Posts: 97
Threads: 8
Joined: Jul 2019
Reputation:
15
10-20-2019, 10:57 AM
(This post was last modified: 10-20-2019, 12:01 PM by admin.)
I still haven't gotten around to doing the video to show this.
Here's what happens...
Level the build plate however you choose to. Now, with no resin in the vat, jog the z axis to home and you'll see that the build plate and carriage both stop at the same time, but as soon as resin is introduced, home the build plate again. What you'll see if you watch closely is that the build plate will stop, yet the carriage will continue to move down slightly.
When the offset is done, the build plate and carriage will go to the trigger point, then the carriage will raise back up to the offset amount.
Not sure on the formula, but the Mars does use a TR8x2 lead screw, so the S0.000625 would be correct. That gcode config file is the original that was just decoded to english and not the chinese descriptions.
Posts: 3
Threads: 0
Joined: Oct 2019
Reputation:
0
10-25-2019, 09:46 PM
(This post was last modified: 10-26-2019, 12:20 AM by tbdad.
Edit Reason: bad formatting
)
had an interesting experience trying the z calibration from the original post (.zip file) here.
i printed a cube in the center and also one near each corner, just to make sure i don't break the calibration anywhere on the plate.
they were mixed, 0.9 to 1.5mm off, but i didn't want to re-level since its been working well, so i just used the smaller 0.9mm to change the zeroing with.
i followed the procedure, printed the .gcode enabling the zero option, removed build plate, homed, moved it up by 0.1 increments 9 times then hit the z=0 button.
re-printed the 5 cubes, and one didn't print. probably due to my lack of calipers (so 0.9 was a bit too much in that corner), which i hope to fix some day. ok, the rest were indeed much better heights, so i figured i'd just adjust back 0.2mm.
i removed build plate, homed, moved _down_ 0.1 increments twice, then hit z=0. reprinted the cubes... they were all back to the original heights, 0.9 to 1.5mm off.
i suspect there is something else going on introducing variability in the Z axis? or does the Home button not go to the last z=0 position?
Posts: 97
Threads: 8
Joined: Jul 2019
Reputation:
15
When you did it the 2nd time and did the first step of homing the build plate, then did the 0.1mm down 2 times, that actually took it .2mm lower then the home position.
What you would have wanted to do is do the homing position, then instead of the 0.9mm up, just do 0.7mm up, then tapped the z=0 to reset.
Posts: 4
Threads: 1
Joined: Oct 2019
Reputation:
0
10-26-2019, 08:41 AM
(This post was last modified: 10-26-2019, 08:43 AM by Marc_H.)
Hello everyone,
my english is not the best but i try my best to explain what i found out about the problem.
I postet two videos on Youtube, where i show the problem.
I mounted a dial indicator to my Elegoo Mars (i bought last week) to check the accuracy and to calibrate the steps/mm.
The Limitswitch to Zero the Buildplate works perfect, the Problem is that there is backlash in the thread, the flexelement (can't understand why they use this) or the motor.
Because of this there is "no" movement for the first 0.4mm as you can see in the second video:
At first i try to remove the flex element and test if this will reduce or remove the problem.
Later i report.
Posts: 4
Threads: 1
Joined: Oct 2019
Reputation:
0
Ok guys i think i found the Problem and it give us the answer why some of us has this problem and the others not.
At disassembling my printer i realized the load on the rail is much to high, the Carriage is moving but it needs to much force and this force pushes back in the flex element and in the motor.
After adjusting this the printer moves very accurate.
Like you can see on this video:
