printing the y-axis too short

[Brit]

My Makergear printer M3-se has been printing the y-axis about 10% shorter on the y-axis than is in the original 3D model and
Gcode.

I also printed the same Gcodes on a M2 printer to double-check; they printed with the correct proportions. These tests included the x-y-z 20mm calibration cube https://www.thingiverse.com/thing:1278865. It printed 20x18x20 mm on the M3 (18 mm on the y-axis).

Any ideas? Do I need a new Raspberry pi. A new Rambo board?

Has anyone heard of this happening before now?

This latest problem is after other multiple problems with this printer going back almost 2 months. It started with the wireless capabilities failing. I have reformatted and redone the raspberry software on the Sd card https://garr-garr.notion.site/MakerGear ... 798ee805eb , wire ethernet, etc. with some intermittent success. Oh...And the machine still stops mysteriously once in a while at the midpoint of the large prints. ... (I probably need a new pi.)

[Brit]

Could the Undervoltage warning be related? I was told by support to ignore it.

[Gwhite]

I wouldn't expect it to, unless it's really low. All the distance results with the printer are controlled by "steps". If the controller is telling the Y-axis to go N steps, and the distance isn't right, there are a bunch of possibilities:

1) The value of N is too small (software problem). If the stepper hasn't changed and everything mechanical is OK, then the software is telling is to go too few steps. There is a firmware parameter buried somewhere that tells the software how to translate Y mm's into N steps. That might be off. Most slicers also have calibration/scale-factor parameters that allow you to fine tune things, and one of those may have gotten set wrong.

Send an M503 commend to the printer to get a readout of the "Steps-per-unit" in the firmware. It will be the line starting with M92. On My M2E, I get:

M92 X88.88 Y88.88 Z1007.70 E471.50

I'm guessing the M3 will also have identical values for X & Y. If Y is ~ 10% less than X, that is your problem.

2) The number of steps being sent to the stepper is correct, but the motor isn't able to produce the correct rotational output. That could mean either a bad stepper, or too much drag some place, or an intermittent cable connection (lost steps). Could be bad bearings, open coil in the motor, belt tension too tight, dirt in bearing or rails.

Is the Y error very consistent? You could make a test print consisting of a series of Y-length segments. If it's software (like a bad scale or calibration factor), I'd expect them to be all exactly the same percentage error. If the error is random, that smells more like an intermittent cable connection, or friction someplace on the mechanical side. You can also check for an intermittent problem by printing something long at a 45 degree angle. If it comes out straight, the error is consistent. If it's got wiggles, something is intermittent.

Good luck!

[Brit]

Thanks so much for your help!

The M3 reads: M92 X80.10 Y80.10 Z1007.70 E471.50 with the M503 command.
So it doesn’t appear to be a software problem. However. I did about 5 tests and checked several other recently printed parts. All are consistently off by about 10% on the y-axis. The triangular 45-degree tests are fine. When finished with the print and the print moves back down, I do hear a grinding noise as if the motor is slipping on possibly the y axis.

It is a 1.5-year-old machine without a huge amount of use but I just put in a brand-new motor for it. It sometimes stalled on the y-axis and it was not that easy to move along the y-axis when off. It now moves back and forth smoothly when off, so the bearings and tension are probably fine.

What a mystery. My wireless stopped working on the Raspberry pi a while ago.I can connect via ethernet, but I have had intermittent problems with it. Could it be a faulty pi?

[Gwhite]

The M92 results look OK, in that both X & Y are the same. It could still be something in the slicer that is applying a scale factor.

The fact that your off the bed with the quick-start app still points to the printer being off. One thing that is odd is that the M92 values on the M3 are about 10% lower than on my M2. If that was the problem, I'd expect the X-axis to be off as well.

Very weird...

Doug White

[Brit]

I'm off the bed on my M2 printer (not the M3). Weirdly, 2 out of 3 of my printers are off on the y-axis at the same time. Although the M2 prints correctly, it is still placing the print a bit down on the y-axis, but the proportions and scale of the prints are correct, so it is usable. Both M2's print the same Gcodes proportionately correctly. These same Gcodes are squashed on the x axis on the M3 printer so it is probably not the slicer.

I'm going to bit the bullet and buy a new Raspberry pi. It has just been acting indeterminantly weird over the last 2 months. Hopefully, that will fix it.

[Brit]

I swapped motors between the M3 and the M2 ..And the cube tests now appear square! ...I'll check again tomorrow.

[Gwhite]

It's possible the motors have different step sizes, and you got one that is mis-marked. If it's was working before & the motor hasn't changed, then it sounds like the drive current from the electronics may be too low. The low-voltage warning on the Raspberry Pi might be important.

[Brit]

I had put the M2 motor onto the M3 when that motor was acting up a couple of months ago. I had forgotten that I put my new motor when it arrived onto the old (2016) M2 and left the other motor on the M3.

The low voltage warning, which Makergear support says to ignore, wouldn't have a bearing on the M2 printer since it doesn't use a pi. So it must be the motors.

Thanks for your help! Your suggestions pointed me to the motors- one short on the y-axis by 10% and the M2 long on the y-axis by about 10%. I would not have thought it since these motors are advertised to work on all M2, and M3 printers (https://makergear.com/products/xy-motor). But these old 2016 motors must have had a different micro stepping.

[Gwhite]

Ah! That all makes sense now. Glad you got it sorted out.