Z height behavior
I am having issue with the Z height changing at the beginning of my print.
In my start script I have a purge line set at Z0.2 Once this is complete the print head moves to start printing the part but lifts up to Z0.7 which is way to far for the first layer to stick properly. I have spent so much time looking through my config files I think i'm getting blind and missing something very obvious. I am using Simplify 3D as my slicer.
A second problem with mesh bed leveling as follows;
I manually set the Z height on all four corners of my bed so that the nozzle is the correct distance above the bed. When I run the mesh bed leveling it shows the right side of the bed is much higher than the left. If I manually lower the right side to get a more level mesh grid, the nozzle is over 2mm above the bed.
Here is a picture of the mesh bed with the bed leveled properly.
What z probe are you using? That height map looks really abnormal.
I am using a BL Touch
Never mind about the first issue of printing at Z0.7. I was so focused on the firmware I did not notice that I had accidentally had a Z height offset in my slicer settings. Ah the silly things one does when they are frustrated and tired. This is one issue a good nights sleep fixed.
Still have the mesh bed issue.
Most BLTouch issues Have been documented here: https://duet3d.dozuki.com/Wiki/BLTouch_Troubleshooting
Give that a look and see how your configuration compares.
How accurate is the heightmap? If you move your nozzle along close to the bed with mesh compensation active does it keep it close to the bed or is it obviously far off?
When I move the nozzel across the bed with the mesh bed leveling on, the nozzle is slowly lifted from the bed. With the mesh bed off the nozzle keeps a consistent distance above the bed.
It would seem that your probe isn't giving you repeatable results leading to an inaccurate height map.
If you do a test print with mesh compensation off does it print well? If your bed is flat enough you don't really need to use mesh compensation at all.
When you measure the trigger height to do you get repeatable results across several tests?
Can you post your G31 and M558 commands from config.g?
dc42 administrators last edited by
Which firmware version are you using?
High amounts of slope in the height map can mean that you are probing too fast. Try reducing the F parameter in the M558 command.
When I do multiple mesh bed compensation the map turns out very consistent every time, so I would guess that the BL touch is consistent.
If I print with the mesh bed on it will push the nozzle into the bed in most areas of the bed. If I do a print with the mesh bed off it is much better but not perfect yet. I am printing on glass but so far have not been able to achieve what I would expect.
Here is the G code:
M558 P9 H5 F320 T6000 ; Set Z probe type to bltouch and the dive height + speeds
G31 P500 X14 Y35 Z0.92 ; Set Z probe trigger value, offset and trigger height
I'm using firmware 2.0(RTOS) (2018-06-05b3)
I Changed in the M558 the F320 to F120 and the result is better. Here is a screen shot.
Running another print with the mesh bed on and it is much better but still puts the nozzle into the bed in a few spots. Overall this is better than without mesh bed compensation but still not acceptable.
dc42 administrators last edited by
Try upgrading to later firmware, such as the 2.02 version just released. There was a known issue in 2.0 that sometimes caused Z height errors.
M558 F100 R0.5
Along with updating the firmware, try these changes. I think you're basically probing way too fast for the BLTouch to settle properly. You may want to reduce your Z axis top speed, jerk, and acceleration also while doing a G29. You want htings to be nice and smooth and stable while doing it.
Got replacement BL touch (First one was apparently faulty) and that has significantly improved things, still not perfect but almost acceptable.
I am going to rebuild with stout linear rails and with busy work schedule should be back at it in a couple months. Hopefully that might improve things.