Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

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

symmetry and position together?

Status
Not open for further replies.

TWJR

Military
Jul 16, 2013
82
I have a fairly unusual situation that I hope to find a resolution for. We have a project going on right now, with some fairly large parts as well as purchased parts of unknown size (until we get them, anyway). The issue is this: in both cases, the engineer simply wants a feature (say, a hole pattern) to be on center. Dimension from one side isn't important, especially since the overall is either very loose or unknown. In the attachment, let's say we want the hole pattern to be on center horizontally, but tolerance is not too close. This pattern has to be a little closer than that from the bottom (datum C), but hole to hole has to be closer than that. Is the feature control frame I constructed legal, or is it totally inappropriate? What I'm trying to say is "symmetric to A within .1, position of hole pattern to C within .030, and hole to hole within .010". If it's not the best way of doing this, what is?
 
 http://files.engineering.com/getfile.aspx?folder=32e410b0-2acd-4b0a-9d80-f078e21c9638&file=symmetry.pdf
Replies continue below

Recommended for you

Not legal -- for now I'll focus on symmetry. The two holes are not symmetric to datum A. The only relationship they have to A is perpendicular. Do you want them to be spaced from one another to the .1 tolerance? Or maybe explain your intention with the symmetry idea.

The other two callouts (position) are also fraught with issues, mostly regarding the datum references and the diameter symbols.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems
 
Oops... made one major mistake which caused you to misinterpret what I'm trying to do. Actually the symmetry is not about A, but B (horizontally). B datum, by definition is center plane, correct?
 
So, what would be the best way to define my goals as stated?
 
Yes, this latest is much better. The distance directly between the holes is the most precise (.010). The centering of those two holes in the left-right direction is .030 relative to center plane B (notice I avoided the word "symmetry"). And the location from C is the loosest tolerance of .100.

A small note... your original post said that you wanted the distance to datum C to be a little tighter than the horizontal position.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems
 
You're right... on second examination I think to express what I want (or the engineer, actually) the middle frame should be "AC" not "AB". Correct me if I'm wrong, but I found out in the meantime that if there is a feature of size, in this case B, and a basic dimension (4.500) is visually centered and a feature control frame references this datum of size, by definition it's on center. I had been taught this, but had not actually seen this stated explicitly in Y14.5M. Perhaps it's there and I missed it.

The tough part here is selling the idea. It's been the practice of a number of folks here to simply throw a center line up on a view with no datums or feature control frames with the implication that something is on center. I'm trying to put forth a better way, but I'm sure I'm going to meet resistance. Not only that, but some of these parts in question are large weldments, and I'll bet anything the average weld shop isn't going to understand it... sigh...
 
Now, the other issue is, what is the case when we're making an altered item drawing, where we want features on center? We don't know what (in this case) the horizontal dimension is, since it we're not making it. Can you have a datum feature of size be a reference dimension, or expressed in some other way denoting we don't know what the value is, but we want the features on center of this datum feature of size?
 
Yes you can derive a datum from a reference feature of size dimension. But make sure to never reference it at MMB or LMB. I do it all the time though there is not an example of this in the standard.

2JL
 
It's been the practice of a number of folks here to simply throw a center line up on a view with no datums or feature control frames with the implication that something is on center.
Well, that's true in a way: common centerlines can indeed imply that things are supposed to be centered. But that says nothing about the allowable tolerance for that centering. (And you can't appeal to a title block tolerance or note.) So it's an incomplete statement.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems
 
Well, that's true in a way: common centerlines can indeed imply that things are supposed to be centered. But that says nothing about the allowable tolerance for that centering. (And you can't appeal to a title block tolerance or note.) So it's an incomplete statement.

Exactly my point. That's what I've been trying to explain to the engineers who do this. Not only that, but depending upon the shape of the part, it may also be ambiguous as to what a centerline is the center of. I'm fighting a difficult battle, though. Battle #2: I'm also confronted with this statement, "well, if we send a drawing with all this GD&T to a weld shop, they're not going to understand it" or, "that will automatically make them quote it higher". Sigh...
 
You could try rewriting the requirements as a note, one that explains the entire dimensioning inspection steps that are required to validate the part. It should duplicate what the symbols are requesting. It would only take a few paragraphs to explain the use of a series of fixed boundaries the holes are not allowed to encroach into.

Best part - no pesky symbols, so no excuse for not understanding or automatically quoting higher.

It still doesn't dodge what is often the real problem - a supplier that doesn't want to be responsible for meeting drawing requirements.
 
3DDave,
That's not a bad idea, although it's not like our vendors are horrible, to me it's a matter of making sure the drawing is clear and can only be interpreted one way. Don't want to get into one of those situations where the drawing could be interpreted more than one way, and then having an issue with a vendor that we can't defend (due to incomplete drawing). Give plenty of tolerance where you can, and hold things down when you need to.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor