Direct Tolerancing Interpretation 1

Keith1029 · Jun 13, 2013

Hi All,
I have a question about direct (+/-) tolerancing and its interpretation (ASME Y14.5-2009). The basic question can be phrased as:

Does direct tolerancing on 2 planes mean that every point on the 2 surfaces must lie within the specified perpendicular distance from each other or only that every point on one surface must lie within the specified tolerance from its counter point along the axis of measure?

I have always been of the belief that it is the former and my reasoning for this follows, but I cannot find the section in the standard that states this completely (other than rule 1).
Attached is a quick sketch showing the problem in its simplest form with a +/- tolerance on the length of a shaft. Given the drawing with case 1 dimension scheme, my interpretation for the functional limits of the part is:
1. The outer diameter establishes a datum axis.
2. The perpendicularity callout establishes a .002 wide tolerance zone perfectly perpendicular to the datum axis A that every point on the surface must lie within.
3. According to Rule 1, all the points on the face of the opposite surface must lie within the maximum envelope established by perfect form at MMC.
4. All the points on the second surface must also lie within a minimum envelope established by perfect form at LMC?

The two drawings for case 1 show a set of simple deviation possibilities.
In the first drawing I am of the mind that the surfaces have deviated as much as possible to still generate an allowable part; if any part of the second face extended past the maximum zone or did not reach the minimum zone this part would be out of specification correct?
For the second drawing, although every set of points on the faces lie within the direct limits of measure from each other along the axis of measure, some points on the face exceed the perfect form at MMC envelope established by rule 1 so this part is definitely out of specification correct?

Finally we come to measurement. Suppose we have a part that is produced like the section drawing in the second attachment. Assuming my interpretation was correct, this part is out of specification and should be rejected. If it is measured with a caliper or surface plate and height gauge as the only measurement, it could be accepted. Measurement by CMM and taking the distance between points along the axis, or measurement on the caliper and surface plate combined with a measure of the flatness of each face would be the accurate way to measure and detect non-conformance in this manner; correct?

Thanks

Keith1029 · Jun 13, 2013

Second attachment

SeasonLee · Jun 13, 2013

File name need to be changed (too many dot) on the 1st attached.

Season

Keith1029 · Jun 13, 2013

Re-posted with shorter file name

powerhound · Jun 13, 2013

For cases 1 and 2 in the top drawing: Datum feature B must be oriented perpendicular to datum axis A in the way you described. The 3" measurement is an independent measurement, unrelated to any datum reference frame. So a simple caliper measurement is sufficient. Your drawing for case 1 (middle drawing) is not correct. The left end of the part will be flat on a surface plate, not held suspended as you are showing.

For case 3 the same thing applies for the perpendicularity callout. The opposite end of datum B is now related to a datum frame though so basically it must be perpendicular to datum A within .010" and no part of the surface can be less than 2.995" or more than 3.005" Also, since case 3 makes use of basic dimensions and GD&T to control size, rule #1 does not apply.

John Acosta, GDTP S-0731
Engineering Technician
Inventor 2013
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

Keith1029 · Jun 14, 2013

If part of a more complex piece I can see a situation where the most convenient way to measure a dimension like this would be with a CMM suspended as shown, but I think what you are saying with the surface plate is that the axis of measure should be perpendicular to the tangent plane of one of the faces and not the initial datum axis A or the theoretical perfect form of one of the faces? Does the standard reference this directly anywhere? Although the original question still to address is:
1. Does each point on face 1 need to be within the distance indicated (along the axis of measure) from each point on surface 2 OR does each point on surface 1 need only be within the specified distance from its direct counterpoint along the axis of measure? (Section 2.7 preamble suggests, to me, that one of these options is correct but if there is one that I am missing please indicate.)
A new second question becomes:
2. What is the correct axis of measure? The three options that I see would be perpendicular to the tangent plane of one of the faces and perpendicular to the theoretical perfect plane of the face.
For now the graphics and I only refer to the dimensioning scheme 1, not the origin or basic dimension (we could tackle the origin dimension in terms of the difference, although I think I am much more clear on the interpretation).
Thanks.

Keith1029 · Jun 14, 2013

For any who are interested, I *may* have answered my own question. Section 2.6.1 of the standard and figure 2.5 lead me to believe that the answer to my question is:
A) Either of the faces can be selected as the origin face and the dimension shall extend from and be perpendicular to the tangent plane of that face
B) ALL the points on the opposing face will lie within the limits proscribed by the direct tolerance along the axis of measurement

This is the best example that I can find in the standard, but to me it means that as long as you have 3 points to establish a tangent plane on one of the surfaces (and the other is flat) your minimum dimension probe dimension is controlled by the flatness of the selected datum face.

pmarc · Jun 16, 2013

Keith1029,
I'd like to throw in my $0.02 to the discussion:

Case #1
While it looks like the easiest way of defining size (length) of the pin, it can easily turn into real can of worms when one wants to analyze direction in which measurements for that size should be taken. It is all because neither Y14.5-2009 nor associated math standard Y14.5.1-1994(R2012) give unambiguous definition of what actual local two-point size is and how it should be determined. Saying that actual local size is "the measured value of any individual distance at any cross section of a feature of size" is so muddy, that I am pretty sure there is no muddier definition in the whole standard.

That being said, the only thing I would take for granted is that such drawing requirement, together with default Rule #1, clearly defines an envelope of 3.005 width that the feature of size must pass through. The envelope, however, is unrelated to any datums shown on the drawing, therefore it actually does not matter how much out of perpendicularity the left side of the part is. Two surfaces must pass through a 3.005 wide gap - that is all.

Case #2
This scheme is different than #1 in that it clearly indicates direction of measurements. Per the standard (fig. 2-5), each portion of right face of the part must fall within two parallel planes that are located 2.995 and 3.005 from tangent plane constructed by highest points of left face of the part. This pair of planes must always be parallel to the tangent plane. I think it is easy to visualize that the direction of measurements, though clearly defined, will be dependent on actual inclination of left face to datum axis A.

Case #3
I believe the clearest and the easiest scheme to intepret. Left face must meet perpendicularity requirement with relation to A. Right face must fall within two parallel planes .010 apart, perrpendicular to A, and having their center located at basic 3.000 from datum plane B. The datum plane B is always perpendicular to datum axis A regardless of how much datum feature B is inclined to A (within stated perpendicularity tolerance). As already mentioned by powerhound, rule #1 does not apply here.

I am not sure if this answers your basic question, but I hope it at least sheds some light on subtle differences between your 3 schemes.

Keith1029 · Jun 17, 2013

Thanks pmarc,

That does address my question and is nearly the same conclusion I came to; the case 1 seems pretty wide open to interpretation. This seems like the classic example to illustrate the necessity of using GD&T wherever possible.

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Direct Tolerancing Interpretation 1

Keith1029

Mechanical

Recommended for you

Keith1029

Mechanical

SeasonLee

Mechanical

Keith1029

Mechanical

powerhound

Mechanical

Keith1029

Mechanical

Keith1029

Mechanical

pmarc

Mechanical

Keith1029

Mechanical

Similar threads

Part and Inventory Search

Sponsor