinion and any properly assembled machine should be able to easily achieve better... I'm not convinced that the same is true for other machines like Cube-x, Leapfrog, or MakerBot's replicators... From what I've seen they really seem to be at their limit at 0.1mm..
Regards,
Troy.
UM 100micron used to be unique. Now everyone is starting to do 100micron. So we are switching our high quality setting to 60microns. There might be some places that still show the UM as 100micron, but we are updating most of those to 60 when we spot them.
We can also do 20 micron, but that's really just advertisement, as going from 40 to 20 micron doesn't yield real visible quality improvement.
I found a surprisingly noticeable difference between 0.1 and 0.08. At 0.1 the quality is great and it feels smooth, but you can quite plainly see the layers... However at 0.08 (which is a very small difference) the layers practically dissapear. Between 0.08 and 0.04 I found the difference only noticeable under magnification.
I haven't tried Cura for a while, I'll give it a go at 0.06. I'm also looking forward to trying out your new slicing engine, but I haven't had time for anything lately.
Cheers,
Troy.
So, is it generally better to have intervals of .02? .14 better than .15 mechanically?
So, is it generally better to have intervals of .02? .14 better than .15 mechanically?
No. It is a theoretically better to go in multiples of 0.001875mm
Zlayers
Thread pitch
3
mm
microsteps per step
8
steps per revolution
200
micro steps per revolution
1600
mm per microstep
0.001875
0.001875 divides nicely into 0.03 so for ease you can go multiples of 0.03 but I'm not sure how much difference it makes. 0.03,0.06,0.09,0.12,0.15,0.18,0.21
I've been mainly going for 0.12 myself and would mainly just go for 0.06 for a finer print.
Finer will always be better though down to a point. Under a certain level the print gets whispy and you can't see the layers and creates more warping.
Oi, them maths!
I like 0.03 as a rule of thumb. 
I'll use these settings to see how it goes.
I'm having better success with .1mm than .2mm. That being said, early on after my build I winged it and printed a material feeder gear in .05mm (Brass/Gold). Here is the result, which I was extremely happy with. I'll try .06 next time.
lol. Well from the top view anything you print with a flat top will be indistinguishable (no better) at .05mm versus .2mm. You should have taken a picture of the side!
Also, if you want to increase resolution, you may want to double check the diameter of your filament.
I usually never do with a .1 print, but on a higher resolution it is easier to see over extrusion.
Pic of gear from the side -
As an example, here's a print I did yesterday at 0.06mm layers. Generally I print at 0.2 - 0.1mm, but recently I've been experimenting with thinner layers.
This was from a 3D scan I did, sliced with Cura 13.06.2, and then printed pretty quickly - 95mm/s, 2 loops (of my 0.65mm nozzle), no infill, no support, 0.6mm top, but with the solid base not printed. I was pleased to see that the overhangs mostly printed fine (just a few droops to cut off under the chin), and even the top of the head closed up perfectly. Total print time was about 2 hours.
For figurative pieces like this, printing without infill is definitely the way to minimize print times, and not printing the base saves a lot of time too, where you can get away with it - and it helps reduce warping stresses by avoiding the solid mass of plastic.
Hi, very interesting, it looks really good. Since I only do figurative stuff, I'd like to try printing with no infill. For small part, infill may not be needed indeed. If I print a character, I probably want to have infill in the legs so they can support the weight. On the other hand, if everything else is hollow, there would not be that much weight to support
But I don't have custom nozzle like you. So you recommend I do 3 loops instead of two?
I am also trying out 0.06mm prints with the cute octopus (http://www.thingiverse.com/thing:27053). But I'm having a problem with the fan side of the print. At such layer heights, the fan side seems to cool too quickly and becomes very "dry". eg. Using black filament, the fan side of the print turns whitish, while the non-fan side is a shiny black.
I'm curious if you have seen such an issue with your print? What temp did you print at, and when was the fan turned on? For the cute octopus, I have not succeeded in getting a good top (the fan side has a hole as the layer seems to cool before being laid down properly)
As an example, here's a print I did yesterday at 0.06mm layers. Generally I print at 0.2 - 0.1mm, but recently I've been experimenting with thinner layers.
This was from a 3D scan I did, sliced with Cura 13.06.2, and then printed pretty quickly - 95mm/s, 2 loops (of my 0.65mm nozzle), no infill, no support, 0.6mm top, but with the solid base not printed. I was pleased to see that the overhangs mostly printed fine (just a few droops to cut off under the chin), and even the top of the head closed up perfectly. Total print time was about 2 hours.
For figurative pieces like this, printing without infill is definitely the way to minimize print times, and not printing the base saves a lot of time too, where you can get away with it - and it helps reduce warping stresses by avoiding the solid mass of plastic.
That print was done at 230ºC, 95mm/s, with the fan on full from layer 2. I've heard of people getting good results with just two loops of a standard nozzle, and indeed yesterday I reprinted the figure with just one 0.65mm loop at 150mm/s and 0.045mm layers, and got pretty good, but not perfect results; some of the overhangs didn't print.
I think two loops would usually be enough for sure... you just need to experiment. Minimum layer time is going to be critical in ensuring that there's enough cooling - although I find that with 0.045 to 0.06mm layers, 2.5s is long enough for it to cool, since there's such a low volume of hot plastic being added on each pass.
I tried yesterday to print with no infill, 3 loops (so 1.2mm shell), 0.1mm per layer, 50mm/s, 210* and it worked fine. I used supports though.
It's super light and seems strong enough for sanding. The model printed in 3 hours. The same model took 7 hours the day before with 20% infill! (it'a a tall cylindrical air tank for my pyro)
I will try with 2 loops next time. If it turns to be reliable, I might only use infill for parts that need to support weight, the legs of a character for instance (however if the rest of the model is so light, it might not even be needed)
Great! I'm glad it turned out good :-) I think that in general, most of us stay dramatically inside the printer's performance envelope most of the time. It's easy to get hung up on folklore and misperceptions, that the only good print is one that is done at 0.04mm layers, 40mm/s, 40% infill, tons of support, etc etc etc. :-)
I'd be the first to caution that all prints are different, and some tricky ones really are going to need really cautious print settings - but in general I think that prints can go much faster than people think.
If I hear about someone spending 20, 30, 40 hours on a print, I'm almost certain that they could print it far faster than that without sacrificing any noticeable quality. The aim always needs to be to get satisfactory quality in the minimum amount of print time - too fast and you get crap, yes - but too slow can cause its own problems, and at the very least, just wastes a lot of time. Probably better to go faster, do a test print, and then tweak the settings based on what you see, to really nail it, than do one print that takes longer than both of those two put together, and still probably could be better if you'd used different settings in the first place.
well, my pyro bust failled after 34 or 35 hours of printing and it took another 5 to 6 hours to finish it, so about 40 hours total. If I had printed with no infill, chances are it would have printed it successfully in one take in a dozen hours
60 microns is what I print the most and I've gone down to 30 microns with no problems. With ABS, the layers become nearly imperceptible at 60 microns, if the bed wasn't ringing or shaking during the printing!
You can use the Cura "Initial Layer Thickness" feature to make the first layer more forgiving or double or triple the flow for only the first layer in netfabb. However, for slicers that don't offer these features or for an alternate and more flexible approach, I just use my Ulticontroller to make realtime adjustments to produce a thicker extrusion on the first layer. After the first layer, I reduce the flow back down to 100%.
Of course it's highly desirable to still get the bed as level as possible (I'm using 5/16" thick Mic6 precision aluminum tooling plate for my heated bed) but Ulticontroller realtime flow adjustments can save you time over tinkering with bed adjustments or scraping off the first layer when the results aren't good. Just adjust the first layer extrusion in realtime to match the actual nozzle height above the bed.
Depending on the toolpath for the first layer, you can even adjust for a non level bed somewhat, though you might want to make realtime adjustments to the bed in this case too. In a few bad cases, I've continued to do realtime adjustments into the second and third layers to "level" the build instead of the bed!
When using expensive Kapton tape (or even worse, tougher, more expensive, but still razor blade sensitive PET tape) I really hate to abort and scrape when I see a first layer problem developing!
UM 100micron used to be unique. Now everyone is starting to do 100micron. So we are switching our high quality setting to 60microns. There might be some places that still show the UM as 100micron, but we are updating most of those to 60 when we spot them.
We can also do 20 micron, but that's really just advertisement, as going from 40 to 20 micron doesn't yield real visible quality improvement.
MariMakes posted a topic in UltiMaker Cura,
ArunC posted a topic in UltiMaker 3D printers,
Recommended Posts
gr5 2,240
I haven't had a need to go below .1mm but here's my guess:
1) bed levelling is super critical if the first layer is less than .1mm. I mean 1/8 of a turn of the screw is about .1mm so you have to have those 4 screws accurate to 1/16 of a turn now? Plus if you still have the acrylic bed - it tends to bow by .1mm easily. So you might need a more level surface and you might not want to use blue painter's tape either. All of this only affects the first layer. You might want to check "add raft" for this reason.
2) clogs/plugs. Cura has a "minimum" feed rate. I assume this is because of plugs but not sure. People have complained with the older hot end that heat slowly travels "up the solid filament" into the peek where if the PLA gets soft it can clog in the peek. I think this is only a problem with retractions maybe? Obviously with retraction you are transferring some heat upwards briefly, and then if you soon retract again, the heat can go up further as the previous retraction heated slightly higher parts of the filament and assembly. I don't know as I've never had this problem. However I know you can go much thinner than .1mm. Anyway you might want to print faster (over 100mm/sec) with thinner layers to avoid this problem? I don't know. I guess I would only print fast if I had ever gotten a clog in the first place.
Check out these tiny frogs - made at .1mm layer and with a .2mm nozzle which use only 1/3 of an inch of PLA! and most of that for the bed:
http://www.printrbottalk.com/forum/viewtopic.php?f=28&t=2755here
Link to post
Share on other sites