Page MenuHomeAleph Objects Inc

Wiping too far to left
Closed, ResolvedPublic

Description

On KT-PR0050NA-0010 the printer is wiping too far to the left. The y-axis is set to the correct distance and the printer is homing just fine.

Event Timeline

MikeR created this task.Feb 28 2019, 8:49 AM
MikeR triaged this task as High priority.

The x-bump stop was also changed on this machine as the first one was over squished. It only helped marginally.

MikeR added a comment.Feb 28 2019, 9:13 AM

During the wipe sequence for the calibration cube, the printer wipes in the center of the pad. Do we have different wipe locations for those two tests?

MikeR assigned this task to marcio.Feb 28 2019, 9:17 AM
logan added a comment.Feb 28 2019, 9:20 AM

@MikeR I know for certain that the calibration gcodes they were given were generated with the latest start gcode, and is working fine on other machines, correct?

MikeR added a comment.Feb 28 2019, 9:26 AM

It sounds like @paulette has been having wipe issues on other machines. I know there is at least one more red tagged machine for missing the wiper pad.

I had noticed this same issue with the difference in wipe location between the tests on another machine, KT-PR0050NA-0011, but it isn't as far off the wiper pad as 0010 is.

logan added a comment.Feb 28 2019, 9:28 AM

@MikeR Is the other machine behaving the same? OK on auto cal wipe but off on vernier start? Are they all off the same direction; to the left of the pad?

MikeR added a comment.Feb 28 2019, 9:30 AM

This might be a cause of the same issue but when 0010 goes to probe, the nozzle almost misses the front right washer all together. It hits just fine on the back right washer.

logan added a comment.Feb 28 2019, 9:31 AM

You said the spacing was correct, did you check the square of Y in relation to X?

MikeR added a comment.Feb 28 2019, 9:31 AM

@logan yes both wiping too far to the left.

MikeR added a comment.Feb 28 2019, 9:34 AM

@logan checking now.

MikeR added a comment.Feb 28 2019, 9:35 AM

Yup its square.

logan renamed this task from Wiping too far too left to Wiping too far to left.Feb 28 2019, 10:31 AM

So I have found the wipe location for the calibration cube and the start gcode are wiping in different locations. This could explain why we are getting different wipes during different tests.

@anolen has updated the start gcode to help with this. Can we get an updated start gcode to calibration to test this since they don't have the most update Cura yet?

@kent @MikeR @oliver @logan

Hey! Please let me know how the new start code works. I have a suspicion it will still be a little to the left and I am working on a fix right now just in case but I don't wanna send it through without a good amount of recent feedback confirming the need. THANK YOU!

MikeR added a comment.Mar 15 2019, 8:35 AM

@anolen Ive been able to test this on one machine so far and it appears to be wiping more centered. The wipes for the calibration cube are still wiping to the left but that may be more of a general firmware thing than gcode.

MikeR added a comment.Mar 15 2019, 9:07 AM

On quiver 14, the left side wipes nice and good, on the right side though it still wipes to the left.

MikeR added a comment.Mar 15 2019, 9:43 AM

Quiver 16 still wipes a little to the left but it is better.

logan added a comment.Apr 1 2019, 12:19 PM

QB21 is also wiping too far to the left, bump stop correct, spacing of Y-axis is correct.
@west any other ideas for ways to solve this or things that might cause this to not be consistent? I looked at this machine and couldn't figure out why it is wiping to far to the left.
@MikeR did you ever find a way to solve this issue?

MikeR added a comment.Apr 1 2019, 12:27 PM

@logan I re-sliced the piano print in a newer version of cura that has some profile updates.

Try this new gcode and see if it makes a difference.

logan added a comment.Apr 1 2019, 12:31 PM

@MikeR I think they are already using that one

This is the gcode that we are using. It may also be worth noting again that this gcode still wipes in different spots than the automatic nozzle calibration. We are not getting good wipes with that, either.

MikeR added a comment.Apr 3 2019, 1:36 PM

This is still an issue with printers going out. This needs to be fixed and soon. @anolen @west @tutley @marcio @Steven

tutley added a comment.Apr 3 2019, 1:38 PM

@MikeR there is new firmware .104 and later that corrects the probe points as some machines were missing the front right corner. This requires new start gcode, which will correct the wipe. I gave @anolen some start gcode that has the wipe corrected, but she also needed to make some edits,so this is in the works

MikeR added a comment.Apr 3 2019, 1:40 PM

Cool thanks @tutley just wanted an update on this.

anolen added a comment.Apr 3 2019, 1:54 PM

I have not made my edits yet but we can go ahead and use what @tutley sent me and we can update with my pattern later.

MikeR added a comment.Apr 3 2019, 2:07 PM

Is it pushed to cura so I can just do a re-slice?

logan added a comment.Apr 3 2019, 2:31 PM

We gave production an updated slice just now. They are testing it out to make sure all of the feedback is included and working well. Once they've printed a few and everything checks out I will push it along with the profile(s), and when the new wipe pattern comes around we can just reslice.

logan added a subscriber: franklin.

@anolen @tutley was there anything in the slice that could assist circular dimensional accuracy? Just checking in case there was something I could include in that slice.

tutley added a comment.Apr 3 2019, 2:54 PM

@logan This is the same start g code as before just with updated wipe locations

logan added a comment.Apr 3 2019, 3:06 PM

@tutley I know.
To be more specific, is there anything I can incorporate in the slicing settings (CuraLE) that would assist with better circular dimensional accuracy? If not, are there any gcode commands that could (or should) be utilized to ensure that backlash is enabled for the entirety of the print or anything else that may help in that endeavor?

tutley added a comment.Apr 3 2019, 3:10 PM

@logan backlash compensation is turned off by default. Hence the M425 S0. This is enabled through Cura settings just like with Mini2.

oliver added a comment.Apr 3 2019, 3:14 PM

@logan This ticket is for wiping to far left, lets test to make sure that the new gcode fixes the problem and close this ticket. T5891 was created for dealing with the circular dimensional accuracy problem.

logan added a comment.Apr 3 2019, 3:31 PM

@oliver T6451 was merged here which included incorporating anything that may help with circular dimensional accuracy, which it looks like can be included by adjusting backlash fading distance in cura.
I will provide them with a slice with backlash compensation enabled, and post results in T5891
But thanks for your input

karrad reassigned this task from marcio to alexei.Apr 3 2019, 3:36 PM
karrad added a project: Restricted Project.
karrad added subscribers: alexei, karrad.

@alexei Ready for review and merge to master, please close once completed.

karrad moved this task from Restricted Project Column to Restricted Project Column on the Restricted Project board.Apr 3 2019, 3:37 PM
alexei changed the edit policy from "All Users" to "Restricted Project (Project)".Apr 4 2019, 6:31 AM
alexei reassigned this task from alexei to karrad.Apr 4 2019, 7:22 AM
alexei added a comment.Apr 4 2019, 7:25 AM

@karrad , I moved branch to T5928-wrng. T5928 branch should be re-based from current master branch, otherwise it fails to merge.

logan added a comment.Apr 4 2019, 8:26 AM
This comment was removed by logan.

Per Adam, we are flashing to .106, and we are using the latest vernier gcode supplied by Logan (v0.7). So far we have noticed a huge improvement on wiping locations on both the nozzle calibration and the vernier print. I have one printer (QB0022) that the left nozzle hits the wiper mount plastic on the right but during the print with the new start gcode wiped on the pads.

karrad reassigned this task from karrad to alexei.Apr 4 2019, 9:55 AM

@alexei Not sure how to rebase, re-assigning to you

@karrad or @logan , Could you attach start g-code that works for you as file here, please?

alexei reassigned this task from alexei to karrad.Apr 4 2019, 10:03 AM
karrad reassigned this task from karrad to alexei.Apr 4 2019, 10:08 AM
jebba reassigned this task from alexei to Steven.Apr 4 2019, 10:38 AM
alexei claimed this task.Apr 5 2019, 9:47 AM
alexei closed this task as Resolved.

OK, merged to master.