Positional tolerance: bi-direction

[Evom]

The question I have is regarding the datum setup and the interpretation of tolerance zones established by them for bi-directional position tolerance frame setup. Attached is the illustrations from ISO 1101:2004 version of standards. ASME seems to have the same interpretation in regards to bi-directional position tolerances.

Typically at my company we use positional tolerance in it's full capacity with a diametrical symbol for a cylindrical tolerance zone when applicable for a cylindrical feature of size but it's eating me up not being able to fully grasp what seems to be a simple answer that I'm missing for this scenario.

Most of the examples I see have 3 datums setup for bi-directional. I understand the tolerance frame setup for cylindrical tolerance zones and 3 datums in both bi-direction..but it seems that using only 2 datums for one of the directions is interpreted the same according to the tolerance zone definition so I'm a bit confused. Is 3 datums necessary when 2 theoretically seem that they would do where one controls the orientation and the other the location with respect to the referenced datum? Or like some orientation tolerances, is the first 2 for establishing the tolerance zone and its orientation with respect to the 2nd datum and the 3rd locates it?

Been thinking on this subject far longer than I should be...

 

[CheckerHater]

Think of your datum reference frame as order of fixturing: lay the part on [C] first, lean against [A] second and against {B} last.

Should you lean against first, your results may be different.
Or, leaning against [A] to take measurement from {A} and then leaning against to take measurement from will bring different result from fixturing against [C|A|B], and THEN taking measurements from [A] and

Confusing?

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

 

[Evom]

No, what you stated makes perfect sense.

I understand the sequence of datums, just had my mind in overdrive and overlooked that very point you made

It's one of those scenarios where your subconscious knows the answer yet can't bring it out.

Thanks for that viewpoint.

 

[Evom]

One more additional question to the top scenario...

A lot other examples I see have datum B to the one control lacking it; how does this change or add anything different that already isn't being controlled with C & A?

I can't see that theoretically serving any purpose as the median line in the direction the dimension is addressing wouldn't gain anything from banking up against a datum B simulator since it isn't controlling location with respect to datum B in that scenario.

 

[CheckerHater]

On the picture shown on your OP imagine part sliding along [A]. Measurement from [A] will still the same, is not relevant, as we don't measure from .

From degrees of freedom point of view, part is not fully constrained, but it still possible to measure.

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

 

[LCform]

I wish someone would explain that order of the datums more clearly and bring it out of subconscious also here

 

[CheckerHater]

I wish someone would explain that order of the datums more clearly and bring it out of subconscious also here

I already recommended this book so many times on this forum, ISO should pay me commission:

http://www.iso.org/iso/publication_item.html?pid=PUB100296

(And no, I am not working for them)

The book explains fundamentals like datums,degrees of freedom, dependency between size and geometric controls, etc., etc.

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