Y-axis short

[exxosuk]

I am printing a circle and it's supposed to be 80.5mm on box XY axis. It is on the X-axis, but on the Y-axis its 0.5mm short and comes out at 80.0mm.

I have calibrated the Axis the printer multiple times and the print is always out anyway.

I have started printing objects out with 0.5mm Y axis increase to try and compensate for the problem. Although I don't really see where the problem is coming from ?

[WF55]

Often this is a mechanical problem. You have to correct this in the printer's firmware.
Before doing so, you can check the STL file with an STL viewer to see whether the problem comes from the printer or slicer.
Unfortunately, Simplify3D only has the option of compensating for a length error for the X-axis and the Y-axis at the same time (horizontal Outer Size compensation).

[parallyze]

Often this is a mechanical problem. You have to correct this in the printer's firmware.

How exactly would you "correct" a mechanical problem in firmware? Before using less than ideal
workarounds like skew correction I'd recommended fixing the problem. Not trying to mask it.

Unfortunately, Simplify3D only has the option of compensating for a length error for the X-axis and the Y-axis at the same time (horizontal Outer Size compensation).

That's absolutely NOT what horizontal size compensation is about, it's not meant to compensate for an axis that doesn't move
as expected.

I am printing a circle and it's supposed to be 80.5mm on box XY axis.

Did you actually check the model? STL files do not contain arcs/circles. And depending on the export settings of the design
software this might be more or less detailed. This also effects dimensional accuracy, so it'd be important to know what you
did and how you took measure.

Also there's many people with sloppy retracts, taking measurements in the area where material flow is still increasing when
printing the perimeter.

I have calibrated the Axis the printer multiple times and the print is always out anyway.

Did you check if the y/x axis do move properly on your printer, maybe by simply using calipers/a dial gauge and using the
menu on the printer itself, ruling out any software/slicing problem?

If that's correct - what's the output of the slicer, are coordinates like they should be within the gcode, is the distance
between min/max x of the circle you're printing the same as min/max y? (Keep segmentation of STL files in mind)

Do you see any problems when printing objects with less segments, like a rectangle or hexagon? Is the offset along Y
always 0.5mm short or is it a percentage off when scaling objects?

[WF55]

Often this is a mechanical problem. You have to correct this in the printer's firmware.

How exactly would you "correct" a mechanical problem in firmware? Before using less than ideal
workarounds like skew correction I'd recommended fixing the problem. Not trying to mask it.

I mean mechanical problems, like inaccurate timing belts, inaccurate pulleys and "e-steps". These can be corrected using the “steps/mm” in the firmware.

Unfortunately, Simplify3D only has the option of compensating for a length error for the X-axis and the Y-axis at the same time (horizontal Outer Size compensation).

That's absolutely NOT what horizontal size compensation is about, it's not meant to compensate for an axis that doesn't move
as expected.

OK, I was probably wrong. Can you explain what it is used for? I got this as a tip for the “cura-slicer” for such a problem.

[parallyze]

I mean mechanical problems, like inaccurate timing belts, inaccurate pulleys and "e-steps". These can be corrected using the “steps/mm” in the firmware.

Depending on the fault I'd recommend replacing pulley with the bores off-center or the like. In the long run those problems will
usually come back and haunt one when expected the least... that's why I'd love to know how he did calibrate the axis.
Without knowing anything about what's done before it's just wild speculation ^^

OK, I was probably wrong. Can you explain what it is used for? I got this as a tip for the “cura-slicer” for such a problem.

Horizontal Size Compensation offsets the outer shape of your object inwards or outwards. But if set to -0.25 to compensate
for an object 0.5mm larger than expected it will also have the inverse effect on openings inside the objects. While the outside
dimension of the object will shrink, openings inside will expand...

Here's 1-2 examples coming within the thread:
viewtopic.php?p=57090#p57090

This is great for finetuning. Or if you wanted to overextrude on purpose but keep outer dimensions. But it is always a fixed value
and therefore often not effective when encountering faulty hardware. Depending on the problem (belt/pulley) the problem might
even be restricted to a certain area on the axis or repeat every x millimeter...

[exxosuk]

Thanks for peoples input.

Did you check if the y/x axis do move properly on your printer, maybe by simply using calipers/a dial gauge and using the
menu on the printer itself, ruling out any software/slicing problem?

If that's correct - what's the output of the slicer, are coordinates like they should be within the gcode, is the distance
between min/max x of the circle you're printing the same as min/max y? (Keep segmentation of STL files in mind)

Do you see any problems when printing objects with less segments, like a rectangle or hexagon? Is the offset along Y
always 0.5mm short or is it a percentage off when scaling objects?

I will try printing a square to see what happens. It be easier to follow than a circle I think. I was printing some squares a few weeks back and had the same problem. I try to take some images if it helps. The size of the object doesn't seem to matter. It always seems to be out by the same amount from what I can tell so far.

I have closely watched the Y axis moving against a ruler and it doesn't seem to stutter at any point. I did try increasing the motor current but didn't help. I have calibrated with a caliper also. It's all spot on.

Its possible the Y belt is simply to tight as the bed doesn't seem to smoothly move manually, even though the motor doesn't seem to have any issues. But with such a small amount being out, its possible "by eye" it simply can't be seen.

The problem with that idea, is I have a second printer (anycubic kobra) while I calibrated that also, its dimensions are out even more than my main printer. So I gave up with that one.

So I am not sure where the problem is. It could be multiple problems. So far I have just been changing the scaling in S3D to compensate for the problem. My only idea to "fix it" is to ignore proper calibration and just set the Y-steps so prints turn out the right size.

[exxosuk]

What I have done is actually tightening up the belt. I have also adjusted the stepper gear because I think it was causing the belt to catch, causing a bit of a "judder" during movements. I've also straightened out the whole bed assembly because I think it was at a bit of a angle (rotated). I also recalibrated the Y steps. Its not 100% percent but 0.2mm over 100mm isn't to bad I think anyway.

I tried a 50x50mm square and X&Y came out at 50,25 each. This could be explained by the Z axis being out now it seems. It was supposed to be 1mm but came out 0.88mm. So I will try upping the "babystep" by 0.1mm to see if it all evens out.

EDIT:

So 1mm height now, X came out at 50.14mm, Y came out at 50,06mm. Considering Y wasn't perfect in calibration there was likely a 0.1mm error or thereabouts anyway. So it basically seems fine to me now. If I keep trying to calibrate the Y-axis I would only likely end up making it worse now.

So the gear adjusting and belt tightening seems to have been the likely solutions. Though the belt wasn't exactly slack to start with.