I feel bad bringing all my troubles here. But I am running into a few weird issues on my new hbot style printer. My 100% infill layers are not the same traveling at 45deg vs 135deg. The fill traveling at 45deg looks pretty good. But the fill traveling at 135deg has gaps between the paths.
Also I get a slight cats-eye effect on circles timed at the 45deg and 224deg positions. I think based on the corexy formula this is where a direction change happens on 1 of the steppers. Print speed seems to have no effect on this issue.
I think these two problems are related… But there is less discussion and calibration documentation for corexy/hbot style printers compared to standard Cartesian or Delta printers.
Any help is appreciated. And if anyone has any other pointers after looking at the pictures for additional improvements I’m all ears.
@Eclsnowman try checking your belt tension first. Always the obvious first. Then take your trusty calipers and check x then y then just for novelty try the 4 corners angle directions by moving x and y equal amounts and use a^2+b^2 =c^2 to determine how far it should have moved. probably over kill but may identify some improper steps per mm calculations. Try with short and long movements
Yeah, eye patterns are sure signs that there is slop in one of the axes. Though Hbot’s combine the two axes together, you can still break down each of the axes to try to determine where the error is coming from.
Yep, now I look at those pictures more closely. It does look like a mechanical issue. Belts should be tight enough to play a tune on. Make sure your steppers are getting enough current. Check for loose screws etc.
@D_Rob thanks. I did make a bunch of calibration squares today. 20x20x5, 20x20x20, 40x40x5, and some 20roundx5 and 40roundx5. But I am unclear how to apply it to hbot since X and Y actually travel at 45 degrees if only one stepper is moving. X and Y work together to move in standard coordinates.
Now that I just typed that out I think I can check the steppers by just rotating my calibration objects 45deg . That way I am only seeing the movement of 1 stepper at a time when measuring. YAY.
The problem is that the infill direction is controlled by the slicer – if your slicer allows you to change directions, great! But otherwise, the lash maze will really help isolate movements to X and Y.
@Eclsnowman Bring all the problems you want, remember your solutions, and share them when the time comes. As a community of makes dedicated to building new things we must also build each other up. Working together we’ll shape the future
Is it an HBot or Corexy? there are significant differences in how the X gantry can move. HBot is subject to racking whereas Corexy is not as much. If you are using HBot configuration then I would suspect this is caused by racking of the X gantry. I just completed my CoreXY, I converted it over from HBot because of the racking. I see very nice diagonals on the infill.
This is the same results I got on the h bots I built. Extremely tight belts should fix it. Be sure to tighten them often until the stretch subsides. Then re calibrate.
@Eclsnowman , as @D_Rob stated, bring all the problems you got buddy! We’ll answer them. You learn from your mistakes. Help others when the time comes!
Honestly, I’m super relieved you’re posting so much about your build, as not much is posted about h bots or core xy. I’m totally convinced that I should go with the core xy now
@Eric_Moy the core xy does look cool but I am hooked on the ultimaker style gantry.i will say this though the extra rods can be a pain because the tolerances are so exact or the axels will show every wobble in the print
@D_Rob , I must admit, I’m baffled why I would not want to use the elegant and effective Ultimaker gantry, but I really like the minimal parts involved in the core xy. The only issue I have is the known issues with h bot, but in the little posts I’ve read, core xy fixes this. This is why I’m hoping @Eclsnowman converts to core xy, so I can learn from his posts. I might be looking to buy an Ultimaker 2 for work, so that should settle any curiosities for me.
Last night I tightened the belts to the point I could play a High E. No effect on cats eye issue. Tonight I will test playing with acceleration settings, and test Cura to make sure it isn’t a slic3r issue. I will report back.
I also noticed a design issue in my build. I think I need to add set screws where my x rods meet the Y gantry. They tend to bow out with the extra belt tension I am using now.
I will also check the current settings. Anyone have a link of the proper method to adjust the pots on the stepper drivers. Also what do the jerk settings in marlin do?
. That way I am only seeing the movement of 1 stepper at a time when measuring. YAY.