Link

I have nothiced something odd with infil, this tube is the same size all the way along its length for example but the infil seems to change is pattern slightly in different points for no apparent reason, this happens with all infil types pretty much. Any idea why this would be ?

could well be, there also seems to be the added complication of the extra infill wall throwing things out, with that enabled nothing seems to keep the seam aligned.

I have been experimenting with the 'sharpest corner' function and it doesn't work consistently on all models as you would expect, despite the corner not changing in the model the location for the seam moves (as seen in the original problem). Also when you manual set the location of the seam it results in much improved travel moves, Cura seems to be able to much more efficiently calculate its paths when you set the seam location. Using sharpest corning seems to make the whole model way more complicated for Cura to handle (or it would appear so)

many thanks - really useful 🙂 Are you aware of the sharpest corner issue being fixed ?

As usual you nailed !, thank you Sir ! it seems to be a combination of the extra infil wall count and the seam not locking as you said when you pick sharpest corner. Removing the extra wall and setting the seam to user specified (even with grid infil) sorts the issue. When you set the seam to user specified do you leave it to the default location it picks, rather than actually type co ordinates in ? Thanks again

thanks UM2_hanger.3mf

Hi, I am printing a basic cylinder and as can be seen from the image at a point near the top the z seam moves and the travels change despite the model being exactly the same over its length. I have tried every option of layer start position and z seam alignment and still at the same point on the model the the travels change and the seam moves. As said the model is the same the whole way up, just a simple cylinder with one flat side and infil. I have also tried all versions of combing and again no change any ideas @smartavionics or anyone else ? thanks !

hi @smartavionics, I am using your new 'not in skin' combing but getting blobs on the outside of the wall at these points (shown below), as mentioned a while back (and the reason i asked you to keep 'only in infil') Cura treats walls as skin rather than just the top and bottom layers and therefore doesn't retract in the walls, hence the material from the non-retract travel is getting deposited in the wall which is often too thin to hide it. In the screen shot below there is a strange 'kick' in the travel at which point the blob appears. Any ideas ?. could the travel be made to not perform this kick, or stick more to the inside of the wall ? i can switch back to 'within infil' and will get retracts in the walls which fixes the problem, but think you mentioned that Cura has issues deciding where infil is and hence the 'only in infil' can sometimes result in travls without retract actually in the top or bottom layers (skin) ? Many thanks !

Hi, Can anyone explain what this feature does, by default there are settings in place for both expansion and removal, so does this cancel itself out ?, its not clear what this actually does and what changing the settings will do. @smartavionics could you advise ? many thanks

hi @smartavionics did this make it into 3.6 ?

ok, thanks. doesn't sound like something worth doing then.

no, smoothers are a HW mode, they are a used to smooth the supply to the stepper motors http://blog.trinamic.com/2018/08/09/tl-smoothers-do-3d-printers-really-need-them/ they are a popular mode on some printers, so not sure if the UM can benefit from them

Does the 2+ come with smoothers already installed from UM ?, if not are they needed ?, they seem to be a popular printer on other printers but not heard about them on UM printers

Thanks, as mentioned in the other thread I now have the Tinker firmware building as he set it up with the make and the simulator. I am also able to build as you described using just the Arduino IDE, for some reason it doesn't seem to build with the latest IDE version, dropping back fixed my problem.

interesting, I would have assumed that the 2+ was the biggest seller still for UM, but guess the other two models are considered newer and maybe need more tuning as the 2+ could be considered optimised. I managed to get the Tinker build working as it was rebased and have managed to change the few things I wanted (heater PID, default retraction etc for the 2+), so I am really happy with what I have now. I am looking to add a cold pull feature and maybe look into how your 2+ code reads retraction from the SD card and include that into the Tinker branch. Thanks for your help, I will obviously help with the Beta when I can (when I can supervise the print), let me know if I can assist in anyway. Cheers

Yea, sorry, just cut it down

I get this list of errors if it try to do that, I am using the latest Arduino IDE and TinkerFirmware Arduino: 1.8.6 (Mac OS X), Board: "Arduino/Genuino Mega or Mega 2560, ATmega2560 (Mega 2560)" In file included from sketch/Marlin.h:21:0, from sketch/Marlin_main.cpp:30: pins.h:1248:41: error: pasting "/* PG2*/" and "_RPORT" does not give a valid preprocessing token #define SDCARDDETECT 39 // PG2

[gr5 removed massive quote of first post of this topic] could you help with this issue ? just about there, I have a compilation error related to the Arduino.h defining the max and min macro. It shouldn't really define those, but has, how did you get around that ? Thanks c:\mingw\lib\gcc\mingw32\6.3.0\include\c++\bits\stl_algobase.h|243|error: macro "min" passed 3 arguments, but takes just 2|

ok, I fixed it #ifdef min #undef min #endif #ifdef max #undef max #endif was needed

thanks, just about there, I have a compilation error related to the Arduino.h defining the max and min macro. It shouldn't really define those, but has, how did you get around that ? Thanks c:\mingw\lib\gcc\mingw32\6.3.0\include\c++\bits\stl_algobase.h|243|error: macro "min" passed 3 arguments, but takes just 2|

Hi, I am interested in making some local changes and if they work out to be useful I would like to become a contributor. Is there a final guide of how to build the Tinker firmware ?, the most recent discussion here seems quite convoluted and not straight forward. I found this link http://www.extrudable.me/2013/05/03/building-marlin-from-scratch/ Does that work ?

Does anyone bother calibrating the extruder esteps on a 2+ ?, or is the default considered near enough ? is there a guide of how do do it (if worth doing) ?

@3rdpig interesting, I recently went from 0.35 (the default in Cura for a Ultimaker 2+) to 0.4 and have to admit the parts look way way better, the top surface looks so much better not to mention the walls etc. All this would indicate that Cura seems to work better with line widths at least the size of the nozzle. My parts never looked as good with a 0.35 line even without actual gaps issue, the top layer looked very rough compared to a 0.4 nozzle.

@3rdpig what line width did you settle on in the end ?, what did you start with 0.35 ?.

looking at the model it would be worth trying changing the 'Seam Corner Preference', I am not clear if setting this to 'hide seam' for instance is having some odd affect in some scenarios, however setting this to 'none' certainly seems to move the seam point and worth a go