Usage of Datums in Profile of a Line. 1

[Madhu454]

Hi All,

Few months back I saw a tip on profile of a line with datum referred. The drawing uses profile of a line with datum A to control the line elements of the surface , also the line elements are located to the datum A with basic dimension. Since each line elements are located by basic dimension to datum A, the entire surface of the part is controlled. It is as good as using profile of a surface instead of profile of a line. My opinion is to use profile of a surface itself instead of using profile of a line and confusing the people.

From the above explanation we can come to a conclusion that, using the datum’s in Profile of a line to locate the line elements can be avoided. Does anyone have different opinion?
Now using the datum’s in Profile of a line to control the orientation of the line elements wrt datum’s, In such a case instead of using the profile of a line, we can use any of the required orientation controls say parallelism, and use the text EACH ELEMENT beneath the FCF. This method will be straight forward instead of using profile of a line with datums to control the orientation.

I heard that using the datum’s with profile of a line is very rare, is that true? Also I would like to know any such practical example where we must go for profile of a line with datums?

I would like to know more about the usage of datum’s with profile of a line ? Can anyone help me on this.

Madhu

 

[Belanger]

Not buying it, Evan
A 1D line is just that -- it goes in one dimension only (hopefully we agree on X, Y, and Z as the linear dimensions of space). So if a line traversing along only the X axis suddenly curves, that means it has deviated into Y or Z (or a little of both). It's now 2D. But it still has no thickness.
I'll pose the same question as before: Given the fancy definitions that you want to impose, what is an example of somthing that's 3D?

I'm not saying the topological stuff is wrong. But certain disciplines have certain rigors to their terminology. And you're trying to blend terminology that might have a place in high-falutin' math into a discipline that deals with real dimensions and tolerances.

It reminds me of Microsoft Word's spell-checker that keeps telling me that the plural of datum isn't datums. If you're a purist when it comes to grammar/spelling, then yes the plural is "data." But we know that in the world of GD&T the plural is "datums."

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

 

[powerhound]

I agree that 2D is not necessarily flat because surfaces, in the context of drafting and IGES files, are 2 dimensional. They can follow various complex curvatures but they have no thickness. For the purposes of defining a tolerance zone, I can't see another way to interpret profile of a line other than flat, 2 dimensional, with the cross section oriented with respect to the datum reference frame.

Powerhound, GDTP T-0419
Engineering Technician
Inventor 2010
Mastercam X5
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[Belanger]

All right... I'm beginning to see why there's a discrepancy.

I agree that a surface itself is not a 3-dimensional thing. It has no thickness. But if that surface wavers at all, it occupies space in a 3-dimensional manner. So both perspectives are correct.

The confusion is that in the world of dimensioning and tolerancing, we are not talking about what the surface is (the ontology of the surface, to borrow a term from philosophy), but we are talking about where the surface lies and what space it takes up.

With this clarification in place, we can say that the standard is still correct to say that a profile of a line tolerance zone is 2-D.

Did I salvage myself?


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

 

[CheckerHater]

axym,

First, Wikipedia is rejected as reliable source even in High School - references to Wiki are not accepted.

Second, if you are so fond of Wiki, search for "2-dimensional" instead. You may be surprised.

Third, topology studies objects that have no shape whatsoever, it is called the study of "connectivity and continuity" - the only thing certain about objects is how they relate to each other. Topology is the area of mathematics that has nothing to do with Euclidean geometry, it also has things like "zero-dimensional space" - very useful in GD&T.
Why not refer to fractals – they have 1.5-dimensional objects as well?

Shortly, "2-dimensional" - I don't think it means what you think it means.

 

[pmarc]

Guys, no offence, but I think you have just prooved your mastery of the art of splitting hairs. The funny thing is that in my opinion all of you are right - you are just trying to convince each other by looking at the issue from different perspectives and using arguments from different disciplines of science.

Whatever the consensus will be, this will not help in finding an answer to a question: "How should cross-sections for profile of a line tolerance measurements be oriented if there is no datum reference in profile FCF or datum(s) referenced in FCF do(es) not define proper orientation of the measurement plane?".

 

[axym]

CH,

I thought you might discredit the Wiki reference - fair enough. It's not that I'm that fond of Wiki, that's just what I happened to find first when I went hunting for things that agreed with my opinion ;^). I had a look at the Wiki entry for "2-dimensional" and it focuses on 2-dimensional planar space and the geometric figures that exist in it - nothing earth-shattering there. 2D space is indeed flat, but that doesn't mean that all 2D surfaces are flat.

I'm not sure what to say about the references to topology - I'm no expert, but I don't think we should dismiss the content of an article that mentions topology just because topology has some arcane concepts that don't apply to GD&T.

Here's a reference from the Y14 world. The "dimensionality" of a surface is discussed in Y14.5.1M-1994 in the context of size tolerances, in which a ball is swept along a "spine". Here's a quote from Section 2.3.1 on page 7:

"A 0-dimensional spine is a point, and applies to spherical features. A 1-dimensional spine is a simple (non self-intersecting) curve in space, and applies to cylindrical features. A 2-dimensional spine is a simple (non self-intersecting) surface, and applies to parallel-plane features. These three types of spines can be more rigorously defined, respectively, as connected regular (in the relative topology) subsets of d-manifolds, for d = 0, 1, and 2."

pmarc,

I realize that we're splitting hairs here. Do I do anything else on this forum? ;^). I suppose that I felt that splitting hairs was more justified than usual in this particular debate. The exact meaning of the term "2-dimensional" determines whether or not a Profile of a Line tolerance zone is restricted to be flat or not. If Y14.5 is going to use a mathematical term in an ambiguous way, with no other supporting text or figures to clarify the intended meaning, then we have to bring in the proper meaning of the term from mathematics.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[DeanD3W]

I think "2D or not 2D..." is just semantics, but still an issue. The standard should be improved to prevent this issue.

The fact is that the tolerance zones MUST be normal to the true profile (the theoretically exact surface, that is) & therefore will tilt out of the cross-sectional plane for any feature which is tapered from one cross-section to the next, otherwise the result will be erroneous. Leading me to...

...CheckerHater - The issue with any tapered feature is shown well enough, I hope, in the file I've attached. If you knew me you would know that any argument I would make about how GD&T should be will have absolutely nothing to do with any type of measurement machine I may be involved in the sales or support of as part of my business, but you don't know me, so maybe that was a fair question to ask .

Dean

http://www.d3w-engineering.com

 

[powerhound]

Man, it's no wonder this thread has taken on such a life of its own. Dean, why would you try to control that profile like that? It just doesn't make sense to me. Why wouldn't you put the profile callout in the left view? I see now why you think the tolerance zone should be normal to the surface but a datum reference frame is made up of mutually perpendicular planes, thus the tolerance zone in this case would be perpendicular to A, parallel to B. The problem your are having with this seems to be based on a really bad way to control a profile. Not illegal, but bad nonetheless.

Powerhound, GDTP T-0419
Engineering Technician
Inventor 2010
Mastercam X5
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[powerhound]

All you're really controlling here is straightness. The straightness tolerance zone would be normal to the surface. I realize that you are only providing an example to illustrate your point. Even though you're showing the implications of maintaining a tolerance zone parallel to datum B, that's the way it is. You can't ignore the DRF just because you don't like what you see. That's why this is a bad way to control this profile. You have to apply the profile callout in the view of the profile being controlled, in this case it's a single straight line. That's also why the J<->K is pretty much meaningless.

Please consider my posts with all due respect. I'm not trying to trash your work. I'm simply disagreeing and telling you why.

Powerhound, GDTP T-0419
Engineering Technician
Inventor 2010
Mastercam X5
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[Belanger]

OK, now my head is starting to hurt! I'm thinking that Dean is on to something, though.

Powerhound, you said that "a datum reference frame is made up of mutually perpendicular planes, thus the tolerance zone in this case would be perpendicular to A, parallel to B." The first half of that statement is true, but not the second. Simply having datum references doesn't mean that the tolerance zones will always be perpendicular to those datums. Plus, the statement in the standard that the tolerance zone must be normal to the true profile nullifies the idea that it will always be normal to the datums.

Overall, the standard could use some work when it comes to profile of a line. That is evident simply by our discussion on these finer points.

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

http://www.gdtseminars.com

 

[DeanD3W]

Powerhound - I would never apply profile the way it's shown in the file I attached... In Y14.5M-1994 Fig 6-18 the cross-sections are cut "parallel to the page" with respect to the view in which the profile of a line is applied... This is the only reason I applied the profile of a line in the right view as opposed to the left... To show the issue with tolerance zones that lie in the plane of the cross-sections this seemed like the best approach. If anyone has a better way to illustrate the problem then I'm all ears & eyes.

I don't expect to ever suggest to anyone that profile of a line is their best approach to controlling anything. This is partly due to the fact that Y14.5 does not define it well enough and partly due to the fact that there is rarely a functional need for a "line control".

The example I attached was purely to address CheckerHater's question about what the big deal was about taper.

Now, on to whether the planes of a datum reference frame should be used to somehow determine the orientation of the cross-sections for profile of a line... Please tell me what the cross-section orientation would be for the two options shown in the attached file. My point is going to be that additional specification is needed in order to make the cross-section orientation explicit and clear, just as it is needed to orient the tolerance zones for line element straightness applied to any non-cylindrical part. Now that we can show X, Y, & Z axes to represent a datum reference frame (per Y14.5-2009), that won't be too difficult, but for now, I think we're missing a necessary element for these specifications.

Dean

http://www.d3w-engineering.com

 

[DeanD3W]

Powerhound - In response to your post about 40 minutes ago... We could show the same issue with more complex features but if the super-simple, but admittedly not realistic, example illustrates the point well enough then that's all we need.

The only purpose of that example is to illustrate why tolerance zones for tapered features cannot lie in the plane of cross-sections when those cross-sections are not normal to the surface. I think it does that in a clear enough manner, but as I said above, I'm all ears and eyes if someone has a better way of illustrating the issue.

Dean

 

[powerhound]

Dean,
About your first example: I do get your point but I still see no justification in changing the orientation of the tolerance zone. By your own admission, this is not realistic and the reason it's not realistic is because it's not right to do it this way. It's kind of like saying there's a problem with perpendicularity because you can't use it to control features that are parallel. There's a right way to control parallel features. Creating an example that shows how perpendicularity can't be used to control parallel features doesn't mean I have an argument.
About your newest example. I don't see an issue with example #1. The zone is perpendicular to A and shaped and located from B by missing dimensions since the drawing is incomplete. Example #2 is the same as before. It creates a problem where none exists. There's a right and a wrong way to control that profile and #2 is the wrong way.

J-P,
A true profile and a surface are two different things. Being normal to the true profile and normal to a surface is not the same thing. My understanding has always been that the tolerance zone is oriented per the DRF (parallel and perpendicular). The flat, cross section generates a 2 dimensional contour. The tolerance zone is then disposed about that contour WRT to the DRF. When you said "Simply having datum references doesn't mean that the tolerance zones will always be perpendicular to those datums." I didn't mean to imply that every aspect of the zone would be perpendicular to every datum. In the second example that Dean provided, the tolerance zone will curve per the basic dimensions in example #1. In example #2 since the true profile is a rectangle, the tolerance zone is rectangular and thus perpendicular to A, parallel to B.

Powerhound, GDTP T-0419
Engineering Technician
Inventor 2010
Mastercam X5
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[Belanger]

PH -- I must have misunderstood your post; sorry. Sure, the true profile and the actual profile are not the same, and that's part of the difficulty we've been mentioning. Datum references are there to help orient the tolerance zones, but the zones could be at some crazy angle to the datums, if we have basic angles to establish such (that was my point). So I think we're on the same page.

Dean -- I'm not clear on the most recent graphic you posted. In the RH view you have "between J and K" yet I think of profile of a line as inherently meaning that we are to take the lines in the direction of the given view.
But if you did that on purpose just to get us to think about the difference between the two callouts, then I would say #1 and #2 yield the exact same tolerance zones (they would be parallel to A, not perpendicular).

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

http://www.gdtseminars.com

 

[DeanD3W]

John-Paul,
I expect others to have opinions that differ from yours regarding the orientation of the cross-sections for options 1 & 2 in the second file I posted. I'll let others comment though, maybe there's something I'm not seeing. The "J <--> K" means the same and is fine to use for both options, and I think the answer is "nobody really knows" regarding the orientation of the cross sections... The only justification for "knowing" the orientation of the cross sections should require citing words, or at least an example, from Y14.5 that support that opinion. I don't think the standard puts anyone in a good position with anything in regard to profile of a line, so I think this can only yield unsupportable answers.

Powerhound,
I've already said this, but I'll just try again... That example is purposely simplistic, just to get only one point across... Tolerance zones for tapered features must tilt out of the plane of the cross-sections that many people associate with profile of a line in order to get a meaningful measurement result. The tolerance zone must be normal to the surface along a given line. If I make the true profile of each of the planar faces that the profile of a line is applied to curvy instead of flat, then the application would look more realistic, but then the simplicity that makes it easy to illustrate the dimensional difference would be lost... It's not intended to be a realistic example!!!!!!!!!!!!!!!!!

BTW - you say it's like straightness... How do you know that it's not more like perpendicularity of each line element? With datum features A and B referenced, one could say that all three rotational degrees of freedom are constrained. What statement in the standard tells you that orientation of each line element is not controlled?

Does anyone think the current definition of profile of a line in Y14.5 is adequate? The general point I've been trying to make is that profile of a line should be avoided since there are various interpretations that many have "read into" the inadequate definition in the standard. The first issue is the orientation of the lines on the considered feature (aka the orientation of the cross-sections). The second issue is treating the tolerance zones for tapered features as not normal to the true profile (where true profile is a theoretically exact definition of the entire surface, not just a line along it).

Any wording that accommodates an "everything lies in the cross-sectional plane" interpretation of the standard needs to be changed, since that interpretation yields misleading results.

 

[powerhound]

Dean,
I get that the example is not realistic which is why it's a problem with me. You are arguing a point using an example that no one would ever use in real life, thus my statement that the example creates a problem where there is none.

Why don't you post a realistic example?

The tolerance zone is only normal to the surface if you put the profile control in the left view, but that's not because the tolerance zone is supposed to be normal to the surface, it's because it's normal to the true profile. The true profile is NOT the surface and it IS a line along it where the cross section cuts through. Look at Fig. 6-18 in the 1994 standard and read what the example says. That zone is normal to the surface because of the direction the cuts are taken. They are taken across the profile, not along it as in your example.

Point taken on the straightness issue. Change that to perpendicularity.

Again, I fully understand, and have understood, that your example was just that, an example. I just don't think that you can support a real world argument using a non-real world example.

Powerhound, GDTP T-0419
Engineering Technician
Inventor 2010
Mastercam X5
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[CheckerHater]

A 2-dimensional spine is a simple (non self-intersecting) surface, and applies to parallel-plane features.

How exactly "parallel-plane features" are not flat?

I am outta here!

 

[axym]

CheckerHater,

The sentence was not meant to say that a parallel-plane feature is an example of a 2-dimensional spine.

The term "parallel-plane features" was meant to describe the general feature-of-size type (i.e. slots and slabs) that the 2-dimensional spine applies to. Strictly speaking, the term should be "nominally parallel-plane feature" since the surfaces of the as-produced part feature will not be perfectly flat and parallel.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[fsincox]

If we really seriously believed in the ASME principles of: “does not necessarily specify process or inspection method”, (cough, cough, RUNOUT!!!). Wouldn’t a Profile of a Line be another generic way of specifying a circular runout tolerance, using the toleranced size dimension and all? Maybe it is not my dear concentricity callout that should watch out about being obsoleted, let’s get rid of runout. (Strategic move here, stop playing defense, take the battle to the other side).
Frank

 

[CheckerHater]

Frank,

I have better idea.

All tolerance zones have mathematical description, so we can teach computer to calculate them.

This way they can be entered directly into 3D model like in the enclosed file.
We will have precise presentation of where part body can and cannot be.

This way we will eliminate not only Concentricity, Symmetry and Runout, but also ALL of the symbology, ALL of the GD&T, and ALL of the Certified GD&T Professionals.

As the data may be passed directly from $10/hr CAD-monkeys to CMM personnel, there will be no need for middle-man.

So, be careful what you wish for.