GRBL $100 & $101 Issue

I’ve created two files, one containing seven concentric Squares sized from 1" to 7" in size and the other with seven concentric Circles, again from 1" to 7".

The photos are only included for reference. The measurements are below.

		SQUARES:

            X-AXIS	    Y-AXIS
DESIRED	   MEASURED	   MEASURED
   7"	    6.999"	    7.013"
   6"	    6.000"	    6.011"
   5"	    5.001"	    5.012"
   4"	    4.002"	    4.011"
   3"	    3.003"	    3.011"
   2"	    2.004"	    2.012"
   1"	    1.007"	    1.011

I don’t think I could be happier with the results of the Square Test. They’re nearly perfect.

However…

Ignore the missing 1" circle. :slight_smile: I had a typo on the depth. And the glare at the top is hiding the hand printed sizes.


CIRCLES

	    X-AXIS	 Y-AXIS
DESIRED	   MEASURED	MEASURED
  7"	    6.871"	 6.877"
  6"	    5.874"	 5.878"
  5"	    4.876"	 4.876"
  4"	    3.881"	 3.873"
  3"	    2.879"	 2.873"
  2"	    1.881"	 1.879"
  1"	    0.885"	 0.883"

As can be seen, both directions are approximately 1/8" smaller than they should be.

The average deficiency is .878.14 on the X Axis and .877 on the Y Axis, I would have to set $100=41.2 and $101=41.3 to correct the circles.

However, if I were to modify the $100/$101 GRBL settings to heal the Circles, I would be ruining the results when cutting squares.

If both the squares and circles were off in one axis or the other, it might make sense to me. But then again, I could be missing something. :stuck_out_tongue_winking_eye:

Anyone have an idea or a suggestion as to why there would be such a discrepancy between the Circles and the Squares?

Any assistance will be most appreciated.

Bill

1 Like

Cutting on path vs outside path?

4 Likes

My money is on Neil’s bet

1 Like

Hey Neil,

Thanks for the note. I used a 1/4" endmill and cut outside the path on both test cuts. If I cut “on” the path would that not just give me 1/8" smaller circles and squares?

Hey Bill,
Can you share the design files? I agree with you that it doesn’t seem as though this should be happening as you describe it.

Yes, it would, but the question is whether you cut outside for the squares and on the lines for the circles. That would produce similar results to what you got.

Hey, thanks for sticking with me guys.

I used a 1/4" endmill and cut “outside” the path on both test cuts. That’s the confusing part. If one is off, both should be off. If one is near perfect, the other should be too.

I use Vectric VCarve Desktop. It generates.CRV source files and they aren’t uploadable here. I’ll be happy to follow any suggestions for the files you need.

Below are two pics from VCarve…

This shows the job setup portion…

This shows the cutting info for the largest circle.

The file for the Squares is identical in all respects except the shape.

Hope this helps. Let me know if there is anything else I can supply.

Thanks again,
Bill

I would import the two files, or create a file which cuts alternating circles and squares, so the classic diamond-circle-square:

https://wiki.shapeoko.com/index.php/Calibration_and_Squaring_the_Machine#Diamond_Circle_Square_Test

1 Like

It looks like your endmill is setup as 1/8 rather than 1/4. Running outside the path, that will put a 1/4 tool 1/16 closer than it needs to be. Measured across the diagonal, that would be an error of 1/16 * 2 = 1/8.

I think that’s what’s going on, anyway.

6 Likes

WELCOME!!! WELCOME!!! WELCOME!!! WELCOME!!! WELCOME!!! WELCOME!!!
WELCOME!!! WELCOME!!! WELCOME!!! WELCOME!!! WELCOME!!! WELCOME!!!

This may be @RickT’s first post, but he was DEAD on. Thanks Rick!!! It was right in my face as I was making the pics for uploading. I ran the Squares Test first and when I created the Circle Test, I entered the 1/8" endmill on the 1" circle and then just duplicated it to the other six circles.

I owe you a beer, Man! I can’t tell you how relieved I am to find that the problem wasn’t with my machine. Also, how embarrassed I am. What a blunder. :crazy_face: I should have seen that myself.

Thanks so much,
Bill

1 Like

Couldn’t wait to get out to the shop and make the corrections.

NEW RESULTS…

           CIRCLES
	    X-AXIS	 Y-AXIS
DESIRED	   MEASURED	MEASURED
  7"	    7.000"	 7.014"
  6"	    6.007"	 6.013"
  5"	    5.005"       5.014"
  4"	    4.008"	 4.012"
  3"	    3.006"	 3.015"
  2"	    2.009"	 2.014"
  1"	    1.010"	 1.014"

More thanks to all,
Bill
2 Likes

I really like this method of calibration. I’ve tried the circle square diamond test and this seems easier to work with and simpler.

I know I’m a few thou off in one direction more than the other. I’m gonna steal this and try it out this weekend.

2 Likes

just zip up the files will a zip file upload here? that might help upload the files…

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.