Oddly shaped holes?
-
@arhi I did all of that. I took start and end gcode I have for my slicer and pasted it in there. Not sure what the issue is. I changed the F value as well because I think it was a little too low. I had 1800 in my setup.
-
@Surgikill I used 10mm/sec as "safe"
.. you can increase .. you can try to increase the E values, maybe I was too conservative and put too low valuesbut it can't be printing in air
the first line after M83 (M83 just set extruder to relative positioning) is
G0X0Y20Z0.25
that's move (G0) to X=0 (X0) Y=20 (Y20) and Z=0.25 (Z0.25)
Z is not set after that so all moves here needs to be on Z=0.25mm
till the last code G0Z20 that moves Z straight up 20mm
The E value is maybe to low and maybe you need to prime it first, something like this:
;bed temp, extruder temp, home.. M190 S60 ; bed 60 M109 S230 ; nozzle 230 M106 S0 ; fan off G28 ; home ;... G21 ; millimeters G90 ; absolute XY moves M83 ; RELATIVE extruder move M200 D0 ; no volumetric extrusion G0X0Y0Z10 ; go to 0,0,10 G1E50F100 ; extrude a blob of plastic in the air to prime the nozzle G1X0Y20Z0.25 ; go to start position, 0.25mm layer G1X10Y10 G1X50Y50F600E2.352 G1X80Y10F600E2.352 G1X100Y30F600E1.176 G1X70Y70F600E2.352 G1X50Y50F600E1.176 G0Z20 M104 S0 ; turn off extruder M140 S0 ; turn off bed M106 S0 ; turn off fan ;... ; shut everything down here -
@arhi This is the start of one of my gcode files. I forget exactly what I had to do, but I think I wrote something special in there because I'm using 4 independent z motors to tram the bed.
M107 M107 M104 S235 ; set temperature G32 ;home and level G28 ;home G21 ; set units to millimeters G90 ; use absolute coordinates M82 ; use absolute distances for extrusion G92 E0 ; Filament gcode M109 S235 ; set temperature and wait for it to be reached G21 ; set units to millimeters G90 ; use absolute coordinates M82 ; use absolute distances for extrusion G92 E0 G1 Z0.200 F7800.000 G1 E-2.00000 F2400.00000 G92 E0 G1 Z0.350 F7800.000 G1 E-2.00000 F2400.00000 G92 E0 G1 X91.390 Y102.905 F7800.000 G1 E2.00000 F2400.00000 G1 F1800.000 G1 X93.026 Y102.689 E2.11451 G1 X117.999 Y102.489 E3.84808 G1 X179.410 Y102.488 E8.11097 G1 X207.724 Y102.689 E10.07646 G1 X208.948 Y102.812 E10.16187 G1 X209.906 Y103.069 E10.23068 G1 X210.823 Y103.463 E10.29994 G1 X211.673 Y103.985 E10.36920 G1 X212.434 Y104.621 E10.43802 G1 X213.273 Y105.598 E10.52746 G1 X215.387 Y108.621 E10.78349 G1 X215.889 Y109.505 E10.85410I'm going to try adding your code in with the prime on the extruder. I usually use a skirt, so that's my priming.
-
@arhi Yea I just ran the updated Gcode and it's still printing in air and not extruding anything. This is what I have for the gcode.
M107 M107 M104 S235 ; set temperature G32 ;home and level G28 ;home G21 ; set units to millimeters G90 ; use absolute coordinates M82 ; use absolute distances for extrusion G92 E0 ; Filament gcode M109 S235 ; set temperature and wait for it to be reached G21 ; set units to millimeters G90 ; use absolute coordinates M82 ; use absolute distances for extrusion G92 E0 G1 Z0.200 F7800.000 G1 E-2.00000 F2400.00000 G92 E0 G1 Z0.350 F7800.000 G1 E-2.00000 F2400.00000 G92 E0 G1 X91.390 Y102.905 F7800.000 G1 E2.00000 F2400.00000 G1 F1800.000 ;... G21 ; millimeters G90 ; absolute XY moves M83 ; RELATIVE extruder move M200 D0 ; no volumetric extrusion G0X0Y0Z10 ; go to 0,0,10 G1E50F100 ; extrude a blob of plastic in the air to prime the nozzle G1X0Y20Z0.25 ; go to start position, 0.25mm layer G1X10Y10 G1X50Y50F600E2.352 G1X80Y10F600E2.352 G1X100Y30F600E1.176 G1X70Y70F600E2.352 G1X50Y50F600E1.176 G0Z20 ;... ; shut everything down here G92 E0 M107 ; Filament-specific end gcode ;END gcode for filament M104 S0 ; turn off hotend M140 S0 ; turn off bed G28 X0 Y0 ; home X axis G1 Z310 ;move Z to max M84 ; disable motors -
@Surgikill said in Oddly shaped holes?:
still printing in air and not extruding anything
the extrusion, you can try increasing the E values in those codes
but "in the air" .. no idea, it must be at z 0.25 .. that is very close to table
.. no clue how it messes up -
@arhi This might help. The circles on the bottom of the picture are closest to Y max, and the circles at the top of the picture are closes to Y min. X min is right side, X max is left side. It definitely look like the circles get worse as it goes towards Y max.
-
Sorry mate, I'm lost here. You really need someone who actually has experience with coreXY machine. With the CNC this type of error is always backlash. No clue how backlash on each motor affect this at all. Normally you can tweak backlash compensation in the firmware of your CNC machine, but AFAIK there is no support for that in RRF (not sure any 3d printing firmware supports it, I know there was try to implement it in marlin but I think they gave up, never entered master branch).
-
@arhi My theory here is that the igus bushings are binding up near Y max. This is causing 2 things:
- I lose microstepping, which causes a loss in resolution
- My belts stretch more (it's a pretty large printer, so the belts are pretty long) and that creates the "backlash" issue
I'll have the LM16UU bearings in this coming week. Those will get installed and hopefully this issue will be resolved.
-
It is a good theory as any. I see even on large printers ppl go with 2mm belts which is weird to me, the only reason to go with 2mm pitch is tight bend radius, but if you are on big printer you can design so that you don't need tight bend radius and go with 5mm pitch where for much less $$ you can get higher quality belts .. but ..
-
@arhi If I still have an issue, I'll re-design for 5mm pitch. I think the 2mm should be fine. I might need to upgrade to a non chinesium belt.
-
@Surgikill even chinesium ones at 5mm work nice, the 2mm prc belts stretch, while a lot of ppl online try to say they don't my experience show different, also no proper reinforcement... 5mm you can chose if you want glass or steel, last a long time, only issue is if you go 180 degrees it has to go over large diameter (18 or 20 5mm pitch teeth is rather large circle) .. so the idea is to design motor to sit on the "straight" path of the belt with some idlers (3 point setup) rather then doing 180 around the motor forcing you to use large diameter driving pulley making 1 step of the motor move too many mm of belt
also, using 5-6mm wide belts on big printers... IMO 5mm is ok on something like 200x200mm print surface, you go over you need to step up to 10mm or more .. if you look at professional machines it's not hard to find belts that are in 10cm width class
... I remember some stratasys printer I was repairing back in the day motor was linked to extruder with belt that was 4-5cm wide, X and Y belts were both 6-7cm wide 5mm pitch belts ... -
@Surgikill said in Oddly shaped holes?:
LM16UU
I think you’re on the right path regarding the igus bushings causing the oddly shaped circles. I bought into the hype of the polymer bushings but had print quality issues. In my case the sticktion could be felt when manually moving the axis with the belts removed. Either they were loose to prevent sticktion and had excessive play or they were tight and caused the oddly shaped circles like you are experiencing. Also found that the rods need to be completely dry and cleaned regularly. Not worth it, went back to LM16UU bearings and the issue went away.
-
@mwolter Yup. I've had the same issues where they're either too loose or too tight. Hopefully the LM16UU fix that issue.
-
Those IGUS bushings have pretty tight tolerances on how they are supposed to be mounted. There is a lot to read on their website.
I have tried them a few times, but as many of you say they either bind or have play.
They seem to be popular in the industry, "lubricated by dirt". -
Igus has plastic bushings that are encapsulated in a hard shell. This way you can clamp it without increasing the binding on the rod. E.g.
https://www.amazon.com/gp/product/B00QZ7YJ7W
Linear bearings are available in double length but I couldn't find double length igus bushings and installing two one after the other creates non co-linearity that increases the binding.
-
@bondus Yes, the Igus bearings are built to quite tight tolerances, and to not like any mounting error or rod misalignment. Industrial equipment, which is what they are designed for, usually is toleranced more closely than 3d printers. I designed a bearing mount: https://github.com/mendenm/hemera_mount_plate which has a self-aligning lower bearing block which avoids this problem.
-
@zapta Yea I was looking at those but I wasn't going to pay 20-30 dollars a bushing when I need 8 of them. I'll just stick with the regular bearings. Are there any recommendations on good bearings? I have all chinesium ones right now, but if I find out that it does end up fixing my problem, I'll most likely upgrade to better bearings.
-
I used to use Thomson Super8 bearings when I was still using 1/2" round guide rails. I was buying them new on ebay for $5-8 each. They allow adjustment of preload and the races can tilt a little to compensate for slightly inaccurate mounting.
Make sure your rails are hardened- ball bearings will cut grooves into unhardened rails.
-
@mrehorstdmd @zapta @mendenmh @mwolter @arhi I just replaced the Igus bushings with the LM16UU I ordered. No more issues. Now I have no idea what to do with all these Igus bushings I have.
-
I'm a little late to the party I know but I started down the carbon fiber/igus route back in 2017 and although it "worked" it was way too finicky. The plain igus bearings MUST be press fit into a very specific hole diameter to be usable with the specified diameter diameter rods. Most of the carbon fiber rods I bought needed tuning with 1000-1500-2000 grit sandpaper to get them to have constant and reasonable friction in the bearings. Some could never be tuned properly because they were too narrow to begin with. Also, while most of the roll-wrapped ones were spot on for straightness, some were not which made trying to get matched sets almost impossible. In the end, I gave up.
