Symmetrical Note on the drawing

[var10]

In our manufacturing drawings we usually have the practice of putting centre line symbol on the view and leaving a note on the drawing saying the part is symmetrical about the vertical centre line or horizontal centre line or both. Sometimes we also say that the part is symmetrical about vertical centre line except for feature x for example or the part is symmetrical about vertical and horizontal cl except for feature A & B.

Is it right to do it this way? Thoughts?

Regards,

Varoon

 

[lifttrucks]

Nereth1,

If you dimension and tolerance from an edge to a center line, then dimension and tolerance other features off that center line, wouldn't you just be introducing additional tolerance accumulation vs. just dimensioning all features from that edge? I don't see any added benefit to such a dimensioning scheme.

Also, without additional notes, that theoretical center plane wouldn't necessarily be perfectly flat. It could flex and tilt to match the profile of the surface it is dimensioned from. Inspection would be exceedingly difficult in my opinion.

 

[Nereth1]

Kenat, Lifttrucks,

Lifttruck, I suppose it could cause stack up issues in some cases. The idea in my head was to place your features symmetrical around a point, then define where that point is. So you have an accurately defined set of symmetrical features, but they are allowed to float on the surface as a group, per the tolerance on the centerline. That's a somewhat contrived case for where it might be useful though, admittedly.

A centerline to show symmetry can also be useful to avoid needing to dimension an obviously symmetrical part twice. For example, you may dimension a set of features on the right hand side of the part, from the right hand edge of the part. Then put a centerline in the middle of the part and visibly show the same features on the left, but don't dimension them. Would that be a more reasonable use of a centerline?

 

[lifttrucks]

What you speak of isn't uncommon. I have seen drawings with dimensions on one side and a note stating "SYMMETRICAL ABOUT CL". However, I believe it is still ambiguous. For example, the features on the right hand side are inspected from the right hand edge, are the features on the left hand side inspected from the left hand edge or the right? In other words, just how symmetrical do the features need to be?

Why not select the part's width as a datum feature, establishing a datum center plane, if symmetry is a functional concern?

 

[KENAT]

"A centerline to show symmetry can also be useful to avoid needing to dimension an obviously symmetrical part twice"

WHAT IS THE TOLERANCE?

How far off centerline can the features, or the origin point of dimensions to the features be?

Posting guidelines faq731-376

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

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

 

[Nereth1]

Don't take me as the either an expert or a crusader for center lines here, I've never done it since my first year or so in the industry, and I don't recall how I did it then - but I would imagine that the tolerances for the left side would be the same as for the right. For example, if the right side is toleranced to the right side of the part, then the left side would be toleranced the same amount off the left side. If the right side is toleranced off the centerline, then the left side must meet the same tolerances off the same plane from which the right side is measured.

Doesn't that seem reasonable? How else would it be interpreted?

Edit: Remember I'm not talking about dimensioning over the centerline, but dimensioning to it, or not at all, such that it just indicates 'do the same thing on the other side'.

 

[Belanger]

Doesn't that seem reasonable?

Honestly, no it doesn't.

How else would it be interpreted?

Many ways. What Kenat is saying is this: What if one little feature on that center line is off center by 4 microns? Would the other dozen or so items follow that thing and go off-center too? Or should they stay in line with the center as derived from the outside edges?

This is a classic GD&T problem. The issue is not the dimensioning, but rather the tolerancing. (Read up on "datum" vs. "datum feature" and it may help.)

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

 

[Nereth1]

I see your point (I think).

I was thinking of it as 'Define a symmetry plane here, then measure off it for the rest'.

But on a drawing, without a note to indicate otherwise, the symmetry plane is based off all symmetrical features. Which is fine, until the real world gets in the way and due to manufacturing variation, all the features have different symmetry planes.