In worst case, the flatness is... 7

[SeasonLee]

In worst case, the flatness on this surface is...
Please see the attached for details.
Thanks

Season

 

[3DDave]

D. Definitely.

 

[Belanger]

0.25 is probably what the quiz question is seeking as the answer. This is from 0.2 traced in one direction (perhaps downward) and a possible 0.05 in the other direction (perhaps upward).
But that notion of straightness in a certain direction is somewhat flawed, because the direction to trace is based on stabilizing the part in a certain orientation (read: datum). Tagging the FCF to a certain view isn't enough.

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

 

[CheckerHater]

C. 0.6

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

 

[CheckerHater]

Here is the same model with straightness of 0 in any cross-section.

Straightness DOES NOT control Flatness. Period.

 

[SeasonLee]

The quiz originated when I'm reading a book says: a feature can have different straightness values specified in different views. Then I’m thinking what the flatness will be in this case, so I’m seeking the answer, thanks for all of your valuable replies and comments.

Dave: Would you pls kindly adv. flatness value if you say D Definitely.
Capnhook: Would you pls kindly adv. how you get the value 42.

Thanks

Season

 

[3DDave]

CH has the correct explanation - I just preferred to hold off on what looked like an academic question.

I've never gotten the interest in trying to interpret one control in terms of other controls, particularly ones that don't perform the same function, such as in this case. Much like speculating as to which makes for a better chisel, a hammer or a vise? I guess as a quiz it is meant to determine if the user knows the difference.

You need to read HHGTTG** to understand how 42 can be the correct answer to almost any question, subject to having a good understanding of what the question means. You'll also need a towel and a package of peanuts.

**Google can find the reference for this.

 

[Belanger]

Here is the same model with straightness of 0 in any cross-section.

CH, I'm not sure that I see that second sketch as having perfect straightness in any cross-section. Do you mean in any cross-section of a given view? Or perfect straightness in all directions?

Because having perfect straightness in all directions would seem to be equivalent to perfect flatness.

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

 

[3DDave]

J-P,

Take a deck of cards and give it a twist. Each card edge remains straight and, twisted correctly, each section perpendicular is also straight.

 

[Belanger]

Dave -- of course, but that deck of card is not straight in all directions. Only in the directions that the cards lay.
Thus my point.

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

 

[CheckerHater]

@ Belanger:

Straightness control itself applies separately in given view:

S specified in front view applies to cross-section parallel to front plane.

S specified in side view applies to cross-section parallel to side plane.

There is no such thing like Straightness "all-over"

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

 

[Belanger]

Fine -- I'll buy that. But I'm curious if anyone caught my point above:
You say "S specified in front view" (or side view). How do we align those sampled lines? Parallel to the face of the part that we see? Or the back face? This is perhaps an inherent problem with the entire idea.

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

 

[3DDave]

The lack of stringent rules concerning the orientation of a part with respect to view-dependent tolerances within Y14.5 means that straightness is something of a fiction for anything but round items where orientation of the part is not material to creating a boundary simulator. At least the trig works to minimize the effects of minor orientation changes.

What's left to the imagination is that, for a given setup, there is a good chance the line segment of interest will not be the complete width or length of the part. It isn't clear if the entire deviation is applicable to the fractional segment or if it is reduced to some proportion of the overall width/length.

This shape has a lot of perfect straightness:

http://mathworld.wolfram.com/One-SheetedHyperboloid.html

Straightness is critical to the creation of the shape, but it is unclear it's a target for a straightness control.

 

[Vishal2015]

CheckerHater is correct.

Straightness does not control Flatness.

Refer to attached image.

 

[UchidaDS]

Vishal2015 (Mechanical):

What does it mean by 'Straightness is maintained in X and Y Direction in zero'?

 

[3DDave]

It's a typo and badly worded. Deviation of straightness in X and Y directions is zero. IOW looking at constant X component the surface parallel to the YZ plane is straight. Likewise for the constant Y component parallel to the XZ plane.

 

[Vishal2015]

That's right 3DDave.

Deviation of straightness in X direction is zero as well as in Y direction is also zero

 

[greenimi]

So, what is the correct answer in the OP? I am trying to learn something from this thread. Please kindly advise.

 

[Belanger]

The correct answer is that there is no flatness control contained within straightness, so answer d (or none of the above, or whatever). See Vishal2015's graphic to see why.

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

 

[pmarc]

The correct answer is 0.6 (assuming that Rule #1 governs the drawing).

Vishal's graphic is great because it clearly shows that straigthness tolerance specified in two orthogonal directions does not control flatness of a surface, but the graphic does not and I believe never meant to prove the correct answer was different than 0.6.

Side note regarding 3DDave's point about "lack of stringent rules concerning the orientation of a part with respect to view-dependent tolerances within Y14.5". ISO solved that dillemma quite reasonably by introducing Intersection Plane concept in ISO 1101:2012. The attached figures show 3D annotation case, but the concept can be applied to traditional 2D drawings as well.

http://files.engineering.com/getfil...-5c4388787d28&file=intersection_plane_ISO.png