Profile callout 2

[dodge78]

We have a callout on a customers drawing as follows;

Profile of .0005 to A.....There is a non basic dimension from A of .300

I see this as Parallelism to A, since the .300 is not basic. Am I correct with this definition.

 

[dingy2]

Frank:

I guess I am not a proponent of this approach since all features are treated equally while in real life, some features are vital to the part's function and mating relationship while others are irrelevant to its function. Jim and I have had discussions on this subject in the past and we, respectfully, stand diametrically opposed on this concept.

This approach also contravenes sections of the 2009 edition such as - 4.15 Functional Datum Features - "Only the required datum feature should be referenced in the feature control frame when specifying geometrical tolerances."

After many discussions here, I ended up developing a bit of a web page on this controversial subject.



Dave D.

http://www.qmsi.ca

 

[fsincox]

Dave,
Thanks, I would love to look at it. I believe I have seen some of those conversations and thank you also for pointing out 4.15. I would tend to agree that they seem to be at odds. Like I said earlier general tolerancing does tend to blend all features together as if they are all equal, this is just as true for any titleblock tolerancing scheme so it is not new.
Frank

 

[dingy2]

If a Designer was opting to use default profile of a surface tolerance, it could be used for the "sluff off" dimensions utilizing a large tolerance with a note under the FCF stating "reference datums as applicable". This now meets the standard. The features that are vital to the part's function and relationship could be shown in separate FCFs with an appropriate tighter tolerance.

Dave D.

http://www.qmsi.ca

 

[ptruitt]

Do adding or removing datum reference frames that are inconsequential to the function of the part affect the amount of tolerance available to the fabricator?

Peter Truitt
Minnesota

 

[MechNorth]

Frank,
Each part would have its own general tolerance value, and drf. This appeases the Y14.5 requirement that all features be controlled. Beyond that, you override with local controls. I.e. appropriate controls on your datum features, tighter or looser controls where needed/warranted on specific features. Some of us who were more experienced would base our values on process knowledge and industry standard such as 2768, while others would pull a thumb-suck from 2768 without understanding the implications. There were various philosophies about selecting which tolerance to put in the general tol frame too; ISO requires the loosest shop capability (or is it the tightest; I keep forgetting) while some designers emphasize the tightest with all looser tolerances being directly applied. To me, it made sense to reduce detailing time by selecting the most common acceptable tolerance. I found that most people then rationalized that they typically were too restrictive with their tolerances on most features anyway, and ended up accepting the general tol as adequate for even more features.

Yup, Dave, we agree to disagree.
Now, on 4-15, these are my thoughts; there is no HARM in adding the datum references even if they are not required for functionality PROVIDING THAT it does induce an additional setup for inspection. I say this because if the DRF is not filled in, then you would need to break down the inspection setup to eliminate over-constraint; I pretty much guarantee that a CMM operator doesn't break down the datums for a new feature. Taking a blend radius as an example, there's no value in the DRF - agreed - but are you going to break down the setup to verify it?
Overall, nothing should be a carte-blanche for design or inspection, but some conveniences can save both parties significant time and effort ... and a general surface profile has proven to be one of them for many situations.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[MechNorth]

Peter,
It may.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[fsincox]

Jim & Dave,
Thanks.