X Axis Travel Accuracy

I created an object in Fusion. It’s basically a rectangle with a negative extrusion so I have a pocket… It was supposed to have a 6mm edge on each side:

(This is a screen grab of the Fusion image.)

(Resulting cut.)

When I cut it, one edge was 7mm and the other was 5mm. Essentially, my pocket was shifted a mm off to one side. A Z-Axis differential I could understand, depending on the accuracy of the bed level. But why would this cut the two sides so inaccurately? How do you calibrate or correct this?


Was this just a pocket you cut out? Or did you cut out both the pocket and the contour (sides)?


Pocket and contour. Otherwise I would have chalked it up to misalignment of the stock.

That’s what I assumed, but needed to make sure.
Is the pocket the size that you created?

It’s a millimeter shy. Was supposed to be a 46mm pocket (width) it’s 45mm. The length seems pretty accurate at 265mm.

I am using a bit I got from Amazon because I wanted a wider cutting tool, to cut down on time. I had to create the tool, in Fusion, manually. I used the dimensions that they advertise. But even if I was off, wouldn’t the error have been symmetrical?

Is this one pass with one bit?
Or multiple passes?
I feel like it has to be a Fusion problem.
I assume you’ve done 3d printing or laser without problems.
Have you had success with fusion before?
Have you tried loading the g-code into camotics or ncviewer.com?

If you want to share your project I can take a look.


And did you cut them out as a single job or as two different jobs? And from your reply I assume, the stock stayed in the same spot during all operations?

What were your feed rates, depth per pass etc. (If you cut out that contour in one pass, there is a significant chance you might have skipped a few steps when starting, but then it would most likely not be aligned anymore at all, so I doubt this is what happened here)

Did you set the origin of both operations (pocket & cutout) at the same spot? Fusion by default adds 1mm (or is it 0.5mm?) to the stock for shape your cutting out (depending on how you defined the stock material)
If you have two cutting operations and set the origin at the “corner” of the stock for one operation and at the corner of your design for the other that might explain it too. But then I would assume the depth is off as well.

When you ask for the project, do you mean the .cnc file, or the Fusion file? I hadn’t heard of ncviewer, but I have, now, opened the gcode file in it. It looks fine, there. It just shows the tool paths and they look symmetrical. I haven’t done any 3D printing, yet. And as for laser, I tried to cut out the sample box, but was not successful in getting through the plywood. Although the image that was engraved looked okay.

As for Fusion success… I don’t have a lot of history. As a trial, I tried cutting a small version (4cm square) with a 7 mm pocket, and that came out okay. But it’s a much shorter object. Less space to wander off path, I suppose:


This piece was pretty accurate with one exception: At the base of the piece there is a small (.5mm) flair out. You can’t see it in the photo. I assumed this was an artifact of the tool dimensions I entered, but I can’t be sure.

It was a single g-code file and, yes, the stock was not moved. I didn’t adjust feed rates, as I am still learning and not really sure how or what t o adjust, so I went with defaults. I assume the origin is the same for both operations. I’m not sure how to change that, especially if they are both in the same setup. I removed all extra stock.

Spindle Speed: 12000 rpm
Surface Speed: 119.695 m/min
Ramp Spindle Speed: 12000 rpm
Cutting Feedrate: 400 mm/min
Feed per Tooth: 0.0166667 mm
Lead-In Feedrate: 400 mm/min
Lead-Out Feedrate: 3000 mm/min
Ramp Feedrate: 100 mm/min
Plunge Feedrate: 150 mm/min
Feed per Revolution: 0.0125 mm

Both paths have the same values.

Multiple passes at 1mm Stepdown.

I would suggest both :wink:

It probably is something in your fusion file, or the linear modules skipped some steps on the exact perfect time to create the shift.

Was the asymmetry on the x-axis or the y-axis?

I think it was on the x-axis. (Across the machine.) I tried another design, using a wider bit that I got from Amazon. I think the first one was with the SM 1.5mm bit, when I think back. But the second one came out pretty darned close to symmetrical. I don’t think I did anything drastically different in the process of setting them both up, mostly just a different tool. I wonder if the SM tool definition is a little off.

Getting ready to try a third design. I am using the larger bit, again.