PR 259205 FCF TP is fetching wrong measured values

huangyhg

FCF TP is fetching wrong measured values
<< Anders Isaksson  --  01/21/09  14:50:35>>
In the attached program, the FCF TP dimensions L&Auml;GE1, 2, 3, 4 (at the end of the program) are showing the wrong measured value for Y in the Summary Table. X is correct, nominals are correct, but the Y values are slightly off.
How can this happen?
If I evaluate legacy TP on the same features, there is no difference between the TP Y and the Feature Y (or the Location Y).
<<END>>


<< Changes made by Anders Isaksson -- 03/10/09  16:57:51>>
Status:  RESOLVED to CLOSED
<<END>>
<< Don Turcotte  --  02/19/09  14:05:42>>
This is because the datums are specified in incorrect order.  They do not correspond to the alignment which is being used.
A2         =ALIGNMENT/START,RECALL:A1,LIST=YES
            ALIGNMENT/LEVEL,ZPLUS,PLAN Z PLUS
            ALIGNMENT/ROTATE,XPLUS,TO,LIN SP&Aring;R UPP&Aring;T,ABOUT,ZPLUS
            ALIGNMENT/TRANS,XAXIS,CIR REF B
            ALIGNMENT/TRANS,YAXIS,CIR REF B
            ALIGNMENT/TRANS,ZAXIS,CIR REF B
            ALIGNMENT/END
Datums constrain degrees of freedom in the order in which they are specified.  The primary datum (plane in this case) constrains as many degrees of freedom as the datum featue type allows.  In this case the primary plane constrains rotation about X and Y and translation in Z (Level).  The secondary datum constrains as many of the remaining degrees of freedom as the datum feature type allows.  In this case, the secondary datum is a line so it constrains rotation about Z and translation in Y (the line defines the orientation of the X axis in the datum reference frame).  This is not what the alignment specifies.  The alignment sets the X,Y origin to the circle (tertiary datum).  To achieve this in the datum reference frame, the circle should be specified as the secondary datum.  As a tertiary datum, the circle constrains only translation in Y (the last remaining degree of freedom).
So if you change the FCF true position to use datums  A,C,B you will get the results you expect which are consistent with the legacy TP using no datums or a Location dimension.
<<END>>
<< Changes made by Don Turcotte -- 02/19/09  14:05:55>>
Action:  Yanhua Huang to Anders Isaksson, Status:  OPEN to RESOLVED
<<END>>
<< Anders Isaksson  --  01/26/09  11:10:33>>
Added another, similar program with the same problem.
<<END>>
<< Changes made by Tim Wernicke -- 01/21/09  08:20:55>>
Action:  Tim Wernicke to Yanhua Huang, Assigned:   to Yanhua Huang, Priority:  to Critical
<<END>>

huangyhg

daily report 2-20-2009
Daily report:
PR259155 - Fixed Porter reported #8 crash issue, set review to Porter again. And filed PR259703 - Detected memory leaks when import a Catia model via Catia 5 DCT, set action to Paul.
PR 259205 - Don T. helped me and gave Ander an answer.

PS: Sorry, my modem was damaged last Friday, so send you my daily report via Email now.
Yanhua Huang yanhua.huang@hexagonmetrology.com
Software Engineer – GD&T | New GPS
PC-DMIS CMM Reporting Group
Wilcox Associates, Inc.
Hexagon Group
86-13681691545

From: Yanhua HUANG
Sent: 2009年1月22日 9:57 PM
To: Yanhua HUANG; Dave Petrizze; Don Turcotte; Steve Barber; Ben Mathews; Don Ruggieri
Cc: Rick Tischler; Porter Bassett; Bill Velle; Joe Zink
Subject: daily report 1-22-2009

Dave,
CHINA’s New Year is coming. Tomorrow morning, I will have a long travel and go back the hometown. So send my daily report now.
Daily report:
1.        Fixed PR259155-Can't change FCF modifier  (Priority #1) in PCDIMS V44B, and now set it review to Don T.
This is not the issue of Inspection Planar, the initial design in PCDIMS is, do not pop out feature MC list until a feature with MC is selected.
Found another issue in IP V1.2B, MC is wrong when edit true position specification with the considered feature = cylinder type and the datum feature = cylinder type. paired working with Porter,  Porter provided a new function as CIPTolerance::GetDatumRefType to get the datum type associated with FCF. Now code is done. Joe, please do more tests on true position of IP V1.2B, thanks in advance.
2.        Working on PR 259230-refactoring V44B\DIMENS\FEATCTRLFRM.CPP, because FEATCTRLFRM.CPP is too big, split FEATCTRLFRM.CPP into AxisInformation.CPP (extract code of AXIS_INFO class, because AXIS_INFO is also using by location dimension), FCFReport.CPP (the functions relative to report axes and report results), GeometricalTolerance.cpp (mainly keep synchronization with GeometricalTolerance.cpp in IP) and FEATCTRLFRM.CPP (remain these core functions), and now set it review to Don T.
3.        For PR256530 - FCF True Position, yesterday BV helped me to look for the essential reason, thanks BV.
PS: I will be on vacation for CHINA’s New Year from Jan. 23 to Feb. 7.  

Best Wishes,
Yanhua