Print not centered on the bed
I have a print that is 240x200. When I slice it in Prusa Slicer it is placed in the middle of the bed. When I print it, the print is offset to the right and to the rear. When I manually move the Core One print head to the center, the X readout says 120mm. It should be 125mm (half of the bed width of 250mm). How can I fix this? The print in the photo prints, but the right edge is at the edge of the bed. The printer really doesn't like this.
RE: Print not centered on the bed
Was the printer factory-built or from a kit?
Is the nozzle/heater block rotated to the left in the print head as it should? One may be inclined to mount it facing straight to the front, which would produce an X offset as you describe. It needs to sit centered in the gap of the part-cooling air duct.
RE: Print not centered on the bed
Or is it layer shift?
I put this test pattern on printables to exercise the whole print area of core one very quickly. Check, whether head or nozzle get stuck somewhere.
Are your belts tightened correctly?
RE: Print not centered on the bed
It understand it to be a static offset, not step loss after dynamic movement, since the OP wrote:
When I manually move the Core One print head to the center, the X readout says 120mm. It should be 125mm (half of the bed width of 250mm).
@mrporcine, I assume that offset also occurs if you do a homing (Control > Auto Home) directly before you move the axes, right?
RE: Print not centered on the bed
Yes, do a home first then move the axis. Goto to all 4 corners of the bed. If the coordinates are still off then your gantry is skewed. If it is skewed this much however, then the Y and homing calibration should fail.
RE: Print not centered on the bed
Yes, do a home first then move the axis. Goto to all 4 corners of the bed. If the coordinates are still off then your gantry is skewed. If it is skewed this much however, then the Y and homing calibration should fail.
How would a skewed gantry cause a significant offset in X?
RE: Print not centered on the bed
Interesting! After auto-homing, the center is at 125 x 110 mm as it should be. When I measured before, it was after I printed top (240x200) that reached the edge of the bed. So my steps to create the offset were:
1) Slice the top in Prusa Slicer. Printer is set to "Prusa CORE One HF0.4 nozzle". The top is 240x200mm and centered in the slicer.
2) Export the bgcode file to my computer (Linux).
3) Move the file to the printer using PrusaLink. Check the box that says start the print after uploading.
4) Push the Print button on the printer.
5) Remove the print when it's finished, then manually center the nozzle. The photo I posted previously is of the print before removing it from the bed.
I am puzzled more than ever!
And before someone notes that I can move the file directly from the slicer to the printer, this worked two days ago but is for some reason is broken now.
RE:
Yes, do a home first then move the axis. Goto to all 4 corners of the bed. If the coordinates are still off then your gantry is skewed. If it is skewed this much however, then the Y and homing calibration should fail.
How would a skewed gantry cause a significant offset in X?
I suppose it wouldn't create that much offset. Mine is skewed -0.15° which is around 0.7mm over the length of the bed. It's measurable though.
In the very beginning of finishing my build I convinced myself that the belts had to be a "perfect" 98/92hz and severely bent the gantry by 10mm. I didn't notice until my squares started looking like a rhombus.
RE: Print not centered on the bed
I put this test pattern on printables to exercise the whole print area of core one very quickly.
Thanks for the link. I printed it using the same steps as above except that I checked the "start print" box before I uploaded the file so the printer started immediately. The printer did a homing calibration before it started the print. And the print finished correctly centered on the bed.
I'll try printing just the first layer of top and see if the offset reappears.
Was the printer factory-built or from a kit?
Is the nozzle/heater block rotated to the left in the print head as it should? One may be inclined to mount it facing straight to the front, which would produce an X offset as you describe. It needs to sit centered in the gap of the part-cooling air duct.
The printer is a kit. At the same time I think you are wrong about heat block rotation affecting offset. The heat block rotates around the nozzle, so the nozzle stays in the same location no matter which way the heat block is facing.
RE: Print not centered on the bed
I think you are wrong about heat block rotation affecting offset. The heat block rotates around the nozzle, so the nozzle stays in the same location no matter which way the heat block is facing.
Duh, you are right of course. Brain fart, sorry...
In that case I really can't see a mechanical or adjustment reason which could cause that kind of offset permanently. Some step loss right at the beginning of your print -- but then no further shift during the remainder of the print? That appears unlikely. Or can there be a re-definition of the origin included in the Gcode? Maybe you could upload your .3mf file (zipped so the forum accepts it)?
RE: Print not centered on the bed
OK, there is definitely something wrong with the gcode in the top file. The printer tries to move the print head off the right edge of the bed, with the motors making a lot of noise. I stopped the print and found the printer thought the center of the bed was 153x107mm! After an auto home, it was back to correct at 125x110mm. The printer did a home before it started to print top, I was watching.
This shouldn't happen. Work flow: FreeCAD --> Prusa Slicer --> Core ONE +. Pretty straightforward. The slicer didn't find anything wrong with the FreeCAD step file.
RE: Print not centered on the bed
I assume the belts are correctly tensioned and not too loose?
RE: Print not centered on the bed
The fact that this offset creeps in once, right at the start of the print, and then the whole layer finishes without step loss, suggests a software issue to me. I'm starting to sound like a broken record, but anyway:
In your Print settings in PrusaSlicer, do you have Layers and perimeters > Quality > Avoid crossing perimeters enabled by any chance? That option appears to be buggy in recent PrusaSlicer versions. It can generate Gcode with impossible moves -- very rapid accelerations and movement speeds which the print head cannot physically execute, leading to step loss.
RE: Print not centered on the bed
The fact that this offset creeps in once, right at the start of the print, and then the whole layer finishes without step loss, suggests a software issue to me. I'm starting to sound like a broken record, but anyway:
In your Print settings in PrusaSlicer, do you have Layers and perimeters > Quality > Avoid crossing perimeters enabled by any chance? That option appears to be buggy in recent PrusaSlicer versions. It can generate Gcode with impossible moves -- very rapid accelerations and movement speeds which the print head cannot physically execute, leading to step loss.
In 2.9.4 this is off by default. But it’s worth checking if it’s inadvertently turned on.
RE: Print not centered on the bed
The fact that this offset creeps in once, right at the start of the print, and then the whole layer finishes without step loss, suggests a software issue to me. I'm starting to sound like a broken record, but anyway:
In your Print settings in PrusaSlicer, do you have Layers and perimeters > Quality > Avoid crossing perimeters enabled by any chance? That option appears to be buggy in recent PrusaSlicer versions. It can generate Gcode with impossible moves -- very rapid accelerations and movement speeds which the print head cannot physically execute, leading to step loss.
It's off.
RE: Print not centered on the bed
If it's not confidential, could you upload the .3mf file (zipped)? I am really curious what might be going on there.
RE: Print not centered on the bed
Me too. I’m in the middle of a 6hr print but when it finishes I’d like to take a look as well
RE: Print not centered on the bed
Here it is: Tray2top.3mf
I'll be very interested in what you find!
RE:
I sliced your project dumped out the g-code in ASCII, pulled it into Visual Studio Code and looked for any print moves outside the printable area, there are none. All X and Y (except for the initial purge line) are within X 0-250 and Y 0-220. I opened the sliced g-code in the g-code viewer and it shows the same. It strongly suggest a mechanical issue is causing this.
Whenever a slice causes the print head to move outside the printable area a huge red warning screen is shown in Preview.
I've attached the sliced g-code output. Perhaps compare with the same on your side.
RE: Print not centered on the bed
Here it is: Tray2top.3mf
I'll be very interested in what you find!
Thanks! My Core One is currently laying down the first layer, nicely centered on the print bed. I just loaded your .3mf, selected my physical Core One (with the standard 0.4 mm HF nozle), pressed Slice and Upload Gcode.
In your photo it looks like the whole print is offset to the side, right from the start, correct? If so, something seems to be different between our printers. Is there anything non-standard in your printer settings maybe? (Although I have no idea what setting might cause such an offset.)
On the other hand, looking at your photo again, it seems that the print is shifted to the left and to the back by ~ equal amounts. So it's a diagonal shift, caused by one motor being off -- which does hint at a physical issue rather than a logical software problem. But then, if it's a step loss, slipping belt or slipping motor pulley -- why would it only occur ahead of the print, and never recur during the print? I did not notice any particularly aggressive moves during the print preparation.
Could you check the Machine limits in Prusaslicer's Printer settings? Maybe an acceleration r speed has been cranked too high? It has happened to me several times: When I just meant to scroll through a settings page, the mouse was hovering over an input field and the scroll wheel was interpreted as a settings change.