PR 238599 Angle Dimension is not converting the quadrant to match the nominal

huangyhg

Angle Dimension is not converting the quadrant to match the nominal angle.
<< Steven George  --  02/21/06  16:04:58>>
In the supplied part program, I inserted an Angle dimension to evaluate the angle between LIN1 and CIR3 in the ZX plane.  The nominal is 7 degrees, and the angle dimension is displaying an actual of 96.985 degrees.  Regardless of what nominal value I enter into the Command Mode Edit Window for that dimension, I am unable to get the dimension to display an actual of anything other than 96.985 degrees.  The Angle dimension is not converting the quadrant to match the entered nominal angle.
<<END>>


<< Changes made by Neil Kay -- 02/02/09  10:29:49>>
Status:  RESOLVED to CLOSED
<<END>>
<< Paola Pallo  --  07/23/08  09:37:10>>
From v42 on the 2D_angle math changed again, and we agreed to apply the geometrical rules to the input features and the workplane. See #255659  for more details .
<<END>>
<< Changes made by Paola Pallo -- 07/23/08  09:37:16>>
Action:  Paola Pallo to Steven George, Status:  REVIEW to RESOLVED
<<END>>
<< Changes made by Neil Kay -- 07/22/08  17:46:51>>
Action:  Bret Naylor to Paola Pallo
<<END>>
<< Wade Burton  --  02/23/06  11:23:54>>
Beginning in V3.7, and #235967, we made it so we don't guess by particular quadrant.  We had found that if we continued that practice, it would be impossible to properly report angles with nominal of +2, and measured value of -2, or the other way around, nominal of -2, and measured value of +2, where the nominal and measured value are close, but not actually in the same quadrant.  Because if it always read which quadrant you wanted, and put the measured value in that quadrant, this situation would always come up wrong.  This would also give problems at every position at the edges of the quadrants... 0, 90, 180, etc.  We had heard this complaint often, and determined that if you wanted the angle 90 degrees from this angle, then it could be done other ways (variables or casting a feature).  In your example, if you want the circle to be checked as if it were a plane, and not checking the vector of the circle like a line, then you can cast the circle to a plane.
So what we do beginning in V3.7, is we check the angles that are truly made by the vectors, and report whichever one is closest to the selected nominal when one of the features is a plane.  For example, if the angle of the plane and the other feature is 95, it can also be viewed as -95, 85, or -85, depending on which vector on the plane is used to determine the angle (it uses a vector on the plane, not the normal of the plane so it has to find a vector on the plane that is the appropriate one).  But it will look at your nominal to determine which one is closest to your nominal.   But with this, it would not be appropriate to expect it to report 5, -5, 175, or -175.  That isn't the same angle that would be represented by the plane.
However, I did find that we hadn't looked at all the options when not using a plane.  When we just have two non-plane features, which means we have just two vectors being compared, that we weren't considering all possibilities where the vector may have been measured in the opposite direction.  I found that in the example above it would only find the 95 and -95, and not the 85 and -85 to compare to the nominal, which are found if the vector is just backwards, meaning a line was measured in the opposite direction.  The two angles that are negative are found by checking the possibility that the feature was measured in the opposite direction.  I made a fix that included that now for the non-plane features.  I don't think this is critical enough to be included in V40R.
Incidentally, when you change your nominal, it won't re-evaluate the measured until the next time the dimension is executed.  This is different than how it was done back when we'd try to determine the measured value by which quadrant it was in.
For review:
V40\DIMENS\DIM_2D_A.CPP
V40\DIMENS\DIM_3D_A.CPP
V41\DIMENS\DIM_2D_A.CPP
V41\DIMENS\DIM_3D_A.CPP
<<END>>
<< Changes made by Wade Burton -- 02/23/06  11:24:25>>
Action:  Wade Burton to Bret Naylor, Assigned:  Paola Pallo to Wade Burton, Status:  OPEN to REVIEW
<<END>>
<< Bill Wilcox  --  02/22/06  07:21:39>>
Wade any comments here?
<<END>>
<< Changes made by Bill Wilcox -- 02/22/06  07:21:41>>
Action:  Paola Pallo to Wade Burton
<<END>>
<< Changes made by Tim Wernicke -- 02/21/06  15:34:18>>
Priority:  to High
<<END>>