gr5

Seems to be always clockwise in Cura 5.0.0. I can't run newer versions of cura so I can't test in the latest. They did add a feature where you can alternate CW/CCW but that was only for every other layer and disabled by default and related to metal prints. Not plastic prints. Maybe try cura 5.0.0?

I know. "Orbit center point not in part" is annoying but you can inspect close up if you do: Try shift+ right mouse drag You can also use a space mouse (6 axis mouse instead of the normal 2 axis mouse)

The main claim - the main feature - in cura 5.0 - is the ability to print thin walls much better. So it's funny that you switched to cura 4.X.

By the way, creality slicer is clearly cura. You could just check all the settings but... it's a lot. At 500 or so settings. Oh wait - it's an older version of cura (I assume). You could also just use cura 4.X. That works just fine for an S5. The final version of Cura 4.X came out something like 4 years after the S5 so it works quite excellent for the S5. There are no major bug fixes or anything that you need in Cura 5 to print on an S5.

If you continue lowering minimum wall line width to 0.1 it prints more. Another layer higher and closer to the edge. Before and after image below. Because of the basic architecture of Cura, it insists on doing two passes on every wall. And you have a .4 nozzle. So that's very difficult to print walls much thinner than 2X the nozzle width. Thinner than 0.8mm. You might want to design your part with rounded edges? Or maybe use the AA 0.25 core (which prints 2X slower). Or you can just set minimum wall line width to 0.1. It doesn't just affect this spot though - it will be printing all pointy ends of walls a little longer. So it may decrease the quality in other places? I really don't know. It may make the other pointy walls better. Another "solution" is to tip the part 90 degrees because now your "walls" are horizontal and with a 0.1mm layer height you can easily print 0.1mm thick "walls" (although they may be delicate). Because they aren't walls anymore. They are ledges. But this will require support material. below is .1

It's hard to see what's going on so it might be best if you post the project file. In cura do "file" "save project..." and post that file. So you may have errors in your model - particularly tiny holes or "inverted normals". Which is particularly common in models created with blender or sketchup. But I'm going to guess you are having problems with thin walls. Make sure "print thin walls" is checked. And look at some of the options in the "walls" section. Particularly: minimum wall line width (walls must be double this I assume) - not that there are 3 features with this or similar name so check all 3 minimum feature size If what I told you doesn't help then post the project file.

The secret word is "combing". You want to disable combing. And you want to make sure it's really really off so there are probably 2 or 3 other combing settings. Combing is the word that describes movements within the walls - movements in areas where you normally have infill or shells. Note that even when you disable combing I think it still combs if the distance is below a certain distance? Or maybe not. So here's how you test it before you print: Go into PREVIEW mode after you slice and scroll throught the layers. Make sure you are looking at it with color scheme set to "line type" and make sure the travel moves (blue) are enabled so you can see them. Now light blue travel moves are retraction moves (this is what you want). Dark/saturated blue travel moves are non retraction moves (bad - creates internal stringing). Make sure you tweak the settings until all (or 99%) of the dark blue lines are gone.

Try to visualize a bow-tie shaped outline for a particular height to slice. Cura gets these as little lines. And they aren't sorted. So cura tries to connect the lines into the bow tie. You may think the (x,y) coordinates of the end points of each little line should match exactly with another line but they don't always match because the STL doesn't hold lines - it holds triangles. and when you "slice" you are intersecting mathematically a plane and a triangle and there are tiny tiny errors. usually smaller than a nanometer. But the points don't line up perfectly. And the STL file is no help as it's not a great storage format because it doesn't tell you which edges of each triangle belong to which other triangle. So you take all these lines and try to connect them into a loop (really potentially hundreds of loops depending on what the print looks like). So there is a tolerance when creating the loops. Do you set the tolerance at 1 nanometer? 1 micrometer? No you have to make it larger or you won't get very many loops at all. I'm not sure what the tolerance is. maybe it's .05mm. Now if the bow tie comes together in the skinny region you might connect the wrong line segment at some point. and now you have two separate loops by accident. This is exactly what you described earlier. I know this is how Cura works because the original author described it thusly long ago on this forum. Now what is the tolerance? I really don't know. Is it a parameter/setting in Cura? I don't know. but that is exactly what is happening. I'm hoping it's a setting and you can make the settings smaller. In addition to finding the loops there is another feature of cura where it discards some points. I had assumed it did this *after* it makes the loops. If it does it before then I would think you would get these "one loop into two loops" problems much more often. I could be completely wrong and your problem is unrelated to the "make the loops" step. By the way, the very next step is to figure out if inside each loop is air or plastic. Maybe the loop is around an empty region (quite often in fact - for example when printing a cylinder there is an outer loop for the outside of the cylinder and a second inner loop to represent the inner wall of the cylinder.

Maybe your printer is leveled so badly that the first 3 layers are squished so much that they are so thin that they are basically transparent. I've seen that before. It would be obvious if you watch it print the first layer and it's transparent.

Another thing that is helpful: Turn off the feature where it drops the part to the bed automatically. Select the part and the move tool and set the Z value to +1mm and use the mouse to orbit around and zoom in and then set the Z value to -1mm. When you do that it is more obvious that the part is above the bed or below the bed. Also I got some feedback from UltiMaker. They said that white line is there on purpose to help you see if the part is on/below/above the bed. In your case it's deceiving. The color of the line unfortunately is different for different colors of filament. I think it does some kind of XOR or Multiply mixing of colors and maybe instead they should just darken.

It's slicing fine for me. I'm using Cura 5.0.0. I get the same thing - that light line. It looks like it's below the print bed but it is not. I know because your part is symmetrical and you can see that it is printing symmetrically: the top and bottom rings are the same thickness/height: Also it's not *below* the bed - I know this because you can see below, the teal colored ring. That's on the bottom layer for sure. That's the skirt. And on that same bottom layer it is printing your part. Maybe there is some bug in your version of Cura. You probably already said but what version of Cura are you using? Could it be that you *had* the bug in the past but now it is slicing fine? Because your screen shots from Cura look perfect. Maybe you had a problem but you fixed it but you are afraid to test? Or maybe you have some buggy version of Cura? I can't run anything newer than Cura 5.0.0. I get some error about libraries or something.

Well you need to post a project file that has a problem. The one you posted was fine. The white line means nothing as far as I can tell. Use those tools I mentioned above and zoom in on a spot and switch modes and you can see it prints the bottom layer perfectly. It's hard to show with 2 photos - easier if you switch between prepar and preview modes. You can't tell but trust me - it prints the bottom layer just fine.

I just looked at some other projects I've sliced and printed in the past and they have a line there as well but the first one I checked had a purple line. I never noticed that before. I don't think you should worry about it.

Oh - I get the white line also. I don't know what that is - never noticed before - I don't think it matters. I think you can ignore it. The part is flush with the plate. When I orbited a little lower so the horizon was a line, everything, including the white line, was above the build plate. It looks fine to me.

I just loaded your project file and sliced it and it looks fine to me. It does not stick below the build plate. Try the following 3 features to look around better: right click drag to orbit around the view shift right click drag to move up/down and get your eye to the build plate plane scroll to zoom in and out

By the way the sensor has nothing to do with mechanical resistance. There is an electric property called capacitance which changes depending on the distance between the flat plate in the print head and the flat plate on the test bed. As the bed moves up to the nozzle that capacitance changes and this changes the frequency of an oscillator. When the nozzle touches the bed the frequency stops changing because the distance stops changing. But if there is a lot of electric noise (like from nearby equipment - like a tesla charging 10 feet away or a mini fridge under the printer or the front fan on the printer) the frequency changes erratically and you get problems.

How old is this printer? Your description of the problem was very helpful. This is what I recommend you do. Well first contact your reseller and let them know. Then you need to go into the maintenance menu and somewhere there is a test to test your sensor (by the way they call it "active leveling"). After the test it will display a number. That number shows the rms noise at the sensor. Values 8 and below are considered a pass. Values above 8 are a fail. Note that this test must be done with the bed heat off. Unfortunately that is not enforced. The true active leveling turns this off. I'm told the results can vary with bed temp (even if the bed is off). Do that test a few times and then the most common fix by far is related to your 3rd fan in the print head which is in the center - in the door that flips open. The best thing to do is to unplug that fan which is not trivial as you have to take the head apart. But you really should just bite the bullet and do that. If that then passes then you can ask for a new fan. There are many many other things that can make the test fail. Bad wires to the middle fan. Bad wires to the sensor (visible when you open the test head door you can see two wires (typically one red and one white) going to the base of the open door. Sometimes you have to replace the circuit board in the print head. Another option is you can turn off active leveling and do manual leveling instead. ONly recommended if you use a small portion of the bed (like divide in your mind into tic tac toe and if you only use one or two squares). This can get you going until you get this problem fixed:

This post is corrected based on Nallath's post. Features appear and dissappear depending on many things. Case #2: This is only available if you have ONLY ONE EXTRUDER enabled. Even if you are using only one extruder, you absolutely have to go back to the PREPARE tab and disable the second extruder. There may be other conditions that hide this feature - maybe you have to have two parts loaded into cura?

By the way this is needed for 2D printers as well. When I'm on the vpn I can't print to my 2d printer at home unless I configure the vpn. So this is a very very common issue and so most (all?) vpn software should let you do this. So you can google about how to do this with "normal" printers for your particular vpn software.

VPN software will let you configure it so that regions of ip addresses don't go over the vpn. So configure the 10.191.X.X address space as never using the vpn (hopefully you don't have any 10.191.X.X devices at work on the vpn). If 10.191.X.X is already needed you can narrow it down to the individual ip address of the printer. Or if you can't figure out how to configure your VPN software - well consider using a different VPN software. But alternatively you can launch putty which you can setup to do port forwarding so you can forward a port on your computer to a different port on the printer and then you can connect to your own computer (hosted by putty) to access the camera.

The UM solution does not use ultrasonics. Just a circulator and I think maybe heat? A lot of people like to use a sous vide machine like this one: https://www.amazon.com/Inkbird-Stainless-Immersion-Circulator-Temperature/dp/B07RNWJZNR/ref=zg_bs_10962055011_sccl_7/139-0445537-1890517?psc=1 Because they provide the circulation and also accurate heat control (stay way below the softening temp of PLA which is 52C so 35C should be safe). However, I have one of those sous vide things (for cleaning pva) and rarely use it and instead just put it in a container of water overnight. The UM one does indeed work better/faster than just using a container with no circulation. Maybe 2x faster? Also the sous vide thing doesn't do as good a job at circulation which tends to circulate mostly just within a few inches of the outflow.

I think you can fix this!!! So when cura "slices" what it really does is mathematically intersect a plane (the current layer) with every triangle in the STL file. This results in an unordered list of line segments and then cura tries to put them back together into "loops" or "islands" and then it tries to figure out which side of each loop is plastic and which side is air. Unfortunately STL files have unordered triangles so you have to guess which triangles are connected to which other ones, or, once sliced, which line segments are associated with which loops. When it is creating the "loops" you can get a crossover as you describe such that one loop (island) becomes two. I'm not sure which parameters help you fix this but I suspect you should look at: maximum resolution maximum deviation And maybe lower those values. Or maybe there's another variable somewhere else that expresses tolerances when creating the loops.

I don't know the answer but you can probably enable a tower and at least it will do this warm up over the tower. I hope. Look for enable prime tower.

C'est Q1 qui contrôle le ventilateur central. Je suppose que votre ventilateur est toujours allumé ou toujours éteint parce que cette puce a explosé !