On 10/20/2023 at 4:34 AM, Slashee_the_Cow said:The maths aren't quite as simple as you tried, because you need to take into account the volume of filament (including layer height) and not just width.
I may be doing it wrong, but I tried doing the math again with volume and got the same result.
Edit
Sorry about the double post, I meant for that to all be in one
Edited by osteracI fooled around with it some more and was able to get some good results. I thing there was a partial clog in my nozzle and changing that seemed to help. It now looks like this:
0.4 mm top/bottom layers on the left (stock flow and density) and 0.6 top/bottom layers on the right (with adjusted flow and density).
I'll attach a project file.
I still think it would be nice to have a bridge layer width setting that modified all these values for you, as it took quite a while to get these results.
Settings:
Bridge Test Simple stairs .6 line.3mf
Edited by osteracIt might be a bit of a hassle, but if it's just the top, you could split it off into a separate mesh and move it right on top of the rest of it - that should allow you to have print settings for it separate to the rest of the model.
Looking at those pictures a bit more closely - in the right (0.6mm) picture it looks like the left side of the frame is bowing a bit. Possibly from the pressure of bridges slamming into it repeatedly.
The way I'd normally suggest to try and fix that is more walls, but you're limited by the thickness of walls beneath you. For some reason your 3mf file doesn't include many of your custom settings, and what I get is:
I wouldn't even know how to do that pathetically thin inner wall with blobs on the corners if I wanted to. It makes it a hell of a lot harder to try and tweak various settings to improve your slice when they're vastly different to begin with.
You can make a feature request in the Cura GitHub repo for different settings for bridge lines, but I wouldn't hold my breath. They're pretty busy and this would be a pretty niche feature.
Personally, I'm actually pretty impressed the results you've gotten so far. But I still think bridging that far with silk PLA is a bridge too far.
(Sorry)
Yeah, it didn't look like that when I sliced it. As for the bowing, I must have removed it before the build plate was cool.
21 minutes ago, osterac said:As for the bowing, I must have removed it before the build plate was cool.
I wouldn't be so sure about that. PLA doesn't even need a heated plate, and I've taken prints off right after they finished a bunch of times and they've never suffered for it.
with a print this thin, it can remain soft all the way through until the build plate cools. It's happened to me before with this model.
2 hours ago, osterac said:with a print this thin, it can remain soft all the way through until the build plate cools. It's happened to me before with this model.
Egads, cool your printing! According to the .3mf I got from you, the regular fan speed is set to 20%? You keep the fan low for the first layer or few so that it's oozy a bit longer to increase adhesion to the plate, but especially with PLA, after that, for your major features (walls/ceilings/floors, which is all you have) you want it up at 100% so that it sets as quickly as possible, since the chance a difference in layer temperature will cause warping is basically nil. A higher fan speed also results in less drooping, because it gives the filament more of a chance to set before it can droop, which gives the next line something more solid to hold on to.
Edited by Slashee_the_CowThe default fan speed for my printer is 100%, yes, but that's with stock cooling. My printer, as stock, doesn't have a layer cooling fan, just a heatsink fan. I have installed a MASSIVE layer cooling fan/duct. Dual fans actually. I could try setting it higher, but I think in the past that has led to poor layer adhesion. Because of the bridging settings the fan speed will, of course, be at 100% during bridging. If I set it to 100% all the time, and tune my printing around that, there will be less advantage to using bridging settings. My max fan speed is 50% also.
Of course I'm not an expert, this is just what I've found by experimenting on my own. There may be other reasons for poor layer adhesion in my prints. I'll try some higher fan speeds.
Also, I appreciate all your help! Thanks for taking the time.
I wrote this because Cura's fan control is rudimentary. Unzip the file and put "AddCoolingProfile.py" into the "scripts" folder in your Configuration Folder (find it using "Help | Show Configuration Folder").
There are options to control the fan "By Layer" or "By Feature" and within By Feature is a Bridge Fan option. The script will be available in "Extensions / Post Processing / Modify Gcode" and then "Add a Script".
1
55 minutes ago, GregValiant said:I wrote this because Cura's fan control is rudimentary. Unzip the file and put "AddCoolingProfile.py" into the "scripts" folder in your Configuration Folder (find it using "Help | Show Configuration Folder").
There are options to control the fan "By Layer" or "By Feature" and within By Feature is a Bridge Fan option. The script will be available in "Extensions / Post Processing / Modify Gcode" and then "Add a Script".
This is wonderful! It should be built into cura. great work.
Thank you. I submitted a pull request for it about 10 months ago. I think eventually it will get rolled in once it is reviewed by the Cura team.
Thought just occurred: have you tried printing with regular (i.e. not silk) PLA? It's possibly more likely to work because it's less slippery and often dries a bit quicker, might help hone in on the best settings.
But:
If you can get it to work with regular PLA, you can change filament mid-print using the Filament Change post-processor (I'm so used to Greg's ones I can't remember if it comes as part of Cura or not), so you can set it to change on the layer it bridges and then add another instance of the post so you can change it back next layer. The adhesion properties of the regular and silk PLA might be a bit different, but it shouldn't be a problem in this case: you're printing walls on top of walls and the regular PLA bridging the gap is trying to adhere to the regular PLA on the other end. Then if your bridging is solid enough, silk PLA will print fine on top (you should probably do it with the lines in a different direction to the bridge), then all you need is some regular PLA roughly the same colour as your silk and hope people don't notice a fraction of a millimetre worth of layer looks slightly different.
Layer adhesion with "Silky's" is not good and it seems to me that it varies by color. Silky Silver is actually pretty good and Silky Copper is really bad with Silky Gold somewhere in between. I generally print them 5° hotter, but there is not a big improvement.
In addition they seem to suck up moisture faster and so you need to use it up or it gets brittle AND doesn't adhere to itself well.
1
1 hour ago, GregValiant said:Layer adhesion with "Silky's" is not good and it seems to me that it varies by color. Silky Silver is actually pretty good and Silky Copper is really bad with Silky Gold somewhere in between. I generally print them 5° hotter, but there is not a big improvement.
In addition they seem to suck up moisture faster and so you need to use it up or it gets brittle AND doesn't adhere to itself well.
I'm guessing it also varies by brand as well as colour (unless there's secretly a monopoly on filament and one company makes it and sends it to resellers). I managed to do one print with half decent bridging (though not as far) with one of the ones that changes colour as you go along, and I noticed mainly because $%*#ing Creality Print for whatever reason decided my thing didn't a bottom so it just printed some pretty sparse infill and then a couple of top layers.
As for moisture, please bear in mind that microwaving your filament is probably not a good idea (I haven't tried). There are dedicated filament dryers (designed more for hygroscopic filaments like TPU or PETG). Possibly also something that depends on brand, though? I once I did a 22 hour print with the aforementioned variegated one, no dryers, nothing to keep moisture at bay, no bridging, plenty of overlapping surfaces, supports where needed... the whole "perfect scenario" idea and it came out fine, even some of the really fine details which I was hoping (but not counting on) coming out well.
But @GregValiant has been at this a lot longer than me so please take his word as gospel and mine as an individual experience.
There are filament driers and then there are ways to dry filament.
I paid no attention at the time but when remodeling our kitchen I bought a new stove. I'll be darned if it doesn't have a "Dehydrate" cycle. It allows low temperatures and the exhaust fan runs. It works perfect but can only be used at night (or someone else might notice what her stove is being used for). The same method can be used to dry silica packs out.
There there is also the "Turn your heated bed on at 60°, set the roll of filament on it, and put a cover on it". People claim that works and although I haven't tried it (got that fancy oven) I don't see why it wouldn't be better than nothing.
42 minutes ago, GregValiant said:There there is also the "Turn your heated bed on at 60°, set the roll of filament on it, and put a cover on it". People claim that works and although I haven't tried it (got that fancy oven) I don't see why it wouldn't be better than nothing.
Isn't the cover both required to keep the warmth in but will prevent moisture from escaping?
Beats the heck out of me. I have the fancy Dehydrating oven.
I use the dehydrate setting on my air fryer
I imagine the cover would be to keep the heat in and uniform, but it would also keep moisture in.
Edited by osterac8 hours ago, GregValiant said:I have the fancy Dehydrating oven.
2 hours ago, osterac said:I use the dehydrate setting on my air fryer
Way to shame me and my whitegoods. Now I feel like the sucker in the room for having bought a filament dryer.
(Cheaper than a a new air fryer, and a lot cheaper than a new oven though)
26 minutes ago, Slashee_the_Cow said:
Way to shame me and my whitegoods. Now I feel like the sucker in the room for having bought a filament dryer.
(Cheaper than a a new air fryer, and a lot cheaper than a new oven though)
on the plus side, your filament won't smell like chicken nuggies
1
1 hour ago, osterac said:on the plus side, your filament won't smell like chicken nuggies
Since when is that a plus?
ArunC posted a topic in UltiMaker 3D printers,
.thumb.jpeg.0b7a05eafc09add17b8338efde5852e9.jpeg)
ArunC posted a topic in UltiMaker Cura,
Recommended Posts
Slashee_the_Cow 320
First: probably more appropriate here than the GitHub repo. That's more about reporting bugs, not asking for help.
Changing the material flow can cause it to make walls thinner (because wall thickness is basically controlled by the amount of material that is usually just laid down on top of existing material so will expand to adhere to that). The maths aren't quite as simple as you tried, because you need to take into account the volume of filament (including layer height) and not just width.
If you want, you can change how closely the walls are printed with Bridge Skin Density. This can be reduced so it will print walls spaced further apart (which can help them cool faster to try to avoid sagging by not printing hot wall next to them straight away) or increased so that it tries to print more lines, closer together (so they'll adhere to each other). But it won't print them any thinner or thicker.
I've personally never had a problem with adhesion of .4mm lines using PLA, excluding shonky PLA (though the worst adhered fine, it just destroyed my printer in the process) but I also probably don't have the exact PLA you have and it's not always one size fits all (especially with the silk PLA, because that usually has added ingredients to make it silky).
Two notes about bridging though:
1) You probably want to do it a fair amount faster than 15mm/s. The auto calculated value in Cura is based on your print speed setting. But the longer it takes to get from support to support (in this case the walls) the more time it has to sag before it gets there.
2) I hope that's just a proof-of-concept model, not one you want to actually try and print. Someone is bound to prove me wrong, but I'd be very surprised if you could bridge 5cm over nothing whatsoever with silk PLA even in ideal conditions.
Link to post
Share on other sites
osterac 1
Thank you, this post is very helpful. One problem I've had trouble with is getting the bridge wall to adhere. This is the absolute best I've been able to do:
Even if that's the best I can expect to do with silk PLA, what settings would make the bridge wall adhere better (in general)?
I didn't realize silk PLA performed so much worse. I suppose it depends on the brand, because I have had much better luck with other silk PLAs. This one (Amolen dichromatic) has been a real pain.
Link to post
Share on other sites