Measurement to depth coding help

[A-machine-insp]

I have a part that has a two identical rectangular ends on it. It calls 1.000 +/-.002" to 1.50 +/-.030 depth and 1.000 +/-.010" for the remainder of depth. We had an issue holding the depth tolerance the last time we made these parts so I have a program that lays a line of points spaced .005 on the top and bottom of the rectangle that starts well before the 1.47 depth and ends well after the 1.53 depth. How can I code the program to tell me at what depth the 1.000 +/-.002 dimension goes out of tolerance so I don't have to find the OOT dimension then find the points and figure out what depth they are at?

I know I did something similar to this in the Level 3 class but that was running a loop until one axis maxed out then started to get smaller (bulged surface on the hex block). Where this is going to differ is that I am not running a loop and I want to run all the points I have laid out, not stop when it goes OOT. Also, this is based on a distance dimension between 2 points, not the location of a single point. The other thing that may be an issue is the trihedron is not on the end face of the part, it is in the middle of the part. I don't know if this will be an issue, I'm just giving as much information as possible.

As stated above, I have a way to do it but it is all manual work and with us running a bunch of them, I am trying to save some time.

Insert crude drawing made in MS Paint here (see export controlled):
depth example.jpg

[DAN_M]

Assuming we're looking down at the part from the Z+...

I would measure circle within the ±0.002 zone, XY origin to it, then measure a point on the bore. Use a while loop to keep measuring the point and lowering it down .005 each time until the radial value of the point is outside of the range of (1.002-0.998)/2). (in F5 Menu under GENERAL tab, "treat theo values as if stored in part coordinates" must be checked off.) Then make a Z origin to the point returned by the loop.

With your new Z origin getting you local to your geometry, measure the depth as you normally would.

[A-machine-insp]

It is a rectangle but I see what you are saying. I could do the same with a midplane instead of a circle.

[JEFMAN]

Assuming the hits of lines are aligned, face to face :
ASSIGN/V1=SQRT(DOT(LN1.HIT[1..LN1.NUMHITS].XYZ-LN2.HIT[1..LN2.NUMHITS].XYZ,LN1.HIT[1..LN1.NUMHITS].XYZ-LN2.HIT[1..LN2.NUMHITS].XYZ)) which is an array of distances hit to hit.
ASSIGN/TOL_P=1.002
ASSIGN/TOL_M=0.998
ASSIGN/V2=((V1-TOL_P)-ABS(V1-TOL_P))/((V1-TOL_P)-ABS(V1-TOL_P)) gives an array of 0 and 1
ASSIGN/V3=SUM(V2) gives the number of hits out of tol +
ASSIGN/V4=LN1.NUMHITS-V3 gives the first hit OOT if the thickness grows up linearly
Do the same with TOL_M-V1 for the minus tol.

[DungT]

Stupid callout

[DungT]

Right, that was my point. The drawing should state "Ø1.000 ±.002 x 1.47 min deep, Remainder Ø1.000 ±..010"

[A-machine-insp]

Umm... That is what it says on the drawing. That's why I typed it in my crude drawing.

[DungT]

I would never flag the depth only size if it's oot and I would add a comment something like this "Ø1.000 ±.002 x 1.47 min, 4 cross sections, .25 apart, last place @1.47 deep"

[tking]

Distance between points?

Report your distances one at a time and the depth "Z" of the 2 points used in each........next distance.