Self-referencing or not 1

[greenimi]

Fig 9-7 / ASME Y14.5-2009

If total runout defining datum feature D is replaced by position callout would you consider that the newly created position to be self-referencing (and hence not valid)?

Reasoning: I am thinking that position is checking the UAME and since D is secondary in the FCF than RAME is the one coming to play.

(Disclaimer: I am not saying that I would apply such callout, but I am just questioning its legality and validity, not its functionality)

 

[Garland23]

I agree that position would be valid, and I agree with your reasoning (the UAME and the fact that C is primary).

 

[CheckerHater]

The good self-check: try to devise a fixture that will control such position.

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

 

[greenimi]

The good self-check: try to devise a fixture that will control such position.

I agree with you CH, but I am not talking about the long shot implications and consequences of the inspection methods.
I am ONLY talking about validity of such callout.

 

[3DDave]

Neither "legal" nor "valid" are anywhere in the '2009 version of the standard used to describe a Feature Control Frame, so neither can be applied to evaluating this.

Generally, it is a bad sign when no effort goes into theorizing an inspection fixture to match a dimensional requirement.

I thought up a similar puzzle. Create a word of only vowels of letters chose from only consonants. Is there a word-creating standard that writes out this is possible or not?

 

[greenimi]

3DDave,

Neither "legal" nor "valid" are anywhere in the '2009 version of the standard used to describe a Feature Control Frame, so neither can be applied to evaluating this.
Generally, it is a bad sign when no effort goes into theorizing an inspection fixture to match a dimensional requirement.
I thought up a similar puzzle. Create a word of only vowels of letters chose from only consonants. Is there a word-creating standard that writes out this is possible or not?

With all due respect you are bringing no valuable addition to this conversation. Just your frustrations came to light. I am NOT Burunduk, so no need to be conflictual and escalate the conversation into an uncharted territory.

I am asking if "red" callout it is allowed or not.

 

[3DDave]

It is not disallowed - no callout is disallowed.

Callouts either have a use or not, but there's no requirement they be useful.

 

[greenimi]

It is not disallowed - no callout is disallowed.

I think some of the callouts are disallowed. For example runout modified at MMC or MMB.
Or you are saying that callout has a syntax error?

 

[3DDave]

The word "allowed" does not appear.

Runout is a tolerance used to control the functional
relationship of one or more features to a datum axis
established from a datum feature specified at RMB.

If the datum feature is not establish at RMB, then runout is not used.

The standard does not say what it means if is is not RMB, but runout is used.

It does seem odd that they say "from a datum feature" when the example uses multiple datum features, where the first reference doesn't have an axis associated with it.

"Syntax" does not appear in the standard. They could have generated syntax rules, but chose not to, and instead relied heavily on examples. Of those examples they did not spend much, if any, effort to create ones that were of disallowed uses as there are no disallowed uses. There are some cases they did not provide definitions for.

 

[pmarc]

greenimi,
Like Garland23 replied, this callout does not fall under the concept of self-referencing exactly for the reason you described.

It would be self-referencing if the position FCF was relative to D only. In such a case, regardless of the actual as-produced geometry of the feature, the measured position error would always be 0 (i.e., check of an axis relative to the very same axis, hence to itself). To me this is an indication if a callout is a self-reference or not.

 

[3DDave]

This is just a more complicated way of saying Perpendicularity to C, but with unnecessary steps.

 

[CheckerHater]

How exactly will we get measured position error other than zero?

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

 

[3DDave]

Lack of perpendicularity. [C|D] is an axis guaranteed to be perpendicular to C; the feature used to create that axis isn't. The requirement is the axis of the feature be within some amount of the true position, which includes the inferred perpendicularity.

 

[pmarc]

This is just a more complicated way of saying Perpendicularity to C, but with unnecessary steps.

Mathematically, I am not sure they always need the be the same. I think in some very special as-produced geometry cases they would not have to be. However, since those cases are theoretical, just like greenimi's question, Perpendicularity to C is definitely the first option to choose.

 

[3DDave]

One day someone will build a piece of software to make such simulations.

 

[greenimi]

Yes, I would confirm that my question is theoretical and for academic purpose only.
On the other hand, the equivalent figure from 2018 standard has been updated to show perpendicularity (instead of total runout shown on 2009) and the secondary datum has been removed from the FCF.

 

[SeasonLee]

On the other hand, the equivalent figure from 2018 standard has been updated

I would choose this callout.

Season

 

[3DDave]

Changing from total runout to perpendicularity controls separate types of variation. Total runout in this case controlled both orientation and form variation; the latter is lost using perpendicularity alone.

I'm sure this difference is explained in the attached "why it was changed" document that always accompanies a new revision. Or not.

 

[Burunduk]

I am NOT Burunduk, so no need to be conflictual and escalate the conversation into an uncharted territory.

Interesting reasoning. So, what does NOT being me change, or mean, in the context of your objection?

 

[greenimi]

So, what does NOT being me change, or mean, in the context of your objection?

Burunduk,

I did not want to start or continue an irrelevant conversation. I leave that pleasure between the two of you guys. I am not interested to get between of you two gentlemen.
I don’t want my thread to go into the perpetual and enduring painful and dire disagreements between 3DDave and you which are happening more often here, on this site, than I would like to.
So, I had to curve the perceived tendency right from its roots. Maybe I overreacted. Sorry about that.