Creating Nominal Center Points of Constructed Scan Segment Arcs

Did you mark the script command in your program? Otherwise it won't execute.

This seems a little suspect:
retval = PCDCommand.PutText (CircName & ".HIT[" & MidHit & ".." & MidHit & "]", REF_ID, 2)

This creates something like CIR1.HIT[1..1], if it is a line it should probably say CIR1.HIT[1]:
retval = PCDCommand.PutText (CircName & ".HIT[" & MidHit & "]", REF_ID, 2)

I didn't put a script command in my program. I just ran it from basic script editor. Do I have to insert a script command in my program?

I did the hit like that because I made a sample program to export in basic so I could see the names of things. When I was constructing the line in my sample program it kept giving me an invalid index error when I changed CIR1.HIT[42..42] to CIR1.HIT[42].

I want the script to essentially do this. I just added the assignment for clarity.

CIR1       =FEAT/CIRCLE,CARTESIAN,IN,NO
            THEO/<0.4724,2.3228,-0.2203>,<0,0,1>,0.5906
            ACTL/<0.4724,2.3228,-0.2203>,<0,0,1>,0.5906
            CONSTR/CIRCLE,SCAN_SEGMENT,BF,SCN1,0.184302,2.2583087,-0.2203124,0.7334256,2.1847235,-0.2203125,0,0,0
            ASSIGN/V1=ROUND(CIR1.NUMHITS/2)
PNT1       =FEAT/POINT,CARTESIAN,NO
            THEO/<0.4724,2.3228,-0.2203>,<0,0,1>
            ACTL/<0.4724,2.3228,-0.2203>,<0,0,1>
            CONSTR/POINT,OFFSET,ORIGIN,CIR1.TX,CIR1.TY,CIR1.TZ
LIN1       =FEAT/LINE,CARTESIAN,UNBOUNDED,NO
            THEO/<0.4724,2.3228,-0.2203>,<-0.1493052,-0.9887912,0>
            ACTL/<0.4724,2.3228,-0.2203>,<-0.1493052,-0.9887912,0>
            CONSTR/LINE,BF,3D,PNT1,CIR1.HIT[V1..V1],,
            OUTLIER_REMOVAL/OFF,3
            FILTER/OFF,WAVELENGTH=0

When I try to do the CONSTR/POINT,OFFSET,ORIGIN,CIR1.TX,CIR1.TY,CIR1.TZ it is creating 2 points in a single point. At least visually on my CAD. Is that the same for you? I feel like doing a CAST point would do the same thing, just the vectors would not be perfect

I could go around my program and paste this with pattern after every scan segment arc but I wanted to learn how to do some cool script haha

CIR1       =FEAT/CIRCLE,CARTESIAN,IN,NO
            THEO/<0.4724,2.3228,-0.2203>,<0,0,1>,0.5906
            ACTL/<0.4724,2.3228,-0.2203>,<0,0,1>,0.5906
            CONSTR/CIRCLE,SCAN_SEGMENT,BF,SCN1,0.184302,2.2583087,-0.2203124,0.7334256,2.1847235,-0.2203125,0,0,0
            ASSIGN/CIRCLE_ID=GETCOMMAND(527,"UP",1)
            ASSIGN/MIDHIT=ROUND(CIRCLE_ID.NUMHITS/2)
PNT1       =FEAT/POINT,CARTESIAN,NO
            THEO/<0.4724,2.3228,-0.2203>,<0,0,1>
            ACTL/<0.4724,2.3228,-0.2203>,<0,0,1>
            CONSTR/POINT,OFFSET,ORIGIN,CIRCLE_ID.TX,CIRCLE_ID.TY,CIRCLE_ID.TZ
LIN1       =FEAT/LINE,CARTESIAN,UNBOUNDED,NO
            THEO/<0.4724,2.3228,-0.2203>,<-0.1493052,-0.9887912,0>
            ACTL/<0.4724,2.3228,-0.2203>,<-0.1493052,-0.9887912,0>
            CONSTR/LINE,BF,3D,PNT1,CIRCLE_ID.HIT[MIDHIT..MIDHIT],,
            OUTLIER_REMOVAL/OFF,3
            FILTER/OFF,WAVELENGTH=0

PNT1 =FEAT/POINT,CARTESIAN,NO
THEO/<0.4724,2.3228,-0.2203>,<0,0,1>
ACTL/<0.4724,2.3228,-0.2203>,<0,0,1>
CONSTR/POINT,OFFSET,ORIGIN,CIRCLE_ID.TX,CIRCLE_ID.TY,CIRCLE_ID.TZ

Does this not create multiple points for you on the CAD? Or just 1?

It creates just one point for me.

The point of doing the offset point is to lock in the nominal value (could use a generic point too). Because on the actual part I'm checking most of my scanned radii are less than 90° of an arc. So this is my way of measuring a small arc circle without using a Fixed Radius circle, aligning to the center, and then measuring the polar radius. There are 350 ish radii on this thing so that would be very time consuming.

What should your nominal value be for these Radii? I was wondering if your method was getting you the correct nominal when going to reporting the length of the line

For the sample program it's this:

DIM LOC1= LOCATION OF LINE LIN1  UNITS=IN ,$
GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
AX    NOMINAL       +TOL       -TOL       MEAS        DEV     OUTTOL
L       0.2953     0.0100     0.0100     0.2953     0.0000     0.0000 ----#----
END OF DIMENSION LOC1
DIM LOC2= LOCATION OF CIRCLE CIR1  UNITS=IN ,$
GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
AX    NOMINAL       +TOL       -TOL       MEAS        DEV     OUTTOL
R       0.2953     0.0100     0.0100     0.2953     0.0000     0.0000 ----#----
END OF DIMENSION LOC2

It's a weird number because it's 7.5mm and I'm in inches.

If you do a Distance Dimension instead, and then select PNT1, and CIR1, then select add or subtract radius option do you get the same nominal value?

Only difference is it's negative

DIM DIST1= 2D DISTANCE FROM POINT PNT1 TO CIRCLE CIR1     (CENTER TO CENTER),SUB_RADIUS  UNITS=IN,$
GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH
AX    NOMINAL       +TOL       -TOL       MEAS        DEV     OUTTOL
M      -0.2953     0.0100     0.0100    -0.2953     0.0000     0.0000 ----#----

Only difference is it's negative

DIM DIST1= 2D DISTANCE FROM POINT PNT1 TO CIRCLE CIR1     (CENTER TO CENTER),SUB_RADIUS  UNITS=IN,$
GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH
AX    NOMINAL       +TOL       -TOL       MEAS        DEV     OUTTOL
M      -0.2953     0.0100     0.0100    -0.2953     0.0000     0.0000 ----#----

Pick Add Radius instead, it may be negative or positive depending on whether you select Add or Subtract. Which feature you select first, and possible which side of the part the circle feature is located. I think only Add or subtract may matter. This will save you from needing to construct lines all over your program if it works for you this way too. And your script would end up less involved, and wouldn't need your variable statements for N_HITS / 2. Don't worry, you can still do a cool script, just an easier one

That would be easier. It did change to positive when I selected "add radius" instead. I'm just not entirely sure that would give me the result I want. Especially the smaller arc-ed radii. If there's only 20° of an arc, the measured circle could be wacky which could make my results look crazy. I really wish they just dimensioned all these pockets as surface profile then I wouldn't have had any of these issues haha. I still am going to report them as profile as a secondary verification.

I apologize for pushing you so far off topic into another subject and method entirely from your original question.

Let's get back to what you would like your script to do. Do you still want to continue using your lines method with Location dimensions for line length?

Yes the line method or create a cast point from a hit on the circle and then measure it back to the offset/generic point nominal center of the circle.

Sorry, I've been trying to get this to work for me properly so I can tell if it is working correctly if I help you with your script. Been a pain in the butt trying to get the scans to do what I want, but I got there. So have you tried executing your script at all since you posted, or changed anything yet?

I tried to put a message box inside the loop, and it will not pop up, so I think there's an issue with the loop.

I think I figured out why, it's a good thing I created this offset point myself. So the Offset point has a required Result = segment in the exported basic script.

So you have to have the 3 Result = for each F_OFFSET segment and can't just use PutText, but need to use SetExpression

You will also need to add Variable --> Dim Result As Long

    retval = DmisCommand.PutText ("PNT1", ID, 0)
    retval = DmisCommand.PutText ("ORIGIN", REF_ID, 0)
    retval = DmisCommand.SetToggleString (1, COORD_TYPE, 0)
    retval = DmisCommand.PutText ("-1.103", F_OFFSET, 1)
    retval = DmisCommand.PutText ("-0.6445", F_OFFSET, 2)
    retval = DmisCommand.PutText ("0.7911", F_OFFSET, 3)
    retval = DmisCommand.SetToggleString (1, USER_DEFINED_THEOS, 0)
  Result = DmisCommand.SetExpression("CIR1.TX", F_OFFSET, 1)
  Result = DmisCommand.SetExpression("CIR1.TY", F_OFFSET, 2)
  Result = DmisCommand.SetExpression("CIR1.TZ", F_OFFSET, 3)

Where did you get this bit of code from? Trying to figure out what creates it

Set CircCenter = PCDFeatCmd.GetPoint(FPOINT_CENTROID, FDATA_THEO, CircName)

Great! Thank you! I'll try that.

Here is where I got my inspiration. I kind of meshed these 2 together:

Measured and Nominal Circle Values

To Points Script

Do you want this script to cycle through each Constructed Scan Segment, ~350 of them. Create the origin point relative to the TX, TY, TZ offset of the circle, then a line to the NUMHITS/2? Or is there more to it than that

Correct!

edit- the problem is this line <If PCDCommand.IsConstructedFeature And PCDCommand.Feature = CONST_SCAN_SEG_ARC Then>

I changed it to <If PCDCommand.IsConstructedFeature Then> I got the message box to pop up at least before I got a mismatch error on the lines where I'm setting the nominal X,Y,Z variables.