Surface location 2

[gabimo]

Relatively new working with GD&T and on this forum.
ASME Y14.5-2009

Does the total runout (as a lonely requirement) shown control the surface location?
Again, the surface LOCATION, not the axis location (coaxiality).

Or the surface location is controlled by the combination between the total runout and size? (not the total runout by itself as a stand-alone callout/ requirement) ?

As we know, runout does not separate form and location deviations, so the amount each of these deviations contribute to the runout deviation will not be known. The verification of a runout tolerance reports the cumulative effects of form and axis offset. But the question is about the surface location.

Y14.5-2009 states:
9.4.2.1 Applied to Surfaces Around an Axis. Where
applied to surfaces, constructed around a datum axis,
total runout may be used to control cumulative variations
such as circularity, straightness, coaxiality, angularity,
taper, and profile of a surface.

 

[powerhound]

I does control location of the surface about a datum axis but it does not control the axial location of a surface perpendicular to the datum axis.

So the total runout in your attachment controls the radial location of the surface but doesn't control where it is axially located.

John Acosta, GDTP Senior Level
Manufacturing Engineering Tech

 

[gabimo]

Thank you John,

My follow up question is: (see revised sketch, attachment)

Can we, NOW, consider that the surface is fully located by the total runout ONLY?
As I stated before, can the total runout, as shown in the attachment, control the SURFACE LOCATION as an independent control? Or the surface location is controlled by the combination between the total runout and size?

 

[powerhound]

If the runout tolerance is smaller than the size tolerance, then the runout tolerance controls the location and form of the feature. If it is larger than the size tolerance, it controls location but not form.

Thus in your drawing, runout controls coaxiality and also the form of the feature. It controls the radial location of the specified surface, not the axial location.

John Acosta, GDTP Senior Level
Manufacturing Engineering Tech

 

[gabimo]

If the runout tolerance is smaller than the size tolerance, then the runout tolerance controls the location and form of the feature. If it is larger than the size tolerance, it controls location but not form.

So, it is safe to make the following statement: regardless of the feature size, if applied to a cylinder, total runout always control the surface radial location ?

 

[mkcski]

So, it is safe to make the following statement: regardless of the feature size, if applied to a cylinder, total runout always control the surface radial location ?

I would be cautious with general statements like this. The design should determine the control applied and the interpretation of the control should meet the design intent. For example: Total Runout is typically applied as refinement of form within size limits with location "tagging" along. If refinement of form is not the design intent, position and/or profile should be considered to control the location.

Certified Sr. GD&T Professional

 

[gabimo]

The design should determine the control applied and the interpretation of the control should meet the design intent

mkcski,
What means: ".....the interpretation of the control should meet the design intent......? "

Why not: The interpretation should meet the standard meaning/ definition?

Also, the question is:
if total runout is the design intent and shown on the drawing, then IS the surface location controlled ? (again, the surface location , not the axis location-coaxiality-)?

 

[axym]

gabimot,

Short answer - total runout does not control the radial location of the surface (assuming that I understand what you mean by "radial location").

One problem here, that Y14.5 perpetuates, is the use of terms such as "location" to describe what is being controlled. As we have found in this thread, the terms "radial location", "axial location", and "surface location" may mean different things to different people.

I have found it very useful to describe constraint in terms of the transformations that are allowed and not allowed. So I would describe a Total Runout tolerance zone in the following way:
-The tolerance zone has perfect form.
-The zone cannot rotate relative to the datums
-The zone cannot translate relative to the datums
-The zone can "progress" (that is, offset away from the nominal surface while keeping the distance between the two boundaries constant)

So the Total Runout tolerance zone in your Surface_location example is two coaxial cylinders 0.02 apart, that are perfectly coaxial to datum axis A. The zone cannot translate or rotate relative to datum axis A. But the zone can progress (the diameter of the two coaxial cylinders is variable - they just have to maintain the radial distance of 0.02). This property is what makes the Total Runout zone not control size. This would be controlled by the 7.780 +.025 -.012 size tolerance.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[greenimi]

One problem here, that Y14.5 perpetuates, is the use of terms such as "location" to describe what is being controlled

Evan,

Look at the attachment. Y14.5 draft
It is stating: “Location of Entire Surface”
I guess the issue is nowhere near to be solved. It is using the word “Location”.
Is the meaning of the total runout going to be changed? Will “finally”/ officially control the “surface location”? I am sure I am missing something, not sure what.
I know your level of expertise and I am sure you can bring this issue to a resolution. What is going on?

Draft states: Where applied to surfaces, constructed around a datum axis, total runout controls cumulative variations such as: circularity, cylindricity, straightness, and location (coaxiality), of a cylindrical surface.

2009 --see above: gabimot post

My notes:

added: cylindricity
removed: angularity, taper and profile of a surface.

 

[3DDave]

An inspector can verify circular runout and total runout without knowing the nominal diameter or diameter tolerance. Therefore one cannot expect that the radius of the surface is controlled. All that is controlled is the radial (for cylinders) variation with respect to a given axis.

Total runout is a control very similar to parallelism and perpendicularity, which also control variation without controlling location.

 

[axym]

greenimi,

Don't worry - I think that this is just a terminology issue.

I don't think that the meaning of total runout is going to be changed. The descriptions are changed a bit, and rightly so, as some of the previous descriptions were misleading. Total runout never controlled the same thing as angularity, and never controlled the same thing as profile of a surface. So I'm glad to see that those terms are being removed from the description.

There is still the issue of what is meant by the "surface location" of a cylindrical surface relative to a datum axis. Some people would say that this means the location of the center - the "coaxiality" aspect, and total runout controls this. Others would say that surface location relates to the distance from the axis to the surface - the "radial location" aspect, and total runout does not control that. Others might say that surface location relates to where the feature is along the datum axis - the "axial location" aspect, and total runout does not control that either. These are all consequences of the FIM (Full Indicator Movement) definition, where there is no absolute setting for the indicator, only the range of readings matters.

3DDave,

I agree with some of what you said. The absolute radius of a cylindrical surface is not controlled - the radial variation is controlled. The total runout can be verified without knowing the nominal diameter or diameter tolerance - these are independent characteristics.

I wouldn't say that total runout is similar to parallelism or perpendicularity though. There is one special case where total runout would give the same control as perpendicularity - a single planar surface that is nominally perpendicular to the datum axis. But other cases would be different. Total runout allows the tolerance zone to progress/offset, and orientation tolerances allow the tolerance zone to translate. This is not the same thing.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[powerhound]

Wow! This has gone about as far off into the theoretical weeds as I've ever seen on this forum. This is like arguing that a light switch doesn't turn off a light, stopping the flow of electrons through the circuit does. If runout doesn't locate the surface then what does? Size? I don't think so. What else is there? Form? Nope, not that either. I get the whole radius versus coaxiality thing but do we really think splitting hairs like that is giving Gabimot useful information?

John Acosta, GDTP Senior Level
Manufacturing Engineering Tech

 

[3DDave]

Evan,

You got me. It's totally different in that when one applies parallelism it requires some other dimensional control to define the limits to the distance to the controlled surface from the given datum reference. Oh. Wait. That's exactly the same description as total runout. So parallelism is to flat things as total runout is to cylindrical things. Which is why I wrote 'similar' and not 'identical,' and further there are cases for perpendicularity that cannot be covered by total runout.

Parallelism vs Total Runout is just a transformation from rectangular to cylindrical coordinate system.

 

[CheckerHater]

@3DDave: If you like analogies, you can also think of Total Runout as Cylindricity with datum.

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

 

[greenimi]

John, Evan, Dave,

I personally like this kind of theoretical discussions. Will force you to “read the letter of the law” (in the actual standard) and I, as individual, learn a lot from those debates.
( Note: Was another debate where we pushed the envelope and has been concluded that the issue was the “definition” (or lack thereof) of the actual local size.

http://www.eng-tips.com/viewthread.cfm?qid=391408

)

Going back to the original thread:

Looks like here the issue is the FIM definition.

Some people would say that this means the location of the center - the "coaxiality" aspect, and total runout controls this. Others would say that surface location relates to the distance from the axis to the surface - the "radial location" aspect, and total runout does not control that. Others might say that surface location relates to where the feature is along the datum axis - the "axial location" aspect, and total runout does not control that either.

The question is why even in the new draft the committee does not act decisively and put an end to this controversy and just muddy the water even more by having “location of entire surface”?
Or maybe the interpretation will be that the total runout will control the “location of entire surface”? Is that the intent?

 

[axym]

powerhound,

If you think that this is discussion far off into the theoretical weeds, you should come to more of our Y14.5.1 meetings ;^).

I defend my hair-splitting in this case. Attention to nit-picking detail is what has made me one of the more successful and wealthy GD&T ... OK, never mind.

3DDave,

You've pointed out an interesting connection that I hadn't thought of before - that parallelism can let distance float in a Cartesian way and total runout can let distance float in a cylindrical way. But I would maintain that total runout and parallelism have more differences than similarities - it isn't just a matter of transforming to a different type of coordinate system.

greenimi,

I agree that statements such as "location of entire surface" muddy the waters. So we need to look past those statements and find the more geometric/mathematical descriptions of the tolerance zone. The new draft also states the following:

"All surface elements shall be within a tolerance zone consisting of two coaxial cylinders with a radial separation equal to the tolerance value specified. The tolerance zone is constrained in translation (coaxial) to the datum axis."

This definition nails it down - it describes well-defined pieces of geometry that behave in a certain way. So we can end the controversy over what "location of entire surface" means. It must mean whatever makes it agree with the geometric/mathematical definition.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[3DDave]

Hi CH!

I have considered that. The proliferation of specialized symbols for essentially identical cases is of some interest. The ones covered here are various forms of profile tolerance; a zone of defined thickness with or without a defined offset relative to a datum reference.

In the case you mention, the transformation is between flatness of a surface and cylindricity, made by changing the nominal surface curvature from zero to a non-zero finite value. Obviously the committee approached this when they 'unrolled' straightness of a center line to become flatness of a center plane. I don't see a rigorous treatment indicating they looked at it this way, but the effect is there.

 

[powerhound]

The tolerance zone is constrained in translation (coaxial) to the datum axis."

Does this not mean location?

John Acosta, GDTP Senior Level
Manufacturing Engineering Tech

 

[mkcski]

All:

Given my level of understanding - more application and interpretation than theory - these "in the weeds" threads are an important learning tool. Please keep it at up!

The tolerance zone is constrained in translation (coaxial) to the datum axis."

As understand it, coaxiality controls 4 DOF - 2 translation and 2 rotation, so wouldn't it be better add "and rotation" to the sentence?

Certified Sr. GD&T Professional

 

[axym]

powerhound,

The answer is yes and no. Again, this is the problem with terms like "location". The tolerance zone is constrained in translation (coaxial) to the datum axis. This means location, but in one sense and not in others. It doesn't mean location in the "radial" sense that some have described. So we can't make a general statement that total runout controls location, or some people may interpret this as meaning that total runout controls the radial location (and thus the size) of the feature.

Personally, I'm not a fan of the "radial location" description. It leads to some odd conclusions, that make things more confusing. One could say that a size tolerance controls the radial location of a cylindrical surface relative to some axis. But I don't think that we want to describe a size tolerance as controlling location. To me, the term "location" in Y14.5 is tied to the idea of translation. If a tolerance zone is allowed to freely translate (as size, form, and orientation zones are), then it doesn't control location.

mkcski,

Yes, I agree that it would be better to add "and rotation" to the sentence. As you say, 2 rotational DOF's are constrained.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca