GregValiant

Reddit has a LotMaxx community HERE. I'm sure someone has run into this before.

You didn't mention which plugin but 5axes is the author of a couple of them. His screen name here is @Cuq and he may take a look at this for you. We are into the weekend though so it may not be until Monday. In the meantime if you could recreate the exact situation in Cura (same model and same settings) and use the File | Save Project command and then post the 3mf file here he will have something to work with.

That was it. It's in that profile. This type of bug has come up a couple times recently in regards to "Remove Empty First Layers". I have written a couple of post-processors and one of them adds the Cura settings to the gcode. I used it and saw that "Infill Line Multiplier" was set to "2". Just to check I went to change it and low and behold it wasn't visible because it is never visible when "Gradual Infill Steps" > 0. So I set Gradual Infill Steps to 0, set the Infill Line Multiplier to 1, set the Gradual Infill Steps back to 10 and the problem went away. So like the "Remove Empty First Layers" bug we have a hidden setting that is being referenced and used. This is not an exact duplicate of the other reports but it is the same sort of behavior. Github bug report #14989 has been closed as the Cura team worked on that bug. @MariMakes I think this one is very similar. Should the OP fill out a bug report on GitHub?

This is an Edit. I can duplicate this. It looks like all of layer 5 is going down twice. I'd call that a bug. This is another Edit. Now I can't get it to do it again. I'm using 5.3.1 for the trial with your settings. I'll keep playing with this to see if I can repeat the problem.

I just saw this. I don't know @DivingDuck, the nozzle should not be re-extruding where it's already been. This looks like it might be a bug. I don't understand the pattern though. It's lightning infill but that outer area makes it look like concentric. @rat_tech I know some time has passed but could you load that model, re-create your Cura settings, and then use the "File |Save Project" command and post the 3mf file here? It may provide a clue as to what is going on.

I call it the "Comb" problem as in hair comb. First, try adjusting your "Minimum Line Width" down. If those fins are 0.58 wide then a Minimum Line Width of 0.29 or 0.28 should fit twice. That will force Cura to do an Out-and-Back extrusion on each fin. If Cura decides that a single extrusion will be sufficient, then that's what you will get...an extrusion out to the end, travel to the next fin, an extrusion back to the base of the fin. That travel across the ends of the fins often leaves a string connecting the ends of the fins. It doesn't do it every time, and it varies from layer to layer. If you turn on the "Starts" in the layer preview I think you will see that there are a lot on the fin ends. Ideally you don't want any. Whether it's possible or not you'll have to experiment. These are .58 wide and the minimum line width is the default .34 for a .4 nozzle. The white marks are the extrusion starts. I think the problem here is obvious. This is at .28 Minimum Line Width. There is one fin that has the normal Z seam on it. The rest are "out and back" extrusions and the nozzle doesn't stop.

You can open a gcode in Cura to preview it. You cannot alter an existing gcode file with Cura because there is no model in the file, just the machine commands.

Load the model. Set Cura up to slice. Use the "File | Save Project" command to make a 3mf file. Right click on the file and (in Windows) select "Send To... | Compressed (zipped) folder". Post the zip folder here.

I don't have a definition file for your printer so the file won't open for me as a project. I did get the model though. Both MS 3D Builder and Mesh Mixer reported errors but they must be minor. Here is the repaired model. When I sliced the model it looked fine. There were no missing layers. I mentioned a clog only because it's some thing that can happen. Until I changed hot ends and the hot end fan my Ender was notorious for clogging about 45 minutes into any print. ExportBox2_fixed.zip

That could be an error in the model or the hot end had a clog that cleared itself. Load the model, set Cura up, use the "File | Save Project" command to create a 3mf file and post the file here.

Easy part first... An outer wall is any wall with air on one side. They are "red" in the Cura preview. An inner wall has model on both sides and they are green in the Cura preview. Anything that is part of a skin is "yellow" in the preview. That would include any extra skin walls. The "Extra Skin Wall" is the border around any skin area. I was playing with the "Skin Overlap Percentage" and if there is a Skin Wall present it moves the Skin Wall closer to the inner wall (or if a negative number then farther from the inner wall). What I noticed is that if the "Extra Skin Wall Count" is 0 then the "Skin Overlap Percentage" will act directly on the skins. Setting the "Skin Overlap Percentage" to 50% (with the "Extra Skin Wall Count" = 0) might help. It looks like it makes a difference in the preview, but there is no shrinkage in the preview. This is with no skin wall and a Skin Overlap Percentage of -50% just to show what it does. This sort of looks like what is going on with your prints. But set to +50% you get this. The Skin lines are mashed into the inner wall lines. Maybe that has some value but you would need to do a couple of tests both with and without the Extra Skin Wall.

The "Gaps" thing. This comes up occasionally on this forum and I tend to agree with other people on this that the root cause is a shrinkage issue that transcends the slicer being used. This might not be what's going on in your specific case, but it is my take on the behavior. If enough time passes between each curved wall extrusion being laid down then shrinkage will occur and the next perimeter doesn't connect to the previous one. This is really hard to fix but making sure your filament is dry can go a long way. If you have a very good layer cooling system and you are running it hard when these extrusions go down (particularly Inner Walls around a small hole) then that can make the condition worse. Most filament comes in a sealed bag with a desiccant pack. That guarantees nothing. I've opened new rolls of filament and they were, well, not sopping wet, but pretty darn close. That is not a popular problem, but it can certainly happen. I have experienced the gap problem myself. I started paying more attention and it usually occurs with rolls of filament that I've left out, or that have just sat on the shelf for too long. I see you have an S5 so the "Flow" should be correct. There are a couple of other things you can try without requiring that the slicer be re-written. Hole Horizontal Expansion - The smaller the diameter of a hole, the more likely that the "Snot Factor" will come into play. That condition happens because the extrusion wants to get dragged in towards the center of the circle rather than stay on the periphery where it belongs. The extrusion WANTS to follow the nozzle and not form an arc. The gap can occur because the inner wall extrusion isn't where it's supposed to be. This is often why small holes don't come out at the design diameter. I did a test and if the hole diameter is over about 30mm then the Snot Factor falls to zero and no HHE is required. Inner Wall Flow - By kicking this up you can force more squish which should tend to keep the inner walls where they belong. Maybe you've thought of those before. Because I haven't experienced the problem lately myself - maybe I'm just out of touch.

@Peter_E6 In addition to Obewan's suggestion you have settings that simply don't apply: Avoid Printed Parts (impossible on this model), Avoid supports (none on this model), and combing set to Not In Skin (there isn't any on this model). Those are going to cause retractions because they affect other settings. "It seems like nothing can stop Cura regularly retracting the filament for no explicable reason" Any printer is a 4 axis (at least) robot. It has to be programmed to go and do what it's supposed to do. Any slicer has to be set up so that the generated gcode does what the user requires. This is the user's responsibility because in the end, the retractions and other things that the printer does is dependent on the input of the user. There is always a reason for any retraction and that is because that's how the software was set up. Now if there is a bug in the software then that is a different story. I do see problems with your settings. I don't see a bug. With your settings there are 182 retractions between ";LAYER:0" and the end of the file. Since the model is 260 layers that is somewhat less than one retraction per layer. Turn off Avoid Printed Parts, Avoid Supports, Retract Before Outer Wall, set Combing to "ALL", and set Retraction Minimum Travel to 10.0. With those changes there are 5 retractions between ";LAYER:0" and the end of the file. That's about 3% of what you are getting with your settings. I hope I made some sense. GV_MicroscopeOccular.3mf

When Cura adds the polka-dot pattern to a model it means the model has errors. That can happen for a number of reasons. In this case I think the 37 separate parts that make up the model assembly have gaps between some of the sub-parts, and the main body. Scaling the model up likely opened up the gaps to the point where Cura can notice them. The result is that if this does slice it will probably be "iffy" to print. Here I fixed a copy of the model using MS 3D Builder. It's pretty obvious which one was repaired. This is the repaired model sliced. The original model sliced after I rotated it on the build plate. There are bug reports regarding that behavior, but in this instance I think it is a model problem.

"I want to add say 2, or 3 or even 5 extra walls (perimeter) lines on every alternate layer, ..... on every Nth layer...?" There are no settings to do something like that. You can alternate a single extra wall I think that thought behind that is to trap the infill at the walls. "...apply only to inner walls, while just leaving a fixed wall count on outer perimeter walls." I'm not sure I understand that one. I'd say that because the answer to the first question is no, that this one doesn't come into play. Wouldn't 5 extra walls every 10th layer result in those extra wall extrusions being printed over air? You would need pretty high infill density to hold them up.

About 3 years ago there was a poster here who built a big pellet machine and he was using Cura. I tried to message him a while ago and didn't get a response. This is the post.

With the Custom setting button turned on and the Setting Visibility turned to All then in the Travel Section you will find "Layer Start X" and "Layer Start Y". There is no such setting for the infill. It will often be as you want it (start near the z seam or layer start point) but for different types of infill patterns it may be different and if you change the Infill Line Directions it may be different.

This is just an FYI... Depending on the "repeatability" of the end stop switches - it is generally not a good idea to home any axis during a print. The switches are going to vary at least a little bit every time you ask for homing. The "0" position this time will likely not be exactly the same as the "0" position last time. When the X and/or Y are involved during a print you can get a minor but noticeable layer shift. When it happens with the Z you can get an over-extruded layer (the new Z0 is a bit too low), or an under-extruded layer (the new Z0 is a bit too high).

There are a couple of bug reports about this on GitHub. Whether "Union Over-Lapping volumes" is turned on or off - they don't union but simply stay separate resulting in what you have there.. Union Over-Lapping Volumes = Off Union Over-Lapping Volumes = ON with Remove Mesh Intersection off Union Over-Lapping Volumes = ON with Remove Mesh Intersection on. This looks like one block is acting as a cutting mesh. There are still interior walls where I would have expected none.

"Incompatible Version" so I can't open the DWG file. I'm working with AutoCad 2000 and Mechanical Desktop 4.0 so you would need to do a Save-As in one of those formats, or put a DXF together.

After adding the shell and the first 2 excluded faces you can spin the model, then edit the shell and add the third face. You need to turn on supports in Cura or the horizontal parts will fail.

I'm an AutoCad guy. Pass me the .dwg file as an ACAD 2000 file (parametric if you can) and I'll take a look (you might have to zip it to post it here). Sometimes when you get too fussy with the "fits" there will be openings that the slicer will notice even though the gap might be on the order of .000001mm. You have those gussets on the two end pieces and they look to have flats where they hit the radii of the tubes. That's the sort of spot that can cause trouble. If the tube is only touching tangentially to the flat then there will be a tiny gap on either side. The gusset should be buried by a mm or so to insure the union works correctly. You can check by putting a fillet there. If it's successful then the union is fine. Computers may be stoopid, then they are accurate.

It's always safe. It's a question of whether it will be acceptable. There are these spots on opposite sides of the print. (You can see it on the right of my previous screenshot.) I have no idea why the flow narrows down so much there. Is it a model feature? There are these divots on the outside of the model. They are going to show up in the print. So would it be safe? Yes. Would it be acceptable? You'll have to decide. For a first effort I would say yes. Those divots have resulted in a few bug reports on Github regarding 5.3.1. Cura 5.2.2 didn't have the problem. It's possible that some of what we see is being caused by the model. BTW does your layer cooling fan work OK? I see in the gcode that it is listed on circuit "1". Typically a single extruder printer will have the fan on circuit "0".

That's a poorly sliced model.

I use the Raise3D slicer occasionally for comparisons with Cura. With an N2 installed in IdeaMaker there are T0 commands in the startup so they are probably alright. My guess would be that the firmware is similar across their product line. That G92 E0 E1 command looks to be contradictory and so it's probably illegal. It's trying to set the extruder location to 0 and to 1 at the same time. Not good. This is the N2 StartUp Gcode from Ideamaker: M221 T0 S94.00 ;set the flow to 94% G21 ;metric M140 S60.00 ;Bed temperature M104 T0 S215.00 ;hot end temperature M109 T0 S215.00 ;wait for hot end T0 ;set the tool M190 S60.00 ;wait for the bed temp G21 ;metric units G90 ;absolute movement M82 ;absolute extrusion M107 ;fan off G28 X0 Y0 ;Home the X and Y if they need it G28 Z0 ;Home the Z if it needs it G1 Z15.0 F300 ;move the nozzle up G92 E0 ;reset the extruder G1 F140 E29 ;purge G1 X20 Y0 F140 E30 ;move while purging some more G92 E0 ;reset the extruder G1 F9000.0 ;set the feedrate M117 Printing... M1001 ;enables the power resume function. There should me an M1002 in the Ending Gcode. I made up the one below. First use the "File | Save Project" command to save your current StartUp Gcode. (If you were to open that new 3mf file in Cura it would bring in your original StartUp Gcode.) After backing up your current startup gcode in the 3mf project file, you can try this by copying and over-writing the current StartUp in the StartUp Gcode box in the Cura Machine Settings dialog. ; <<< StartUp >>> M221 S100 ;reset the flowrate M220 S100 ;reset the feedrate G21 ;metric units G90 ;absolute movement M82 ;absolute extrusion M140 S{material_bed_temperature_layer_0} ;start the bed M104 S{material_print_temperature_layer_0} ;start the hot end G28 X Y ;home the XY G28 Z ;home the Z G1 X0 Y0 Z15.0 F600 ;move the nozzle up so the purge has somewhere to go. M109 S{material_print_temperature_layer_0} ;wait for hot end M190 S{material_bed_temperature_layer_0} ;wait for the bed temp M107 ;fan off G92 E0 ;reset the extruder G1 F140 E29 ;purge G1 X20 Y0 F140 E30 ;move while purging 1mm more G92 E0 ;reset the extruder G1 F9000.0 ;set the feedrate M1001 ;enable power resume function M117 Printing... ; <<< End of StartUp >>>