Size versus UOS default specification

[Kedu]

If a feature of size tolerance is large enough, can it vary in form to a degree such that the angle between surfaces can be larger than an "UNLESS OTHERWISE SPECIFIED" block angle tolerance allows…?

In addition to Rule #1, can/must form also be limited by block
angle tolerance? Another way I've posed the question is to consider a cylindrical feature with a liberal size tolerance. Does a block angle tolerance impose a cylindricity-like refinement due to the limits on the implied 90° angles depicted on the drawing?

 

[Burunduk]

General angular tolerance is not a good way to limit form variation.
In theory, it could limit the taper-shaped form error of the opposed elements of the FOS, but that would only have any value if all elements - the two opposed ones and the face at right angle to them, came out ideally straight. But what if the form error is waisting or barreling? The angular tolerance has no value in limiting that because in the ASME standards, there is no definition of how a toleranced angle dimension should be measured, let alone between imperfect planar features or line elements.

 

[drawoh]

Kedu,

If your drawing has a note on maximum angle errors, then each feature must meet that requirement. An FOS tolerance can be sloppy enough that the angle specification applies.

--
JHG

 

[greenimi]

If I understand the OP's question correctly, it is a rather direct question:

Given:
Rectangular plate dimensioned with ± for size (generous and liberal tolerance)
ASME driven drawing

Question: Could the left wall's perpendicularity (or right wall's perpendicularity for that matter) to the bottom surface be limited by the default angular tolerance such as:
Unless Otherwise Specified tolerance on angles = ±1°

IS the perpendicularity limited by those DEFAULT tolerances applied to the implied 90°?

 

[drawoh]

greenimi,

As per ASME Y14.5?

Yes.

--
JHG

 

[Belanger]

Greenimi, I think the questions is still a little different from what you have shown, because on your green part the size dims are still perfect. But the OP mentioned having a large size variation.

The sketch below shows how a part can utilize the size variation across a FOS to maximum effect, thus creating angular error.
Using my sketch, would the 4 corners shown in the "produced part" still have to comply with the general tolerance for angles? I would say yes.

It's the same answer you guys were saying, but I just wanted to put a different spin on the illustration of the question

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

 

[3DDave]

I have never seen anyone inspect the default angle tolerance. I suppose if a nominally right angle was far out they might, but it never happened in a way I could ask for historical measurements, a distribution vs process, nada. As I have mentioned I saw a DoD program that had no default angle tolerance on any drawing - no one noticed or cared.

 

[TheTick]

If it's important enough, use a compound tolerance.

 

[Burunduk]

TheTick
What do you mean by "compound tolerance"? That's not a standardized term. Could you give an example?

 

[greenimi]

TheTick
What do you mean by "compound tolerance"?

Yes. I am not understanding that either.

 

[greenimi]

Greenimi, I think the questions is still a little different from what you have shown, because on your green part the size dims are still perfect. But the OP mentioned having a large size variation

J-P, I agree with you that my "as imagined" case is different than the one described by the OP in the initial post. Looks like the problem is more generalized than I've could ever guessed or even imagined.
One interesting fact is how the industry "survived" so many years without a clear definition?
I think UOS is at least as muddy as the "actual local size". Don't you think?

 

[Belanger]

I think UOS is at least as muddy as the "actual local size". Don't you think?

Yes, I agree!

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

 

[Burunduk]

I think UOS is at least as muddy as the "actual local size". Don't you think?

Not "UOS", but general plus-minus title block tolerance. UOS followed by a general profile of a surface tolerance note is the opposite of muddy if done right.