Radius as a datum 3

[ImnotfromMars]

Hello,

Can a radius be a datum?

 

[KENAT]

Depends, most folks interpret 1.3.17 of ASME Y14.5M-1994 1.1.17 as meaning that to be a feature of size the radius needs to be more than 180° in order to have opposing points.

Feature of Size.One cylindrical or spherical surface, or a set of two opposed elements or opposed parallel surfaces, associated with a size dimension.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[Belanger]

In theory, a datum is a plane, axis, or point (or a combination of those). So as long as the radius can repeatably derive an axis, yes, it can be the datum feature. If you have the Y14.5 standard, an example would be datum B in Fig. 4-29.

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

http://www.gdtseminars.com

 

[pmarc]

So as long as the radius can repeatably derive an axis, yes, it can be the datum feature

Radius shown in fig. 4-29 does not derive an axis. It constrains rotation of two mutually perpendicular planes associated with datum axis A derived from datum feature A.

 

[CheckerHater]

OK, a datum is a plane, axis, or point.
Radius shown in fig. 4-29 does not derive an axis.
So what IS datum B on 4-29?

 

[bxbzq]

I would say datum B is still an axis, but it is a different datum axis from datum axis A. It clocks down the mutually perpendicular datum planes derived from datum axis A. Precisely, it derives mutually perpendicular planes, but in DRF we only see the one passes through both axes, the one coincides with one of the datum plane derived from datum axis A. We don't see the other one.

 

[TheTick]

What kind of radius? Turned part? Hole? Fillet?

 

[Belanger]

Datum B sure seems to be an axis.
Pmarc -- sure, it constrains the rotational degree of freedom, but if it's not an axis, what does the 28 mm dimension originate from?

Maybe we're getting sidetracked on another academic thing. ImnotfromMars -- can you provide more info about your situation or a sketch?

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

http://www.gdtseminars.com

 

[Belanger]

Wait -- my mistake, pmarc. The 28 is going from datum A to B (I was looking at it backwards). That still doesn't solve the dilemma: according to the standard a datum must be a plane, axis, point, or combination thereof. Which of those is datum B, if not an axis?

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

http://www.gdtseminars.com

 

[pmarc]

I see at least two reasons why it is not 100% correct approach to consider datum feature surface B in fig. 4-29 as a feature used to derive a datum axis from it (although I admit that bxbzq's interpretation makes sense in this paricular case):
1. Explanation given in para. 4.11.4. Notice that datum axis/center plane interpretation applies only to features-of-size type of datum features [subparas. (d) and (e)]. When it comes to the surfaces, there is no such interpretation, which sounds reasonable to me because of... see point 2.
2. What if surface B was an irregular basic contour (like for example a set of basic radii tangent to each other), so that it would be impossible to clearly identify a center of datum feature simulator B? Would you search for a datum axis or a datum plane in that case? Would this be convenient interpretation at all?

So trying to answer to CH's question -- I am leaning towards saying that THERE IS NO datum B in fig. 4-29 - at least this is how I understand the letter of the standard. There is just datum feature B that is serving (through its datum feature simulator) as a rotational constraint for 2 planes derived from datum axis A, but that is all.

Does it sound reasonable at all or should I prepare for attack?

 

[fsincox]

pmarc,
Well, you know I don't likre it!

 

[Belanger]

Interesting stuff, pmarc. Although the foundational items you give make sense, the conclusion drawn from there doesn't make sense: There is a datum feature B on the part but no datum B?

Turn the page: There is no datum B in Fig. 4-30 either? But most of us would say that there is a datum B (a plane) for that part.

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

http://www.gdtseminars.com

 

[pmarc]

Well, unless I am missing something, fig. 4-30 and accompanying paragraphs say not a word about datum plane in this case.

 

[CheckerHater]

I find the idea of datumless datum features fascinating, but…
From Para. 3.4.3 Y14.5-2009:
“Where more than one datum is required, the datum feature reference letters are entered in separate compartments in the desired order of precedence”
As long as letter B has its own “separate compartment”, there is a datum B – that’s the letter of the standard.

 

[Belanger]

Well, unless I am missing something, fig. 4-30 and accompanying paragraphs say not a word about datum plane in this case.

Right, so I'm asking what you would say that datum B is in Fig. 4-30. Or is it another case of no datum B?

I admit that the radius in Fig. 4-29 is not a feature of size. But there has to be a datum (else why have a datum feature?) and that datum must conform to the definition of a "point, axis, line, plane, or combination thereof" (para. 1.3.13).

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

http://www.gdtseminars.com

 

[pmarc]

Okay, I agree that the most intuitive interpretation of datum B in fig. 4-30 is to name it as datum plane, but then I repeat my question: what if the surface was not nominally flat, but a kind of really irregular contour? Would we be able to clearly identify datum?

 

[CheckerHater]

According to 2009 you can have “mathematically defined” feature. (Para 4.31)
To me surface that has axis and the radius is mathematically defined enough.

 

[pmarc]

I did not say that mathematically defined contour could not be used as datum feature. I asked if it is possible to clearly identify datum derived from that irregular datum feature. Figs. 4-29 through 4-31 show really simple features (regular single radius and flat surfaces) used as secondary datum features. In such simplified cases datum axis or datum plane interpretation works nice, or like I said before, is the most intuitive method. But if the datum feature is of more complex geometry, so that is difficult to clearly indicate datum point/axis/plane of datum feature simulator, this approach does not give satisfactory results. At least in my opinion.

 

[CheckerHater]

Actually the uglier the feature the better.
Standard recommends using coordinate system for “disfigured” features.
So if your drawing (or better yet CAD data) shows origin with X, Y, and Z sticking out; what could be easier than derive point, axes and planes from that?

 

[bxbzq]

If an irregular hole can be a datum feature and derive a datum axis, so can portion of irregular hole, like an oval hole or the one shown in 8-19. But indicating the origin of DRF should be necessary.