Plus Minus Tolerances

[drawoh]

I am updating notes on my website on dimensions and tolerances. I have a question.

You are inspecting parts fabricated from the attached drawing. The fabricator has barely achieved the specified perpendicularity, i.e. the sides are 89[°] from the base.

How do you measure to the holes -- from the side, or from whatever corner extends the furthest to the outside? Note that I have not called up datums, although I have called up ASME Y14.5-2009.

Am I correct in assuming that the holes at the top are about 2mm out of nominal position?

I did the drawing here at home using QCAD, and I have not figured out how to drag the hole diameters away from the holes. I am not serious about the drawing. I think it illustrates the importance of datums and positional tolerances.

JHG

 

[MintJulep]

I don't see any specified perpendicularity.

 

[drawoh]

MintJulep,

As per paragraph 2.1.1.3 in ASME Y14.5-2009, perpendicularity is implied, and controlled by the tolerance note.

JHG

 

[MintJulep]

How do you measure to the holes?

Exactly.

With no data and no basic dimensions and positional tolerances the inspector is free to measure any which way he wants. To the point sometimes of the way things are measured making the difference between "good" and "bad".

Am I correct in assuming that the holes at the top are about 2mm out of nominal position?

No, I don't think so. That assumption would seem to include assumptions about how the part was fabricated. Just because the "block" may be a rhombus doesn't mean that the hole pattern has to be a rhombus too.

 

[axym]

drawoh,

I think that the use of the term "perpendicularity" in a generic (i.e. non-Y14.5) sense is causing the confusion. MintJulep is correct that there is no specified perpendicularity. As per 2.1.1.3, there is an implied 90 degree angle on the drawing (in fact there are several). The tolerance on this angle is governed by the general angular tolerance note, in this case +/- 1 degree. A plus/minus tolerance on a 90 degree angle is not the same as Perpendicularity. So it would have been more clear to say "the fabricator has barely achieved the specified angular tolerance between the sides and the base". I hate to be the terminology police (well, truth be told, I don't hate it) but in this case it's necessary.

As far as how to measure the holes when the actual part is parallelogram, I don't think there's a right answer. The specification is open to different interpretations.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[desertfox]

Hi drawoh

Well the holes should be checked dimensionally from the edge that you have given dimensions from and the general tolerance of +/-0.1 should apply to them.

desertfox

 

[drawoh]

axym,

A perpendicularity FCF as per ASME Y14.5 specifies a tolerance zone which controls straightness as well. My drawing imposes no such control -- a good case for profile tolerances if there ever was one.

I think we can still assume a 2mm error at the top edge, approximately where the top holes are measured from.

JHG

 

[pmarc]

drawoh,

I think your example is very good for presenting shortcomings and ambiguity of coordinate (plus/minus) dimensioning when considering such applications. As MintJulep and axym said an inspector is free to measure the part any way he wants.
In my opinion to eliminate any assumptions, GD&T (position tolerance) is the only reasonable solution in this case. Of course which features will be assigned as datum features should depend on the function of the part.

 

[fcsuper]

Doesn't Rule #1 come in to play regarding how to approach this at MMC?

Matt Lorono
Lorono's SolidWorks Resources & SolidWorks Legion

 

[drawoh]

Doesn't Rule #1 come in to play regarding how to approach this at MMC?

I don't think so.

When I started my GD&T course, I was under the impression that perpendicular +/- tolerances defined a tolerance zone, controlling profile and perpendicularity. My instructor said "No.". Perpendicularity is controlled by the tolerance note on the drawing.

The width controls only the distance between the two specified faces.

JHG

 

[fsincox]

2.7.3 Relationship between individual features. The limits of size DO NOT control the orientation or location...... bla, bla bla. ASME Y14.5-1994, pg. 27