Coaxiality and general tolerance

[Andera]

ASME world

If there is no implied coaxiality tolerance (co-axiality tolerancing of features that are shown to share a common centerline/centermark), but there are implicit zero basic dimensions (both linear and angular, implied by simply showing multiple diameters on a single, common centerline), then why the general tolerance (assumed specified) could not take care of the missing (coaxiality) tolerance? Example: if a general profile is shown on the drawing similar to fig 5-4 from ASME Y14.5-2018 and the coaxial diameters are defined for size only (plus minus dimensions ---- as they are feature of size) then why coaxiality between those four diameters is not defined?
Or is it?

Please provide your thoughts.

 

[Burunduk]

The general profile tolerance is a surface control, and it requires basic dimensions for the tolerance zone definition. Since the diameters are directly toleranced, there are no sufficient basic dimensions for the true profile definition, thus the general profile can't control neither the size nor the alignment between the diameters. The basic zero between centers is not enough, because profile acts on the surface, not on the axis.

 

[Belanger]

Asa burunduk mentioned, a profile tolerance requires the size dimensions in that case to be basic (not ±). So you may have to clarify your question.
But the general idea you're referring to comes from fundamental rule (L) in paragraph 4.1 of the Y14.5 standard (or paragraph 1.4(j) for those who only have the 2009 standard).

I think the issue is with the first part of your question: you can't have implicit zero basic dims aligning the axes if there is no geometric tolerance that subsequently controls the coaxiality.

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

 

[greenimi]

J-P,

With all due respect I found it a little double standard your replay.
On the discussion below you seemed to advocate that using plus/minus dimension in combination with profile tolerance is kosher when we are talking about a regular feature of size.

Both cases (this discussed case and the one from the thread below with CH's pictures) are referring a regular feature of size therefore should be judged with the same unit of measurement.

I am sure I might have missed something so I thank you for clarification and cleansing.

https://www.eng-tips.com/viewthread.cfm?qid=470702

 

[flash3780]

Although there are implied basic dimensions for orthogonality, co-axiality, co-planarity, etc., there is no default geometric tolerance on that co-axiality.
But, you can certainly use a general note to apply a global control of position tolerance on your drawing.

For example:
UNLESS OTHERWISE SPECIFIED, POSITION TOLERANCE OF .020 INCHES DIAMETER RELATIVE TO DATUMS A AND B APPLIES TO ALL CIRCULAR FEATURES.

 

[Belanger]

Greenimi... obviously that was a pretty hefty discussion over on that other thread, but yes I think there is a key point that you're overlooking.
I've always insisted that profile requires basic dims for those dims that define the profile for its own form. The OP clearly mentioned a diameter, so the form tolerance is part of the size tolerance; thus the diameter must be basic (not considering dynamic profile and other special circumstances).

At the same time, I equally insisted in that other thread that it's not required for profile to use basic dims for dims that locate the profiled feature back to its datum reference frame. (See Fig. 11-32 of the standard for an example of plus/minus being used with profile in terms of location.)

Thus, if I'm hearing your question properly, it's not a double standard -- notice the words that I've underlined.

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

 

[Burunduk]

Belanger, greenimi,
My opinion on this slightly differs than Belanger's.
I think that it is datum-less profile tolerances, which act as form controls, when the need for basic dimensions for the true profile definition comes under question.
For example, profile of a surface can act like flatness on a planar feature, and it is known that flatness and other tolerances of form don't require basic dimensions, but instead use assumptions about the feature's "true form".

On the other hand, when it is part of the function of the profile tolerance to locate a surface relative to a datum reference frame, basic dimensions, explicit or implied, become mandatory. In figure 11-32, the profile of a line tolerance does not locate the feature. It works in a way that resembles a straightness tolerance, but with a DRF-related orientation of line sampling "each line element" orientation control. The location of the surface relative to the bottom face, but not the location of the true profile/profile tolerance zone, is derived from the directly toleranced dimension*. The profile tolerance could be found accurately conforming, but the top surface location relative to the bottom could be off, and that would be indicated only by the measurement of 80+/-0.2.

*As a side note, I would say that the case of two parallel opposed planar surfaces spaced by a directly toleranced dimension - which is a size control, is the only one when a directly toleranced distance dimension can be meaningful for location - mutual location between the opposed faces, that is.

 

[Belanger]

Hi Burunduk,
A profile tolerance is applied to a true profile. And a true profile is defined by basic dimensions (or math data, etc.).
Thus, from my perspective, every profile involves some sort of basic dimension. Even if the profile is not controlling a location (Fig. 11-32), there is an implied basic dimension to define the elements of that top surface as being perfectly straight. I guess that's where we differ.
Think of it this way: If the elements of the top surface were intended to be bowed just a tiny bit, then we'd have to show the radius as a basic dim. Well, since it's not meant to be bowed, that's still a basic dim being communicated.

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

 

[greenimi]

J-P and Burunduk,

....At the same time, I equally insisted in that other thread that it's not required for profile to use basic dims for dims that locate the profiled feature back to its datum reference frame.

So, where will be the conflict if you are interpreting the OP question as follow:
Size is to be controlled by the plus-minus
Location(s) of the cylindrical features by the general profile callout?
And yes, there are implied basic dimensions (zero) to the datum reference frame to each of those 4 cylindrical features...

 

[Burunduk]

Belanger,
I agree that there are implied basic dimensions in fig. 11-32, but in my opinion these are a zero degrees angle of the top face relative to datum feature A, and zero degrees angle of sampled lines relative to datum feature B. I'm not sure if I follow you regarding the implied basic dimension that makes the line elements nominally straight. If it was intended to be bowed, I think we both would agree that a basic radius would be required for the true profile definition. Are you saying that "straight" is an implied basic zero radius (R0 basic)?

 

[Burunduk]

Location(s) of the cylindrical features by the general profile callout?

I would say no.
The general profile callout is a surface control. Without a basic link from the surface to the DRF, it can't locate.

 

[greenimi]

Without a basic link from the surface to the DRF, it can't locate.

Well, here is the conundrum. IT IS a basic link from the surface to the DRF. It is zeo basic implied.

Per the discussion:
There is, in fact, implied coaxiality in ASME Y14.5. There is no implied coaxiality TOLERANCE. There are implicit zero basic dimensions, both linear and angular, implied by simply showing multiple diameters on a single, common centerline. What is not implied is the tolerance of that coaxiality. The tolerance must be specified by the designer.

Which tolerance, I think, the OP wants to be specified by the general profile/ general title block.

 

[Burunduk]

greenimi,
Without basic diameters, the basic link is partial and only useful for a position tolerance - which controls the axis.
With profile, we are left with a tolerance zone equally distributed about a diameter of 'øX', centered to the datum axis, but we don't know the value of the 'øX'. So, we don't have a defined tolerance zone for the feature to fit into.

 

[Belanger]

Burunduk... I say yes. We can think of a flat/straight surface which is subsequently toleranced by profile as having basic R0, in a sense. Perhaps it's a gray area (does flatness then have an implied basic dim?) but to me it helps square all of the Y14.5 requirements.

Greenimi...The conflict is that such a profile tolerance is not applied to a true profile; you'd have to drop the plus/minus. Otherwise it's not in conformance with Y14.5.
There are two exceptions to my claim. One is if the dynamic profile modifier is used, and the other is if the profile tolerance is applied to a cone with no datum references (that's because the profile zone can scoot back and forth longitudinally, so it doesn't get nailed down to a specific diameter).
But I'd prefer to avoid discussion on that exception and stick to the OP's dilemma about location.

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

 

[3DDave]

It would be curvature of 0. A radius of 0 has infinite curvature. It would be nice if CAD software makers used curvature as a controlling value, but that's an argument for another time.

The trick with overall coaxiality is "to what?" If it is between every possible pair and between each feature and every possible combination of other features it becomes really inconvenient - inconvenient mostly in that for all those combinations no one will do the tolerance analysis, no one will inspect all the combinations, and no one will be able to figure out if there is a problem in the machine just how this default callout directly affected that result.

 

[Belanger]

Good point... an infinate radius.

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

 

[greenimi]

Without basic diameters, the basic link is partial and only useful for a position tolerance - which controls the axis.

Why the presence or the absence of the basic dimension has an influence with the applicability of the profile?
If a surface does not have a location control then the general profile will take care of it (of the missing coaxiality control) regardless if that surface is dimensioned FOR SIZE with plus minus or basic. I am not sure I understand applicability (of the profile) or lack thereof and why this has any effect on the coaxiality control.

 

[Burunduk]

greenimi,
To establish a tolerance zone for profile, a True Profile is required. A True Profile is defined by basic dimensions, and the tolerance zone is tied to it. Because profile is a surface control, the True Profile is the basically defined surface, not a basically defined axis. No basic dimensions for complete definition of True Profile - no tolerance zone. No tolerance zone - no control of size OR LOCATION by profile. As Belanger mentioned, the exception to this is the Dynamic profile per the 2018 version of Y14.5 (then, the size of the true profile can be adjusted to perform "fitting" to the actual as-produced feature and only form, orientation and location can be controlled).

"If a surface does not have a location control then the general profile will take care of it"

How can it, when there is missing input for the establishment of the tolerance zone?

 

[CheckerHater]

If I understood correctly, in order to have basic relationship between "multiple diameters" they must be basic themselves, and the only way to have basic diameter is to control diameter via profile.

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