profile on trimmed edge 8

[DGN1975]

does anyone have idea or references how to profile tolerance a complex trimmed edge on say a vac formed part? I don't really see any good examples in the standard to other places. Would it make sense to just point to the trimmed edge with a 'ALL OVER' profile tolerance with a note below that states "trimmed edge". using "all around" wouldn't work to me as it is controlled only in the view its shown in and a trimmed edge would go all around in 3d.

 

[3DDave]

Except for missing helical case, Tandler's is the exhaustive list of geometry constraints for a feature. All others are reducible to these and these aren't reducible any further. The better charts list the specific # of rotation and translatiion constraints to show the pattern more clearly.

Where you are failing is mistaking my complaint about the ambiguity of the example for a complaint about the abstraction; it is clear you don't understand abstractions because you repeatedly fall back to creating example puzzles to be specifically evaluated in some mentally satisfying way, but then cannot accept other examples to accomplish the same with less encumbrance.

Line-in-plane is any linear extrusion and does not need to bisect the angle - it can be any angle to the faces and the two faces are considered one feature, not separate faces. It can be a group of holes. The axis need not be centered. That fact is obscured in the (f) example by implying a symmetry is part of the determination.

Really - the entire diagram is backwards - the datums are chosen and the datum features are defined from them.

As often before, you have helped uncover a larger flaw than I initially noticed.

 

[Burunduk]

There really is no use in such charts other than for identifying the datum being derived for the simplest and most obvious cases. Once the geometry becomes more intricate there are many possible combinations of planes, points, lines and axes, often interchangeable.
It is ironic that you insist that there can only be 6 datum possibilities and that 7-3 (g) could only show a shape fully converging into a point to get that plane, point, axis datum - and yet you say that I'm the one that "cannot accept other examples".

"Where you are failing is mistaking my complaint about the ambiguity of the example for a complaint about the abstraction"

Where you are failing is remembering what you complain about - apparently because you complain so much. Let me refresh your memory, then. You didn't complain about ambiguity of the 7-3(g) example. You complained that it doesn't show in the orthographic views what you considered the only correct type of geometry that would produce that datum (talk about defining the datum feature from the datum), match the description in the text and what you thought the trimetric view shows (hopefully only until I pointed out the error to you). Tandler's figure doesn't have a surface converging into a point in what is probably his version of the "complex" type (Point-on-Line-in-Plane) - why are you fine with that?

Let me try to make sense of the following by filling in some words:

"Line-in-plane is (the datum obtained from?) any linear extrusion and (the plane?) does not need to bisect the angle - it can be any angle to the faces and the two faces are considered one feature, not separate faces. It can be a group of holes."

Considered one feature and labeled how? The "common datum feature" concept as I suggested uses a single TGC and results in a single datum, so it doesn't matter you consider the two faces as one feature or talk about them in plural. Also, a group of holes can be associated with the Line-in-Plane datum, but it doesn't have to. It would be more straightforward to describe the datum as a group of axes - since this is the direct result of using the datum simulating pins for the group of holes.

 

[3DDave]

Line in plane is the constraint, not the datum.

A group of holes can be associated with the Line-in-Plane datum constraint, and, as primary it does have to when identified as a datum feature. Where the plane and line are shown are wherever it makes sense to show them.

 

[Burunduk]

It says "The Six Possible Datums" not "The Six Possible Constraints". How is a combination of theoretical geometric elements a constraint?

 

[3DDave]

Call up Tandler and maybe he can explain it to you.

 

[Burunduk]

Tandler agrees on this with the definitions in the Y14.5 standard:

"constraint: a limit to one or more degrees of freedom."

"datum: a theoretically exact point, axis, line, plane, or combination thereof derived from the true geometric counterpart."

Then Line in plane is the datum, not the constraint.

 

[3DDave]

You called Tandler?

 

[Burunduk]

"Line in plane is the constraint, not the datum." - your statement, not Tandler's. Then it's up to you to support it. Don't shift the responsibility onto Tandler.

 

[3DDave]

A datum is constrained by - constraints. There are a limited number of constraints that are derivable from infinite datums as derived from datum features and vice versa.

I feel constrained by your lack of math skills and cannot fill that gap. Machine design and mechanism design courses discuss constraints.

You should be able to derive this and also why it was a bad idea for the ASME Y14.5 committee to drop this chart in place with no further explanation based on your engineering degree courses.

 

[Burunduk]

"A datum is constrained by constraints" is one of your brightest moments. Very informative and makes a lot of sense. Did your math skills help you to come up with this explanation?

What a mess. I should have known that I'll have to clean up afterward:

A 'constraint' takes out a degree of freedom, in other words, it immobilizes the part in a defined direction. Constraints are achieved through the interaction between the datum features and the (physical) datum feature simulators or (theoretical) true geometric counterparts.  

The datum feature simulator or true geometric counterpart simulates or defines the 'datum' - such as the axis of a collet that holds a cylindrical datum feature or of the minimum circumscribed perfect cylinder, respectively. Datums in general are theoretical axes, lines, points, planes, or for the more intricate datum features - combinations of these elements, such as the "Line-in-Plane" in question. They are the links from the datum feature simulators or true geometric counterparts to the datum reference frame (a set of 3 perpendicular planes and 3 perpendicular axes) which is established based on the datums and used as the zero origin for the location and orientation of tolerance zones.

GD&T training companies should pay you a percentage of their profits. The more someone that takes your explanations seriously tries to make sense of them, the more desperately that person needs their services to undo all the misleading, hence better business for them.

 

[greenimi]

Guys,

I guess we need constructive discussions here and attacking one another does not help........just saying.

 

[Tarator]

Hi all,

Referring to the image showing 6 possible datums:

There are only 3 possible datums: point, axis, and plane.

It is important not to mix datums with datum features and/or simulators.

 

[Burunduk]

There are only 3 possible datums: point, axis, and plane.

Not quite. According to ASME Y14.5 there are actually 5 possible datum types. These are the 3 you mentioned and two additional ones:
* line (note the differentiation between line and axis).
* any combination of lines, axes, points planes or some of these elements.

 

[3DDave]

Oddly, "datum line" is only mentioned as a term in the definitions section and never used in the text**, except to suggest a case where it is identical to an axis.

Probably it was meant to reflect the idea from Y14.5.1 of a spine, but it doesn't matter. Just another loose end to tie training material to.

Ever notice that a datum reference frame is:

datum reference frame: three mutually perpendicular datum planes and three mutually perpendicular axes at the intersections of those planes.

but the related figure also includes a datum point?

Is the datum point part of the datum reference frame as given by the definition or is this another loose end?

How can a document created by so many experts be like this if it isn't intentional?

**What garbage that the diagrams are not created with searchable text. For a document that should be a quick reference and not a word-for-word, picture-for-picture eidetic memorization exercise for Rainman levels of recall, this is contemptible.

 

[Tarator]

Burunduk,

1. What is the difference between a datum axis and a datum line?
2. For a conical feature (identified as a primary datum feature), would we have 2 datums (1 datum axis, and 1 datum point) or 1 datum (datum axis and datum point counted as 1 datum)?

 

[Garland23]

Tarator, here's an offshoot question to yours: Is a datum axis a single datum, or is it comprised of two intersecting datum planes?

 

[Tarator]

Garland23,

A datum axis is a single datum, but can be thought as the intersection of the 2 mutually perpendicular planes.
Same thing as a datum point, thought as the intersection point of 3 mutually perpendicular planes.

 

[3DDave]

... or a datum point is the intersection of an axis and a plane - but the standard hand waves around the question.

For example, a cylinder as primary to a cross hole. Until the hole is drilled, the axis can be determined from the cylinder, but not the orientation of intersecting planes. Those aren't fixed until the feature that references them exists - the feature is what orients the DRF even though the cross-hole, in this case, is not a datum feature.

 

[Tarator]

3DDave,

If you think a datum point as intersection of 2 mutually perpendicular planes, those planes do not have to be fixed, they can freely rotate around the intersection axis, until the secondary, and tertiary datums come into play.

Even if you have just a cylindrical feature (datum feature A), your DRF would be | A |. You would still have a coordinate system (3 planes plus the origin). However, the planes would be floating/rotating, and the the origin would not be fixed, but it would be somewhere on the datum axis, once you have your secondary (and tertiary) datums in your DRF.

 

[3DDave]

The standard hand waves around the question. It's not " those planes do not have to be fixed", it is those two planes cannot be fixed in orientation; they aren't simply free to move - they cannot be used as a reference at all, defeating calling it a reference frame.