Is there a setting to adjust the Findnoms search area?

[MrComment]

SCN9       =FEAT/SCAN,FREEFORM,SHOWHITS=NO,SHOWALLPARAMS=YES
            EXEC MODE=DEFINED, NOMS MODE=FIND NOMS,CLEARPLANE=NO,SINGLE POINT=NO,THICKNESS=0
            FINDNOMS=1,SELECTEDONLY=YES,USEBESTFIT=NO,PROBECOMP=YES,AVOIDANCE MOVE=NO,DISTANCE=0,CAD Compensation=NO
            DIR1=FREEFORM,
            HITTYPE=VECTOR
            INITVEC=0,0,1
            DIRVEC=0.9743702,-0.2249507,0
            POINT1=143.5777,-47.78939,0
            MEAS/SCAN
            BASICSCAN/FREEFORM,SHOWHITS=NO,SHOWALLPARAMS=YES
            <143.5777,-47.78939,0>,<181.86933,38.8552,0>,DirVec=0.976563,-0.2152316,0
            InitVec=0,0,1,THICKNESS=0
            FILTER/NULLFILTER,
            EXEC MODE=DEFINED
            BOUNDARY/
            HITTYPE/VECTOR
            NOMS MODE=FINDNOMS,1
            ENDSCAN
            ENDMEAS/

I'm using a freeform scan to scan the perimeter on the top of this groove shown in the picture.

I use freeform because the Perimeter is thin and not perfect (of course) therefore the probe wants to fall off the edge surface. My solution to that is to first scan using a Linear Open on the inside of the groove wall. Then export out the measured data to Excel, offset X and Y by the wall thickness and reset my Z, I, J and K values to measure a plane normal to the surface. These values then get loaded into the freeform scan shown below. The probe then scans exactly on the center top of the groove wall of the part.

The problem is unless the CAD and Part match (which they never do) findnoms "falls" off the edge of the part if the deviation is greater then the CAD thickness of this wall. (Which it is, and also why a Perimeter Scan doesnt work.)

I thought about creating and merging a large unbroken Plane on a separate CAD layer in PCDMIS as a potential work around, before I go to all of that trouble though I'd try to get other potentially easier options first. (i.e. a setting to change the "search dia or thickness" of the Findnoms Findnom-er

Thanks

[Boomer]

would it be possible to do a relearn scan around a perimeter and then do a best fit vector alignment to that scan...then do the type of scan you are doing???