Implied Centerlines/Symmetry

[KENAT]

This topic has come up before most recently in thread1103-239885, but on a quick look in ASME Y14.5M-1994 & a search of this site I can't find the text/section in the standard that says explicitly when using datum feature center planes you get implied symmetry/centerlines.

It's shown in figure 5-4 amongst others but I'd like to refer to the actuall text that says when you show a pattern or similar centered to "datum center planes" you don't have to add an extra dimension centering it.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

 

[DesignBiz]

ASME 1994 Y14.5m does reference ASME Y14.2m 1992 (Line conventions and lettering). One of the references is in regard to symmetry. I don’t have a copy of that standard and was wondering if the centerline on a drawing which is common to multiple features is covered in the Y14.2m standard. Does anyone have access to this standard?

I would note that I as I read thru this thread, I did wince a time or two. At the beginning there were comments as to the importance of using correct terminology, however as I continued to read there were comments that made me wonder if the distinction between "symmetry" and "symmetrical relationship" as defined by the standard are being used correctly. A few comments (particuarly referencing fig 5-4) would be referring to a positional callout in regard to "symmetrical relationships". Others like the one referring to why board drawings only were delineated with half of the part, lead me to think of a "symmetry" callout.

It seems as if the "shared centerline" on a drawing may have highlighted that when discussing such a callout or drawing delineation, that we should be careful to distinguish whether the design intent requires a "symmetry" or "symmetrical relationship" callout. Clearly there is a distinction made by the standard.

(Axym,
I used the word centerline because the standard also uses this word when referring to the delineation on a drawing. I understand that in the realm of the physical part we would be referring to center planes and/or axes. Sorry if I was premature in presuming that you would comment on this particular terminology, however I am betting it would at least come to your mind )


DesignBiz

"Quality is in the details"

 

[drawoh]

MintJulep,

This attached file shows what I meant. I just slapped it together. It is not a real part.

Presumably, this is a cover that is inserted into a pocket, and retained with M6 screws. The M6 screws are positioned to a Ø0.3mm diameter, with a little perpendularity error. It is symmetric across the vertical centre. Otherwise, the fit inside the pocket is not critical.

I have shown a side dimension as a reference. Ordinarily, I would not do this. Because the outside width is sloppy, I have not used it as a datum. The side with the 10mm dimension could be 3mm inside the MMC state, and the opposite side could be close to MMC. The reference dimension is potentially not very useful.

JHG

 

[ringster]

Drawoh,

If I were to receive this drawing for a check, my initial reaction woud be to send back and say "lets try again".

Somehow it seems disconnected.

 

[MintJulep]

drawoh,

I don't see how the hole pattern is located or controlled wrt anything in the left/right direction.

Each individual hole is located wrt to other holes, but the pattern is entirely unlocated and uncontrolled.

 

[DesignBiz]

Datum B being the pattern of holes is what the part profile is centered about. It is my understanding that the DRF origin would be at the center of the hole pattern. In this case, a vertical center plane at half the basic dimension 80 and a horizontal center plane at half the basic dimension (100-10). The profile callout would center the part profile around the hole pattern.

As mentioned the drawing needs modifcation to pass check, however I don’t believe Datum C is needed in this dimensioning scheme. A major element of the profile callout dimensioning would be to tie the basic profile dimensions back to the hole pattern. The pattern as Datum B along with the surface as Datum A locks down all DOFs.

See attached file for DRF.


DesignBiz

"Quality is in the details"

 

[CheckerRon]

1. When I use symmetrical datums (feature of size) as in Fig. 5-4, I am perfectly comfortable with not using 1/2 dimensions off a centerline of athe symmetrical datum. They are redundant.

2. Thanks, Mechnorth for that reference to 1.4(k) in the new 2009 spec. This really does clarify it. I am, however, contracturally tied to the 1994 spec for now.

3. Drawoh: Like ringster, I have an immediate problem with your testing.pdf example. If I got a drawing with that dimensioning method to check, my reaction would be, "now let me see...there's got to be a better way to do this".

 

[ringster]

DesignBiz,

i think you need a review of your graphics. I do not believe that an axis will provide you with 2 datum planes B as you have indicated.

 

[DesignBiz]

If one doesn’t use "half dimensions" to derive the origin of a DRF from a pattern of features that are used as a datum, then how do you arrive at the DRF origin from which all dimensions are measured from (ref: pg 73 4.2.2.1)in regard to callouts referencing that particular DRF?

How do you reconcile 4.5.8 Pattern of Features if you do not determine the axis of the feature pattern which is established by "all of the holes" in regard to the DRF axis if half dimensions(in this case)are not used?

Ringster,

I believe my graphis are correct. Yes an axis is at the intersection of 3 planes which define a DRF, however the definition of a Datum Reference Frame is 3 mutually perpendicular planes. Again pg 73 4.2.2.1. No planes, no DRF.



DesignBiz

"Quality is in the details"

 

[fcsuper]

The half-dims are needed to provide the complete specification, otherwise the pattern floats without control. The only "centerlines" that are mentioned in Y14.5 are datums that use the full size of a feature to determine its location. In this case, it is not "implied"; it is explicit since it is directly stated in the standard.

Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group

 

[drawoh]

CheckerRon,

Would a vertical centre line have made it work for you?

My actual point is that I do not want to use a sloppy feature of size as a datum. Locating DesignBiz's centreline B-B can only be done using the width, which I have shown with an admittedly exaggerated sloppy tolerance of +0/-6mm. There is no vertical symmetry, so the location of the horizontal plane is not critical.

I could have set datum[ ]C as the bottom edge, which would have provided the same clocking as the second hole. The bottom edge still is a sloppy feature.

JHG

 

[ringster]

Kenat,

Where are you when we need you? Can you either justify or clarify this statement? Yes, an axis is at the intersection of 3 planes which define a DRF, however the definition of a Datum Reference Frame is 3 mutually perpendicular planes.

 

[KENAT]

Ringster, I'm right here, I was contemplating posting my own diagram of the couple of cases where what I'm talking about in my OP is, in my opinion, justified/correct. However, it seems this thread has deviated a bit from my original intent and I wasn't sure it was worth the effort to try and bring it back onto my intended track.

I just looked again through and I don't see where I made any such statement to justify, that was you responding to Design Biz.

Neither Drawoh's or Design Biz's illustrations are classic, simple examples of what I'm talking about.

Drawoh's looks more along the lines of figure 4-8. Assuming B is meant to be the single hole not the pattern I don't see the obvious symmetry/common centerline. I can see the part is nominally symmetrical but it's a bit too subtle for my liking and combined with potential confusion over if B is the individual hole or a pattern I probably wouldn't allow it.

Design Biz is to me a, perhaps understandable, misinterpretation of Drawoh's sketch. That said,it looks like it follows the idea of hole pattern identifide as a datum fig 4-22. Figure 4-22 shows how 2 planes can be developed from the 'B' axis.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

What is Engineering anyway: faq1088-1484

 

[drawoh]

KENAT,

My assumption is that a centreline does not exist on an actual part. To inspect the part, you must fixture to a datum feature. The datum feature may or may not define a hypothetical centre. The symmetry is nothing more than the nominal state of the part.

On a shaft with a precision diameter, you have a feature that reliably defines a centre, and that you can fixture to.

On my example, the centre is defined by an inaccurate feature, the outline. This is equivalent to a sheet metal base with four folded edges. My intent was to make the part symmetrical. On the actual part, if the holes line up with the mating holes, and the outline fits in a pocket, we are done. I arbitrarily picked two holes as my secondary and tertiary datums, to control position and rotation. I am looking at Figure[ ]4.22 just now, and it does show what I was trying to do.

On my outline profile, I ought to have indicated datums[ ]B and[ ]C at MMC. If I had done this, the inspection GO[ ]gauge would be a pocket with six pins inside, all features representing MMC. I can inspect LMC with NOGO[ ]gauges for the holes, and a NOGO[ ]shim to be used all around the outside.

All the symmetry does is simplify the drawing. There is no need for the inspector to test for it. As per my drawing, the outline is allowed to be 1.5mm off centre.

JHG

 

[DesignBiz]

Ringster,

I re-read my post and might not have worded it the best way possible. I should have said that there are 3 axes at the intersections of the 3 planes of a DRF. I did not mean to imply that all 3 planes defined 1 axis.


CheckerRon,

If your comments were in reply to my jpg regarding a vertical centerline, then I should say that I only posted a jpg for the DRF to support the jpg of the drawing that you posted. You said the drawing was lacking and this was hopefully to clarify to MintJuleps response when he replied to your drawing where he didnt see the centering control from left to right.

Kenat,

I posted earlier regarding symmetry and symmetrical relationships which the standard clearly makes a distinction here. In DrawOh's jpg which is a clear case of symmetrical relationships using position and profile to center the part profile to the "hole pattern". The symmetry symbol is not used in this case, and if it where used can only be used at RFS. Symmetry is inspected differently as well.

The DRF demonstrated in my jpg is correct and is similar to an axis thru a cylindrical feature, however it is oriented in this case by the implied geometry and the dimensions. I have listed the refereneces for what is stated in the standard regarding datums and a pattern as a datum. There is no opinion here it is simply what the standard defines as the datum. The axis of a pattern is established by all of the holes collectively. This axis is the datum and is intersected by 2 planes according to the standard.

I dont really understand where "opinion" is required here. This is defined very clearly in the Datum section of the standard. If opinions are equal to stated standard definitions, then there probably isnt anything more that can be discussed.

If I am in error then I would gladly accept correction if i were directed to specific statements in the standard. If we go off into the realm of opinion when it comes to the basic concepts and definitions of the standard then I dont see any constructive value in it.

I hopefully do not come across as angry, however it is disheartening when opinions are regarded equal to the written standard. Of course opinion on "how to apply" the standard is obviously neccessary.

DesignBiz

"Quality is in the details"

 

[ringster]

Our discussions sometimes, often perhaps, get misdirected by carelessness, in not thinking through the the proper terminology. For example LINE versus AXIS. Line verses Line segment. Datum versus datum feature. And one of my favs is establishes, with regards to the mutually perpendicular planes.

If more thought were applied to the content, perhaps, just perhaps, we might have a better understanding of the intent
of posts in this forum.

Design,

I dont think you came across as angry.

I hope I have thought this one out.

 

[KENAT]

DesignBiz, it wasn't entirely clear to me if Drawoh was using the pattern of holes or the one hole as secondary and the "C" as tertiary. Even from his response I'm still not entirely sure.

This was the basis for my post.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

What is Engineering anyway: faq1088-1484

 

[DesignBiz]

Thank you KENAT,

I understand. I looked at it from the context of this thread that the pattern (Datum B) was the the Datum used to center the profile. In that case Datum C was not neccessary to me.

In addition I re-read your post regarding the jpg i posted and believe we are on the same page. Appologies for mis-reading you. Ringster has commented regarding miss use of terms, not reading a post closely, etc; and I agree with him.

DrawOh,

According to the standard, one can not use MMC to define symmetry.

Ringster,
I agree with what you wrote, regarding accuracy of posts and comments to them.

I am a little curious regarding, "And one of my favs is establishes, with regards to the mutually perpendicular planes". I'm not quite sure I understand your point.


DesignBiz

"Quality is in the details"

 

[drawoh]

KENAT,

I used the bottom left hand hole as datum[ ]B and the bottom right hand hole as datum[ ]C. It did not occur to me that I could use the whole pattern as a datum. All I was trying to do was use an accurate feature as a datum.

If I had left off datum[ ]C, I would have something close to Figure[ ]4-22, except that I am not sure where the horizontal datum would be. Unless my holes are symmetrical in two axes, I think I prefer to datum my way.

DesignBiz,

On my drawing, symmetry was the nominal state, not the final functional requirement. My datums are not located on the axis of symmetry. With all features at MMC, the part is symmetric. As the part deviates away from MMC, it is potentially not symmetric, but it still works.

I am kind of puzzled at how important symmetry is to people. There is a form control for symmetry, but I cannot imagine a situation where I would use it. At one level, symmetry is a convenient layout for drafting. At another level, a symmetric feature is an appropriate feature-of-size datum. At yet another level, you do not care about the width or the parallelism or the flatness of the two sides, but you specify accurate symmetry.

JHG

 

[KENAT]

Drawoh, that's how I interpreted your sketch originally, which I why I mentioned figure 4-8 which is where using holes for secondary & tertiary is illustrated.

At risk of frustrating Ringster & others through not using quite the right terms, your sketch relied on a 'common center' between the horizontal pairs and the overal width etc. However, to me this wasn't clearly stated or perhaps implied is the right word and so there is in my mind an argument that it's not completely clear.

This thread was really just trying to see if anyone could come up with a decent reference in the standard of why cases like figure 5-4 are OK.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

What is Engineering anyway: faq1088-1484

 

[drawoh]

KENAT,

As I noted above, this is more a matter of format and clarity than of a drafting standard.

I claim that there obviously is only one way to interpret Fig[ ]5-4. The part is symmetric. I have no problems with it.

Now, we move out onto the slippery slope. I have attached a modified version of Fig[ ]5-4 which is less clear, due to an asymmetric feature. I still think there is only one way to interpret this, I consider this to be standards compliant, but there is a good chance of a phone call from the fabricator to verify it.

JHG