burtoogle

So, I have to put my hand up here and admit to committing this crime. I recently created a new setting called "Max Comb Distance With No Retract" that forces combed travel moves to use retraction if they are longer than some threshold distance (which is the value of the setting). Now, I had two choices here as to how this feature is enabled, I could either add another checkbox to make the setting active or not or I could choose a value that denotes that the setting is not active. One possibility is to make the setting default to such a large number that all combing travel moves would be shorter than that distance, say 100000000mm (or whatever). Personally, I don't like doing that because it looks ugly and you may not choose a value big enough (people are making some large form factor printers these days). So I plumped for zero and I made it explicitly obvious in the blurb for that setting that the value has to be greater than zero for the setting to be active. If the Cura guardians wish to add a checkbox to explicitly enabled/disable this setting I won't complain but I won't be doing that myself.

There was an issue whereby if you have different acceleration or jerk values specified for travel and for printing support and/or infill it would insert zillions of change accel/jerk M codes. This possibly got fixed for 3.4 so you could try the beta and see if that is better. Alternatively, either turn off accel/jerk control (if enabled) or ensure the accel/jerk values are the same for travel and for support/infill. If it isn't this that's causing the slowdown, I have't a clue! Hope this helps.

I think I have already fixed that in https://github.com/Ultimaker/CuraEngine/pull/776 (if you're lucky, the fix will be in Cura 3.8 ?) This is what I see when I slice your project:

It should apply to all combing moves irrespective of what they are moving over or between.

And whilst we're blowing our own trumpets ?, there's a new Travel setting called "Max Comb Distance With No Retract". When set to a non-zero value, combing travel moves that are longer than that distance will use retraction. I believe this has been a requested feature for some time now.

It sounds like you are using the concentric skin pattern. Try using lines or zigzag instead?

You're welcome, happy printing!

The dialog you pictured above is controlling the setting visibility, not the setting itself.

You have coasting enabled: Without coasting it looks like this:

OK, well can you please attach a project (.3mf) file that has the problem so I can take a look, thanks.

Looks like you have enabled the experimental coasting?

No, it doesn't already exist. The new bridge code does some acceleration stuff to try and stop underextrusion after a bridge wall line but there's no general ability. Some printer firmwares (Marlin, RepRap) have the pressure advance feature that modifies the extrusion rate to try and avoid fat corners due to low acceleration but that probably doesn't work when there's travel moves and/or retractions between the print moves so maybe an experiment is worth trying.

Hi @gr5, following on from what you are saying there, I noticed recently something that quite surprised me. I was printing a first layer where some wall outlines were printed slower than others and I was repeatedly getting an underextrusion at the start of a wall that was printed quite a bit slower than the wall outline that preceded it. As we know, the extruder cannot do a step change in extrusion rate so this underextrusion could be expected but what surprised me was that there was a retraction and prime between the printing of the first wall and the printing of the second. So that experience tells me that under/over extrusion due to step changes in extruder rate can occur even when a retract/prime occurs at the rate change point. Think what that means if, say, you are printing infill before walls and the infill print speed is a lot higher than the wall print speed. Even with a retract/prime between the infill and the first wall, it's going to overextrude when it starts that wall. If you want quality output there's a good argument to print all walls and infill at the same speed (probably skin as well).

Yeah, you're right, the end result is very similar to what you started with. Glad it's working. I find that with PLA some sharp corners do tend to curl up and adding a skirt that's almost a brim is a great help. My theory being that it's something to do with the temperature gradient on lower layers and having the brim helps keep the temperature uniform across the bottom of the part (this could be complete rubbish but that doesn't stop it being effective ?)

Ack! it doesn't work because the skirt hugs the part's hull not the outline! What you could do would be to print your own brim that gets close to the part's outline.

Can you post the STL without the brim? EDIT: don't bother, I'll just sink it into the buildplate!

Here's an example and it doesn't even use a very small gap and it is effective at stopping the corners of the part lifting due to them cooling.

I don't know what you are doing wrong but it sliced for me using 0.2mm layers: However, I wouldn't do this myself as I think it will make a really bad mess to the edge of the part trying to remove that brim. What I would do is use a wide skirt with a really small gap (say, 0.1mm) so that it looks like brim but only just touches the side of the part. I use this technique for PLA parts that I don't want to warp and don't want to clean up! YMMV

OK, I understand now. It's useful to be able to specify the number of skin layers independently from the number of walls because, sometimes, you want your printed part to have specific physical properties, i.e. strength, quality of surface finish and, so on. For example, if you are printing with a very low infill density but still require a good finish on the top of the model, you would probably want to increase the number of top layers so that the gaps between the infill don't cause the top surface to have dimples. The same model may only require a relatively thin bottom skin so you wouldn't need so many skin layers on the bottom. It would be possible to have a single setting that specifies the thickness of the model's outline (walls and skin) but a lot of the time it would not satisfy the user's requirements of beauty/strength/cost/etc.

For each layer, the walls are the lines that go around the perimeters of the model and separate the enclosed skin/infill from the air. The region inside the walls can be either skin or infill. Skin is always 100% filled (it's solid) and most models have one or more top and bottom skin layers. Regions within the walls that are not skin will be filled with infill whose density varies depending on the user's requirements. Sorry, I don't understand your second paragraph as walls and outline are the same thing, aren't they?

Ah, I failed to read your post correctly, I thought you were worried about the skin not being complete. If you want the pillars solid you will need to set the infill density suitably high. If you don't want everything filled to the same density you will need to learn about infill meshes. As to your second point, the min infill area feature (currently broken) only applies to infill regions within skins, not to infill regions within walls so even if it was working as expected, it still wouldn't make your pillars solid.

Hello @Lemaru. Yes, the Minimum Infill Area settings does appear to be broken. However, all is not lost. You can achieve your desired aim by setting the Skin Expand Distance to something like 2mm. Have a play with it!

Two well known printer firmwares (Marlin & RepRap) already implement this feature. Arguably, modifying the extrusion rate depending on impending/preceding nozzle speed/direction changes is best done in the printer as it has a better understanding of its capabilities and limitations.

One distinct possibility is that it is underextruding. So try slowing the inner walls down to the same speed as the outer walls and see what happens. You could also increase the temp a little. As a general principle (of mine), printing the inner and outer walls at different speeds should be avoided as it introduces a step-change in extruder rate which often produce either bumps or hollows. Hope this help

I am still working on handling bridge skins better. There are some changes already done and some I have yet to consider. I will bear your suggestion in mind. Cheers!