Profile related to DRF

[powerhound]

Hey guys,

Another question:

Do you see anything wrong or questionable about the attached drawing? Assume all the dimensions are there, it's just teh profile related to the DRF that I'm asking about.

Thanks,

Powerhound, GDTP S-0731
Engineering Technician
Inventor 2010
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[Belanger]

A-OK!

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

http://www.gdtseminars.com

 

[powerhound]

My concern was that reference to C was a self referencing datum. I guess not.

Powerhound, GDTP S-0731
Engineering Technician
Inventor 2010
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[CheckerHater]

My concern was that reference to C was a self referencing datum.I guess not.

Agree.

For purists datum target can be used for C as it only does the clocking.
This way it will be clear that C does not apply to the entire surface.

 

[pmarc]

Powerhound,
It is good that you do not feel comfortable with this callout. If you have access to "Advanced Concepts of GD&T" by Alex Krulikowski, look to chapter 22 and to what he calls "Profile datum rule". The conclusion is that such approach results in ambiguous specification which can be differently interpreted by different drawing users. Let me know if you need a screenshot from the book.

 

[powerhound]

Yes pmarc, I'd like to see the screenshot. Thanks!

Powerhound, GDTP S-0731
Engineering Technician
Inventor 2010
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[pmarc]

Here you are:

http://files.engineering.com/getfil...b450-e71412505c7f&file=profile_datum_rule.JPG

 

[Belanger]

Yes, I spoke too soon. I thought your concern was that profile of a line is only 2D and yet you were referencing 3 datums. That question often arises about the profile of a line symbol.


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

http://www.gdtseminars.com

 

[powerhound]

Thanks pmarc. I printed it out and gave it to the QA manager. He lovingly gave me a bird in return... Maybe it had something to do with me highlighting the "common sense rule" part of the paragraph.

Thanks also to all who responded.

Powerhound, GDTP S-0731
Engineering Technician
Inventor 2010
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[CheckerHater]

I feel a bit suspicious about “common sense” rules (AKA “not really standard”)

Also, as I mentioned, single point contact (datum target) could be enough, thus we will not have “surface referencing surface”.

Altogether, it can be fixtured, it can be measured, and it’s not ambiguous. I agree it’s “exotic”, but if function of the part requires it to be done this and only this way, it’s do-able.

 

[powerhound]

Perhaps datum targets might change the overall consensus but this is a real world situation going on right now and datum targets are not used, thus the question. I agree about the common sense thing. It's really subjective BUT it is on teh page so I thought I'd use it.

Powerhound, GDTP S-0731
Engineering Technician
Inventor 2010
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[gwiz42]

If the profile tolerance was unilateral into the part, would this call-out be ok? The Profile Datum Rule is specified for a bilateral tolerance zone.

 

[pmarc]

I would first re-consider the need of tertiary datum for the profile callout for this particular geometry. If we assume that basic dimensions fully relate part's contour to datum axis B (like Powerhound said), I do not think C is required at all.

 

[axym]

I'm not comfortable with the callout either. In general, "self-referencing" FCF's that reference the considered feature as a datum feature are rarely, if ever, functionally necessary or appropriate.

I don't agree that the callout is ambiguous, but I do agree that it is likely to be interpreted differently by different users. So I agree with the results of Alex's Profile Datum Rule, that the self-referencing specifications are confusing and should be avoided. But I would say that the meaning of the tolerance zone is clear, if one follows the rules of how datums are established and how Profile tolerance zones are constructed around them. It's just that the meaning doesn't correspond to function.

I agree with pmarc that we need to confirm whether the C feature actually clocks the part - in other words, does contact on the C feature constrain the last rotational degree of freedom? If it does not, and C was chosen as a datum feature arbitrarily, then the reference to C should be removed from the Profile FCF. If the C feature does actually clock the part, then it needs to have its own Profile tolerance that references A and B.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[powerhound]

The hand sketch is a highly generalized version of a real part. There are holes on the actual part and the geometry is different, but including them on the sketch served no purpose. On the actual part, datum feature C serves no real function. It was chosen as a clocking measure only. My recommendation was going to be to reference everything to A and B and let simultaneous requirements do the rest but ran into a little resistance, thus the OP.

Powerhound, GDTP S-0731
Engineering Technician
Inventor 2010
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[pmarc]

Evan,
How can a callout having two or more different interpretations, depending on who looks at it, be unambiguous? (Of course assuming that the readers are equally well educated in GD&T).

Powerhound,
I somehow felt that your sketch was showing a simplified situtation. But even for more complex geometry, you do not have to use simultaneous requirements approach. You can still keep datum feature C as it is. If we assume for a moment that there is another hole which has to be located relative to hole B and clocked to the bottom surface, all you have to do is to get rid of C reference from profile all around callout (according to "Profile datum rule") and apply positional tolerance to the second hole wrt to |A|B|C|. In such case C reference will be absolutely legal, clear and will not violate any rule. Moreover, if Y14.5-2009 was in charge, even (M) or (L) modifier could be put right after C datum reference, depending on the functional requirement.

 

[powerhound]

I want to go with simultaneous requirements because it is the simplest solution, in my opinion. I wasn't implying that it was the only solution.

Powerhound, GDTP S-0731
Engineering Technician
Inventor 2010
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[pmarc]

It is your choice and I am not saying it is wrong. I just wanted to mention the other option.

 

[Belanger]

I couldn't find it by searching the forum, but haven't we discussed in the past the use of a "self-datum"? I recall Jim S was a fan of this, and we had discussed an example of a window or cut-out that had an all-around profile, yet it referred to one side of the cut-out as a datum.

The thinking was that the true datum is of course a perfect plane, and the actual surface could still be checked to that perfect plane. I suppose along that edge it becomes similar to flatness.

I really like how Evan stated it: "The meaning of the tolerance zone is clear, if one follows the rules of how datums are established and how Profile tolerance zones are constructed around them."

But let me know if anyone finds that other thread that I am thinking of.

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

http://www.gdtseminars.com

 

[fsincox]

JP,
Yes, we have, and I too was also under the same impression, I guess things around here are like the wind, always changing. The committee should clarify this, in my opinion, particularly as they have now added the all over option and discuss the option of using general profile tolerances. My suspicion would have to be that possibly they don't all yet agree themselves.
Frank