TP with or without a Diameter Symbol

[LargeNCharge86]

How I've been working with GD&T for 5 years and not understood this like I thought I did is beyond me but our GD&T guru is out until next week so he cant explain it to me.

Say I've got a slot with a TP of the entire feature of 0.5mm at MMC back to the datums. How is it measured differently if a diameter symbol is in front of the 0.5 vs. if it isn't?

 

[Belanger]

Dave, the example you cite clearly gives a direction to the intended tolerance (one for radial and one for angular). That is quite different from the graphic I posted, where there is only one FCF. In my case, why would you not resort to the boundary interpretation, which makes the tolerance zone conform to the shape of the nominal feature?

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

 

[3DDave]

John-Paul,

The boundary in 7-34 isn't a correct interpretation of the virtual condition for the feature - it's another reinterpretation of position as if it's a composite profile. The actual boundary cannot have full radius ends. As a result it doesn't match the limitations on position that the mid-plane interpretation gives.

 

[Belanger]

Dave -- not sure I understand your last post. I don't get how that would have any relation to a composite profile.
But if I hear you, you're claiming that the only true boundary in Fig. 7-34 is the 12.5 x 6.75 straight-across dimensions (and not the radii)? Their picture sure makes it look like the bumpy radius of an actual slot is restricted from violating a radius boundary of position also.

However, I think that's all irrelevant to the question posed in my graphic, which I still can't get a straight answer on:
A hole is a single, regular FOS (so very different from Fig. 7-34), and if that hole has a position tolerance as I showed (without the diameter symbol), would it be interpreted by the standard as a cylindrical boundary, or rejected as an illegal callout?

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

 

[3DDave]

John-Paul,

By convention it should be rejected as a typo/spelling error. If the zone is cylindrical it should have a diameter modifier. If it doesn't have the modifier then the zone won't be cylindrical and is undefined.

Look what happens to the boundary when the position tolerance is equal to or larger than the hole, leaving a non-existent or imaginary boundary. This is where the zone interpretation is needed. Where the boundary is existent it needs to match the result the zone allows for.

I'll make an example picture of the reason the example boundaries are incorrect.

 

[axym]

Hi All,

Here are some thoughts.

I would question the use of the diameter symbol for anything other than a cylindrical feature. So in LargeNCharge86's drawing, I wouldn't apply the diameter symbol. I agree with J-P that we don't know exactly what would have to be within a cylindrical zone, in other words there is no definition of the "center" of an oblong slot.

J-P,

Your NoDiameter graphic with the cylindrical hole and no diameter symbol is interesting - you may have found a case where Y14.5's rules are not entirely self-consistent. I think that the drawing makes sense and should be legal - the surface interpretation is well defined. Without the diameter symbol, there would be no axis interpretation. However, I would also say that the drawing would likely be misinterpreted - the reader might think that it's part of some kind of bidirectional spec and the other FCF is missing, or that the diameter symbol was unintentionally omitted.

We may need to be more rigorous on terminology here. Where you said "no diameter symbol, so the position zone isn't cylindrical and isn't rectangular", I believe that you meant the position boundary that the surface would have to conform to. But because you used the word "zone", some may have thought you were referring to a tolerance zone that the center geometry would have to conform to.

3DDave,

I see your point about the larger diagonal movement on a square or rectangular shape. This is because of Y14.5's rule that tolerance zone boundaries preserve sharp corners, is that right? Even in Fig. 8-24, the cutout could move further in the direction towards the sharp corner than it could in the direction towards the R10 corner. This is assuming that the cutout doesn't rotate at all. Defining and controlling the "center" geometry of irregular shapes quickly becomes a mess - I don't think that J-P was suggesting this. It's really only possible on regular features of size.

I'm not sure that I fully understand your reference to composite profile in the context of Fig. 7-34. I would agree, though, that the way that the boundary is calculated in 7-34 has some dodgy aspects. Offsetting in by 0.75 on the rads gets rid of some of the material from the flats, so I agree that this relaxes some of the requirement on the centerplane. The oblong slot is a special case that Y14.5 defines and the "blended tolerance" boundary isn't fully defined where the rads meet the flats.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[3DDave]

At this point I have to say that Inkscape takes a bit of getting used to.

Since the only feature for which a zone vs positional boundary interpretation are identical is circular, it makes sense to eliminate the use for anything except circular features and come up with an alternate, such as limiting the translation of the MMC shape and disallowing encroachment of the LMC shape; basically create an outer offset within which MMC movement is restricted rather than an inner boundary that can become degenerate.

The current position boundary interpretation is identical to the profile composite, supplying a uniform offset that does not restrict changes in position to the tolerance allowed for those features where the profile curvature is greater than the tolerance. To clarify, any sharp corner has infinite curvature; curvature of radii = 1/radius.

 

[3DDave]

No comments?

 

[Belanger]

Sure Dave -- see this attached graphic for a twist on the idea. Instead of a slot having two FCFs, suppose we use one FCF that doesn't care about horizontal vs. vertical. I've just combined the size dimensions into one line item, and then applied a position tolerance to the entire feature (irregular FOS, to be specific).

Does it still result in the reddish-colored area from your Inkscape graphic? Or would this be interpreted as smoothing out the full boundary to become a slot-shaped reddish zone?

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

 

[Belanger]

Graphic shown in this post, since it didn't seem to go in my previous one.

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

 

[3DDave]

The standard offers insufficient and self-contradictory guidance about what such usage means. It only works for circular holes and only almost works for round end slots. There is no generalized case that can be drawn from that. It is much better to make a basic dimensioned area that excludes part material explicitly than to have to guess what is supposed to happen.

Frankly it seems like an academic exercise that doesn't seem applicable to an actual product.

 

[Belanger]

So you're saying that the boundary idea isn't practical except for a regular FOS (that might be what Evan was saying too).
True, it is mostly an academic discussion now. Profile is the go-to control for these irregular shapes.

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

 

[greenimi]

Profile is the go-to control for these irregular shapes.

AMIN. That I was saying a week ago when this discussion started.
Why using "problematic" callouts (position, Boundary) when a more robust way of defining this feature is already offered by the standard.