Centerlines needed or not? 3

[kingsboy55]

Hi, I would like to know if there is any definitive method for whether or not centerlines should be shown when using features of size.
The attached file shows a simple angle bracket with two slots in one leg, positioned using implied symmetry to datum A (the length of the bracket) and one slot in the other leg, similarly positioned to A. I sketched red centerlines in to represent the plane of datum A. Some would argue that the red centerline is meaningless and just adds clutter.
The same thing applies to the angle leg with one slot in it, should the centerline be shown through the 2" width of the leg or just locally through the width of the single slot?
We often have animated discussions about this type of stuff, for the record I want to omit the lines in red!

Thanks.

http://www.eng-tips.com/

P. S. As an old guy myself I really enjoyed this thread about centerlines, I can so relate to making holes in paper, chisel points etc...thread1103-248576

 

[dgallup]

I see no need for the red centerlines.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[SwinnyGG]

In my opinion, centerlines are only necessary if you are dimensioning to them. Otherwise they are just clutter.

 

[KENAT]

While I tend to favor 'less is more' on this kind of annotation this is a situation where I could happily go either way.

Fundamentally I don't think the CL are required but for 'implied' symmetry (enforced by applicable FCF) then it can be useful - more so on a more complex part.

Posting guidelines faq731-376

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

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

 

[SwinnyGG]

I see your point on symmetry- but isn't symmetry implied unless dimensioned/otherwise specified?

It is definitely a point that is arguable either way.

 

[drawoh]

kingsboy55,

If I were checking that drawing, I would accept with or without the centre-line. If it were my drawing, there would be a centre-line. It explicitly shows design intent.

--
JHG

 

[CheckerHater]

If it was my drawing, it would have a center line AND I would dimension from center line.

There is no such thing as "implied symmetry"

"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

 

[looslib]

Centerlines on a part are theoretical and do not exist. The dimensioning should come from a hard edge that can be established and controlled.


"Wildfires are dangerous, hard to control, and economically catastrophic."

Ben Loosli

 

[CheckerHater]

@looslib: do you mean we have to dimension to the edge of the hole to establish position?

"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

 

[JNieman]

I agree with KENAT's post, as well as drawoh mentioning the 'design intent' shown by the centerline, even if it isn't /directly/ dimensioned to/from.

I disagree with looslib regarding all dimensions coming from a hard edge. Not important. Centerlines of 'features of size' are INTANGIBLE, certainly, but not impractical to establish and work from, whether the work is manufacturing of inspecting.

 

[kingsboy55]

@CheckerHater, with regard to a dimension from the centerline - what about the leg of the angle with only one slot in it? That slot is centred on the plane established by datum A so the dimension, if there was one, would be zero which is a no-no. The center of that slot has to be 'implied / assumed' congruent with plane of datum A within the tolerance band.
On the angle leg with two slots I could show a centerline and a 1" dimension from the centerline to one of the slots but who is to say what that centerline represents? Is it the plane of datum A (that is implied) or is it an imaginary mid-way 'line' between the two slots? If there were other features on this part symmetrical about datum A, any indicated centerline could also be interpreted as being the centerline of any of those symmetrical features.

I could move datum A to be one side of the bracket instead of centred on the 3.0 dimension and put a dimension from that side to one of the slots and a dimension to the center of the one slot but I don't want to!

Thanks for all the replies and comments.

 

[SwinnyGG]

I could move datum A to be one side of the bracket instead of centred on the 3.0 dimension and put a dimension from that side to one of the slots and a dimension to the center of the one slot but I don't want to!

Why, specifically, don't you want to?

This will probably catch some flak, but in my opinion a centerline should not be used as a datum feature except as an absolute last resort.. datums in the GD&T world are only useful if they are surfaces which can be used for gauging. That's the point. Gauging from the centerline is much more difficult than from an edge or face of the part.

 

[kingsboy55]

@jgKRI - I don't want to use an edge datum in order to keep the slots central to the width. The width tolerance is +/-.08" and with the positional tolerance on the slot we want to make sure it looks tidy. We would use edge datums if the features were not symmetrical but we've had parts made that look bad when they're centrally offset that much. That's the only reason really.
Thanks.

 

[drawoh]

kingsboy55,

As per ASME Y14.5, you cannot use a centre-line as a datum feature. You must call up real features.

You can call up the width as your datum feature, but it is a feature of size (FOS). You have to account for your width tolerance. This means calling up the datum at MMC/B, or having some procedure for fixturing your part centred. I try to avoid using inaccurate FOSs as datum features.

--
JHG

 

[KENAT]

Wow, at least 3 posters above I disagree with, one of them got starred.

While it's not 100% stated as I recall Y14.5 does give examples where hole patterns centered on a part are not dimensioned 'to the center'. So long as the drawing clearly expresses features are nominally centered/aligned (i.e. their 'basic' location is centered) and then have a tolerance on that level of symmetry/alignment then you can get away with implied symmetry. The key point is having a tolerance on how 'symmetric' which using position tolerance with appropriate datum scheme does. The usual problem with 'implied' symmetry or alignment is the lack of tolerance on it.

Center-planes derived from real features e.g. 'width' are often functionally appropriate and fully supported by ASME Y14.5.

Posting guidelines faq731-376

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

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

 

[SwinnyGG]

I understand cosmetics being a factor- I deal with that sometimes too.

But the problem remains- you're calling out Datums A, B, and C as faces on the part, which makes fixturing and gauging easy- then you have two centerlines on the drawing which have no dimensions on them or referencing them..

So in this case, what do you gain by putting centerlines on this drawing?

If I was designing this part and less-than-ideal appearance was a concern because of the wide tolerance on part OAL causing the slot(s) to not be centered properly, I would just tighten the OAL tolerance and dimension from the edges.

If you're cutting this piece of angle from stock, which I have to assume you are, .080" is a very loose tolerance.

 

[KENAT]

jgKRI, datum A is not the face of the part in the OP's sketch.

Datum A is the center plane derived from the 3.0 width.

The OP's tolerance scheme is technically valid, whether if is functionally appropriate I don't know for their application.

Posting guidelines faq731-376

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

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

 

[kingsboy55]

@jgKRI - hi, datums B and C are on the faces of the part but datum A is established the centre plane of the 3" dimension, whether that dimension is actually 3.08" or 2.920" does not matter, datum A plane is at the center of that actual dimension hence the visual aspect of my comment above. I do not gain anything by putting the red centerlines on the drawing as they are are not dimensioned to (neither should they be) or referenced in any way, they are just clutter. This is why, per my original post, I do not want to put them on the drawing but some here say they should be present and I wondered what the community thought about it. The two red centerlines are of course the same representation of plane of datum A, just in different views.

 

[SwinnyGG]

Ugh. Whenever I see that centerplane as datum callout I forget it exists because I dislike it so much.

I understand the chain of logic but still think they are unnecessary.

I would echo drawoh's comment and say that if I was checking that drawing I would approve it either way, but if it was my drawing the red centerlines would be omitted.

 

[drawoh]

kingsboy55,

So far, I have been responding to this on whether or not the centre-line is appropriate.

You do not have a good datum scheme. Using a sloppy FOS as your primary datum feature is not good, functionally or from a fabrication and inspection point of view. Your primary datum feature generally should be your primary mount face.

Without knowing the function of your part, I would guess that the angle long face would be a better primary datum feature, and the short face a better secondary datum feature. I would use one of your vertical slots as my tertiary datum, and I would apply sloppy profile tolerances to the sides. A centre-line still would explicitly indicate that the side faces are centred on the slots. If you really, really want symmetry, you could apply a sloppy width tolerance, and an accurate position tolerance.

--
JHG