Profile of a Surface 1

[James Orlando]

I'm hoping someone can help me answer the correct meaning of the Profile of a Surface call out is on this drawing. We have two individuals with differing opinions and I would like to understand the true intention.

The part has an overall dimension of 220.12. The side surface (B) has Profile of a Surface of 0.25 mm.

Does this mean that the overall dimension would be +/- 0.5 mm or +/- 0.25 mm?

Thank you in advance for the help.

 

[drawoh]

James Orlando,

Your profile tolerance applies all the way around the outline. It controls size, perpendicularity and parallelism of the various features. An allowable extreme profile variation at each end could result in a length measurement of 220.37.

Converting profile tolerances to [±][ ]tolerances is an approximate process at best. It is better to understand the profile tolerance. Your part would be more inspectable (as opposed to functional) if datum feature[ ]C were the secondary datum.

--
JHG

 

[CheckerHater]

Interesting question though - if datum is derived not from one but from two surfaces, is it still self-referencing?

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

 

[James Orlando]

Thank you

 

[drawoh]

CheckerHater,

Effectively, the 220.12 dimension is a feature of size, and perhaps ought to be dimensioned that way. Since this is the secondary datum feature, the 22.35 lengths control rotation of the part, as well as the position. Would you pick up two points on one side and one point on the opposite?

thread1103-405622

From a practical point of view, as opposed to the standard, I think FOS datum features ought to be very much more accurate than the features controlled by the datums. For example, a hole and slot for dowel pins could be datum features[ ]B and[ ]C. The alternative is to call up the datum at MMC/B, which also is messy, especially in the OP's case.

--
JHG

 

[Belanger]

Max/min distance across would just be +/- 0.25 because the tolerance zone is assumed to be equal bilateral, but then it wraps around the entire part. IOW, each side could expand 0.125.

CH, did anyone mention that it's self-referencing?

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

 

[3DDave]

and [C} are redundant/self-referential in the profile FCF. There are bigger problems than what the overall width might be.

 

[drawoh]

Belanger,

I went quickly through that thread on self-referencing datums. I don't see a problem, especially if I specify an edge as the datum feature, not the width. On the secondary datum edge, the profile would control straightness only. On the tertiary edge, it would control perpendicularity to the primary and secondary datum features, only.

When I look at datums, I think about how to fixture the part for fabrication and inspection. I don't worry about conversion to [±][ ]tolerances.

--
JHG

 

[CheckerHater]

CH, did anyone mention that it's self-referencing?

The confusion arose because the ASME standard didn't envision a situation where someone would wrap a profile tolerance around some surfaces that are the very datum features referenced in that profile tolerance.

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

 

[AndrewTT]

May I ask why you did not mimic figure 8-5 (see attached)?

 

[Belanger]

CH -- First, in your initial post were you REALLY looking back to that old thread? Um, isn't it customary to discuss the thread at hand unless someone specifically links back to an old thread?

Second, if you had followed that discussion, you would see that it was talking about datums derived from surfaces. This thread uses datums derived from a FOS. Not really the same, now are they? However, 3DDave brings up a good point in that it's still not wise.

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

 

[CheckerHater]

John-Paul, you are getting too little too personal.

Yes, That old thread was REALLY the first thing I thought about as soon as I saw the original post.

Second, that old thread started from this discussion:

http://files.engineering.com/downlo...-e671-4c63-8243-8678066fd906&file=drawing.pdf

Third, my suggestion to OP would be to drop datums B and C altogether, because profile to A will do just fine



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

 

[Belanger]

Fine, CH. But realize that I can't get inside your head to know that you were continuing a discussion from two years ago! That's the only reason I asked why the self-referencing datums would pertain to the OP's question.
That said, I still stand by both of my comments -- it's legal but perhaps confusing.

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

 

[3DDave]

On this concept - can I have a single hole in a plate, nominally perpendicular to a face identified as datum feature [A}, with the hole identified as datum feature , and the hole positioned to [A|B]? I think that would also qualify as legal but confusing.

 

[CheckerHater]

3DDave, I like you more and more. In your example position to [A-B] may be legal, but seriously confusing.

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

 

[Belanger]

Agreed -- confusing!

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

 

[3DDave]

I like to think of self-referential references of this type as "Wherever you go, there you are."

 

[axym]

Hi All,

I have a couple of comments.

First, there is the question of whether or not datum features B and C are features of size (resulting in datum centerplanes) or single surfaces (resulting in datum planes). The drawing has some conflicting indications, but I would say that by Y14.5 rules B and C are not features of size. There needs to be a directly toleranced dimension, and the 220.12 and 36.83 width and height are basic dimensions (and hence not directly toleranced). The designer's intent may have been to make B and C features of size with centerplane datums, but the drawing does not get there.

Second, I would agree with 3DDave that the "self-referencing" tolerances are legal but confusing. I believe that the case of the single planar datum feature has been addressed - the generally accepted result is that only half of the profile tolerance zone is available for the datum feature (this can be derived from the rules on how the tolerance zone and datum feature simulator are established).

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[CheckerHater]

Evan, I find your post disturbing. But so is the idea of using self-referencing datums.

About features of size.
By Y14.5 rules "directly toleranced dimensions" are not limited to sizes, but include dimensions controlled by FCF. Features B and C are controlled by Profile and Profile controls size.

About only half of the profile tolerance zone being available.
We have total of 4 features defining datums, and they cover most of the outline. Do you mean that most of the profile must be "half-profile"?

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

 

[pmarc]

Widths B and C are not REGULAR features of sizes, but IRREGULAR (type A).

I agree that controlling datum features with profile tolerance that references datums derived from these datum features is confusing at best. In some cases it may work quite well (e.g. datum target application), but sometimes it is just like asking for troubles.

In this case, for example, it leads to different opinions on what is the allowable "size" variation of datum widths B and C. In my opinion it is +/-0.25.