Making X and Y axis perpendicular

I just finished building a repstrap and found a big flaw in my design. The printer works by moving the bed in the Y axis and the X axis riding on the Z axis. The way I made all the parts, the Z/X can be completely removed from the Y axis without removing any screws (so the z axis is pretty much just sitting in the table around the Y bed). Is there any way to ensure the X and Y are perpendicular besides any major redesigns?

If it's hard to ensure that X and Y are perfectly at right angles while still keeping the Z/X system removable, then you could use orthogonal axis compensation in the firmware to correct it. See [reprap.org].

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Not easily unfortunately. i have a machine made from extrusion, milled to the exact length. This helps insanely when getting my machine square and true. I use the offcuts of extrusion to position the bed, as I know each piece is exactly 80mm wide. I then place my guides next to this, clamp it down and do similarly to the other side before fastening it permanently. You could use calipers to similar effect, so that you know the Y axis is the exact same distance from the edges of the frame at each end. Or, without moving the hot end position, move your bed by 100mm and measure the distance from the nozzle to the edge of the bed. if it is the same distance at each side, its square.

Basically, start with a tool or frame etc that you know is true, and use it to measure/gauge the rest of the build.

First, let me say congratulations on thinking about your printer in sufficient depth to question a thing like orthogonality. Not many people think about it. You will no doubt end up with a well calibrated printer that can do more than print Yoda heads and Pokemon.

I designed a test cube and wrote a companion spreadsheet to calculate the error and tell you how to correct it. It will let you check and correct nonorthgonality in X, Y, and Z axes. See: [www.youmagine.com]

It is based on the diagonal measurements of the printed cube. If the diagonals are equal, the axes are orthogonal.

3D printer firmware (and the slicer) makes certain assumptions about the printer, including an assumption that the axes are orthogonal. If your printer has dual Z axis motors, you may have problems keeping the X axis orthogonal to the Z axis. If the X axis alignment shifts, which happens easily in dual motor setups (someone can manually turn a screw, some binding in the mechanism can cause one screw to lose sync with the other, driver problems, cable problems, or even powering off with the motors positioned away from full step positions), your printed parts may not fit together properly. The only way to be certain that the X axis remains orthogonal to Y and Z is to drive both screws with a single motor.

Edited 2 time(s). Last edit at 07/31/2016 09:08PM by the_digital_dentist.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Quote
the_digital_dentist
3D printer firmware (and the slicer) makes certain assumptions about the printer, including an assumption that the axes are orthogonal.

Except RepRapFirmware, which allows non-orthogonality to be compensated. But of course it's better to get the axes orthogonal in the first place, because this removes the need for the calibration step.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Doesn't RepRap firmware make the assumption that the alignment of the axes is fixed? With two Z axis motors, the alignment can change.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Quote
the_digital_dentist
Doesn't RepRap firmware make the assumption that the alignment of the axes is fixed? With two Z axis motors, the alignment can change.

Yes, it assumes they are fixed.

Although I agree with you that a single Z motor and belt drive is a better solution in many cases, I have been looking at the issue of keeping the Z motors in sync because I have customers who want to use them. The main issue is that the motors can go out of sync by 4 full steps at a power cycle. I can see a couple of solutions:

1. On the Duet WiFi we have a power monitor, so what I plan to do when power down is detected is to raise the Z motors to the the next multiple of 4 steps, to correspond with the position they want to adopt when next powered up.

2. On systems using three independent Z motors to do bed levelling, it doesn't matter if the motors get out of sync during power up, because the bed levelling routine will sort it out.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

On dual motor Z axes you could put a zero sensor at each end of the X axis and run the two motors until both have tripped their respective z=0 sensors. You might also have to add rotation sensors to ensure that both screws are turning or the system could do some damage if one screw is jammed and the other keeps turning.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Quote
the_digital_dentist
On dual motor Z axes you could put a zero sensor at each end of the X axis and run the two motors until both have tripped their respective z=0 sensors.

Yes, I have already been asked to support one Z homing switch for each motor in the firmware.

Quote
the_digital_dentist
You might also have to add rotation sensors to ensure that both screws are turning or the system could do some damage if one screw is jammed and the other keeps turning.

I think that's an excellent idea. A loose stepper motor connection or failed stepper driver (not that I've ever had one) could have the same effect. The drivers on the Duet WiFi can do stall detection, so with any luck I can use that to check for rotation.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].