Coaxial Datum features and their relationship...

[Tenkan]

Question about Coaxial Datum’s and their Features:

What controls the relationship between the two separated cylindrical features that establish the coaxial datum axis? Would the coaxial datum be defined as the simulated axis of the two features regardless of position, if say the two datum features were positioned as far apart as possible within tolerance?


Bonus: can this also apply to a datum center plane established between two separate features? Say using position instead of runout…

reference example attached.


lightweight, cheap, strong... pick 2

 

[fsincox]

There is no definition of possible eccentricity specified, people have allways relied on the processing method for parts like this to produce parts that will be "good enough". This practice is not acceptable ASME, but it is universal, and the battle that those of us who care have fight every day.
Either position or runnout of the datums to the common axis is acceptable practice, I would say prefered, if it mirrors a functional requirement. Position or runout of one to another is possible and just a 2x position to no datum is another option.
Frank

 

[CheckerHater]

Fsincox,
What do you mean it’s not supported by ASME? Clarify please.
Koda, yes, you can create datum plane using 2 separate features.

 

[KENAT]

CheckerHater, I believe fsincox is essentially referencing section 2.7.3 of Y14.5-1994 regarding relationship between individual features. If you don't add some kind of control to the coaxiality then there is no specification of how coaxial the 2 diameters are.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[fsincox]

CH,
It is not supported because they say all features should be defined.
Frank

 

[pmarc]

Koda,
Frank and Kenat are right - there is no relationship between the two cylinders defined on your print.
The relationship can be defined by at least half a dozen of methods, depending on what is really required. See the following link as a reference (Subject: Total Runout Question):

http://www.geotolmeadows.com/newsletters/2012/nov2012.htm

A method not mentioned there has been already mentioned by Frank - "2x position to no datum" as shown in fig. 4-24 of Y14.5-2009.

 

[Tenkan]

Pmarc, in your reference link describes how I could define the relationship. Thank you.

All, good to know I was correct in identifying undefined features.
Sometimes I beat myself up trying to define features that are commonly relied on by the processing method as “good enough” because some of my drawings have been rejected on this basis, and I struggle putting my initials out there on an incomplete drawing. I’m not perfect, and make mistakes but when I learn something I’m moving forward. Its hard when sometimes I’m rejected because an end user might not understand it or they don’t want to add inspection cost to the part….


lightweight, cheap, strong... pick 2

 

[pmarc]

Since the initial question seems to be answered, I would like to sidetrack the discussion a little bit, if you don't mind, without starting a new thread.

I think the question is addressed to Frank, however other votes are also really welcomed, as usual.
Frank agreed that there is no geometrical relationship between datum features marked as A and B on Koda's print. Now, if we go back to the 100+ thread where we debated about backdoor ability of parallism tolerance to control location of a feature with relation to other features just because the feature had been assigned as a datum feature (

http://www.eng-tips.com/viewthread.cfm?qid=315082),

don't you think that following this logic we should say that THERE IS mutual relationship between the cylinders on Koda's drawing? The cylinders are assigned as datum features, so per what some guys claimed in the thread in the link implied basic 0 linear dimension that exists between datum feature simulators A and B defines the relationship.

Any comments?

 

[CheckerHater]

1. Yes, absolutely positively there is basic zero between [A] and
2. Basic dimensions cannot exist without FCF.
3. In absence of FCF basic zero dimension between [A] and is as good as non-existent.

 

[pmarc]

So if B was controlled to A with parallelism callout, would you say that there is basic zero between them?

 

[pmarc]

Ad. #1. What exactly creates this basic zero relationship?

 

[CheckerHater]

What parallelism has to do with basic dimensions?

 

[pmarc]

I do not know (or actually I know what it has to do with basic angular dimensions).
This was the conclusion that came out of the other thread -- if a feature is controlled by parallelism tolerance (or any other orientation tolerance) to A, AND the very same feature has been assigned as datum feature B, there is a locational dependency between A and B because paragraph 4.5.2 of Y14.5-2009 says that datum feature simulators shall have basic location to other datum feature simulators for all the datum references in a feature control frame, unless a translation modifier or movable datum target symbol is specified.

 

[KENAT]

Am I missing something, saying something is parallel is not the same as saying it's coaxial.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[CheckerHater]

I didn’t try to dig that deep.
I was simply trying to make the point that basic dimension is meaningless unless relationship that requires basic dimension is specified. Specify position B wrt A and tell me there is no basic dimension between them

 

[Tenkan]

What I got from Franks reply was that there was no relationship between datum features appearing on the same axis, and that one was required if a coaxial datum is required

lightweight, cheap, strong... pick 2

 

[pmarc]

I wish this was so simple, CH.
I can't wait to see what others will say about it.

 

[drawoh]

Koda94,

A crude assumption about your drawing would be that your datum features are round, but not cylindrical. If this were my design, the cylindrical datums each would be short, and undercut. In this state, your two features would define a centre axis.

If this is close to the proportions of a real design, you could use datum targets to define where on the cylindrical features you would pick up the diameters.

--
JHG

 

[fsincox]

pmarc,
Are you alluding to my keyway "orientation for location" type of thing?
Frank

 

[pmarc]

Yes, I am Frank.
I am not going to give up quickly on this.