Inkscape & HeeksCNC: Un-smooth curves

[cvoinescu]

Would it be possible to take the individual points in the G-code and plot them in excel?

Yes. If you can use a text editor (or small Python script) to extract the pairs of X and Y coordinates in all G0, G1, G2 and G3 commands in your file, and have them one pair per line with a tab between them, that can then be imported into Excel. The X-Y plot can then show them.

However, there are a number of path visualizer and milling simulators, for example OpenSCAM.

[msg558]

That gives me quite a few things to test.

I do get an error from G-code sender from the first 6 or so lines of the G-code. I've been ignoring them because they seemed irrelevant. Will post when I have the chance.

[msg558]

These are the lines of G-code that are causing G-code sender to say "error. Unsupported statement"

N10G10L1P1 R1.500Z15.000
N20G10L1P2 R3.000Z63.500
N30G10L1P3 R3.000Z63.500
N40G10L1P4 R1.500Z63.500
N50G10L1P5 R7.990Z63.500

[WillAdams]

NB - the code which I posted will do a bizarrely distorted ``J'' since it is drawn using only straight lines.

[msg558]

Here's a picture of what I get when I rotate the J by 90 degrees. Comparing this to before I'd definitely say this is due to flexibility in the Y-axis, because now the accuracy is crap in the direction opposite than what it was before. The shape of the top part of the J is quite unfortunate haha.

J2.JPG (125.07 KiB) Viewed 2558 times

[cvoinescu]

There are a few things you can do to improve accuracy:

• Tighten the V-wheels on the X rail;
• Increase the spindle speed (RPM), and/or use a cutter with more flutes;
• Decrease the depth of each pass;
• Reduce the feed rate;
• Raise your workpiece to get it close to the gantry, so that you can mill with the Z axis higher up.

There's also a modification to the machine that requires no extra parts and reduces flex a little: you can flip the X MakerSlide around, so that the V rails are toward the front of the machine. Of course, you'll have to take the carriage apart and reverse the order of the parts on the four bolts connecting the two plates together. The improvement isn't tremendous, but it's essentially free.

To check that it's flex around the X axis, and not racking because of the undriven Y, try to mill the same pattern twice, once near the idle Y end, and once near the driven Y end. You should be able to see the difference immediately.

[WillAdams]

There's also a modification to the machine that requires no extra parts and reduces flex a little: you can flip the X MakerSlide around, so that the V rails are toward the front of the machine. Of course, you'll have to take the carriage apart and reverse the order of the parts on the four bolts connecting the two plates together. The improvement isn't tremendous, but it's essentially free.

And to think I struggled (in vain fortunately) to disassemble mine which is thus, so as to match the assembly plans.

To check that it's flex around the X axis, and not racking because of the undriven Y, try to mill the same pattern twice, once near the idle Y end, and once near the driven Y end. You should be able to see the difference immediately.

Great tip!

[msg558]

Thank you both for your assistance. I appreciate the help and I look forward to being a part of this community!

Since this is an issue relating to hardware and has already been discussed before, this thread is closed.