re:G02 - G03 - an aside
Posted by
ballendo@y...
on 2000-08-21 15:58:03 UTC
Art,
You will probably discover that IJK is MUCH easier to program (as a
control writer) than R word. Remember this G code standard was
developed before digital computers were anywhere near commonplace!
Means that the people figuring out how to do it used the simplest
means possible, since they had such limited facilities to work with.
Fred Smith gave you some good advice, I'd like to expand on it.
The R word is only reliable for angles of 0 to 180 degrees. Beyond
this, the location of the center point becomes less and less
determinate as you move towards 359.9 degrees. Keep in mind the
numbers you have to work with: Current XY, Radius, and ending XY. At
360 degrees you now have only two! Current XY, Radius (since the end
XY is the same as current)So how are ya gonna tell where the center
of the circle is??? It can be located ANYWHERE on a circle R distance
from currentXY!
Most current commercial CNC controls use positive R for 0 to 180, and
-R for 180 to 360.
Now to the second question floating around. What do these controls do
when the numbers in a circle/arc dont compute?
Many, like EMC, just flag it as an error and set an alarm.(described
in previous posts)
Some (older Fagor and some Fanuc) create and execute a line segment
at the end of the arc to the specified ending location.
Nearly all(including most of the commonly used p port S/W mentioned
on this list) have some setting for an error range. This allows for
the rounding and binary to decimal conversion errors someone (Jon or
Tim, I think) mentioned earlier.
Hope this helps.
Ballendo
You will probably discover that IJK is MUCH easier to program (as a
control writer) than R word. Remember this G code standard was
developed before digital computers were anywhere near commonplace!
Means that the people figuring out how to do it used the simplest
means possible, since they had such limited facilities to work with.
Fred Smith gave you some good advice, I'd like to expand on it.
The R word is only reliable for angles of 0 to 180 degrees. Beyond
this, the location of the center point becomes less and less
determinate as you move towards 359.9 degrees. Keep in mind the
numbers you have to work with: Current XY, Radius, and ending XY. At
360 degrees you now have only two! Current XY, Radius (since the end
XY is the same as current)So how are ya gonna tell where the center
of the circle is??? It can be located ANYWHERE on a circle R distance
from currentXY!
Most current commercial CNC controls use positive R for 0 to 180, and
-R for 180 to 360.
Now to the second question floating around. What do these controls do
when the numbers in a circle/arc dont compute?
Many, like EMC, just flag it as an error and set an alarm.(described
in previous posts)
Some (older Fagor and some Fanuc) create and execute a line segment
at the end of the arc to the specified ending location.
Nearly all(including most of the commonly used p port S/W mentioned
on this list) have some setting for an error range. This allows for
the rounding and binary to decimal conversion errors someone (Jon or
Tim, I think) mentioned earlier.
Hope this helps.
Ballendo
Discussion Thread
Art Fenerty
2000-08-21 10:23:46 UTC
G02 - G03
Kevin P. Martin
2000-08-21 10:51:01 UTC
G02 - G03 - an aside
Fred Smith
2000-08-21 12:18:57 UTC
Re: G02 - G03
ballendo@y...
2000-08-21 15:58:03 UTC
re:G02 - G03 - an aside
Ray
2000-08-21 17:32:50 UTC
Re: G02 - G03
ballendo@y...
2000-08-21 17:59:04 UTC
Re: G02 - G03
Alan Marconett KM6VV
2000-08-21 22:43:27 UTC
Re: G02 - G03