PR 257415 Bad angle result between two lines. Changing nominal, the measured

huangyhg

Bad angle result between two lines. Changing nominal, the measured value doesn't match.
<< Antonio Cuenca  --  10/02/08  19:20:41>>
The attached partprogram has an output from the angle between two lines, and customer needs output of the two possible angles between the lines. But select the other angle is impossible, depending on what value is written for theoretical. My explanation it's not so clear, so please whatch the video attached to clarify the problem.
Steps to reproduce:
See attached video capture.
The meas value (comp to 180) is 124.52 degrees. This is shown right if I write as a nominal an angle between 122 and 123.9, but the nominal of the angle is 124.50 Writing the nominal ok, the measured value changes to 55.4 (the angle in first quadrant) So, it's impossible to show the right measured value writing the right nominal.
Results:
Measured value in report in wrong quadrant if I write the right nominal value for the angle.
Expected Results:
Right angle displayed.
<<END>>

<< Antonio Cuenca  --  10/09/08  10:22:24>>
Tested in V42R Built on Oct 8th. Works right. Now user can select the angle he wants very easily.
<<END>>
<< Antonio Cuenca  --  10/08/08  20:24:34>>
I will test in 4.2R tomorrow build.
<<END>>
<< Don Turcotte  --  10/08/08  10:00:03>>
Merged into V42 MR2 and V431 (V43 MR1)
Files inserted to server
------------------------
V42R\DIMENS\DIM_2D_A.CPP
V431\DIMENS\DIM_2D_A.CPP

<<END>>
<< Changes made by Don Turcotte -- 10/08/08  10:00:13>>
Action:  Don Turcotte to David Petrizze, Status:  OPEN to REVIEW
<<END>>
<< Tim Wernicke  --  10/08/08  08:30:01>>
Hi Don, please go ahead in V42R, V431 per my e-mail.  Thanks.
<<END>>
<< Changes made by Tim Wernicke -- 10/08/08  08:30:13>>
Action:  CMM Group to Don Turcotte, Status:  REVIEW to OPEN, Priority:  Critical to Stop Rel.
<<END>>
<< Antonio Cuenca  --  10/06/08  10:01:34>>
Tested on 43B with some customer partprograms. It seems to work much better. I believe 5 degrees is better than only one degree of margin. I attach a partprogram for verification. Please download "AngleVerif.prg", open it and find "ANGLE VERIFICATION". Look how now it's much more easy reach the desired angle result.
<<END>>
<< Don Turcotte  --  10/03/08  09:25:19>>
When PC-DMIS tries to figure out which quadrant to use it compares the entered nominal with the nominal calculated from the nominal vectors of the two features.  There are four possible answers (quadrants):
    vector1 to vector2
    vector2 to vector1
    vector1 to -vector2
    -vector2 to vector1
PC-DMIS picks the quadrant that matches the entered value within some epsilon.  The epsilon used was 1 degree.  I have increased this to 5 degrees since the entered nominal was varying by more than 1 degree from the nominal calculated using the nominal vectors of the features.  When no match is found, the quadrant used is the quadrant derived from vector1 to vector2.
   
Fixed in V43B, V44B.
Files inserted to server
------------------------
V43B\DIMENS\DIM_2D_A.CPP
V44B\DIMENS\DIM_2D_A.CPP
<<END>>
<< Changes made by Don Turcotte -- 10/03/08  09:25:48>>
Action:  Don Turcotte to CMM Group, Status:  OPEN to REVIEW
<<END>>
<< Changes made by Tim Wernicke -- 10/02/08  14:22:29>>
Action:  Tim Wernicke to Don Turcotte, Assigned:   to Don Turcotte, Priority:  to Critical
<<END>>