Page MenuHomeAleph Objects Inc

Need Quiver compatible X Axis leveling routine
Closed, ResolvedPublic


The current X Axis leveling routine in Marlin does not work on Quiver. We need to come up with a better axis leveling sequence, or make the decision that Quiver will not have or need that capability.

Have we attempted leveling to the top?

Event Timeline

marcio created this task.Oct 10 2018, 10:36 AM
west added a subscriber: west.Oct 18 2018, 1:40 PM

The left side z upper has a switch that sits a little further down than the right and the x motor mount has a peice of plastic that sticks up higher than the idler to touch the switch

west added a comment.Oct 18 2018, 1:40 PM

I am printing parts to test homing to the top

tutley added a subscriber: tutley.Oct 18 2018, 2:21 PM

Over the past few weeks my quiver has had a few probe fails, which causes it to level the x axis like mini2 by skipping the motors on the z lowers. Due to the added torque from the planetary gearbox, the motor skipping (at normal current) is causing damage to the belt teeth.

Normal belt teeth

Damaged belt teeth

Sorry if the photos arent the best quality

tutley added a subscriber: Steven.Oct 18 2018, 2:22 PM
tutley added a subscriber: anolen.Oct 18 2018, 2:41 PM

I just confirmed that Mini 2 is not causing damage to the belts when leveling. I looked at a belt off of one of @anolen 's machines since they are often leveled.

west added a comment.Oct 23 2018, 10:41 AM

Quiver 1 is set up to level at the top. First couple rounds havnt been loud and the motor is skipping not the belt so this may be a good solution

i still think we will need to lower the current for skipping the geared motors

west added a comment.Oct 23 2018, 11:06 AM

I have successfully leveled the x at the top with the current at 800 mA. Anything lower can possibly bind

west added a comment.Oct 23 2018, 11:50 AM

@marcio I think this will work

G28 Z
G1 Z292
M211 S0
M906 Z800
G1 Z299
M906 Z960
M211 S1
G28 Z

west added a comment.Oct 23 2018, 11:55 AM

@Steven wanted to test the belts to make sure they weren't getting torn up

@west yeah we should put fresh belts on a machine. then run the leveling thing over and over for 24 hrs or something then check the belts for damage

nickp assigned this task to west.Oct 24 2018, 9:49 AM
Steven changed the edit policy from "All Users" to "LulzBot Hardware Products (Project)".Oct 25 2018, 11:52 AM

@tutley @west @samantha do you know if this was ever done?

@Steven no this hasnt been done yet. We need to figure out some leveling g code and get some fresh belts on a machine to test this

Looks like only software endstops were being disabled and not hardware, so I added a M120 and a M121. However, this code still seems to just gently bump the endstop and stop.

M121 ;hardware endstops off
M211 S0 ;software endstops off
M906 Z800 ;current 800
G1 Z299 ;move past top
M906 Z960 ;current 960
M211 S1 ;soft endstops on
M120 ;hardware endstops on
west added a comment.Nov 12 2018, 10:28 AM

@tutley is running the x axis level test on the printer at his desk

G28 ;home all axes
M121 ;hardware endstops off
M211 S0 ;software endstops off
M906 Z725 ;current 725
G1 Z312 ;move past top
M906 Z960 ;current 960
M211 S1 ;soft endstops on
M120 ;hardware endstops on

This has seemed to work pretty well.
I will have the leveling results by ~2pm today.

The code ran for 24 hrs and completed about 10,800 leveling routines, After this test the belts were removed and compared against belts from a machine that has been in normal use for a few months.

Normal Use:

10,800 leveling routines:

The leveling routine is not causing excessive damage to the belts

@karrad this leveling code can be added to cura

Thanks @tutley I will get this included on T1868 as soon as we cap c.32 today

karrad closed this task as Resolved.

@marcio can we get this added to the quiver firmware