Perpendicularity tolerance as refinment of Concentricity tolernce 2

[waqasmalik]

Is perpendicularity tolerance is a refinement of concentricity and symmetry tolerance?
To me, perpendicularity tolerance is a refinement of position tolerance, run out and profile of surface.

 

[pmarc]

Evan,
While I also don't see anything conceptually wrong in |CON|Dia 0.1|A|B|, I am afraid the committee would not be comfortable with it - at least per what para. 7.6.4.1 of Y14.5-2009 states. It the middle of this paragraph it says: "The specified tolerance can only apply on an RFS basis, and the datum reference can only apply on an RMB basis." No "the datum references", just "the datum reference".

CH,
The scenario proposed by Evan would be perfectly legal per ISO (see following link:

http://www.eng-tips.com/viewthread.cfm?qid=354287).

This is the very same configuration I tried to describe in the last sentence of my reply from 19 Nov 13 15:49. Then applying perpendicularity to A (using Evan's datum features nomenclature) in addition to the concentricity callout would be perfectly legal. You are right though, in ISO this is called "coaxiality" (since it is 3D case), and not "concentricity" (2D case).

 

[CheckerHater]

pmarc,
Given that ISO concentricity/coaxiality is very close mathematically to Position RFS, I agree Evan's set-up will work.

I was more concerned about ewh's example. His post from 11:55 shed some light though. I just still think that in ewh's example specifying larger perpendicularity would not be completely illegal, but definitely useless.
So we can agree that we will have some sort of "quasi-refinement"

 

[fsincox]

CH,
I am with you. I think in the ISO world all bets are off! It is a different world! they are just using it to say: "hey the dimension is zero, not missing"
Frank

 

[axym]

Hi All,

What if we omitted the reference to datum feature B, and just kept the reference to planar datum feature A? Like this:

CON|Dia 0.1|A

What do you think of that?

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[fsincox]

Even to me your case takes a leep of faith, but my bet is, In the ISO wold it would be fine. Kind of like orientation controlling position (spacing?). I should have done my keyway in ISO!!
Frank

 

[pmarc]

That is too much for me, Evan.

Frank, even for ISO that would be too much, I believe. When you take a look at definition of concentricity in ISO 1101, it requires a datum axis.

 

[axym]

pmarc,

I knew it would be too much, but I wondered what reasons people would give as to why. Really, I'm just stirring the muck here.

How about making the toleranced feature datum feature B, and referencing B in the Concentricity FCF? The feature would have to be concentric to its own oriented AME. Technically, the requirement for a datum axis is satisfied.

I'm not suggesting that anyone do this in practice, but I wonder if it violates any rules.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[fsincox]

Wait, How about if done in a case like an "interuppted diameter"!! I think we purposefully make these evamples too simple for our purpose, and that the real world is filled with parts that are not that way. It has to work in the real world or it is really useless to us.
Frank

 

[pmarc]

Evan,
In my opinion making the toleranced feature datum feature B, and referencing B in the Concentricity FCF does not violate any rules.

Frank,
Could you clarify your concern about an "interrupted diameter"? I am afraid I am not following. Thanks.

 

[axym]

Frank,

I'm not sure what you mean about the case of an interrupted diameter. Do you mean if Concentricity was applied to a cylindrical surface with an interruption in it? Or something with the <CF> modifier? Or something to do with the datum feature being different if the feature was interrupted (the A-B thing) ? Please clarify.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[fsincox]

Evan,
Sorry, I meant a plain perpendicular datum surfaces referenced concentric with (2) diameters separated by a groove. A situation one might now use "CF", but since ISO, so "CZ" would be more likely.
Frank

 

[fsincox]

Evan,
I am getting the two threads mixed togeather here, sorry.
Frank

 

[desertfox]

Hi

Concentricity is not a refinement of perpendicularity, Concentricity controls the centreline of two features relative to each other.
So if you imagine you weld to cylinders together at right angles they could have a perpendicular tolerance but would not need a Concentricity tolerance.

Desertfox

 

[CheckerHater]

Actually this thread was about Perpendicularity being a refinement of concentricity, which is (kind of) the other way around.

 

[CheckerHater]

And I guess, we agreed that, although it's not the "real" refinement as tolerances use different datum references, perpendicularity may narrow the area where the feature is allowed to reside, hence - the refinement.

 

[asmeY145Manswer]

Gentlemen,

Is it possible that a cylindrical feature could be cylindrical wrt a datum axis established by that datum feature at say MMC,within a controlled tolerance, but still not be perpendicular to a plane, or primary datum A, let's say? So what then would control the perpendicularity of that datum axis to A?

And imagine now that this cylinder further, needs a projected tolerance zone such that it can mate with a plate that has a hole of sufficiently toleranced diameter!

But the mating direction is specified perpendicular to datum A? Do you see a problem with this?

Just asking!

 

[pmarc]

"cylindrical wrt a datum axis"? Cylindricity never uses datum references.

Could you clarify? I am afraid I am not following you.

 

[asmeY145Manswer]

True.

But let's say that the cylinder is a secondary datum feature B. And you only specify a cylindricity for that cylinder surface.

In other words, imagine this cylinder feature is protruding from a flat plate which represents my primary datum A.

What's to say that it doesn't protrude out at some obtuse angel from the plate, which it will do in reality?

There is noting yet that controls the perpendicularity of Datum axis B to Datum A.

Make sense?

 

[asmeY145Manswer]

And my other question is - why did you specify a cylindricity, when a total runout might have worked just as well, and it would have been much easier to measure with a simple dial indicator.

 

[asmeY145Manswer]

Or, you could have simply specified a positional tolerance for the feature, which would have made it far easier to measure with a hard guage!