Controlled Spherical Radius (ASME Y14.5 2009, 2018)

[AndrewTT]

Does anyone have a reason why a controlled spherical radius could not be invoked as an extension of principles (extension of the principle of a controlled radius)?

Thank you.

 

[3DDave]

Might as well. Of course at enough magnification any surface is subject to reversals, so Controlled Radius is something of a fiction that everyone can simply say is good enough. Crank up an electron microscope and a planar crystalline surface looks like heavy moguls. Above that are scratches and other surface textures, which the standard does not exclude from consideration.

 

[chez311]

Andrew,

Without sufficient additional documentation, I agree with 3DDave that CR as written in the standard is not realistic. The reasons why the committee decided to use such ambiguous and imaginary language such as "fair curve without reversals" has always mystified me. Without additional definitions for these terms one can imagine applying any number of interpretations - as noted above at a high enough magnitude no real surface will be perfectly "smooth" as the language seems to suggest. At least the standard does suggest that their definition is incomplete when it says "It is recommended that the CR be further defined with an engineering control specification."

I would say most cases would benefit from a profile tolerance if the radius is that critical. If this is not sufficient, one must either use an existing addendum or create one to make CR inspectable and in this case you could flesh out any additional requirements or clarification for spherical CR.

 

[Kedu]

Chez311,

We just talk about a solution to fix the CR ----Scott Neumann----

https://geotol.com/core-programs/2018-standards/

"Dynamic Profile

The Dynamic Profile Tolerance modifier is the shiny new symbol in 2018. Profile has always been a powerful symbol, able to control size, form, orientation, and location based on datum references. By default, on a non-planar feature, profile will always control size and form of the surface. When this modifier is added, the profile controls the form of the feature without size. The offset between the profile boundaries remain fixed but the size of the boundaries is now variable. Profile with no datum references and the dynamic modifier would be the equivalent of cylindricity on a hole but for more complex shapes (think trianglicity and hexagonicity). Datum references may also be used with this concept to allow profile to control form, orientation, location without the size. This is usually used as a refinement of a regular profile specification. On another note, this modifier could be the solution for fixing the definition of CR."

 

[chez311]

Kedu,

I considered recommending dynamic profile, I think its a viable solution but it doesn't do *exactly* what someone might desire out of CR. Putting aside the ambiguity of the as written definition, the tolerance zones are not identical and CR has the additional tangent requirement.

 

[SeasonLee]

Author's comment regarding CR from Alex's Fundamentals of GD&T 2018 SAE.
Edit[highlight #204A87][/highlight]: So I don't think there is a stress concentration problem on a spherical ball.

Season

 

[AndrewTT]

Thanks for the input everyone.

I agree that the CR and proposed CSR have their own issues. I am putting together a list of options for defining a critical feature. CSR is going to be one of them ...... followed immediately by "do no recommend", lol.

feature: hemispherical concave feature (less than half of a sphere, so I'm not considering it a regular FOS)
Goal: dis-allow a convex "bump" at the bottom of the SR indentation. This bump is currently being considered something other than a surface texture issue.

Surface texture requirements will be considered after all options of geometric techniques have been identified.

My current list:
CSR
circularity
Profile (surface and line)
Composite/Multiple Single Segment Profile (surface and line)
Dynamic Profile
Non-Uniform Profile
Note/ENGS Spec (this is basically identical to CRS)
addition of a blind hole where the "bump" can be present

 

[Dean Watts]

I agree with Scott. The dynamic profile modifier is the well-defined tolerance to replace CR in all cases. While surface texture specs are important, as far as "Flats and reversals" go, once we get to the more microscopic realm we have no means of specifying which way any ups, downs, or discontinuities go, so I agree with 3DDave on this too.

I would never recommend CR for any purpose. It should be deleted from Y14.5. Profile with the dynamic modifier is a very good method though.

Dean

http://www.validate-3d.com

 

[Dean Watts]

I should have said that profile with a dynamic modifier is the well-defined replacement to address the "flats and reversals" portion of CR's intent. To control the radius either profile of a surface or a directly toleranced radius would need to be applied along with the profile-with-dynamic-modifier refinement.

Dean

http://www.validate-3d.com

 

[Kedu]

Dean,

Let me ask: do you think that a directly toleranced radius (±) COULD to be applied along with the profile-with-dynamic-modifier refinement? Or if dynamic profile modifier is used than the radius shall be basic? I would say the latter, but I am open to counter arguments. Just curious what do you think?

 

[Dean Watts]

Hi Kedu,

I am not aware of any reason why a basic dimension, added to define true profile for the profile of a surface with dynamic modifier, cannot exist with a directly toleranced radius dimension.

My preference would be to use profile instead of a directly toleranced radius, but there is no prohibition of a basic dimension to be applied for one particular reason along with a directly toleranced dimension for some other reason. I also don't see a need to prohibit this.

Dean

http://www.validate-3d.com

 

[Kedu]

I am not aware of any reason why a basic dimension, added to define true profile for the profile of a surface with dynamic modifier, cannot exist with a directly toleranced radius dimension.

My preference would be to use profile instead of a directly toleranced radius, but there is no prohibition of a basic dimension to be applied for one particular reason along with a directly toleranced dimension for some other reason. I also don't see a need to prohibit this.

Interesting.
I am here to learn, but I've never seen basic radius and direct toleranced radius (±) applied to the very same feature. Not even sure it is a right thing to do so.

Could you, please, expand a little more, maybe even with some examples/ figures from Y14 set of standards (Y14.5, Y14.8, Y14.41, Y14.45, etc)

 

[Dean Watts]

Hi Kedu,

I am here to learn too, and to help also if I can. In this case, I will have to ask if you have ever found a rule that prohibits a basic dimension for one purpose and a directly toleranced dimension for another purpose on a particular feature. I don't know of any existing rule or any reason for such a rule.

There is no need to find an example in a Y14 standard that shows this. The only need is to determine if a written rule prohibits this, or maybe whether there is a practical reason that makes this a bad idea. I don't think there is either one.

I would generally prefer that this practice was not followed, but I would still argue against a rule that prohibits it, unless someone points out a reason that I am not thinking of that creates a problem with this practice.

A similar situation is when a profile applies to two features and a directly toleranced dimension is also applied to address the distance between the two features to serve as a process control dimension. The process control dimension is not related to the function of the part and it should usually be easily measured on the shop floor. In that case a basic dimension to define true profile and also a directly toleranced dimension may apply to the same feature. No harm no foul, as far as I know.

Dean

http://www.validate-3d.com

 

[Kedu]

Dean,

...if you have ever found a rule that prohibits a basic dimension for one purpose and a directly toleranced dimension for another purpose on a particular feature

That is clear out of the box thinking. Not sure if that is the committee's intent however.
My example (regarding dynamic profile modifier):
On Y14.5-20xx draft (public draft before 2018 release) fig. 11-19 had dynamic profile modifier and Ø30 was basic. Then, during one of the technical changes the commitee decided to remove "delta" symbol (from 11-19) AND they changed Ø30 basic to Ø30 ± (plus-minus). This is the important part: they changed the basic to ± (for Ø30)
For me that is a clear intent that "delta" shall be used only with a basic dimension.

Were you part of that discussion?


Another example, just to underline the importance of clear rules: are there any rules that prohibits application of flatness to a cylindrical surface? If no, would you apply such?

 

[Dean Watts]

Hi Kedu,

Your last sentence is kind of silly, isn't it? I get your point, but you need a better analogy in my opinion. Yes, there is an unwritten rule against applying flatness to a cylindrical surface. It is obvious why such a rule would be written if someone attempted to assert that this should be a practice. There is no reason that I know if why a rule should prohibit application of a basic dimension and a directly toleranced dimension to one feature. Again, I wouldn't choose that option, but we don't need to prohibit a practice just because you don't like it.

I'm only one member, but I am the member that presented a proposal at the Nashville meeting in Oct 2011 that led to the addition of the dynamic profile modifier in the 2018 standard. I bring this up only as part of my assurance to you that I understand its application and limitations. In that presentation I said that the name and symbol can be whatever the committee comes up with, but I called it the "Expandable Zone" <EZ> modifier just for that presentation. It was later named the Dynamic Profile Modifier, which is better since it doesn't imply expansion only.

What I am saying is not out of the box thinking in my opinion. To assert that a particular practice cannot be done should be justified by an existing rule or some specific reason why that practice is a bad idea. Just because you don't like the the looks of it isn't going to convince me that the practice should be prohibited.

Dean

http://www.validate-3d.com

 

[3DDave]

I'd argue total runout applied to a nominally flat surface with an infinite radius applied to it should not cause any objection. I think controlling a nominally flat surface by making it parallel to itself should also qualify.

It does get tough to deal with putting an MMC position tolerance on the depth of a counterbore, but I've run across those who say the depth has a size and is therefore a feature of size. It gets worse from there.

My favorite was a guy who wanted to use datum features on a casting that were machined off during manufacture to center up the machined feature in the original over-size cast part.