Datum Reference Frame Requirement for Holes on a Round Part 1

[Keith1029]

Good Morning All,
(As relating to Y14.5-2009)

My question really extends to any parallel axis features/reference frame but to illustrate the point as simply as possible:

By way of example, lets assume a cylinder with flat ends perpendicular to the axis of the cylinder. Let Datum A be the center axis of the cylinder, let Datum B be one end face of the cylinder and assume that all the required dimensions and tolerances to constrain the part be applied.

Now the goal is to drill a hole down the center of the cylinder. The tolerance on this hole will be a +/- size dimension to establish size and form boundaries of the feature. The location of the hole is then defined by a positional tolerance (which would apply to the axis or the actual boundary depending on the interpretation but should not change the outcome of the question). My interpretation is that the only reference datum required in the FCF is datum A relating the 2 axis. Even if the hole was offset from center (or a pattern) located with a polar system (radius and angle spacing between holes on the pattern) the only datum reference in the FCF would still be the datum axis A. Is this correct? Or is a callout to datum B required for some reason that I am not seeing/interpreting properly?

Figure 7.5 in Y14.5-2009 shows an example of this with the hole pattern tolerance only to Datum B (the cylindrical datum feature in the image) as is my interpretation of the requirement.
Figure 7.6 in the std. shows an example of a C'bore pattern of similar holes with the single datum reference described applied to the C'bore also, but for the thru hole pattern directly above it, Datum features A and B are referenced. Is the reference to A in the CFC superfluous in this figure or does it somehow modify the control?

Thanks.

 

[greenimi]

You did not tell us, how the part is oriented in the assembly. Is it primary oriented to B and then to A or is centered on A and then make contact with B?

 

[Keith1029]

I'm not directly concerned with which scheme to use and when, just what the theoretical difference in meaning would be between them.

From this question however, I assume you are alluding to the fact that datum precedence in the callout could effect the shape/size of the tolerance zone, as the ultimate zone would then have a different stack up relationship to the true deviation of the datum features?

 

[greenimi]

Quote: "just what the theoretical difference in meaning would be between them"

So, what are our options? Theoretical difference between what schemes?
Could you have a drawing attached?

What are you trying to control? The ID?

 

[dgallup]

You don't need datum B.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[3DDave]

You need to control two directions of location and two rotations to locate and orient a cylinder.

A reference to the cylinder provides two location and two rotation controls. Any further controls are redundant.

If the end plane was used as primary it would fix two rotations and one location, but the location doesn't apply to the hole because the location fixity is perpendicular to the axis of the hole. Then a cylinder reference can be added to fix two locations that do apply to the hole while ignoring the two rotation controls from the cylinder reference that are redundant with the plane reference.

 

[CheckerHater]

I assume you are alluding to the fact that datum precedence in the callout could effect the shape/size of the tolerance zone, as the ultimate zone would then have a different stack up relationship to the true deviation of the datum features?

Something like this

"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

 

[greenimi]

Thank you CH.

That is what I was talking about in my post " Is it primary oriented to B and then to A or is centered on A and then make contact with B?"

I cannot see or attach any picture/ drawings from work anymore

(stupid policy)...................

 

[Keith1029]

Thanks CH, that is the image that is exactly the graphic that I was thinking about but I am struggling with what it means from a datum reference frame/position set up.
Situation 1 I am clear on:
When controlling a hole position in the sample part (or any feature with an axis), if function required only parallelism to datum A (with no regard to face B relationship) then the callout would be "POS|x.x|A|" and the axis of the basic axis of the hole would then be identified as perfectly parallel to the axis A ("primary") as shown in the top view, the true axis of the hole would then be allowed to vary within the x.x tolerance zone around the basic perfect location and form. When calling out "A" as then primary reference in this manner, any further datum references to B are not required and do not refine the callout further.

Situation 2 where the hole is required to be perpendicular to B mostly makes sense to me but the setup of the datum reference frame is what I am unclear about. For this second requirement, the callout would be "POS|x.x|B|A". The first datum reference would establish the primary plane shown in the lower image and the basic axis of the hole would then need to be perfectly perpendicular to this. This then controls the basic axis orientation of the hole (as parallel to the axis in image 2 as compared to the axis in image 1). So the orientation of the hole is now controlled, but its position is still free to move around on face B and must be constrained which is done with the secondary callout to A. My point of confusion stemmed around the fact that the secondary reference to A actually relates the position of the hole axis to the "secondary" datum shown in the lower image.

This means that a single datum feature callout does not necessarily establish a single theoretical datum, but could establish a set of theoretical datums, the selection of which depends on the precedence of the reference frame selected as well as the actual deviation of the datum features from perfect form? In fact if the same drawing were to have multiple features, one with the callout referencing A only and one referencing B|A, the datum axis used for A could be a different theoretical axis depending on the feature being inspected? If this is the case, then this is the piece of the puzzle I was misinterpreting and everything is now clear. (Of course if this is wrong then I am worse off, because now the wrong answer makes intuitive sense).

 

[CheckerHater]

@Keith:
The datum is nothing more, nothing less than the origin of your measurement.
Look at something like Fig. 2-5 in ASME Y14.5-2009: You measure different way - you get different result.
There is no need to look for some mysterious meaning to it.

"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future