Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations SSS148 on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Compound datum from 2 planes (non-parallel, non-perpendicular) 3

Status
Not open for further replies.

Billykeys

Automotive
Sep 8, 2021
2
I'm trying to agree on a datum system between my engineers and a supplier...
The part has two primary functional planes, unfortunately not perpendicular. Normally one plane would be primary, slide against the second.
What is the proper way to establish a coordinate system from these 2 planes? (The tertiary will be a hole drilled into the A plane, that's more straightforward)

My thought is to establish a plane with the B feature (This is as-cast, I think I would call it Y instead, to align with X-Y-Z as-cast norms), translate it down to A, the line generated from the A-B intersection actually becomes the B datum (rotate the "Y" plane 10° where it intersects A, call that B.

Any thoughts? or what is a good way to use two non-perpendicular planes as datums? (My engineer just wanted to leave it with the A and B as shown on the attachment - It is not possible to have two non-perpendicular planes as the datum system subsequently referenced in feature control frame GD&T, is it?) I saw a reference to using two planes as a "compound datum", but I think that's only when they a parallel, and a 'mid-point' plane becomes a "A-B" datum, but two oblique planes?
 
 https://files.engineering.com/getfile.aspx?folder=8bec5130-94c0-4827-a9f4-ec734fada146&file=bracket_gdt.jpg
Replies continue below

Recommended for you

This is described in the ASME Y14.5 standard - Fig. 4-7 Inclined Datum Features for the 2009 version. The primary plane is used to establish two orientation controls and one location control. The second datum feature plane is intersected with the first datum plane - the second datum plane is located on that intersection line and perpendicular to the primary plane.

The callout for the datum feature B is correct. It has to be applied to the surface or an extension line from the surface.
 
And I would use angularity to "connect" B with A.
 
What is the proper way to establish a coordinate system from these 2 planes?

Just my .02 cents, it often is more useful for me personally to think about constraint of degrees of freedom, than establishing a discrete coordinate system. What you really care about is are your datum features accurate, and do they constrain your part in a manner that is repeatable and reflects design intent (there may be some other concerns such as ease of inspection and fixturing - a purist would say thats your quality/production department's problem, unfortunately we live in the real world with real concerns like inspection/production costs). The only time a coordinate system becomes of personal concern to me is when defining a customized DRF, and thats a whole other complex topic.
 
Thanks all. So my "physical interpretation" would be to have a block with a flat plane that would rest on -A-. In most GD&T examples, you would then have a 90° plane that you would slide against the secondary datum. In this case, that second surface would be at 100° instead, you'd slide it against the -B- feature, but the -B- datum plane itself would be an 'imaginary' plane perpendicular to -A- extending through the intersection corner of my -A- face and the 100° plane, yes? We then have a throat opening which makes sense as a centering feature as the tertiary...

**greenimi - <use angularity to "connect" B with A> -- I was considering using a profile of a surface to define that -B- feature. This would encompass it's true angle as well as form (flatness) right?
 
Why not just define the angle between the two datum faces A nad B as basic and apply profile tolerance with no datum feature references and with 2X prefix to the faces? Then, any other feature on the drawing that would require a control relative to that compound datum would be controlled relative to A-B.
 
pmarc,

Are you saying that fig 7-27/2018 approach could be used? Otherwise stated, instead of "zero basic implied" (as shown in the figure from the standard) will be a "basic angle" between A and B.

 
Pmarc,

Thank you for your answer. Just to make sure I understood you correctly: are you proposing to keep "my angularity" or remove it?

greenimi said:
greenimi (Mechanical)8 Sep 21 18:27
And I would use angularity to "connect" B with A.

 
greenimi,
I was thinking of having just profile.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor