0

When pushing the limits of the bed size (235x235 mm) for two Creality Ender 3 V2 printers running the same mriscoc firmware and the same config and upgrades. I noticed that despite using the same G-code for these two fairly identical printers resulted in offsets that was not accounted for.

I sliced and printed a test model 175mm_bed_center_calibration_crosshair.stl from Thingiverse.com and when slicing in Cura it is perfectly centered as shown below:

test center model

However when printed on each printer and measured, the position in mm for both printers (namely Owl and Fox) are off as shown below:

measurement of center position for two printers

I would like to setup the printers fairly similar, I only print from OctoPi over USB to these printers, one Pi for each printer, so that I can slice once and use the same G-code on both printers. How do I calibrate this without recompiling the firmware? Can I setup a start or end G-code in OctoPi per printer to set a proper center offset? And moreover, why does this happen for identical printers? Shouldn't the tolerance be much smaller than 3 mm difference horizontally and 2 mm vertically (shown from above)?

Additional measurements

For reference, based on request in the comments some measurements of the end stops relative to the frame of the XY-axes of both printers.

Measurement of the Z-axes are irrelevant for this purpose.

Owl Y-axis Endstop (85mm) Owl Y-axis Endstop

Fox Y-axis Endstop (85mm) Fox Y-axis Endstop

Owl X-axis Endstop (45mm) Owl X-axis Endstop

Fox X-Axis Endstop (45mm) Fox X-Axis Endstop

I also measured the size between the bed and the frame as shown below and it's precisely 7mm on each side of the bed on both printers.

Space between bed and frame

I think these are the only measurements that make sense, but correct me if you have other suggestions for measuring it.

Also, the only offset that is reported after running the M503 G-code in G-cod M851 it not relevant for centering, the Z-offset. For Fox M851 X-41.50 Y-7.00 Z-1.13 ; (mm) and for Owl M851 X-41.50 Y-7.00 Z-1.06 ; (mm).

Bob Ortiz
  • 1,053
  • 1
  • 3
  • 29

2 Answers2

3

Perhaps the hot-end or end stops of one of the printers are located in a slightly different location. That's why there is a calibration guide to be able to correct that.

agarza
  • 1,559
  • 2
  • 15
  • 33
Miguel Risco-Castillo
  • 186
  • 3
  • Thanks, I will try to do additional measurements later to try and find out the difference and report here if I found it. – Bob Ortiz Jan 07 '24 at 21:15
  • I've added endstop measurement photos in the question and they are basically identical for the X-axis and Y-axis of each printer, so is the bed location. – Bob Ortiz Jan 08 '24 at 18:45
1

I was surprised to find this out as I haven't heard or read anything about this before in any 'calibration instruction' or 'calibration guide'.

However, as I just found out, the mriscoc firmware gladly has a very easy way of changing this from the menu as documented on the mriscoc GitHub Wiki.

As I already took measurements as shown in the question and as my bed is 235x235 mm. I simply used the LCD menu to go to Advanced > Physical settings and then for printer 'Owl' I changed it to:

X_MIN_POS = 5
X_MAX_POS = 240
Y_MIN_POS = 5
Y_MAX_POS = 240

For 'Fox' I changed it to:

X_MIN_POS = 2
X_MAX_POS = 237
Y_MIN_POS = 7
Y_MAX_POS = 242

I also set the bed size to 235x235 for both printers. I reprinted the test file on each printer and now it is perfectly and equally aligned for each. The question still remains how this could be so far off the center and off identical printers.

Bob Ortiz
  • 1,053
  • 1
  • 3
  • 29
  • What offsets have you set for each printer? Look at the response of a M503 in a console (in OctoPrint e.g.) to see what is listed behind M206. – 0scar Jan 06 '24 at 10:46
  • The only 'offset' that is reported after running M503 is in G-code M851 (not M206). Does that make sense? And it's for Fox `M851 X-41.50 Y-7.00 Z-1.13 ; (mm)` and for Owl `M851 X-41.50 Y-7.00 Z-1.06 ; (mm)`, so except for the Z-offset no difference. – Bob Ortiz Jan 06 '24 at 21:04
  • Are the 2 printers exactly the same, it sounds like the bed or the endstops are mounted differently. – 0scar Jan 07 '24 at 09:10
  • I guess there is too little space in the frame for the bed to be mounted so differently. But I have to do some additional measurements to try to spot differences. It looks not very different to me, but I'll measure it when I get to it. – Bob Ortiz Jan 07 '24 at 21:14
  • It it a very strange thing happening, it puzzles me. Hope you have some additional information. – 0scar Jan 07 '24 at 23:12
  • 1
    Me too, and I also want to understand why so I will post more info when I have time to measure more things. – Bob Ortiz Jan 07 '24 at 23:34
  • I've added endstop measurement photos in the question and they are basically identical for the X-axis and Y-axis of each printer, so is the bed location. So, still in unknown. – Bob Ortiz Jan 08 '24 at 18:45
  • I never like the way the bed is sized according to e.g. that firmware configuration and others use. Basically, the corner of the bed should be (0,0), not (5,5) for Owl. We used to express the minimum positions with negative values to indicate there is space between the endstop and the origin. If you do this, the bed itself is always 0 to 235 and not 5 to 240. Now that you added measurements, you still need to look into the actual position of the build surface mounting relative to the endstops. – 0scar Jan 08 '24 at 22:06
  • That won't be too hard for the X-axis as it is static to the build surface but the Y-axis is a moving bed, how would I measure that relative to the endstops? With the bed at the max. other end? – Bob Ortiz Jan 08 '24 at 22:54