Cone Circle Constructed Point

[Lukas]

What is the correct type of constructed point to intersect a cone and a circle? I believe the correct type of constructed point for a line/cylinder is a pierce point (please correct me if I am wrong), but that doesn't seem to work for cone/circle.


See pic. I am trying to intersect the radius with the 1.5° taper cone.asdf.PNG

[RIDER]

You can measure a line down the cone at the same Y location as the circle. Then construct a pierce point between the line and the circle.

[Lukas]

You can measure a line down the cone at the same Y location as the circle. Then construct a pierce point between the line and the circle.

I have done that before. Sorry, I completely forgot about that. Wish there was a a way to do circle and cone.

[Schrocknroll]

I have done that before. Sorry, I completely forgot about that. Wish there was a a way to do circle and cone.

Construct a circle from the cone using the nominal diameter of the circle as the gage diameter. Then in the command window change the gage diameter to CIR4.D if the name of the circle is CIR4. This will use the measured value of the circle to create the gage circle. Then you can dimension the height of the circle instead of a point.

CIR1       =FEAT/CIRCLE,CARTESIAN,IN,NO
            THEO/<8.9134,0.4921,-1.0352>,<0,0,1>,0.5
            ACTL/<8.9134,0.4921,-1.0352>,<0,0,1>,0.5
            CONSTR/CIRCLE,CONE,CON1,DIAMETER,CIR4.D

[JEFMAN]

If the circle is in Z workplane, X right and Y bottom, then :
ASSIGN/V1=DEG2RAD(CON4.A/2)
ASSIGN/V2=MPOINT(-SIN(V1),COS(V1),0)
ASSIGN/INT_POINT=CIR4.XYZ-CIR4.R*V2
Then create a generic point.

[Lukas]

Construct a circle from the cone using the nominal diameter of the circle as the gage diameter. Then in the command window change the gage diameter to CIR4.D if the name of the circle is CIR4. This will use the measured value of the circle to create the gage circle. Then you can dimension the height of the circle instead of a point.

CIR1 =FEAT/CIRCLE,CARTESIAN,IN,NO
THEO/<8.9134,0.4921,-1.0352>,<0,0,1>,0.5
ACTL/<8.9134,0.4921,-1.0352>,<0,0,1>,0.5
CONSTR/CIRCLE,CONE,CON1,DIAMETER,CIR4.D

Interesting. But I'm trying to dimension the location of that point in the other axis. If you're looking at my picture as an 2D Cartesian coordinate system, then I'm trying to dimension it in the X direciton, not Y. I'll give JEFMAN's idea a try next. I would like to sort this out because I'm not a fan of taking a line on a turned part.

[Lukas]

If the circle is in Z workplane, X right and Y bottom, then :
ASSIGN/V1=DEG2RAD(CON4.A/2)
ASSIGN/V2=MPOINT(-SIN(V1),COS(V1),0)
ASSIGN/INT_POINT=CIR4.XYZ-CIR4.R*V2
Then create a generic point.

JEFMAN, it didn't work. I'm not sure what I did wrong. Please see pic and code. The X nominal is suppose to be .170. I'm not sure where the .2306 and -3.6293 are coming from. Note that I am looking for the intersection between CIR4 and CON3 (not CON4 as shown in my original pic). It shouldn't make a difference though.

            ASSIGN/V1=DEG2RAD(CON3.A/2)
            ASSIGN/V2=MPOINT(-SIN(V1),COS(V1),0)
            ASSIGN/INT_POINT=CIR4.XYZ-CIR4.R*V2
F1         =GENERIC/POINT,DEPENDENT,CARTESIAN,$
            NOM/XYZ,<INT_POINT,0,0>,$
            MEAS/XYZ,<-3.6293,0,0>,$
            NOM/IJK,<0,0,1>,$
            MEAS/IJK,<0,0,1>
DIM LOC2= LOCATION OF POINT F1  UNITS=IN ,$
GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
AX    NOMINAL       +TOL       -TOL       MEAS        DEV     OUTTOL
X       0.2306     0.0050     0.0040    -3.6293    -3.8599     3.8559 <--------
END OF DIMENSION LOC2

snip.png

[JEFMAN]

INT_POINT should give 3 coordinates, so you have to use INT_POINT.X, INT_POINT.Y and INT_POINT.Z in the generic point. (and the code is given in zplus workplane...)

I see now that CON3 or CON4 change everything in the formula, because the code gives the tangent point between CON4 and the torus !!!!!!
With CON3, it's not a tangent...

[JEFMAN]

So it's a little more hard...(here in the Y workplane !!!!!!!)
ASSIGN/V1=CON3.XYZ (should be the summit of the cone)
ASSIGN/V2=DEG2RAD(CON3.A/2)
ASSIGN/V3=MPOINT(-COS(V2),0,SIN(V2))+V1 gives a point on a generatrix of the cone (if the part is leveled on CON3)
ASSIGN/CC=(V1.Z-V3.Z)/(V1.X-V3.X)
ASSIGN/DD=V1.Z-CC*V1.X
ASSIGN/FF=DD+CC*CIR4.X-CIR4.Z
ASSIGN/DELTA=(2*FF*CC)^2-4*((CC^2+1)*(FF^2-CIR4.R^2)
ASSIGN/X1=(((-2*FF*CC)-SQRT(DELTA))/(2*(CC^2+1)))-CIR4.X
ASSIGN/Z1=(CC*X1+FF)-CIR4.Z
F1 =GENERIC/POINT,DEPENDENT,CARTESIAN,$
NOM/XYZ,<X1,0,Z1>,$
MEAS/XYZ,<X1,0,Z1>,$
NOM/IJK,<0,0,1>,$
MEAS/IJK,<0,0,1>

Here it's a little late, any error is possible

[Lukas]

So it's a little more hard...(here in the Y workplane !!!!!!!)
ASSIGN/V1=CON3.XYZ (should be the summit of the cone)
ASSIGN/V2=DEG2RAD(CON3.A/2)
ASSIGN/V3=MPOINT(-COS(V2),0,SIN(V2))+V1 gives a point on a generatrix of the cone (if the part is leveled on CON3)
ASSIGN/CC=(V1.Z-V3.Z)/(V1.X-V3.X)
ASSIGN/DD=V1.Z-CC*V1.X
ASSIGN/FF=DD+CC*CIR4.X-CIR4.Z
ASSIGN/DELTA=(2*FF*CC)^2-4*((CC^2+1)*(FF^2-CIR4.R^2)
ASSIGN/X1=(((-2*FF*CC)-SQRT(DELTA))/(2*(CC^2+1)))-CIR4.X
ASSIGN/Z1=(CC*X1+FF)-CIR4.Z
F1 =GENERIC/POINT,DEPENDENT,CARTESIAN,$
NOM/XYZ,<X1,0,Z1>,$
MEAS/XYZ,<X1,0,Z1>,$
NOM/IJK,<0,0,1>,$
MEAS/IJK,<0,0,1>

Here it's a little late, any error is possible

Thanks JEFMAN! I'll give it a try ASAP. Man, it looks like you know your stuff haha.

[Paul Sarrach]

Construct a circle from the cone using the nominal diameter of the circle as the gage diameter. Then in the command window change the gage diameter to CIR4.D if the name of the circle is CIR4. This will use the measured value of the circle to create the gage circle. Then you can dimension the height of the circle instead of a point.

CIR1 =FEAT/CIRCLE,CARTESIAN,IN,NO
THEO/<8.9134,0.4921,-1.0352>,<0,0,1>,0.5
ACTL/<8.9134,0.4921,-1.0352>,<0,0,1>,0.5
CONSTR/CIRCLE,CONE,CON1,DIAMETER,CIR4.D

When constructing a circle for a gage point it gives a height and that is what it is asking. You need to construct a circle, from a cone and give the height and the output will be the circle at that height?

[UKCMM]

Another method you could use is .

Measure the cone and circle/radius

Construct 2 circles at the ends of the cone by

Select the measured cone
Click the radio button ( cone )
Select Type ( Height )
Select Value 0 or the length of the cone to get the circles at either end
Select Point ( cone start )
Select Vector ( cone vector )

Once you have the 2 circles create 2 generic points using data from the circles for the measured values

So the measured data would look something like

CIR2.X.MEAS,CIR2.Y.MEAS,CIR2.Z+CIR2.R


Construct a line from these points and intersect with the measured circle to get the point you want

You may have to change the selection order to get the line in the correct direction to intersect where you want.

HTH

[JEFMAN]

Another way again :
(assuming that the cone axis is the X level, Y and Z origin)
Construct a circle / cone / height with CIRC4.X as the height (so you obtain the diameter of what I call CIR_CON)
So you have :
ASSIGN/H=CIR4.Z-CIR_CON.R
ASSIGN/D=H*COS(DEG2RAD(CON3.A/2))
ASSIGN/ALPHA=DEG2RAD(CON3.A/2).........................alpha = CON3.A/2, and D=R.COS(BETA) is also true
So :
ASSIGN/BETA=ACOS(D/CIR4.R)
So you can construct a generic point with :
ASSIGN/COORDINATES=CIR4.XYZ+MPOINT(-CIR4.R*COS(BETA+ALPHA),0,-CIR4.R*SIN(BETA+ALPHA)

[Lukas]

I did what UKCMM said. It worked and it makes sense to me.

JEFMAN, your methods seem interesting, yet harder. I don't know if I am experienced enough to follow along. I tried your post #9 method and I got stuck on the ASSIGN/DELTA line. It turned red and I couldn't find any typos. I am going to attempt your post #13 method. I think it will make more sense to me because I can refer to the diagram. What do you think of UKCMM's method? Do you think there would be any difference in end result?

[measurestuff413]

Above are some great answers.

A more simple way ( and the way I'd approach this and have with success in the past):

Since it looks like you don't have much of that radius as real physical surface......

1) Switch to appropriate workplane
2) Measure the rad with 4 points using FIXED RAD
3) Origin to the fixed rad
4) Measure a new rad using LEASTSQU at X0Z0 (or whatever...depending on the workplane you're in..)
5) Measure a line. USE POINTS and CONSTRUCT them into lines. PC DMIS IS A DILLHOLE and doesn't like autolines sometimes for some reasons (which nobody can figure out why so I never use them).
6) Put a point at the intersection of the line and the LEASTSQU rad

**NOTE**
The above will only work if the features intersect. The vector of the line must be pointed AT the end of the radius otherwise they technically don't intersect.

Now that you have the one point...you can copy & paste with pattern if you want to do this in multiple spots/use the points for advanced constructions