Flatness and parallelism dependability

[kolvos]

Hi,

A very simple question. Is the picture attached sensible? (Parallelism of 0.05mm for surfaces that both have 0.05mm flatness tolerance)
Either way - please try to explain why.

http://files.engineering.com/getfile.aspx?folder=c22be41c-e1c1-4e73-bf93-49347073df69&file=Doc1.docx

 

[weavedreamer]

The second flatness callout with the parallel callout is redundant, the flatness would need to be a refinement of the parallel specification.
If the milled surface is parallel to A within 0.05, it is already flat within 0.05 as an independant feature.

 

[pmarc]

Flatness attached to parallelism callout is redundant, regardless of whether the drawing is per ISO or ASME standard.
Parallelism tolerance 0.05 will inherently keep this surface flat within 0.05.

 

[kolvos]

Hi Pmarc/weavedreamer.

Can you explain this logic? If I define parallelism of 0.05 to a surface of flatness 0.05, won't my flatness tolerance be 0.1mm?

 

[JNieman]

Parallelism is just Flatness at a specific orientation. Parallelism doesn't get added to Flatness; Parallelism further defines the feature.

Parallelism is a tolerance zone of two planes .05 apart parallel to the Datum. Flatness is a tolerance zone of two planes .05 apart.

If ASME:

6.2 ORIENTATION CONTROL
An orientation tolerance controls parallel, perpendicular,
and all other angular relationships. Note that
an orientation tolerance, when applied to a plane surface,
controls flatness to the extent of the orientation
tolerance. When the flatness control in the orientation
tolerance is not sufficient, a separate flatness tolerance
should be considered. An orientation tolerance does not
control the location of features. When specifying an orientation
tolerance, consideration must be given to the
control of orientation already established through other
tolerances such as location, runout, and profile controls.
See Fig. 7-8.

_________________________________________
NX8.0, Solidworks 2014, AutoCAD, Enovia V5

 

[JNieman]

To clarify, when I speak of parallelism in my previous post, I speak strictly of parallelism OF A SURFACE. Context was assumed, but I wanted to be clear.

So you can specify a flatness that EXCEEDS the parallelism, if required, but if the parallelism is equal or more-constraining than the required flatness, then the flatness is redundant.

_________________________________________
NX8.0, Solidworks 2014, AutoCAD, Enovia V5

 

[fsincox]

jNieman,
This will be true, only, if you use an orientation tolerance specifying a tangent plane modifier. The actual surface, then, need not conform to the orientation tolerance.
Frank

 

[JNieman]

@fsincox,

I do not understand your statement. Is not parallelism a tolerance of orientation and thus my statement is adequate? I believe specifying a tangent plane modifier would modify the statement I made but does not negate it. A tangent plane modifier would instead look at a plane resting on the peaks of the surface rather than a standard plane measurement, correct? All else would be equal? I'm not very experienced with the tangent plane modifier as I've never used it or had anyone else use it on parts I've come across.

Or maybe I misunderstand which part of my original post you disagree with.

_________________________________________
NX8.0, Solidworks 2014, AutoCAD, Enovia V5

 

[pmarc]

kolvos,
Basically the statement that these two tolerances do not get added plus the bold text in the quote offered by JNieman pretty much clarify everything.

Putting it differently, if entire surface must fall between two parallel planes spaced 0.05 apart that are parallel to datum plane A, there is no physical possibility that the very same surface will have flatness error greater than 0.05, thus flatness tolerance 0.05 or greater does not make sense.

 

[pmarc]

JNieman,
You said: "So you can specify a flatness that EXCEEDS the parallelism, if required, but if the parallelism is equal or more-constraining than the required flatness, then the flatness is redundant."

I believe this is the statement that Frank has objections to. And I agree with him. Nobody can forbid you to specify flatness tolerance greater than or equal to the parallelism tolerance, but there is absolutely no chance it will be ever required, because geometrically it makes no sense.

 

[JNieman]

I think there is confusion in my word choice. I agree with what you said, pmarc. However there is a sense to specifying flatness flatter than the parallelism - I can think of very good reason for it, but I won't get into that.

_________________________________________
NX8.0, Solidworks 2014, AutoCAD, Enovia V5

 

[pmarc]

Of course it functionally makes sense to specify flatness tolerance that is tighter than parallelism tolerance. No question about it - especially that it is clearly stated in the quote you provided.

 

[kolvos]

Actually I will like to get into that - in what occasions would I specify flatness tighter than parallelism? Won't that make parallelism redundant?

 

[CheckerHater]

Be my guest:

http://www.eng-tips.com/viewthread.cfm?qid=238967

http://www.eng-tips.com/viewthread.cfm?qid=332250

http://www.eng-tips.com/viewthread.cfm?qid=217299

http://www.eng-tips.com/viewthread.cfm?qid=314585

http://www.eng-tips.com/viewthread.cfm?qid=312392

Also, the picture that is being around for a while:

http://files.engineering.com/getfil...bf33-484c-ac71-fd99ff92ad94&file=FLATNESS.bmp

 

[kolvos]

how about a summary?

 

[weavedreamer]

Check out the picture link provided by CheckerHater, it sums it up rather succinctly.

The tolerance zone is what the refinement of a parallel zone must reside within, while any additional refinement of flatness must reside within the parallel zone.