How to control features separately 3

[bxbzq]

Take fig 7-54 in '09 std as an example, the four holes are treated as two separate patterns. If I want each hole to be controlled separately, how do I specify on the drawing?

 

[pmarc]

This time I really give up.

 

[axym]

Hi All,

For what it's worth, I agree with pmarc on all counts. His explanation of separate requirements, and the additional freedom it allows, is correct and accurate. This guy knows what he's talking about, and I hope that people appreciate the high level of knowledge being demonstrated here.

The objections that are being raised seem to relate more to how one might choose to utilize the additional freedom in manufacturing and inspection (or choose not to utilize it). But hypothesizing that manufacturing or inspection might choose to ignore part of the specification does not make it worthless. At this point I'm not sure if the objections are fully genuine or are just arguments for argument's sake. The dismissive and sarcastic language doesn't help.

The specification of separate requirements allows each characteristic to be inspected in different candidate datum reference frames. In other words, inspected in different setups. Candidate DRF's occur when the degrees of freedom between the datum features (i.e. the part) and the datum feature simulators (i.e. the inspection fixture) are not fully constrained. Examples include the following:
[li]Datum features of size referenced at MMB (usually called datum feature shift, translational and rotational DOF's are only partially constrained)[/li]
[li]Rocking on planar datum features (rotational DOF's are only partially constrained)[/li]
[li]Certain degrees of freedom left completely open (not usually called datum feature shift but it's the same thing)[/li]
The distinction between simultaneous requirements and separate requirements only becomes significant when the degrees of freedom are not fully constrained, so to keep directing the discussion back to fully constrained cases is pointless.

Also, the concept of candidate DRF's is general. It doesn't depend on the geometry of the considered feature. So it doesn't really matter if we have a hole pattern or not, it's just that the examples illustrating separate requirements usually have hole patterns in them. But we can have the same thing for other features with profile controls as well.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[greenimi]

[This guy knows what he's talking about, and I hope that people appreciate the high level of knowledge being demonstrated here]

I agree wholeheartedly.

 

[gabimo]

There are not enough stars on the sky the give to pmarc for his contribution on this site and on linkedin.
A lot of people here promised him lunches, booze etc. for the ABSOLUTELY FREE help they got from pmarc.
THANK YOU PMARC

 

[CheckerHater]

The distinction between simultaneous requirements and separate requirements only becomes significant when the degrees of freedom are not fully constrained

Did pmarc agree with that statement? Because that's all I wanted from him.

 

[Belanger]

Perhaps it boils down to this: If there is any datum shift available, does that constitute having all degrees of freedom fully constrained?

IOW, if Fig. 7-54 didn't use any MMB modifiers, everyone would agree that all DOF are constrained. But with the MMB modifiers, is it "fully constrained"?

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

 

[CheckerHater]

Looks like a matter of terminology.
To look from the other side: can we call it a "datum shift" if datum is not referenced at all, so we can "shift" indefinitely?

 

[greenimi]

Probably you already know.
AGI: Datum shift explained: page 33 and up.
Found this:

http://www.conceptmachine.com/files/zeiss/GD&T_Handouts.pdf

Same thing pmarc explained in his sketches.

 

[powerhound]

I understand the question as follows: If we have a planar primary datum (the face of a plate), an axis for a secondary datum (hole in center of plate) and a centerplane for a tertiary datum (slot in edge of plate) and all are referenced RMB then the part is considered fully constrained, but if the secondary and/or tertiary datum are referenced at MMB then the part is no longer fully constrained. If that's the question then my opinion in regards to the datum features being referenced at MMB is that is it as constrained as it needs to be. No one would argue that if only the primary and secondary datums were referenced the part would be unconstrained. Would one really go so far as to say if the tertiary datum were referenced at MMB that the part is still unconstrained?

John Acosta, GDTP S-0731
Engineering Technician
Inventor 2013
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[axym]

J-P,

Your questions bring up a subtle issue, that is unfortunately a major digression in a thread like this.

The issue could be described as "partial constraint". Y14.5 does not directly address this - the standard describes degrees of freedom as being either constrained or unconstrained. But when there is shift or instability between the datum feature and its datum feature simulator, the constraint is somewhere in between.

With MMB modifiers, there is a high probability of datum feature shift between the datum feature and its fixed-size simulator. I think we can safely say that the DOF's are not fully constrained.

With planar datum features, there is a small (but significant) probability of rocking and instability. It would be most prevalent on primary datum features and possible on secondary. I would have to say that the DOF's are not fully constrained.

With RMB modifiers, there is a small possibility of datum feature shift even with a snug-fitting simulator. This issue could conceivably occur on primary cylindrical or width datum features, when a tapered condition exists. I would say that the DOF's are not fully constrained, but many others would disagree with this one. Almost fully constrained, but not quite ;^).

CH,

If a datum is not referenced at all, meaning that one or more degrees of freedom are completely unconstrained, then I would say yes - we can shift indefinitely. It's generally not described that way, but the effect is the same as datum feature shift on an MMB datum feature reference (only with no limits).

powerhound,

I'm not sure if the above comments address your questions or not - what do you think?

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[powerhound]

Yes, I think you addressed the question. I've never even considered that a FOS datum feature referenced at MMB did not constitute a fully constrained part though. If all datum features are referenced at RMB, it can completely immobilize a part but I've never really considered "immobilized" and "fully constrained" as the same thing.

John Acosta, GDTP S-0731
Engineering Technician
Inventor 2013
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[CheckerHater]

I am glad to see “fair and balanced” approach to the issue.
Would that be safe to say that the benefits of separate requirement are more significant when the datum(s) are allowed to shift / translate (use your favorite terminology) and less obvious when part can be “fully immobilized” (thanks powerhound)?

 

[pmarc]

Evan, greenimi, gabimot,
Thank you very much for these kind words. I am really flattered.