After the 15° flat is taken I will take manual hits on the inside top Ø as well as the bottom inside Ø. I will take these two Ø's and project each one out onto the 'X' flat which I had taken and also leveled to.
How do you project a circle onto a non-perpendicular plane? That sounds like part of your alignment woes. 2d circles have no axis vector, so they cannot be projected accurately.
If you are asking for a simple assign function to square up your alignment, on an organically-shaped part, positioned/fixtured on the CMM in who knows what method.... You are asking for the impossible, WITHOUT FIRST establishing a secondary feature from which to derive the 15° angle from. Hear me out:
Taking a 4 point plane on the 15° angle flat on the right end. I level this plane to 'X' plus, and also make it my 'X' origin. I will offset the angle 15° to 'X' plus about 'Y' plus. (my trihedron is fixed like it is in the original CAD state)
--This initial 15° angle is relative to the MACHINE coordinates, AND rotation, and absolutely varies significantly each time you fixture this type of part on the CMM.
For you to think that your "I will offset the angle 15° to 'X' plus about 'Y' plus" is relative or accurate to the part at this point, is error within your alignment, as this offset you are basing that 15° from, is not clocked to the part, it's clocked form the machine's "startup" alignment.
You need to establish an axis line for the part, then you need to level to that axis line. if you want to "square up" your alignment, to be about the 15° angles as a basic, you need that part's axis line, and a rotation datum first.
Here's my suggestion for your method of measure:
Take one hit someplace top of first 15°plane, origin xyz to it. Turn on dcc. create autofeature ID circle at top with a higher than normal prehit/retract (like 0.15"). ORIGIN YZ to that.
Then take one hit on bottom of your first 15°plane, Origin X to it. create another autofeature ID circle at bottom, (same high prehit/retract).
Now construct a 3d line between both ID circles. Origin YZ to bottom circle centroid, and rotate Z+ about the 3d line. Construct a midpoint between the two circles origin YZ again after the rotate. make sure you do this in a separate alignment as the rotation. --You can get flipped vector issues if you don't.
NOW measure your 15° plane.
Recall startup alignment. Level X to new plane, rotate to Z+ about your X+ axis with the 3d line. NOW offset your basic 15° and origin X to plane, YZ to centroid of the two circles.
Repeat two circles again.
Do all of the same on opposing side of part.
Construct an axis line between the two centroids of the secondary sets of circles. Construct a plane from the 4 circle centroids as a rotation plane (averages out any twist in the part)
Pierce the two planes with the axis line.
Level to axis line. Rotate about your plane from the 4x circles, origin to one end of the axis line.
Now, change workplane to y, dimension 2d angle between Plane 1 and axis line. Dimension 2d angle between plane 2 and axis line.
I'm presuming one will be 15°and the other is going to be something silly/inverse like -165°
I would do an assign function to average the two 15°values out. something like this:
NEWANGLE ASSIGN\(DIMANGLE1.M.MEAS+(180-DIMANGLE2.M.MEAS))/2
That ideally will take your 15.x and14.8x and average the two out.
Then you can rotate to that and measure your length from axis/plane pierce point 1 to axis/plane pierce point 2, parallel to X.
(DO a CTL+ALT+A align, add a rotation of 15° as you always have, click ok.
WIthin edit window replace the 15 value with the "NEWANGLE" assign id.)
YZ again after the rotate. make sure you do this in a separate alignment as the rotation. --You can get flipped vector issues if you don't. NOW measure your 15° plane."
I guess if you want to refine the plane vectors better, you can take a third hit on it someplace and level to the initial plane and rotate the line about z before re-measuring the new plane.
YZ again after the rotate. make sure you do this in a separate alignment as the rotation. --You can get flipped vector issues if you don't. NOW measure your 15° plane."
I guess if you want to refine the plane vectors better, you can take a third hit on it someplace and level to the initial plane and rotate the line about z before re-measuring the new plane.
