visu-al

is it possible to just add a stronger heatunit that uses its on power supply ? (i mean obviously it is but without changing something else just plug in the allready existent cables/plugs)

i just say because i usually set the fans in material settings in the printer and leave the cura setting default.. and for testing i set them always on 100% because that works best on the older um2 as well

here is one cube from my new series its a 0.2mm sliced at workstation1: https://www.dropbox.com/s/d846ud3twwmv5ou/w1z2.gcode?dl=0https://www.dropbox.com/s/d846ud3twwmv5ou/w1z2.gcode?dl=0

did you use the fans on 100%? because i had to on the 0.05mm layer calibc cube

i like that cause i hate it sometimes @ infill printing it starts wobble and that creates some peaks of material were it goes over next time and wobbles even harder and creates higher peaks until the whole thing souns terrible but most of the time this doesn't destroy my prints and creates a nice finish still at the end..

Well here i got some pictures to discuss: Orange pieces are from my older UM2 and the red ones from the new 0.2mm Layer Cura home 0.2mm Layer Cura work1 0.2mm Layer Cura work2 0.2mm Layer Cura home 0.2mm Layer Cura work1 0.2mm Layer Cura work2 0.2mm Layer different (regular) gcode same other angle 0.1mm Layer Cura work1 0.1mm Layer Cura work2 Comparisons: as you can see at some angles the effect is very visible and on some not I cutted two pieces to compare the infill Its clearly visible that the RED part show a lot stronger pipes (i know the pictures aren't very awesome^^) I FINALLY i printed the 6mm/s cube to proof gr5 that it doesn't change the outcome and..: It looks better! LOL that took hours to print.. sorry but no picture yet What is there that i need to slow down 90% to have same quality then my older UM2?? And why this does not happen with 0.1mm layer at much higher extrusion rates???

OK thanks for the research but don't you forget that i printed with 25% (12.5mm/s)speed to test and i also printed cubes with 0.4 and 0.8 shells wich looked very similar.. i also printed with 240° once.. and to say this: on my other printer i print with 0.5, 0.6 and 0.9 shells with absolutely not having this problem...? But anyway ill go to print a cube went wrong again 10% speed to see if its true.. If you would have read my posts you would know that my newly sliced cubes with 0.8 shells still looks the same on new machine and the same cube printed on my older um2 looks fine and they have 25mm/s speed.. It must be something else i guess cause my two machines behave very differently and they can do nearly the same extrusion test result 13mm3/s and 15mm3/s

Ill upload some pictures tomorrow if i find enough time.. Actually im testing the new sliced cubes at my home machine and compared to the other machines prints they are again nice expect a strange other effect. until now all new 0.2 cubes printed on the newer um2 are showing the line seperation effect and all cubes printed on the older um2 look good so far and 0.1 layers print well on both machines still And because i choose half speed @ 0.2 layers it should be the same extrusion amount per time wich means not more resistance to print then the 0.1 right?

@Nicolinux Yes the second one i sliced to test very small layers. unfortunately the 0.1 before was overwritten^^ I saw the top layer of the new print were i used regular gcode and it still looks ugly and the same.. i will upload my results soon..

@gr5 thanks im printing the one with regular gcode (0.2 lh) right now and it looks better, the infill lines are more like a wall then the stacked pipes before.. also allready printed new sliced 0.2 layer from both curas and both have still the stacked-pipes effect even if the surfaces for some reasons look better(but still not perfect) filled now..

@LePaul Thanks for the picture i would probably need a bit more but will be hard to find time to model this week^^

@printerfan No, i don't send you my girlfriend haha Maybe ill try to scan something else in the next days and provide the file here. I use blender.

Ok here are my gcodes i used for the tests of last weeks: https://www.dropbox.com/s/k8nfmllrrjayuk9/calibc.gcode?dl=0 https://www.dropbox.com/s/4u69yj7ivxw0q1t/calibration_cube.gcode?dl=0

Gern geschehen. Der Extrude-Test ist bei 230°C gültig und du hast ihn auch bei 220° geschafft, das ist sehr gut! Ja, seit ich den Feeder von Robert montiert habe konnte ich alle Filamente drucken, mit dieser Führung zum anklipsen auch flexibles PLA, Filaflex und Ninjaflex. Noch viel Spass beim Drucken! Gruss visu-al

I dont know but i usualy move the sensor around focusing on the object. the software i use tracks the cameras position and accumulates some kind of textured pointcloud data wich is calculated to a mesh object you can work with..

can you show me some pictures of the belt your speaking about? If the geometrie isnt too complicated you could measure some apects for me and i could model from the pictures with the measured sizes you give me a clean file probably..

if you have windows 8 and kinect for windows then you can use the "3D Builder" app from the windows app store it has mesh edit and export functions its even optimized for multitouch use I use that as well but its in early stage and i had some hard times properly scanning objects and find good settings but i am very convinced that with the kinect for windows v2 sensor i have there will be very good solutions to scan all kinds of objects that aren't tiny.. with the old kinect i saw some pretty nice scans.. but as i said i guess its useless for smaller and tiny objects and this belt stuff i guess is bit to small you would surely need to mesh editing to make it properly in the cm range..

by the way thank you all for the great support here! well i will do the following steps: upload my tested gcode file/s and post the link here slice 7 new cubes Cura @home 0.1mm layer (Ultigcode) 0.2mm layer (Ultigcode) 0.2mm layer (regular gcode) @illuminarti wich in Cura? RepRap (Marlin/Sprinter)? [*]Cura @work1 0.1mm layer (Ultigcode) 0.2mm layer (Ultigcode) [*]Cura @work2 0.1mm layer (Ultigcode) 0.2mm layer (Ultigcode) testprint 14 new « cubes » : one set of 7 on the older UM2 @home one set of 7 on the new UM2 @work 0.1mm layer-height-settings i will use : First layer height 0.1 50mm/s (25mm/s @ bottom layer) 220°C @ 100% flow and 2.85mm 0.8mm shell thickness 0.7mm bottom/top-thickness 20% infill /speed 0 Disabled retraction 0.4 nozzle size Travel speed 150mm/s Outer and inner shell speed 0 0.2mm layer-height -settings i will use : First layer height 0.2 25mm/s 220°C @ 100% flow and 2.85mm 0.8mm shell . 1.2mm top-bottom 20% infill /speed 0 Disabled retraction 0.4 nozzle size Travel speed 150mm/s Outer and inner shell speed 0 Cura will be reset to default before So then all combination of my printer and Cura installs on both layer height will be tested to compare and i will also upload the files and pictures then for you guys to check

Ok i will but the other printer (and the sd card) is located at my workplace and i cant go there over the weekend. If i dont remember wrong it was also on the new um2 with gcodes i created with the same cura/computer at home with my other um2 were it prints well. And i printed files from the other cura i use with the new printer at home wich resultet in nice prints. Wouldn't that exclude that possibility? Of course i also installed an all new unofficial cura on another computer and sliced without any other changes. Is it impossible, that somewhere between the gcode and hardware steering its misscalculating something somehow by ignoring the layer height or something? i dont have any experience with this arduino electronics i just know the basic concept of it..

I allready printed several rolls to the very end on my older um2, with roberts feeder (thank you robert) it is no problem at all maybe its really the teflon isolator but why is there no resistance with 0.1 at double speed and why does the flow set to 200% work nicely with 0.2 ? and why i get a good extrusion test? Anyway the spools are new, all of them so no special curvature effect im sure.. for me it just looks like 50% Extrusion on 0.2 layers also because i see infill like stacked pipes instead of nice walls with the 0.2 setting on just 100% flow, i agree that its underextrusion but not related to mechanical problem because its like a switch on 0.2 it turns to half of the needed material not speed or temperature related Roberts feeder is really great as soon as i printed it for my old um2 i was able to print flexible PLA, Filaflex, Ninjaflex (with the guide) without any other change or oil. everything with the same tension, i even printed conductiv filament wich is just not so funny^^ And i reached full 15 mm^3/s at 240° on the extrusion test

Sorry, i forgot to mention that i exchanged the feeder with roberts version2 like i did on my other machine because i was thinking about underextrusion at first and i can run extrusioin test up to 13mm^3/s @ 240°C wich is enough i guess. Also i allready have an older UM2 wich prints all that stuff absolutely perfect compared to this and i couldnt find a reason just that at 0.1 it prints fine and it looks like the problem of nicolinux and the others and for some reason it extrudes just half of the amount it should at 0.2 Also to mention that at first i had more kind of backlash sepearation between lines wich i kind of fixed a bit. @gr5 no the red print looks fine to me and everything i print with 0.1 looks normal to me like on my other machine (i have some of my prints on the older machine in my gallery) I have some experience with fixing underextrusion my older machine didn't always worked as well as it does now and i tried all the stuff on actual machine as well i have low resistance of the filament all the way trough the nozzle at normal temperature after some Taulman Nylon 618 atomic pulls it also didn't change and the tip was nice.. it's nearly impossible that this is an (mechanical feeder/bowden related) underextrusion problem in my eyes..

Funny because i got a brandnew Ultimaker 2 and have exactly the same problem if i make more then 0.1 Layer height. Allready checked all kinds of solutions that where discussed.. one rod is slightly excentric but wouldn't that affect also other layer heights? Here are my pictures of some tests: The last pictures shows that the effect is constant from layer 2-3 If i print with 0.1mm Layerheight its just fine. If i set the diameter to half or set flow to 200% it prints 0.2 layers fine as well^^ hmm..

Hallo Dorian Am Feeder kann man ja eine Schraube mit dem Imbus von oben drehen, da kann man die Spannung erhöhen. Wenn er zu oft skippt ohne das eine Verstopfung vorliegt ist er zu wenig gespannt aber achtung wenn du zu stark spannst frisst er sich ins Filament weil der skip nicht ausgelöst wird. Das Ultimaker blau ist sehr gut. Es muss steif sein, wenn du mit halten auf 8 kommst bedeutet das wahrscheinlich, dass du zuviel Widerstand am Eingang hast. Es gibt hier z.B. einen Filament-Guide den ich benutze https://www.youmagine.com/designs/filament-guide-um2 Falls du sicher die 10er Grenze erreichen willst, was nicht notwendig ist (schöne Drucke meist bei4-7) dann empfehle ich dir den Original-Feeder mit diesem Feeder von Robert auszutauschen https://www.youmagine.com/designs/alternative-um2-feeder-version-two falls du Fragen dazu hast, einfach Fragen :wink: wenn du den Feeder nicht zu drucken schaffst, dann kann ich dir einen schicken.. Gruss visu-al

Ach ja: Wenn du mit schlechterer Qualität drucks bedeutet das eine grössere Schichtdicke und somit mehr Materialfluss, da aber dein Filament nicht schneller durchkommt erhöht sich nur der Druck und der Feeder macht einen "skip"

Hallo Dorian Das ist so ziemlich normal und bedeutet nur, dass das Filament nicht wiederstandslos vorgschoben werden kann. Das zurückspringen ist dazu da, das Filament bei zu hohem Druck nicht gleich durchzufressen. Das kommt meistens von Verstopfungen in der Nozzle, welche man mit der "Atomic-Methode" lösen kann. Meistens hilft es schon die Temperatur zu erhöhen von den Standart 210° in 220-230° um den Widerstand zu verringern. Hier ist ein nützlicher Link für diverse Probleme: http://umforum.ultimaker.com/index.php?/topic/6574-a-visual-ultimaker-troubleshooting-guide/ Und noch der zur Atomic Methode: http://umforum.ultimaker.com/index.php?/topic/6339-atomic-method-for-various-materials/ Gruss VISUAL