Bed physical location offset from what it should be

[DennisM]

New build of a 2.4r2 300mm model. I used my slicer to slice a bed leveling model, and scaled it so it was fairly large, and the skirt ended up being close to the bed edges on the slicer (Prusa Slicer). Then I edited the gcode file and removed the model and left only the skirt, for speed and easy checking of the bed level in general.

The resulting print is offset from the bed and I'm wondering if there are settings in the printer.cfg that will help fix this. Some parameters:

• You might initially say, "just move the bed back" but...
• I moved the toolhead gantry as far back as I could, and set the hall effect end stop to call that "300" - you know, the end of the travel
• I then moved the toolhead down and placed the z-endstop so that the nozzle would contact it properly.
• I moved the bed to within 1-2 mm of the endstop

With those steps in mind, I don't think I can move it back any more because the nozzle just won't reach the z-endstop if that's the case.

So I'm hoping there are some parameters in the printer.cfg that I can edit instead.

Thanks!

Edited September 7, 2022 by DennisM
Added slicer screenshot

[Simon2.4]

I think adjustments are done with a combination of these parameters:

#####################################################################
#   Homing and Gantry Adjustment Routines
#####################################################################


home_xy_position:267,350

##  Gantry Corners for 350mm Build
gantry_corners:
   -60,-10
   410,420

#####################################################################
#   X/Y Stepper Settings
#####################################################################

[stepper_x]

position_endstop: 345
position_max: 345


[stepper_y]

position_endstop: 350
position_max: 350

 

[DennisM]

OK, discourd thinks that I need to modify position_min and position_max under the [stepper_y] section but it was unclear how. I guess I'll try shifting them 5mm to see what happens, not sure which way. I'll try something like min from 0 to 5, and max from 300 to 305.

Still open to suggestions if anyone knows for sure.

 

Here's what I am using now for the X, Y in printer.cfg:

 

##                  __                       
##    _____   _____/  |_  ___________  ______
##   /     \ /  _ \   __\/  _ \_  __ \/  ___/
##  |  Y Y  (  <_> )  | (  <_> )  | \/\___ \
##  |__|_|  /\____/|__|  \____/|__|  /____  >
##        \/                              \/

#####################################################################
##   X/Y Stepper Settings
#####################################################################

##  B Stepper - Left
##  Connected to MOTOR_0
##  Endstop connected to DIAG_0

[stepper_x]
step_pin: PF13
dir_pin: PF12
enable_pin: !PF14
rotation_distance: 40
microsteps: 32
full_steps_per_rotation:200  # set to 400 for 0.9 degree stepper
endstop_pin: PG6
position_min: 0
position_endstop: 300   # for 300x300 build
position_max: 300       # for 300x300 build
homing_speed: 25        # Max 100
homing_retract_dist: 5
homing_positive_dir: true

[tmc2209 stepper_x]
uart_pin: PC4
interpolate: false
run_current: 0.7
sense_resistor: 0.110
stealthchop_threshold: 0


##  A Stepper - Right
##  Connected to MOTOR_1
##  Endstop connected to DIAG_1

[stepper_y]
step_pin: PG0
dir_pin: PG1
enable_pin: !PF15
rotation_distance: 40
microsteps: 32
full_steps_per_rotation:200  #set to 400 for 0.9 degree stepper
endstop_pin: PG9
position_min: 0
position_endstop: 300   # for 300x300 build
position_max: 300       # for 300x300 build
homing_speed: 25        # Max 100
homing_retract_dist: 5
homing_positive_dir: true

[tmc2209 stepper_y]
uart_pin: PD11
interpolate: false
run_current: 0.7
sense_resistor: 0.110
stealthchop_threshold: 0

 

 

[Simon2.4]

 

 

I would try this since you want the endstops further outside the print area, you will probably need to shift your Z probe coordinates up by 5 too

 

[stepper_x]

position_endstop: 305

position_max: 300

[stepper_y]

position_endstop: 305

position_max: 300

 

I'm 75% sure on this one but it is what I'd try

 

[DennisM]

Thanks, I don't think I'll touch the stepper_x since the left/right positioning is close enough for me.

But for the Y, I'll try what you suggested, and also make this change:

from:
home_xy_position:204,300
to:
home_xy_position:204,305

I think that's what you are suggesting (?)

[Simon2.4]

2 minutes ago, DennisM said:

I think that's what you are suggesting (?)

yup

[Simon2.4]

did it work?

[DennisM]

It worked! Here's what I ended up doing:

[stepper_y]
...
position_min: 0
position_endstop: 310   # for 300x300 build
position_max: 310       # for 300x300 build

And

From:
[safe_z_home]
...
home_xy_position:204,300

To:
[safe_z_home]
...
home_xy_position:204,310

And lastly I needed to update my autoz.cfg and euclid.cfg since the z-stop location changed, as well as the dock position relative to the x,y coordinates. I can still do some fine tuning, including shifting the X to the right a little but this method seems to work.

Thanks for the help!

[Buurman]

can be the photo, but if you move your gantry to the front (free motors, and slowly move it, or better take off cables), do both sides touch at the same time? or is there space on one side?

Can be the photo that it looks like its off....

[Simon2.4]

39 minutes ago, Buurman said:

can be the photo, but if you move your gantry to the front (free motors, and slowly move it, or better take off cables), do both sides touch at the same time? or is there space on one side?

Can be the photo that it looks like its off....

You mean check here to make sure gantry is square?

[mvdveer]

1 hour ago, Simon2.4 said:

You mean check here to make sure gantry is square?

Yes indeed. A good reference is:

 

[Simon2.4]

8 hours ago, DennisM said:

position_min: 0
position_endstop: 310   # for 300x300 build
position_max: 310       # for 300x300 build

did you have to put position_max to 310 for this to work? It's my understanding that this is effectively increasing the print capacity by 10

[DennisM]

I first assumed I should be using 300x300, so I tried position_min: 10 but that didn't work out -- too many things were failing, so I changed it back.

But when I move the head to Y310 the nozzle is off the bed (basically over the z-stop) but at Y300 it's on the bed. So I think the 300x300 print area still applies for me.

 

[Buurman]

12 hours ago, Simon2.4 said:

You mean check here to make sure gantry is square?

yes, that was what I meant, since Dennis is asking multiple questions at ones and I looked at the picture, I coudnt help wondering if the basics are setup right...

 

[DennisM]

I did check the squareness very carefully during my build, and also saw the video you mentioned some time ago. I'm good as far as the squareness of the gantry. I actually had verified all of this as well using a large framing angle tool as well, just to be sure.

[claudermilk]

17 hours ago, DennisM said:

I first assumed I should be using 300x300, so I tried position_min: 10 but that didn't work out -- too many things were failing, so I changed it back.

But when I move the head to Y310 the nozzle is off the bed (basically over the z-stop) but at Y300 it's on the bed. So I think the 300x300 print area still applies for me.

That should be fine. For my Trident 250, my position_max for Y is 260, and the endstop pin is also defined at 260. That places the nozzle over the Z endstop pin perfectly. In the slicer, max size is set at 250, so it won't allow placing objects in that space behind the physical bed.

[DennisM]

Also, I wanted to mention I also followed Andrew Ellis' squaring tips on his site:

https://github.com/AndrewEllis93/Print-Tuning-Guide/blob/main/articles/voron_v2_gantry_squaring.md

[ianryeng]

Old thread but had to do some troubleshooting based on the information provided so updating if anyone else runs into this issue. Had prints running off the back of the bed due to the z-probe location. You need to update the z-probe by however much you are extended past the back of the bed (8mm in my case) but you ALSO need to update the y_max position to match or you'll get an error.

position_endstop: 357 #was 350
position_max: 357 #was 350
home_xy_position:232,357 #was 350

Hope that helps and thanks for the original content in the thread, was very helpful, just confused when i was getting klipper errors

[DennisM]

 

Going to reanimate this thread -- also I am the OP so that's neat

So I have a Voron 0.2 that has the opposite problem - the print is too far forward. But for me I don't think you can go into negative numbers for positions (or can you??) so I'm trying to figure out how to move the print BACK towards the back of the printer. My original problem here was that I needed to move the print forward.

Any thoughts from folks?

[Dirk]

Wow, what an old thread

I think the most detailed and correct answer is given to you, years ago, by @Simon2.4.

When your print is not centered,

On the 2.4 it has to do with the gantry_corners definition, which lives in the [quad_gantry_level] section. With these coordinates, you define two points, (left front edge and the right back edge). These points are measured from the belts (on a 2.4 from Z, and Z2). And tell klipper exactly where your bed starts and ends, compared to the z-belts. Which it then can calculate once you do a home-x-y.

For the trident it is in the [Z_TILT] section and you specify it from the pivot points. And it has 3 points for tilting. 

The V0 and the switchwire, do not have this autoleveling beds / gantrys. So you set the beginning and end of the bed location, with your stepper settings:

- position_endstop,
- position_min,
- position_max.

 

So on your V0, you need to check all your position_min position_max (Position_endstop is probably a sensor) and take your nozzle tip to position x=0,y=0 and position x=120 and y=120 see which direction you need adjusting with your position min and position max.

 

I don't own a v0 (or a trident) so let me know how it folds out for you

 

 

[claudermilk]

On the V0 it should just be the position_endstop and position_max at 120. Then sensorless homing runs the tool head into the back right corner to position itself. From there it ought to know where center is. Are your files sliced with the parts at the center of the bed? (dumb question, but I have to ask)