The cut is always off by approx 12 percent

Discussion about the arduino based g-code interpreter, grbl
Post Reply
kristenanne77
Posts: 432
Joined: Thu May 08, 2014 11:34 am

The cut is always off by approx 12 percent

Post by kristenanne77 » Sun Aug 10, 2014 7:15 pm

I am using CUT2d and the shapeoko post processors for inch and mm.
For the usb connection to Arduinio i am using Universal G code sender.

With the ShapeOko inch post processor for CUT2D
I modeled a 6 inch square but it cut an approx 6 3/4 inch square.
I modeled a 3 inch square but it cut an approx 3 1/4 inch square.

With the ShapeOko mm post processor
I modeled a 100 mm square but it cut an approx 110 mm square.
I modeled a 52 mm square but it cut an approx 58 mm square.

These were all quick measurements with a ruler, but they appear to be definitely off by a certain amount. Almost the same percentage
I also posted to Vectric CUT2D forum, but I think this might have something to do with the gcode from the post processors ???

Any ideas on how to correct this so what i see is what i get?

thanks
K

cvoinescu
Posts: 4442
Joined: Thu Jul 19, 2012 6:50 pm
Location: Camberley, UK
Contact:

Re: The cut is always off by approx 12 percent

Post by cvoinescu » Sun Aug 10, 2014 9:30 pm

Does your Shapeoko move the correct distance when commanded manually (that is, when you type G-code commands in the G-code sender)?
Proud owner of ShapeOko #709, eShapeOko #0, and of store.amberspyglass.co.uk

kristenanne77
Posts: 432
Joined: Thu May 08, 2014 11:34 am

Re: The cut is always off by approx 12 percent

Post by kristenanne77 » Sun Aug 10, 2014 10:01 pm

When I set inches (G20) and moved manually it looked liked it moved about an inch, and when I set mm (G21) it moved a much smaller distance so I assumed it was mm.
But no I didn't measure exactly.
I will take some measurements and post back my results.

(Also, now that I think about it, I believe it cut deeper than expected also (so the z plane was exaggerated as well)

Thanks
K

cvoinescu
Posts: 4442
Joined: Thu Jul 19, 2012 6:50 pm
Location: Camberley, UK
Contact:

Re: The cut is always off by approx 12 percent

Post by cvoinescu » Sun Aug 10, 2014 10:26 pm

Well, six inches and three quarters is "about" six inches. :) Do measure manual movements more accurately, because that'll tell you whether it's the G-code or the machine configuration.
Proud owner of ShapeOko #709, eShapeOko #0, and of store.amberspyglass.co.uk

kristenanne77
Posts: 432
Joined: Thu May 08, 2014 11:34 am

Re: The cut is always off by approx 12 percent

Post by kristenanne77 » Mon Aug 11, 2014 12:53 am

I used command G20 to get inches and manually incremented by 1 step at a time in inch increments and then measured later with a caliper.
For 1,2,3, inch I measured 1.109, 2.25, and 3.36 inch,
I did for both x and y and got same results.

Also did command G21 to switch to mm and took some measurements.
For 10, 20, 30 and 40 mm I got 12.25, 22.4, 33.87 and 44.68
My measurements aren't perfect but probably a little better than the ruler measurements I gave earlier.

Also I did a rectangular outline which was supposed to be 10 x 9.5 but came out 11.37 x 10.75.
It appears that the measurements I give get scaled up somehow.
What can I do here besides precalculating what I want and then muliply by a scaling factor to get desired results?

Thanks
K

kristenanne77
Posts: 432
Joined: Thu May 08, 2014 11:34 am

Re: The cut is always off by approx 12 percent

Post by kristenanne77 » Mon Aug 11, 2014 2:27 am

Also, I got this information from typing $$ in the command line .
Do you think it will help solve the scaling mystery?

**** Connected to COM3 @ 9600 baud ****

Grbl 0.8c ['$' for help]
>>> $$
$0=45.000 (x, step/mm)
$1=45.000 (y, step/mm)
$2=320.000 (z, step/mm)
$3=30 (step pulse, usec)
$4=500.000 (default feed, mm/min)
$5=500.000 (default seek, mm/min)
$6=156 (step port invert mask, int:10011100)
$7=255 (step idle delay, msec)
$8=200.000 (acceleration, mm/sec^2)
$9=0.050 (junction deviation, mm)
$10=0.100 (arc, mm/segment)
$11=25 (n-arc correction, int)
$12=3 (n-decimals, int)
$13=1 (report inches, bool)
$14=1 (auto start, bool)
$15=0 (invert step enable, bool)
$16=0 (hard limits, bool)
$17=0 (homing cycle, bool)
$18=0 (homing dir invert mask, int:00000000)
$19=25.000 (homing feed, mm/min)
$20=250.000 (homing seek, mm/min)
$21=100 (homing debounce, msec)
$22=1.000 (homing pull-off, mm)
ok

jarretl
Posts: 135
Joined: Mon Feb 24, 2014 2:50 pm
Location: Edmonton, AB

Re: The cut is always off by approx 12 percent

Post by jarretl » Mon Aug 11, 2014 2:46 am

Your $0 and $1 may be too high if you have a stock shapeoko 2 from inventables, I think the wiki lists 40 steps/mm for both. Try setting to 40 and retest.
-J.
Shapeoko 2 #4043; DW660

CastIrony
Posts: 650
Joined: Mon Jul 21, 2014 10:21 pm
Location: Portland, OR

Re: The cut is always off by approx 12 percent

Post by CastIrony » Mon Aug 11, 2014 2:48 am

jarretl wrote:Your $0 and $1 may be too high if you have a stock shapeoko 2 from inventablees, I think the wiki lists 40 steps/mm for both. Try setting to 40 and retest.
-J.
45 being 40 + 12.5%, this sounds plausible. :D
Shapeoko 3 #192 - Complete! Needs limit switches.

kristenanne77
Posts: 432
Joined: Thu May 08, 2014 11:34 am

Re: The cut is always off by approx 12 percent

Post by kristenanne77 » Mon Aug 11, 2014 5:21 pm

For X and Y, I changed the 45 to 40.2 as recomended here.
http://www.shapeoko.com/wiki/index.php/ ... or_X_and_Y:
Things seem to be going better.

Thanks
K

cvoinescu
Posts: 4442
Joined: Thu Jul 19, 2012 6:50 pm
Location: Camberley, UK
Contact:

Re: The cut is always off by approx 12 percent

Post by cvoinescu » Mon Aug 11, 2014 5:28 pm

40.2 is not recommended there. It's just an example. The theoretical value is 40 exactly, and you should use that unless you've actually measured it to be different.
Proud owner of ShapeOko #709, eShapeOko #0, and of store.amberspyglass.co.uk

Post Reply