How do i measure these datums correctly?

[pcdmisstudent]

Hey all!
Ive posted parts of the drawing here so you can look at it. Because thats much easier than explaining it in words.
How do i measure this correctly?

for Reference C, do i measure two circles and construct a line between the circles and make that line Ref C ? (then do the same for ref D).
But then how do i do the C-D? Construct a line between the two lines?

For Reference B im not really sure how to measure it. Do i measure it as two planes on opposite sides and construct a midplane between them and set that as ref B?

For Ref A, do i measure it as a cone?

And regarding the symmetry dimensioning. From what ive heard symmetry shouldnt be dimensioned for circles or holes. Only planes: should i redo the dimensioning as position instead?

[Matthew D. Hoedeman]

All of those callouts look bogus to me.

[pcdmisstudent]

Whys that?

[Matthew D. Hoedeman]

What is the "A" datum? That drawing sure doesn't define it for crap. The "B" looks to be one side of the shoulder, but it is the same axis as "A" so which is which and what do each do?

[JEFMAN]

The intersection of round slot with main surface make me think that the part is a double cone, so "Ø" are missing on dimensions ?
Am I right ?

If so, A is a circle constructing on the cone at the right diameter.
B is the mid plane between both sides.

Assigning a primary on a circle is a little strange, but here, it could make sense if you constuct the circle from the cone axis and not from the alignment.
But I wouldn't draw it like this !!!!!!!

[louisd]

Symmetry to A is bogus, because the section is implied to be rotated normal to C-D. How do you assess symmetry of the cylinder, back to the centroid of the diameter, without either like JEFMAN mentioned A as an axis line instead of a diameter, or without implying the section view is orthogonal to any other feature than C-D plane? even then, it's not relevant to A then, that symmetry would be relevant to C-D plane.

Symmetry to B technically works. construct a midplane for B datum, and evaluate the symmetry of the cylinder axis back to the B datum Midplane.

Perpendicularity to C-D I would personally argue, as the method of constructing a plane along two SLOTTED axis lines is ambiguous, especially if the parts have any bow,, warping, or parallelism errors to the C & D axes. Depending upon where along the C or D axis lines, you measure, to derive the centroid points to produce a mid-plane, will significantly affect the measured value.

[DungT]

I know I'm gonna get yell at but wadahell

Hey all!
Ive posted parts of the drawing here so you can look at it. Because thats much easier than explaining it in words.
How do i measure this correctly?

for Reference C, do i measure two circles and construct a line between the circles and make that line Ref C ? (then do the same for ref D).
But then how do i do the C-D? Construct a line between the two lines?
Measure C as round slot 2 places top and bottom, then do the same for D.
PC-DMIS sees 4 slots as points, change work plane and then construct a best fit plane then do perp to that plane

For Reference B im not really sure how to measure it. Do i measure it as two planes on opposite sides and construct a midplane between them and set that as ref B?
That's what I would do

For Ref A, do i measure it as a cone?
I would measure cone and then construct a circle cone/plane both sides and then construct a midpoint between 2 constructed circles.

And regarding the symmetry dimensioning. From what ive heard symmetry shouldnt be dimensioned for circles or holes. Only planes: should i redo the dimensioning as position instead?
Yes, I would do it as TP.

If I remember correctly, ANSI Y14.5 1973 symmetry interpreted as TP. ANSI Y14.5, 1982 eliminated symmetry and suggested to use TP instead. ASME Y14.5, 1994 reinstated symmetry with different set of definition.

[pcdmisstudent]

Hey thanks for clear reply.
A stupid question tho. I have PC DMIS 2011 so theres no autoelement for rounded slots. Do i need to do a "measured slot" and manually measure it? or can i construct a slot out of two auto-circles?

[louisd]

create two circles, then construct a midpoint of the two circles.

[louisd]

If I remember correctly, ANSI Y14.5 1973 symmetry interpreted as TP. ANSI Y14.5, 1982 eliminated symmetry and suggested to use TP instead. ASME Y14.5, 1994 reinstated symmetry with different set of definition.

Symmetry has always been fundamentally defined the same way through all revisions of ASME/ANSI Y14.5 & 14.5M, until it was removed entirely with 2018 rev.
It has most certainly never been considered equivalent or anywhere near as robust as tolerance of position (TP). In very specific instances you can control symmetry the same way by using TP, but they have never been synonymous or interpreted as the same.

I don't have the 1973 or 1982 spec's, but here are the last two revisions:

1994

Symmetry is that condition where the median points of all opposed or correspondingly-located elements or two or more feature surfaces are congruent with the axis or center plane of a datum feature. Where design requirements dictate a need for the use of a symmetry tolerance and symbol, the method shown in fig 5-61 may be followed. The explanation given in para 5.12 applies to the considered feature(s) since symmetry and concentricity controls are the same concept, except as applied to different part configurations. Symmetry tolerance and the datum reference can only apply on an RFS basis.

2009:

Symmetry is that condition where the median points of all opposed or correspondingly located elements of two or more feature surfaces are congruent with a datum axis or center plane. Where design requirements dictate a need for the use of a symmetry tolerance and symbol, the method shown in Fig. 7-66 may be followed. The explanation given in para. 7.6.4 applies to the considered feature(s), since symmetry and concentricity controls are the same concept, except as applied to different part configurations. Symmetry tolerance can only be applied RFS and the datum reference can only be applied RMB.

[DungT]

I have 1982, 1994 and 2009. I don't have 1966 and 1973 but I have seen both