Unequal tolerance for a negative nominal

[Belanger]

Here's an off-the-wall question that came up this past week, and since Y14.5 doesn't directly address it, I'd like some opinions...

Suppose a zero/zero baseline is called out in each direction from a tooling hole, and positive and negative locations go out from that 0,0 position. If a designer then adds a +.000/-.010 to one of the negatively designated locations, how is that interpreted?

I guess a picture would help, but I hope you get the gist: Suppose another hole is dimensioned to the left of the 0,0 hole, and has a nominal distance of -2.250. Do you think the above tolerance would yield:

-2.250 max
-2.240 min

or

-2.260 max
-2.250 min

My first response would be that it should be avoided, because of the confusion! But if pressed for a specific answer, I would probably say that the negative sign is merely for direction, so the tolerance is meant to apply to the absolute value, resulting in the first max/min set given above.

Or better yet, tell the designer to add a note "all dimensions are treated as absolute values, whether positive or negative," but that sounds goofy. Thoughts?

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

 

[dingy2]

The negative sign is direction only and the actual specification limits are -2.250 and -2.240. I think that anyone who has worked with drawings for a while should be able to figure this out.

To make it clearer, just place a nominal dimension with a bi-lateral tolerance so, instead of the first option, have a requirement of -2.245 +/-.005.

Dave D.

http://www.qmsi.ca

 

[CheckerHater]

I have very strong feeling that when doing baseline dimensioning "the dimensions coming from the base line are
not labeled positive or negative", but I don't have a copy of 14.5 with me right now.

Will greatly appreciate if somebody could prove or disprove the statement by referencig actual book.

 

[fcsuper]

I honestly don't know why there would be a negative linear dimension on a drawing. It's not like the part is being manufactured in negative space. The only time I can see this happening is on a cooridinate system where 0,0 is somewhere in the middle.

Matt Lorono
Lorono's SolidWorks Resources & SolidWorks Legion

http://groups.yahoo.com/group/solidworks & http://twitter.com/fcsuper

 

[Belanger]

Yes, Matt, that's what they did, despite the confusion they caused. (Instead of placing 0,0 at the corner of a rectangular part, think of the origin as being a pilot hole in the middle of a component.)

I agree with Dave's response, but my point is that the standard doesn't actually say that. So someone could push the idea that from a purely mathematical stance, a unilateral negative tolerance on a negative nominal dimension pushes the feature farther to the left!


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

 

[SteveMartin]

I could see ending up with negative dimensions in a tabular dimensioning scheme off some manner of central hole.

I don't think there is an unambiguous interpretation of the tolerancing you describe. Whoever drew up this print should have specified the tolerance explicitly (and, if still available should be made to revise it). Failing that, I believe the only recourse of anyone using such a print is to call up whomever they got it from and ask for clarification.

If I HAD to interpret such a drawing without recourse to the responsible party, I would assume negative means 'towards origin'.

 

[KENAT]

Pretty sure this question came up hear some time back.

I think we decided no '-' sign because of the potential confusion.

However, I'm not sure of an explicit directive in Y14.5M-1994

Posting guidelines faq731-376

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

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

 

[fcsuper]

If the negative dimension is a coordinate, then the tolerances would apply off the location where the coordinate lies. You wouldn't be subtracting a negative from a negative. You'd be subtracting from a location that is expressed as a negative where the negative symbol only identifies direction and not an actual negative number. The negative tolerance would still be minus the location itself (the minimum). HOWEVER...

The only statement that supports this in ASME Y14.5-2009 is 2.2(a). The problem with 2.2(a) is that is only applies to Limit Dimensioning, and not Plus and Minus Tolerancing. So, ambiguity is still present. THEREFORE...

Either convert your dimensions to limits or add a statement on the drawing as to how to interpret tolerances on the negative dimensions.

Matt Lorono
Lorono's SolidWorks Resources & SolidWorks Legion

http://groups.yahoo.com/group/solidworks & http://twitter.com/fcsuper

 

[PeterStock]

If I was a vendor forced to accept this dimensioning scheme, I would try for +/- the smaller value. But as that is 0 in this case and not possible, I would produce a set of drawings with limit dimensions based on all the schemes given. I would send them to the customer asking for what one is what he wants. If I did not get an answer I would use +/- half of the larger value (+/-.005 in this case). That way I would have at least some tolerance zone in the range the customer wants, but does not correctly state.

If I was checking the drawing, it would be rejected in a second.


Peter Stockhausen
Senior Design Analyst (Checker)
Infotech Aerospace Services

http://www.infotechpr.net

 

[MechNorth]

All dimensions are to be +ve with ordinate dimensioning, never -ve. There is a precedence / statement in one of the ASME standards, but I don't recall which anymore; one of my former colleagues found it. Logically, a basic dimension can't be negative as it is a theoretically perfect value, not a vector. As an extension, no dimensions are vectors, therefore they do not indicate direction, therefore they cannot be negative. Tolerances are a different story; they are the permissible variance from the dimension.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[drawoh]

Belanger,

I agree with not using the minus sign on your dimensions. It is obvious from the drawing which way the dimension goes.

If there was any ambiguity about a +.000/-.010 tolerance, I would switch to the limit tolerances you provide in your example.

JHG