Relationship Between Parallelism and Flatness

[[Za...]]

Hello everyone,

My question is, as can be seen at the example figure below, is it possible to check 0.05 parallelism when you have two planes both with 0.05 flatness tolerance? I feel like, in a situation like this, parallelism tolerance should be 2 times flatness value but I could not figure out it.

Thanks in advance.

Parallelism.png

[[Da...]]

Please sign in to view this quote.

If it's possible? Yes it is. Does it make sense? Absolutely not. Parallelism would make sense when its tolerance is bigger than tolerance of flatness.

[[Er...]]

Depending on the manufacturing processes it can be easy or hard to make to specification.
Technically all three dimensions could be at maximum on the same part.
In practical terms the tighter manufacturing holds the flatness the more room they have on parallelism. In the end it depends on what processes are chosen and how the part is set up.

Knowing the manufacturing process and use helps me know what measurement strategy to use.
For instance, a punched hole getting a bolt through it. 🤣

[[Ja...]]

Flatness of .05 on the datum A makes sense. But, if you must be within .05 parallelism on the other side, then you will be within .05 flatness of the other side. The .05 flatness under the .05 parallelism is unnecessary. I would just give them what they want though.

[[Da...]]

Since A is the datum , it should be LSQ. That would remove form errors from A . Flatness of A would be a separate characteristic.
Parallelism is from A (lsq) to H , including the form (Flatness) of H . At the tolerances stated, this would include flatness of H. So if the parallel characteristic of H is good then the flatness of H is by definition also satisfied. I'm sure there is room for debate, but that's the way I see it.
The flatness call out at H could be eliminated and not affect the quality of the part. IMHO

[[Da...]]

Please sign in to view this quote.

I agree LSQ for datum A in ASME, just wanted to point out that ISO requires Chebyshev (minimax).

[[Ma...]]

Another way to look at this is that the flatness spec is unnecessary. Flatness deviations will only be accepted up to the parallelism tolerance anyway.

[[Ja...]]

What?? Maybe I need more coffee to understand what you mean, but I don't think a datum should be LSQ and have form errors removed.

[[Ow...]]

[[Ma...]]

In my opinion, for the flatness evaluation of A "Minimum" fit should be used. An "Outer Tangential" fit should be used for A as the datum for evaluating the parallelism spec.

[[Jo...]]

Please sign in to view this quote.

If ASME it requires Datum simulator or True geometric counterpart for Datum Feature, as Mark described O.T.E . It is customary to qualify each Datum. In the case of the Primary , flatness is limited to .050. It could be a refinement of Feature of size (rule #1) requirement as well.

The flatness of Datum H is qualified as well at .050 .

The orientation of Datum H includes only that feature form error because Datum A is a perfect geometric counterpart.

A secondary Datum could have a perpendicularity requirement to either of those. This part otherwise could be a rhombus. That is why their relationship must be limited by tolerances.

[[Ma...]]

Please sign in to view this quote.

Owen, I don't think that Item 3 in this example is correct, at least in a GDT sense where parallelism includes all the form error.

[[Br...]]

The 0.05 Flatness tolerance applied to the surface with the 0.05 Parallelism tolerance is incorrect. A fundamental principal of GD&T is that Orientation controls From. So if you have an orientation tolerance, any additonal form tolerance should be a refinement. In other words, the flatness tolerance must be smaller that the parallel tolerance(The parallelism tolerance doesn't need to be 2X larger or anything like that.). From an inspector's standpoint, its just sort of redundant. If the parallelism tolerance passes, the flatness tolerance will pass necessarily.