neute

Hi Nallath, thanks for your reply. First, sorry for yet another post of a user venting frustration about things they don't understand, I see those quite often around here. I think the latter is the issue here. There are no qualities for these specific iglidur materials, since I've created them from scratch with new GUIDs and everything. I've played around a little with loading different materials between the 4 extruders and disabling single extruders and I do now understand more or less how it behaves, as well as understanding the formula that calculates the average shrinkage between the active extruders. I just need to remember that I have to check on this setting if I'm using more than one material or have more than one extruder active at a time (which the whole intention of creating material and printer profiles in the first place was to not have to go through the whole settings pane before every print job) How do I know if a setting is shared between extruders, is it just the "settable per extruder" parameter? Re my question "am I doing this right" - where should the setting "material shrinkage" be included for a material; it seems that it is not intended to be in the material XML. Should it be in a quality profile even if it's theoretically the same over all hypothetical qualities? Intuitively I'd want it to set only once/in a single place in this case. thanks again, have a nice week Best, Niklas

Hi all, so I've bitten the bullet and started to deep dive into printer and extruder definitions, material profiles, variants etc. because I'm using self-built printers or the printers don't have profiles provided for them in Cura (yet). So far so good, I learned JSON and XML (rant: I wish someone came up with a fool-proof GUI for this...) and have things going pretty okay. Now I'm trying to set up material profiles for the filaments I use the most. Of course, in the filament .xml.fdm_material at first only basic properties are being stored, basically the ones that you can change in the "materials" settings pane in Cura (rant: why do I have to install a plugin to access more than 6 filament settings that are definitely related to the filament's properties?) --> Because my filaments are a bit more complex than this (e.g. I want to include settings like extra retraction amount, horizontal expansion, size compensation scaling etc.), I've then installed the "Materials Settings Plugin", created new filament profiles and altered them to see how the settings would reflect in the .xml.fdm_material file. For some reasons, the settings are not longer tagged as <setting> but <cura:setting> and wouldn't be included otherwise. --> Now comes the tricky part: say I created a new profile and modified it inside the material settings pane, having more settings selected by the plugin, counter-checking it in the XML file to see how the settings are reflected, then remove it from Cura and import it again, the settings change?? In particular, it is the "material_shrinkage_percentage" setting that I'm concerned with right now (rant: why are the variables named differently than their label in Cura??) example; see the igus iglidur i180 XML file (attached, remove .zip), values are exagerrated for demonstration/debugging purposes <cura:setting key="material_shrinkage_percentage_xy">100.2</cura:setting> <cura:setting key="material_shrinkage_percentage">102</cura:setting> this it what shows in the material settings pane after a clean import (Material Settings Plugin activated): ...but when I check the setting in the parameters section, this is what I get and what I presume is actually used for slicing: ...now I *felt* that it wasn't only this setting that is being altered upon importing, but it was the only setting that I could reproduce this behaviour with. Could it be a bug or is it related to how the values are calculated by the printer definition? Because there seems to be some calculating going on, I didn't take the time to figure out what the formulas mean, but I have the feeling there's something fishy going on... (rant: there are some really stupid questionable standard settings in the fdmprinter definition that are also used for the Ultimaker profiles, such as - when 100 % infill (lines) is selected, bottom layers is turned up to 99999 so that effectively the part only consists of bottom layers - I might want to use different line widths, speeds etc for the infill, why would this be the default behaviour???) nevertheless: here's the code for the scaling (from fdmprinter.def.json), what's going on there? might this be the reason? "material_shrinkage_percentage": { "label": "Scaling Factor Shrinkage Compensation", "description": "To compensate for the shrinkage of the material as it cools down, the model will be scaled with this factor.", "unit": "%", "type": "float", "default_value": 100.0, "enabled": true, "minimum_value": "0.001", "minimum_value_warning": "100", "maximum_value_warning": "120", "settable_per_mesh": false, "settable_per_extruder": false, "resolve": "sum(extruderValues(\"material_shrinkage_percentage\")) / len(extruderValues(\"material_shrinkage_percentage\"))", "children": { "material_shrinkage_percentage_xy": { "label": "Horizontal Scaling Factor Shrinkage Compensation", "description": "To compensate for the shrinkage of the material as it cools down, the model will be scaled with this factor in the XY-direction (horizontally).", "unit": "%", "type": "float", "default_value": 100.0, "enabled": true, "minimum_value": "0.001", "minimum_value_warning": "100", "maximum_value_warning": "120", "settable_per_mesh": false, "settable_per_extruder": false, "resolve": "sum(extruderValues(\"material_shrinkage_percentage_xy\")) / len(extruderValues(\"material_shrinkage_percentage_xy\"))", "value": "material_shrinkage_percentage" }, "material_shrinkage_percentage_z": { "label": "Vertical Scaling Factor Shrinkage Compensation", "description": "To compensate for the shrinkage of the material as it cools down, the model will be scaled with this factor in the Z-direction (vertically).", "unit": "%", "type": "float", "default_value": 100.0, "enabled": true, "minimum_value": "0.001", "minimum_value_warning": "100", "maximum_value_warning": "120", "settable_per_mesh": false, "settable_per_extruder": false, "resolve": "sum(extruderValues(\"material_shrinkage_percentage_z\")) / len(extruderValues(\"material_shrinkage_percentage_z\"))", "value": "material_shrinkage_percentage" } Final question; am I doing this right? Or do the more advanced settings like horizontal expansion etc. have to be included in a quality config file and would this solve my problems? (rant: yet another preset file, yet another scripting language?? WTF) reference 3MF project enclosed, remove .zip thanks for taking some time, best, Niklas igus_igliduri180_1-75.xml.fdm_material.zip demonstration.3mf.zip

@Noppi do you have coasting enabled? It stops extrusion a bit before the actual end of a line to release pressure in the nozzle, this is mostly relevant for bowden extruders to prevent stringing. It shows in the GCode just like it does in your screenshot.

Dann bin ich leider auch überfragt. Du könntest mal versuchen, dir beide Gcode-Dateien nebeneinander zu öffnen und vergleichen, vielleicht siehst du dann, welches Verhalten zu dem Problem führen kann. Und diese auch mal hier hochladen, damit evtl. andere sich das anschauen können.

Nebenbei: Combing ist etwas schwierig bei Cura. Wenn man es ausstellt, macht es fast nach jeder Bewegung eine Retraction. Wenn man es einstellt, versucht es ewig lange Umwege zu fahren, dabei im Bauteil zu bleiben, um eine Retraction zu vermeiden. Beides ist auf Ultimaker-Drucker hin optimiert, da die nicht die beste Filamentkontrolle haben (aufgrund des Bowdens), daher Retractions eher zu vermeiden sind; jedoch aufgrund des leichten Druckkopfs schnelle Bewegungen machen können. Bei anderen Druckern und besonders bei Direct Drive ist das aber nicht so schlau. Bei sehr langen "Combing"-Moves im Bauteil, d.h. ohne Retractions über das Bauteil fahren, läuft jedoch trotzdem sehr viel Material aus der Düse, dass dann beim erneuten Ansetzen der Druckbahn fehlt. Meine Lösung sieht so aus: ich wähle für Combing die Einstellung "Not in Skin" und zusätzlich stelle ich bei der versteckten Option "max comb distance with no retract" den Wert 5-10 mm ein, dann macht er für kleine Wege einen Combing-Move, bei größeren aber trotzdem eine richtige Retraction und es läuft kein Material aus der Düse.

prüf mal die Reihenfolge der Wände. Cura 5 hat aus einem unerfindlichen Grund die Reihenfolge der Wände zu "Außen nach Innen" als Standardwert geändert. Die Löcher sehen so aus, als wäre nach einem längeren Verfahrweg nicht ausreichend Material in der Düse, um direkt mit einer sauberen Bahn loszulegen. Wenn die Reihenfolge "Innen nach Außen" ist, merkt man das nicht, da die außenliegende Wand direkt nach der innenliegenden, d.h. ohne vorherigen Travel-Move, begonnen wird und damit ohne das o.g. Problem.

let's agree to disagree instead of engaging in a pointless keyboard fight*. For metric users, you're quoting non-issues. Engineers will use millimeters, woodworkers will use centimeters, builders will use meters or (decimal) fractions thereof, but everybody understands each other without issue, and calculate between units (nobody needs to remember how many inches are in a foot, how many ounces are in a quart, how many bananas there are in a bathtub - am I doing this right? 😄 ). Nobody uses decimeters, but they aren't forced upon anyone, they're just there because how the decimal system works. And if the whole world had agreed to use metric instead of the whole world minus one or two countries, this thread wouldn't exist. *and that's all I'm going to add to this before I have to face allegations of nourishing the online goblin

yeah that's probably where this analogy doesn't work any more, since I can't really find any downsides to the metric system. Maybe that kids don't learn fractions as early on as they probably do in the US, but then, why should they...

LOL american egocentrism is really remarkable. It's one thing to stick to horseriding when the rest of the world has agreed to use cars (there might be "good" reasons for it - having the infrastructure or just doing it because it's always been done)! But to tell oneself and others, that it's the only way and everyone else is an idiot (for driving a car because they have to buy gas and can't even grow their own horsefeed) or they're only doing it because they're not Free(TM) and forced by their government - that is truly, truly remarkable.

ich habe bisher bei jedem der drei Ultimaker (3, S3, S5) einmal das Sensorboard getauscht. Wenn das "tut", dann läufts auch - allerdings muss ich immer das Z-Offset plugin aktiviert haben und einen Offset von 0,1 mm bei jedem Profil drin haben, sonst ist die erste Schicht stets zu dünn - weil der Ultimaker das so für richtig hält.

Moin, damit das Auto-Bed-Leveling mit diesen bescheuerten Sensoren vernünftig funktioniert, müssen beim manuellen Nivellieren die Bettschrauben auf eine bestimmte Weise angezogen werden. https://support.ultimaker.com/hc/en-us/articles/360013133699-Active-leveling-errors For the Ultimaker S5, the thumb wheels should be fully tightened. Then, loosen with 2 full turns clockwise. ...also voll anziehen, dann zwei Drehungen gegen den Uhrzeigersinn wieder heraus, und dann Feintunen mit der manuellen Nivellierungsprozedur. schöne Grüße, Niklas

Hallo Lothar, ich finde die Lösung für das, was sie ist, gar nicht so schlecht. Ich behelfe mir wie folgt: - das neue Silikonteil von unten durch beide Bleche drücken, dass die Düsenbereiche vollständig sitzen - die schmalen Stege zwischen bzw. unterhalb der Düsen (die T-förmig angeordnet sind) drücke ich mit einem Schlitzschraubendreher von unten durch, sodass die dicke Stelle, die das Silikonteil hält, durch das innere lose Blech durchgedrückt wird - mit einer Pinzette die beiden Laschen außen um das Blech ziehen, damit das Silikonteil nicht mehr nach unten wegrutscht. viel Erfolg!

das war ein super Tipp, ausgezeichnete Unterhaltung 😄

Hi, I also have the problem with all my Ultimaker printers, that without any fix, the auto bed leveling that is done before each print is too close to the build plate. This results in a shitty first layer (not quite as bad as in your case, where the nozzle seems to be touching the build plate). What has helped me is the "z offset plugin" by ahoeben. For the Ultimaker S-line printers, you have to have "extensive G-Code processing" enabled, iirc because the z coordinate is not relative in the gcode or something like this. It's just a workaround, but it works well. I have 0,1 mm of offset for my UMS5 and UMS3 which results in a better first layer. You might need even more. FWIW I think the bed leveling sensor is not a very good design, albeit cheap to manufacture since it's just a pcb. In my opinion, a mechanical switch probe would be the best solution. Of course this is difficult to implement since it has to stick out below the nozzle, but hey, Ultimaker has some talented engineers so I think it should be possible to find a solution. Viel Erfolg! LG, Niklas

thank you guys, I hadn't thought this far but it makes a lot of sense. I will try to adjust the minimum line width setting. ...I do think that the new engine brings a lot more flexibility but we'll have to get used to it and learn all the parameters to become arachne masters! Best, Niklas

Hi all, the title pretty much sums it up. I haven't found a way to restrict the maximum line width that Cura chooses. While I love the new variable line width feature, it's become a bit frustrating at times because especially with thin walled prints, Cura chooses one over two lines a bit too often for my taste. Aside from that, there are some more technical materials that I use, which don't like to be extruded at 200 % nozzle diameter and/or the extruder just can't keep up with (looking at you especially, Ultimaker S5 bowden extruder). It seems as if the variable line width goes up as much as 200 % of the nozzle diameter printing a single line wall is a neat feature, but often two wall lines is still preferred, if the wall thickness allows it some materials can't be extruded at higher nozzle diameter/LW ratios, especially filled materials (haven't tried with CF though) I could just add a feature request on Github, but it seems as there are a lot of users at the moment that mistake their 3D model's or computer's problems with the slicer's problems, and I don't want this to go under. Also I thought I put this up for discussion and see if other users have the same issue and share my feelings. thanks, best regards Niklas

this is old Cura's weird way of filling gaps between walls. - select "nowhere" for the setting "fill gaps between walls", you'll get rid of the weird infill but will have two separated walls that are not connected - try the new Cura 5 beta, it has a new slicing engine that can vary the wall thickness to perfectly fill these gaps with circular lines, as you would expect

Hi, these news were certainly unexpected. And completely unthinkable some years back, as Joris Peels put it. It will be interesting to see how this moves forward. IMO Ultimaker is in a very interesting position, or was before yesterday, as they have arguably the best or most comprehensive software offering that enables digital workflows on an industry level. Add to this the propably most extensive material portfolio (as it represents a big chunk of the materials market). In contrast to this, the UMS5 leaves quite a bit to be desired, especially compared to more recent offerings in the same ballpark. If taking on Makerbot also means the adoption of IP, developers and a bunch of cash in the process, it will hopefully mean a new, competitive hardware lineup in the not too distant future!

Hi Lorenzo, unfortunately I just now noticed your campaign. Are you still looking for feedback on this or have you already implementet all learnings into Cura 5, and are done with it? If it's not too late; I work a lot with Cura, slice for Ultimaker and other machines as well. I design parts myself and also do a lot of multimaterial printing (most capable machine is an E3D Toolchanger with 4 toolheads). I have some strong opinions on this topic in particular, as I feel that multimaterial handling in Cura leaves a lot to be desired. CAD integration is another topic I can say a few words about.

Hi all, from my experience, Arc Welder and Ultimaker printers don't play nicely together. My impression was that the firmware doesn't recognize arcs at all, but what @ahoeben says makes more sense. FWIW you won't get too much more quality out of your general purpose print profile, for me it's a big advantage to have Arc Welder (use it for my other printers, such as the E3D Toolchanger) because I print a lot of small circular bearings, which make the Ultimakers stutter. Another nice side effect is that Gcode becomes a lot smaller in size which makes saving and uploading faster.

it might be that in 4.13, Cura included a back & forth path for each comb tooth. In 5.0, it might only do a single line, retract and move to the next line, which would be the next tooth right next to it, unretract and start again. the "bridge" that you are seeing is probably stringing from the retraction, aka a printer problem. Possibly cura is doing a very thick extrusion, it seems to do that since you can't specify a maximum line width IIRC, which would make clean retractions more difficult since it is pushing more plastic than usual. It might help to set combing to "all" so that the nozzle will travel back over the already printed line and start at the beginning of the next tooth and not at the end. take a look at the gcode preview to validate this.

printing over USB is so 2014 and there are so much better ways of bringing the GCode to your printer (Octoprint should be the easiest and best for your use case). Take a look at this solution and be amazed. "Wer nicht mit der Zeit geht, geht mit der Zeit..."

if you ever want to discuss and get end-user feedback to find out what the users would like these changes to look like, I'd be happy to join this conversation.

After the development of arachne and its implementation in Cura 5 (edit: and all the great improvements that came with both), the handling of print, printer and material parameters is my biggest issue with Cura that's left. There are profiles, intents, qualities that seem way more focussed towards filament suppliers in terms of creation, maintaining and finetuning. For the end-user, it seems that they have to dive into what setting is stored where and learn xml in the process, because the Cura interface doesn't seem to have a function for this (all custom settings are still stored in custom profiles, right?). It seems as this is historically grown and started out with just a profile for each set of parameters, and was expanded upon with "intents, qualities" etc. For the end user, a system like in Prusaslicer or Superslicer that has separate profiles for the print, the printer and the material, it seems way more intuitive and easy to use. I'd still wish for more dependencies, e.g. having a base (print) profile that can continuously be fine tuned, and basing different profiles for different layer heights, nozzles etc on that which inherit settings from the base profile. FWIW my use case involves a multimaterial printer with 4 different toolheads. This is a breeze in Superslicer to setup and a pain in Cura, because you have to maintain a profile for each combination of filaments, layer heights etc., as opposed to just being able to select a print and material profile from the existing list for each print head. Cura adds to the confusion (just for example) by relying on parameters such as surface tension and adhesion tendency to determine which materials and in which order to choose for the creation of the prime tower, parameters which are normally hidden and - as mentioned above - are aimed towards material suppliers to be determined in the creation of the print profiles that go onto the marketplace. I've voiced this concern on this forum already, but since it would mean a complete remodel of the print profile system I don't really see a chance for a change... I hope to be proven wrong some day. edit 2: I wish that some people around here would appreciate the amount of value that they receive for free in the form of a great software, instead of complaining in a rude and not respectful manner about problems that mainly arise from their own choice of hardware/software. A general rule should be: if I'm using (open source) software that I got for free, I either contribute to the improvement by pointing out bugs and issues in a respectful and orderly manner, e.g. on github, or STOP COMPLAINING!

if this filament is PA based and/or the support is PVA, these strings might be related to moisture. You need to thoroughly dry your filaments and keep them dry, this will make the filament's flow more predictable and it will ooze less. you might also want to tune your retraction settings if not already. best, Niklas