Im having a problem with using a hole to show surface deviation...

[Jeff Hunt]

Our customer wants us to use our hole measurements to show deviation of the surface. The problem is, the numbers from the hole do not match a
surface hit right next to the hole. Here is an example:

Surface hit next to the hole

SURF1 =AUTO/VECTOR POINT,SHOWALLPARAMS = YES
THEO/4585.119,-378.738,1801.165,0.8168293,0,0.5768794
ACTL/4585.072,-378.738,1801.132,0.8168293,0,0.5768794
TARG/4585.119,-378.738,1801.165,0.8168293,0,0.5768794
THEO_THICKNESS = 1.295,RECT,SNAP = YES,$
AUTO MOVE = BOTH,DISTANCE = 5,$
FIND NOM AXIS = NONE,UPDATE VEC = 0.8168293,0,0.5768794,$
REPORT VEC = 0.8168293,0,0.5768794
DIM LOC188= LOCATION OF POINT NET A10
AX MEAS NOMINAL +TOL -TOL DEV OUTTOL
X 4585.072 4585.119 0.000 0.000 -0.047 F 0.047 <--
Y -378.738 -378.738 0.000 0.000 0.000 0.000 -#-
Z 1801.132 1801.165 0.000 0.000 -0.033 L 0.033 <--
T -0.058 0.000 1.000 -1.000 -0.058 F 0.000 -#-
END OF DIMENSION LOC188

Hole with 3 sample hits

H4268M0004 =AUTO/CIRCLE,SHOWALLPARAMS = YES,SHOWHITS = NO
THEO/4580.593,-385.359,1807.572,0.8168293,0,0.5768794,13.999
ACTL/4580.81,-385.282,1807.244,0.8234862,-0.0019719,0.5673329,14.397
TARG/4580.593,-385.359,1807.572,0.8168293,0,0.5768794
THEO_THICKNESS = 1.295,RECT,IN,CIRCULAR,LEAST_SQR,NORM,ONERROR = NO,$
AUTO MOVE = BOTH,DISTANCE = 5,RMEAS = None,None,None,$
READ POS = NO,FIND HOLE = NO,REMEASURE = NO,$
NUMHITS = 12,INIT = 3,PERM = 3,SPACER = 3,PITCH = 0,$
START ANG = 0,END ANG = 0,DEPTH = 0,$
ANGLE VEC = 0.5768794,0,-0.8168293,$
PUNCH = 0.8168293,0,0.5768794,PIN = 0.8168293,0,0.5768794
DIM LOC189= LOCATION OF CIRCLE H4268M0004
AX MEAS NOMINAL +TOL -TOL DEV OUTTOL
X 4580.810 4580.593 0.000 0.000 0.217 B 0.217 -->
Y -385.282 -385.359 0.000 0.000 0.077 I 0.077 -->
Z 1807.244 1807.572 0.000 0.000 -0.328 L 0.328 <--
D 14.397 13.999 1.000 -1.000 0.398 0.000 --#
END OF DIMENSION LOC189

Now, I was expecting the X deviation of the hole to match the T deviation of the point, being that 3 sample hits were taken. Why is this not the case?

[Matthew D. Hoedeman]

You should be able to use 3 sample hits to measure the hole, then report the S (surface) axis. This used to work, not sure if it does anymore. It should report the surface deviation, based on the 3 sample hits plane, where the center of the hole is.

[Jeff Hunt]

Do you have any idea why I couldn't just referance the axis itself (x in this case)? Could I use T to report the deviation?

[Matthew D. Hoedeman]

"T" will report the approach deviaiton for VECTOR, SURFACE, and EDGE points. When using EDGE points, the S will report the Surface deviation, it also used to report the Surface deviation for holes, slots, sqaure slots, etc. As for reporting the individual X, Y, or Z axis, that will only work IF the hole has a vector of 1,0,0, in any combination. If the surface the hole is in is NOT 100% square to an axis, reporting just the axis will not be correct.

[Jeff Hunt]

Thanks for the explination Matt. Now I have a new problem. When I dimension the hole in XYZD&T, everything comes out right, but when I send the stats to datapage & report on it in dataview, the T value of the hole will be correct, but the automotive deviation letter will be wrong. It is the same thing that use to happen on edge points before I learned about the GAPONLY command.

What I need to do is take the Holes dimensions, save them under a different feature name, & make the X,Y, & Z deviations 0, but have the T value still retain its value. I hope that is clear. Is there any way I can do this? Something like this:

ORIGINAL:
hole1
X 4580.902 4580.593 0.250 -0.250 0.309 B 0.059 -->
Y -304.631 -304.699 0.700 -0.700 0.068 I 0.000 -#-
Z 1807.200 1807.571 0.700 -0.700 -0.371 L 0.000 #--
D 14.393 14.000 0.500 -0.500 0.393 0.000 --#
T 0.039 0.000 0.250 -0.250 0.039 B 0.000 -#-

NEW:
surf_hole1
X 4580.593 4580.593 0.250 -0.250 0.00 0.0-->
Y -304.699 -304.699 0.700 -0.700 0.00 0.000 -#-
Z 1807.571 1807.571 0.700 -0.700 -0.0 0.000 #--
D 14.393 14.000 0.500 -0.500 0.393 0.000 --#
T 0.039 0.000 0.250 -0.250 0.039 B 0.000 -#-

All this is because of an issue of PCDMIS not passing the IJK values through stats. To figure out what auto deviation letter to use, it jus looks for the axis with the greatist error & uses that.

[Jeff Hunt]

There has to be a way to do that with a subroutine, but Im not smt, I mean smart, enough to figure it out.