Holding multiple surfaces parallel within 1 common tolerance zone 2

[SpaciouS]

How do you call-out Parallelism over multiple surfaces that are controlled within 1 common tolerance zone.

Can you simply call-out "x surfaces" under a Parallelism FCF and it's understood that all indicating surfaces are controlled simultaneously (similar to Flatness of multiple surfaces defined by a Profile FCF with such "x surface" notation?

Or would that indicate separate tolerance zones per individual surface?

Or, since I have never seen that example in any ASME standards, is that notation not appropriate when using Parallelism?

Is so, can I simply use Profile instead of Parallelism while referencing a datum and adding "x surfaces" beneath the FCF?

Or is it something else?

Thanks,
Sean

 

[Belanger]

Either option would work, but with a slight difference between them. Profile can be used but without a basic dimension for the distance to the datum. This would require both surfaces to stay coplanar to one another, and parallel to the datum, but the distance to the datum would float within a larger location tolerance.

The other option of using parallelism (with "2 SURFACES" notation) would require each surface to be parallel to the datum on its own terms, but not require the two to be coplanar to one another.

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

http://www.gdtseminars.com

 

[fsincox]

ISO or ASME?
Coplanarity (Profile) is the common way in ASME. Now after 2009, ASME offers another option note: " CONTINUOUS FEATURE" "CF" in a hex symbol, and it would cover you.
Frank

 

[SpaciouS]

Belanger,

That was my gut instinct but I'm a little confused about the following.

Reading from my reference chart, "basic dimensioning must be used define the true profile", does this not apply to what you're saying?


Thanks,
Sean

 

[fsincox]

Yes, the basic dimension is "ZERO". The standard provides for this.
Frank

 

[SpaciouS]

Frank,

I'm not sure what yes means. I know what a basic dimensions represent but I'm still confused as to why a basic dimension can't be used in the case mentioned earlier.

Thanks,
Sean

 

[fsincox]

"basic dimensioning must be used define the true profile", does this not apply to what you're saying?

Yes, it is applied automatically by reference to the ASME standard.
ASME Y14.5M-2009, section 1.4 (k), page 8.
Frank

 

[DeanD3W]

While some assert that leaving a basic dimension off will affect what profile will control, there is nothing explicit in the standard to back that up... In fact if that were true it would not be a good trait of the standard... Offsets are often zero, so the basic dimension would not be shown. We are also moving towards annotated 3D CAD models and querying to find basic dimension values, so basic dimensions would not be shown at all.

The good options to gain the coplanarity and parallelism are to use parallelism with a Continuous Feature modifier, or profile of a surface with a customized datum reference frame that deletes the translation constraint that would locate the features. Either of these tolerances would be placed below a profile of a surface with a larger tolerance zone that controls the location of the surfaces.

I hope this helps.

Dean

http://www.d3w-engineering.com

 

[SpaciouS]

Frank,

Thanks, I know what you are referring to.

But in the case where I do 'not' want to hold the location of the 2 surfaces as tight as the co-planer relationship between each other and the referenced datum, it seems what Belanger has mentioned, would be the only other alternative. This is what I trying to clarify.

Thanks,
Sean

 

[Belanger]

Sean, it's true that basic dims must be used to define the true profile. That means the actual shape of the surface in question. And flat surfaces have a true profile with a basic dimension of zero (no curvature) so the basic dim is implied.

But the relationship of the true profile to the datum doesn't have to be basic. So that distance is different than the "true profile" itself.

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

http://www.gdtseminars.com

 

[SpaciouS]

Okay got it. Just needed that clarification.

Thank you very much.

Thanks,
Sean

 

[DeanD3W]

Sean,
Leaving the basic dimension between the feature and the datum you are referencing off does not make profile of a surface control only parallelism and coplanarity... It would just mean that a basic dimension is then missing (awol) from the drawing. If you reference a datum feature that is capable of constraining the profile tolerance's location, then the profile tolerance controls the considered feature's location.

I do expect this to create controversy since some trainers in industry have included in their books and training materials, and are to this day telling people that the presence or absence of a basic dimension will affect the control provided by profile... This is not backed up by the standard and as I said in my post above, if this were stated in the standard then this would cause significant issues. I don't think anyone will be able to point out where Y14.5 says that the presence or absence of a basic dimension affects the control provided by profile of a surface. If they can then I will, of course, stand corrected.

Figure 8-27 and section 8.8 of Y14.5-2009 are illustrating and explaining a profile of a line application that I believe can be used well enough to explain one issue. What if this case were for an annotated 3D CAD model with all basic dimension values to be queried from the CAD model? Wouldn't the basic dimension then be "present"? If so, then the profile of a line tolerance zones are located and will stack up to provide the same control as profile of a surface rather than the desired control? Am I the only one seeing a problem will everything shown in figure 8-27..? The control provided should have been achieved in with parallelism of line elements on the upper surface. Even the depiction of the size tolerance limits is not correct, since it incorrectly implies perfect form at LMC.

Either way, nothing anywhere in Y14.5 says that omitting a basic dimension will affect the control provided by profile of a surface. The fact that a profile of a surface tolerance zone is orientation constrainable and location constrainable, combined with the constraint capability of the referenced datum features will determine what profile of a surface will control. That control can be modified by using a customized datum reference frame per Y14.5-2009 sections 4.22 and 4.23, but not by leaving a basic dimension off of the drawing.

This issue is the greatest misnomer regarding GD&T in my opinion. I am very hopeful that the next version of Y14.5 will leave no question regarding this issue. The control provided by a given profile of a surface tolerance dictates the basic dimensions that will be needed, not the other way around.

Regarding the desired control described in the original post, I'm re-stating.. The good options to gain the coplanarity and parallelism are to use parallelism with a Continuous Feature modifier, or profile of a surface with a customized datum reference frame that deletes the translation constraint that would locate the features. Either of these tolerances would usually be placed below a profile of a surface with a larger tolerance zone that controls the location of the surfaces unless something less common, like a size tolerance provided the control of the feature's location.

Dean

http://www.d3w-engineering.com

 

[fsincox]

Dean,
If I understand what you said, above, that is real news to me. I am curious what the last statement in Y14.5M-2009, section 8.2, page 158, means to you? "Where used as a refinement of a size tolerance, created by toleranced dimensions".
Frank

 

[DeanD3W]

Hi Frank,
I see several problems with that statement... The profile is not refining "size", so I think that is misstated... In Fig. 8-27 profile of a line is being used to refine the orientation and form of line elements on the top surface... Why would profile of a line be used to do this? Especially now that we have the CF modifier to help with those cases alignment such as coplanarity is desired, and customized datum reference frames to better handle some other more involved part geometry, we don't need profile of a line to be used in this role. What if that top surface happened to be aligned with the origin of the datum reference frame? If so, the basic dimension would be zero, so it wouldn't be shown. Is it then "present" or "not present"?

I think that last sentence of 8.2 describes an odd, unneeded, and problematic application of profile. Since there are problems associated with the notion that the presence or absence of a basic dimension affects the control provided by profile, I think Fig. 8-27, Figs. 8-17, and 8-18 sections 8.2, 8.8, and maybe some other items need some rework. Without explicitly stating it, they tend to imply that the presence of absence of a basic dimension affects profile. Due to the problems associated with that practice, I don't ever use profile that way and my opinion is that we should firmly shut the door on that practice. There is some implied justification of the practice, but we have other tools that accomplish what is needed without the zero basic and queried CAD body problems.

When some of these things are cleaned up a bit, maybe this will all be not as much fun..? .

Dean

http://www.d3w-engineering.com

 

[fsincox]

Dean,
Please don't mistake me; I am not a fan of toleranced dimensions, by any stretch. We are still actually using 1982 at work, I don't have "CF" and the other options, even composite profile is a new concept for most around me. I really think we are in danger of making the standard more irrelevant to the real world, particularly, if we can't trust what is shown and written in it.
Frank

 

[Belanger]

Dean, I'm surprised to hear you say this (I presume your comments were in response to my explanation).

First, just because the standard doesn't explicitly state or show the precise example being analyze, that doesn't mean that we can't venture into that area. Recall that GD&T is a language, and we can use the building blocks of the language to create a control for any conceivable design, as long as we don't violate an obvious rule of the language.

Second, why it is so problematic to use profile without a basic distance to the datum? Fig. 8-27 obviously does such, so why attempt to explain it away? There is a valid point in mentioning the CAD model, which would obviously be a basic dimension. If we wish to unlock the location aspect of profile, however, then we simply display a toleranced dimension on the height. Problem solved! You stated that parallelism should have been used, but what if the top is a curved surface by design? The only option would of course be a profile control.

Third, we have the descriptions given in paragraph 8.2. You dispute that profile is a refinement of "a size tolerance created by toleranced dimensions," but realize that it does refine the orientation portion of the size tolerance. A refinement does not mean that every aspect of the original control must be trumped -- consider how a FRTZF is a refinement of a PLTZF.

Here's the best way to think of it: Profile is, at heart, a form control. If we take Fig 5-7 and simply use the profile of a surface symbol instead of flatness, it would have identical meaning.
But, if desired, profile can be elevated to the next level of the GD&T hierarchy, which would be orientation -- if we take Fig 6-2 and simply use the profile of a surface symbol instead of parallelism, it would have identical meaning. (Apparently, this is where you disagree?)

Finally, if desired, profile can be elevated to control location. That is the only way to control location of a surface, but I'm saying that we shouldn't insist that profile is incapable of doing those intermediary levels of control (orientation), if that is what the function requires.

This has been discussed before on the forum, but it boils down to this: is profile best described as a lower-level control that can be promoted to higher things? Or should it be described as a higher-order control that can be lowered by omitting things? Using logic, wouldn't the first idea be the only way? I can tell you that I live in the US (a low-level piece of information), and that means that I could be in Texas, Oregon, or Illinois; you don't know. But if I say I live in Chicago (the highest-order piece of information), you immediately know that I live in Illinois and that I live in the US. The only common information between them is the country. So the information I give is, at its heart, the country. If I so choose, I may elevate the information to give you more detail.



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

http://www.gdtseminars.com

 

[Belanger]

Sorry -- one more tidbit to my mini-rant

I have a copy of a GD&T Pocket Guide created by a very well-known GD&T instructor, and it drives me nuts that the symbols table on the front cover shows profile as a "location" control, alongside position.
As I tried to show, profile may control location, but it is incorrect to present profile as location control by its very nature. (I don't think anyone in this thread has claimed that, but I'm just illustrating the "middle ground" capability of profile, lying between form and location.)

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

http://www.gdtseminars.com

 

[fsincox]

Dean,
I am glad that you mentioned directional datum controls as an option; I had not actually thought of that approach, it does seem an overly complicated solution. The use of “CF” (if allowed on non-FOS, which seems accepted here but is not really clear in the text of the 2009 standard) is the easy solution. Even using composite profile would be radical in my world. In the world I am from, people do not see/understand the problems with “implied tolerance dimensions” that people here, and the standard committee, are very concerned about.
Frank

 

[DeanD3W]

John-Paul,
Leaving a basic dimension off a drawing to change what profile will control is a method that may be implied in Y14.5, but it has never been explicitly and clearly stated. This is a method that can be applied in specific, simple cases but that does not make it a robust method.

Please see the attached file.

There are better ways to specify the type of control the OP (Sean) is after. We don't need to implement a problematic method to provide the desired control.

Dean

http://www.d3w-engineering.com

 

[DeanD3W]

Attached file wouldn't open correctly when I tested the link... Here's attempt #2.