Orientation of a center plane of a tapered feature 2

[Burunduk]

A designer applied perpendicularity feature control frame with the leader pointing to a centerline representing the center plane of an internal taper feature (a "pocket" with symmetrical non-parallel opposed surfaces). I don't consider this specification valid according to ASME Y14.5-2009 but I'm struggling to provide a good explanation of why perpendicularity shouldn't be used this way. The one use of perpendicularity I know when it is applied on a virtual, derived geometry (as opposed to an actual surface) is when a center plane/axis of a feature of size is controlled. It doesn't seem right in the context of a tapered feature, not associated with a size dimension - but I can't form a good argument why. A valid point is that a center plane can be derived from a tapered feature (for example, a datum plane derived from tapered datum feature), and I suppose that a way to evaluate the derived plane orientation relative to a DRF can be found. I need your help, please.

 

[Burunduk]

Yes.

My previous post was based on the assumption that the horizontal bottom face (now identified as datum feature B) is the only datum feature reference of the profile tolerance. I probably should have stated that assumption more clearly.

Your assumption was correct. Initially I didn't think it through and didn't see that an additional datum reference (or simultaneous requirement) is needed to control the "height" of the inclined face.

Through this example, I now understand more clearly why you stated (at 20 Dec 19 03:11) that size for a conical feature is only meaningful when the surface is considered in relation to the flat surfaces (the boundaries of the conical feature). chez311 made similar statements too. Just like datum feature C in my example that allows to limit the height of the inclined face in relation to datum feature B, a flat face at the end of a cone referenced as a datum feature and the location tolerance of the opposing flat face relative to it may limit the size of a cone at its extremes. This is why a profile control applied on the surface without referencing datum features as in chez311's example (11 Jan 20 15:30) doesn't control that size.

Is it always necessary that a feature or an element must be able to be controlled for size when considered in isolation, to fit in the criteria of features of size? For example, fig. 8-18 in Y14.5-2009: is the circular element of the specified diameter 24+/-0.2 a feature of size? The size of it would be meaningless if not the basic dimension 18 to datum feature B.

 

[pylfrm]

Is it always necessary that a feature or an element must be able to be controlled for size when considered in isolation, to fit in the criteria of features of size? For example, fig. 8-18 in Y14.5-2009: is the circular element of the specified diameter 24+/-0.2 a feature of size? The size of it would be meaningless if not the basic dimension 18 to datum feature B.

The standard doesn't really discuss 2D elements much, so you probably won't find a solid answer to these questions. If you define the feature to be a nominally circular 2D element of the conical surface in a plane with a specified relationship to the flat surface, then it might be reasonable to consider it a feature of size. Note that the relationship between the plane and the flat surface is an integral part of the feature definition, so it would still be involved when the feature is considered in isolation.

Also note that ASME Y14.5-2009 Fig. 2-21 implies that the toleranced dimension in Fig. 8-18 would actually control the entire conical surface, not just a 2D element.

Does it matter whether such a feature is a feature of size?


pylfrm

 

[Burunduk]

pylfrm,
I noticed in the thread that chez311 linked to at the beginning of this thread ( thread1103-448819) that you summarize the feature of size concept as:

"Feature of size: A feature with an unrelated actual mating envelope." (8 Feb 19 01:54)

Is a regular feature of size always required to be of a defined actual mating envelope? I ask because it isn't mentioned in the definition in Y14.5-2009:

"1.3.32.1 Regular Feature of Size. regular feature of size: one cylindrical or spherical surface, a circular element, and a set of two opposed parallel elements or
opposed parallel surfaces, each of which is associated with a directly toleranced dimension."

Only irregular features of size are explicitly required to be contained by an AME, and even there, it is not specified that the AME must be unrelated. A circular element is listed as a regular feature of size. What is a circular element without a location that depends on some other feature? What does it mean about the feasibility of a UAME for this RFOS?

Does it matter whether such a feature is a feature of size?

I'm trying to estimate if position tolerance is applicable to this type of feature.

 

[chez311]

Just like datum feature C in my example that allows to limit the height of the inclined face in relation to datum feature B, a flat face at the end of a cone referenced as a datum feature and the location tolerance of the opposing flat face relative to it may limit the size of a cone at its extremes. This is why a profile control applied on the surface without referencing datum features as in chez311's example (11 Jan 20 15:30) doesn't control that size.

We are in agreement here. Though the preceding statement seems to me more like location than size (location to the flat face) one could say it limits the size of the circular elements derived from the cone, so I won't gripe about that too much. I think we are mostly aligned here.

It looks like the mathematical definition (Y14.5.1M-1994) doesn't go much further than "contact or simply touching" either:

The definition for AME in Y14.5.1-1994 is almost an exact repetition of the definition in Y14.5-1994. As such I would not consider this a "mathematical definition" but simply a restatement of the relevant and compliant standard at the time of release - it adds literally nothing to the definition from a mathematical/geometric standpoint. I can't speak to why thy chose not to include "at the highest points" - other than that they are identical.

(a) For an External Feature. A similar perfect feature counterpart of smallest size which can be circumscribed about the feature so that it just contacts the surface.

(a) For an External Feature. A similar perfect feature counterpart of smallest size that can be circumscribed about the feature so that it just contacts the surface at the highest points.

As such, the Y14.5-2009 definition should be utilized for all 2009 compliant drawings - the fact that the Y14.5.1-1994 definition differs is not a conscious decision by the math standard committee but a fact that it lags behind the newer standard by over a decade (and the latest release Y14.5-2018 by over two decades).

The Y14.5-2009 is notable in that it specifically refers to "contraction/expansion" instead of "circumscribed" and trades "just contacts" for simply "coincides". I think the elimination of the word "just" is a notable change for as you pointed out "just" could imply simply touching the surface and place less emphasis on the minimization/maximization of the envelope.

envelope, actual mating: this envelope is outside the material. A similar perfect feature(s) counterpart of smallest size that can be contracted about an external feature(s) or largest size that can be expanded within an internal feature(s) so that it coincides with the surface(s) at the highest points.

The new draft of Y14.5.1-20xx which will be Y14.5-2009 compliant does not provide a separate definition for AME and simply states "See: ASME Y14.5-2009".

I can see how the main pivot point for you would probably be "largest size" (for internal features) and "smallest size" (for external surface), but what defines how this size should be determined for envelopes of irregular features of size (consider for example the pocket from fig. 8-19)?

This was discussed at length - see the multitude of posts above about pylfrm's definition of "maximum offset".

As you don't ascribe to this definition, I'll provide an alternate explanation:

IFOSb does not have a single actual value for size, however the concept of "size" is still applicable ie: as discussed previously, a profile control applied to Y14.5-2009 fig 8-19 controls both form AND size). A uniformly expanded envelope would have a larger "size" and therefore an envelope expanded to its maximum extent would be of maximum size, the inverse would be true with uniform contraction. There is no need (or ability) to determine a single actual value for size of this envelope - one can certainly determine what is therefore the envelope of maximum/minimum size.

The definition brought by Kedu involved normals to the surface, nothing about 2D cutting planes. The surface is in 3D space. How can 2D elements isolated from the 3D space have opposed points?

Cutting planes are the method by which 2D elements are derived from a 3D shape in Y14.5.1

1.4.8 Cutting plane. A plane used to establish a planar curve in a feature. The curve is the intersection of the cutting plane with the feature.

The answer is contained in your question. How can 2D elements isolated from the 3D space have opposed points? Exactly that - they are isolated in 2D and the normal vectors under consideration would be 2D vectors existing on said cutting plane. Same with the tolerance zones which are applied to a 2D element - they are a subset of the cutting plane which is used to extract the 2D elements. For example, regarding straightness of line elements:

6.4.1.2 Straightness of Surface Line Elements
(a) Definition. A straightness tolerance for the line elements of a feature specifies that each line element must lie in a zone bounded by two parallel lines which are separated by the specified tolerance and which are in the cutting plane defining the line element.

I also don't think I saw any real attempt at reconciling the below put forward by pylfrm. The constrained L2 and envelope of minimum/maximum size for an actual produced feature (ie: real world variation) will almost never be the same.

The procedure I described is based on minimizing the separation between the envelope and feature

... which for regular external features of size always equivalent to minimizing the size of the envelope, and for irregular features of size equivalent to scaling down the envelope.

Assuming we're still talking about minimizing the sum (theoretically the surface integral) of the squared separations, that is not true for a nominally cylindrical or nominally square hole. I'm pretty sure it's just generally not true.

For a counterexample, imagine a nominally cylindrical hole that's actually produced more like a short slot with two flat sides connected by tangent cylindrical ends. If the flat sides are slightly non-parallel, one of the ends will have a slightly larger radius and will end up coincident with the maximum inscribed cylinder UAME. On the other hand, the least-squares cylinder will be smaller, contacting the flat surfaces roughly in the middle and equalizing the separation with the two ends.

 

[Burunduk]

chez311,
Following the recent communication in this thread, I no longer stand by the position that the size of a cone is defined in isolation from other features. Nor do I think it is the least squares which should determine the UAME. pylfrm's counterexample was spot on.

However, I am still not sure that "largest" or "smallest" is not determinable for an envelope that interacts with the feature.
Consider Y14.5-2009 fig. 2-21. A radial tolerance zone of 0.03 is shown. The text in the figure says that the "size of the tapered section" is controlled. If it can control the size of a feature why can't it be a "scale" to determine the size of the theoretical envelope that contacts the feature at the high points? Doesn't that radial tolerance zone allow different sizes of envelopes with the taper of 0.3:1 to reside within it? Even if "resizing" until contact with the feature's high points is technically done by translation and free rotation ("unrelated" in all DOF) of a fixed envelope useful for any cone in the world defined with the same basic taper, isn't it equivalent to scaling (contraction) within the boundaries of that tolerance zone (with the focus only on what happens at the extent of the actual produced feature)? It is surely not as convenient as having an envelope with a single size but:

There is no need (or ability) to determine a single actual value for size of this envelope - one can certainly determine what is therefore the envelope of maximum/minimum size.

 

[chez311]

Does it matter whether such a feature is a feature of size?

I noticed in the thread that chez311 linked to at the beginning of this thread ( thread1103-448819: Position of a taper) that you summarize the feature of size concept as:

"Feature of size: A feature with an unrelated actual mating envelope." (8 Feb 19 01:54)

pylfrm - I'm sort of thinking along the same lines as Burunduk. I would say there are several points in the standard which specifically state that a certain tolerance (for example - position) applies to a FOS so categorization as a FOS would technically be required to apply said tolerance without a special note or additional instructions/clarification.

Even without that, it was my understanding from your definition (the relevant portion posted above by Burunduk) that the existence of a UAME went hand in hand with definition of a FOS - perhaps in my responses above I put a little too much emphasis on whether or not a cone or similar feature was a FOS and I guess thats less intrinsically important than whether or not a feature has a UAME.

Also note that ASME Y14.5-2009 Fig. 2-21 implies that the toleranced dimension in Fig. 8-18 would actually control the entire conical surface, not just a 2D element.

I'll be honest, I hadn't given 2-21 too much thought and after further consideration the method shown in that figure seems somewhat problematic - if I had to guess this seems like carryover from previous practice. After having given it some thought, the best interpretation I've come up with is that the conical feature must be within a tolerance zone bounded by the conical true profile intersecting (containing) a 50 dia (basic) circular segment located on a plane parallel to the flat face at either extreme of the 9.9/10.1 directly toleranced location. This combination of directly toleranced location and basic 50 dia imparts a sort of orientation requirement because any angulation of the true profile would result in an elliptical conic section instead of circular one required by the basic 50 dia. Where the issue for me comes in is that the figure implies that these tolerance zone bounds must be coaxial - I don't see anything which suggests this other than the way the figure is drawn. To me it seems like the conical surfaces which bound the tolerance zone can shift in location relative to one another ie: they can be coaxial, or shifted until they touch on one side, or anywhere in between - the tolerance zone is sort of amorphous. This is in contrast to the profile tolerance applied in 8-17/8-18 which imparts the requirement that the bounds of the tolerance zone must be coaxial (or maybe more generally, equally disposed around the true profile).

My interpretation of 8-17/8-18 would be similar namely:

8-17: The conical feature is bounded at its extremes by the true profile intersecting (containing) circular elements of size 30.2 and 29.8 located at basic 0 distance from the flat face which imparts a sort of orientation requirement on the bounds as outlined above for 2-21. Between those bounds, the form is also subject to a 0.02 datumless profile tolerance which since it is datumless may shift in location/orientation relative to these bounds as well as the flat face. The inclusion of this 0.02 profile tolerance provides a control on form that the bounds of 2-21 are not able to provide.

8-18: The conical feature is bounded at its extremes by the true profile intersecting (containing) circular elements of size 24.2 and 23.8 located at basic 18 distance from the flat face which imparts a sort of orientation requirement on the bounds as outlined above for 2-21. Between those bounds, the form is also subject to a 0.02 profile tolerance to |A|B| which is fixed in basic orientation/location to A and B - though since the basically located 24+/-0.2 diameter is directly toleranced the effect is essentially the same as if the location were directly toleranced to B so that the tolerance zone may shift axially within the bounds outlined above. Obviously the inclusion of the 0.02 profile tolerance not only controls form above and beyond 2-21 for the same reasons as 8-17 as well as provides orientation/location constraint to the specified DRF.

 

[chez311]

I am still not sure that "largest" or "smallest" is not determinable for an envelope that interacts with the feature. Consider Y14.5-2009 fig. 2-21. A radial tolerance zone of 0.03 is shown. The text in the figure says that the "size of the tapered section" is controlled.

I see 2-21 as rather problematic, and I personally would not use the method shown on an actual part/drawing to control a taper. See above in my response to pylfrm for my issues with that figure.

That said, to the issue at hand I don't see what value it adds to consideration of your stated inquiry if we assume said inquiry to essentially be "is the largest/smallest envelope determinable for a cone considered in isolation". I would not consider 2-21 to be a conical feature in isolation - indeed its right there in the description: "the basic diameter controls the size of the tapered section as well as its longitudinal position in relation to some other surface."

 

[Burunduk]

chez311,
The following is my interpretation of fig. 2-21:
First, for a complete unambiguous definition I would add the dimension origin symbol instead of the dimension line arrow at the side of the flat face.
Then, consider two theoretical circular elements of basic dia. 23, one at an offset of 9.9, the other at an offset of 10.1 from a tangent plane to the flat face, both parallel to the tangent plane and with their center points located at a common axis. The conical tolerance zone boundaries would both have a basic taper of 0.3:1 and each would be coincident (intersecting?) with one of the basic dia. circles. This way the two boundaries can't be anything but coaxial.

As agreed the size of a conical feature is not meaningful for a cone in isolation but it can be controlled by dimensions also related to other features as in the said fig. 2-21. However, arguably an envelope proposed to be the UAME can still be considered in isolation (remaining unrelated) and can be evaluated for size within the bounds of the radial tolerance zone that controls the size of the feature.

 

[chez311]

Burunduk,

That would be a modification or suggested change, not a direct interpretation of fig 2-21 wouldn't it? Regardless, I don't see how the addition of the dimension origin symbol adds any requirement that the boundaries be coaxial - perhaps less ambiguity about how the directly toleranced axial location should be established, but I see no reason that the center points of the circular elements of basic dia 23 located at 9.9/10.1 must lie on a common axis with or without the dimension origin symbol. Where do you see this requirement?

As agreed the size of a conical feature is not meaningful for a cone in isolation but it can be controlled by dimensions also related to other features as in the said fig. 2-21. However, arguably an envelope proposed to be the UAME can still be considered in isolation (remaining unrelated) and can be evaluated for size within the bounds if the radial tolerance zone that controls the size of the feature

I inquired about this previously, pylfrm provided the below insight which I think makes intuitive sense and I agree with:

In general, determination of a feature's UAME (if it exists) or RAME(s) doesn't have anything to do with the tolerances that are (or are not) applied to that feature.

A feature's UAME is not constrained by the tolerance(s) applied to a feature. For an illustrative case, imagine a cylinder with rule #1 overridden - there can be large discrepancies between the size of the feature and the size of the UAME.

 

[Burunduk]

I see no reason that the center points of the circular elements of basic dia 23 located at 9.9/10.1 must lie on a common axis with or without the dimension origin symbol. Where do you see this requirement?

Where there are two points, a theoretical line or an axis that passes through them can always be found. In addition, since the 9.9/10.1 located 23 diameter circles are theoretical (the limits of location of a physical 23 size circular element) I don't see why these elements should be skewed relative to the common axis that connects their centers and to the flat face from which they are located.

I agree that the UAME is not constrained by the tolerances applied to the feature, but if a tolerance zone can describe the size of a feature (regardless if the dimensioning of the feature is "in isolation" or involves relationships with other features) then it seems to me that a way to describe the size of the envelope that contacts that same feature on the highest points can also be found (even if we don't involve the feature's tolerance zone in it) - for example diameters of cross sections of the envelope at the section surrounding the actual feature can be measured at fixed locations relative to where the feature starts or ends.

 

[pylfrm]

I noticed in the thread that chez311 linked to at the beginning of this thread ( thread1103-448819) that you summarize the feature of size concept as:

"Feature of size: A feature with an unrelated actual mating envelope." (8 Feb 19 01:54)

I wouldn't call that a summary of the feature of size concept as defined in the standard. The purpose of that definition (and the UAME definition presented with it) is explained in the surrounding paragraphs of that post.

Is a regular feature of size always required to be of a defined actual mating envelope? I ask because it isn't mentioned in the definition in Y14.5-2009:

If you want to take the standard it its word, you've basically answered your own question.

A circular element is listed as a regular feature of size. What is a circular element without a location that depends on some other feature? What does it mean about the feasibility of a UAME for this RFOS?

I think answers to these questions are generally undefined, or at least not sufficiently defined to be useful. For example, the minimum circumscribed circle of a set of points is not a valid concept in three dimensions.

I'm trying to estimate if position tolerance is applicable to this type of feature.

My usual approach: If the tolerance would have a well-defined meaning, feel free to apply it. If not, don't. Ignore the arbitrary classification.

I would say there are several points in the standard which specifically state that a certain tolerance (for example - position) applies to a FOS so categorization as a FOS would technically be required to apply said tolerance without a special note or additional instructions/clarification.

For the sake of argument, suppose there exists a feature that is not a FOS but does have a well-defined UAME and UAME axis. If a position tolerance without MMB or LMB modifier is applied to such a feature, I think the only reasonable interpretation is that the tolerance defines a zone within which the UAME axis must fall. That's exactly the normal meaning, so I don't see a need for a special note.

I'll be honest, I hadn't given 2-21 too much thought and after further consideration the method shown in that figure seems somewhat problematic - if I had to guess this seems like carryover from previous practice.

I suppose I hadn't given it all that much thought either. Except "because we say so", I don't see any particular reason that the scheme should mean what the figure says it does. For instance, why does it create a tolerance zone for the conical surface instead of for the flat surface? I suppose that aspect might make more sense if the dimension origin symbol had been used.

I think a more natural meaning would be that the basic dimensions and the 10.1 limit define a maximum material boundary consisting of a plane and cone, the basic dimensions and the 9.9 limit define a least material boundary consisting of a plane and cone, the relationship between the two boundaries is unconstrained, and the actual planar and conical surfaces must not violate either boundary.

I agree that the UAME is not constrained by the tolerances applied to the feature, but if a tolerance zone can describe the size of a feature (regardless if the dimensioning of the feature is "in isolation" or involves relationships with other features) then it seems to me that a way to describe the size of the envelope that contacts that same feature on the highest points can also be found (even if we don't involve the feature's tolerance zone in it) - for example diameters of cross sections of the envelope at the section surrounding the actual feature can be measured at fixed locations relative to where the feature starts or ends.

The tolerance zone in Fig. 2-21 describes the allowed size of the conical surface relative to the large flat surface. That's the sort of size that would be relevant for a RAME, not a UAME.

Perhaps you should take another shot at the question at the end of my 4 Jan 20 03:48 post. It may help to clarify exactly what you have in mind for measuring diameters of cross sections.


pylfrm

 

[chez311]

For the sake of argument, suppose there exists a feature that is not a FOS but does have a well-defined UAME and UAME axis. If a position tolerance without MMB or LMB modifier is applied to such a feature, I think the only reasonable interpretation is that the tolerance defines a zone within which the UAME axis must fall. That's exactly the normal meaning, so I don't see a need for a special note.

I agree with this, like I said I probably put too much emphasis on whether or not a feature is a FOS/IFOSa/IFOSb when the focus should have been whether or not the feature has a defined UAME. I know you said for the sake of argument but I can't imagine such a feature - that said I ascribe to your definition which directly ties classification as a FOS to whether or not it has a UAME, if a feature has a well defined UAME I would have no issue treating it as a FOS and applying controls to it without a note.

I suppose I hadn't given it all that much thought either. Except "because we say so", I don't see any particular reason that the scheme should mean what the figure says it does. For instance, why does it create a tolerance zone for the conical surface instead of for the flat surface? I suppose that aspect might make more sense if the dimension origin symbol had been used.

I think a more natural meaning would be that the basic dimensions and the 10.1 limit define a maximum material boundary consisting of a plane and cone, the basic dimensions and the 9.9 limit define a least material boundary consisting of a plane and cone, the relationship between the two boundaries is unconstrained, and the actual planar and conical surfaces must not violate either boundary.

This is an excellent interpretation, I'm intrigued. I guess I was so focused on the conical portion due to the way the figure is drawn I hadn't even considered that. I would agree that without the dimension origin symbol the tolerance should apply to both surfaces instead of originating at one or the other. Even if the dimension origin symbol is applied to the side of the dimension originating at the flat face, would you agree that there is still no requirement that the boundaries be coaxial?

What if we were to consider the inverse case, say if the dimensions applied were a basic 10 axial location and directly toleranced 23+/0.1 diameter. It seems to me that it would still control form in a similar manner with two conical boundaries, and a similar behavior between these conical boundaries namely that they are not required to be coaxial (ie: some plane containing circular elements 23.1/22.9 and the true profile intersects/contains said circular elements, however these elements can shift in x,y location relative to each other, there is no requirement they share the same center) - HOWEVER the basic 10 only defines that there should be some plane containing the said elements, not how this plane should be constrained to the flat face so location of the conical feature is unconstrained and only form is controlled. If we're to make some pretty big assumptions, perhaps one could say theres an "implied datum feature" (I know, just saying the words "implied datum" makes me uneasy) and that the basic 10 should originate from the tangent plane contacting the flat face - I don't think I'm comfortable making such assumptions though.

What if the basic distance were a specified or implied zero - like in 8-17? Or is the meaning of 8-17 even more ambiguous than that? In my response (21 Jan 20 02:38) I outlined interpretations of 8-17/8-18. I initially thought the implied zero basic of 8-17 provided some constraint of the tolerance zone, though I see now I may have been wrong - I think only in the case of 8-18 where the profile tolerance is held to |A|B| is that location constraint valid. Perhaps the size dimension in 8-17 means something less well defined - I'm not even sure where to begin if thats the case. What do you think?

 

[Burunduk]

My usual approach: If the tolerance would have a well-defined meaning, feel free to apply it. If not, don't. Ignore the arbitrary classification.

That's an interesting practical approach that I should probably bear in mind.

The tolerance zone in Fig. 2-21 describes the allowed size of the conical surface relative to the large flat surface. That's the sort of size that would be relevant for a RAME, not a UAME.

That's a statement that I don't understand (that is not to say that it's incorrect). First, how can size be "relative to" something? The location of where size is measured is meaningful (in this case and in any case where size is not uniform), but the size is not of one thing relative to another. It is always a characteristic of only one feature. Second thing, I don't know what the relevance of RAME versus UAME has to do with how the feature is dimensioned (in terms of the dimensional relationships between different features). The considerations between RAME and UAME as far as I know are completely different; Examples of where the UAME is relevant: 1. Primary datum feature simulator. 2. A considered feature controlled by position tolerance at RFS. Example of where the RAME is relevant: a simulator for a secondary datum FOS referenced RMB.

Perhaps you should take another shot at the question at the end of my 4 Jan 20 03:48 post. It may help to clarify exactly what you have in mind for measuring diameters of cross sections.

The question was a request to describe the UAME determination procedure for a conical feature. Suppose that an external cone is considered. My suggestion is 1. Establish a theoretical envelope of the specified basic included angle. 2. Mate the envelope with the actual produced cone so that it contacts the conical feature on the highest points. If there are multiple solutions for this, refer to the default stabilization algorithm per the version of Y14.5 you are using. 3. You have now established the "the smallest size envelope" for the actual produced feature. This smallest size envelope can be indicated by the diameters along the portion of the simulator which contains the feature, these will vary from one actual feature to another. If for some reason a numerical value needs to be documented, the following practice is suggested: 4. Determine the 2D cutting plane normal to the envelope axis which is the closest to the small flat end of the cone but intersects only the conical surface (it does not intersect the surface of the flat end). Record the local diameter of the envelope at that cutting plane. Determine (possibly by calculation with the knowns of the above mentioned diameter and the basic included angle) another local diameter of the envelope at a fixed axial distance from the first cutting plane. This axial distance can be approximately half of the nominal feature length. This latter diameter represents the "mating size" of the actual feature. The smaller the conical feature in size (the non-uniform size which is controlled by a radial tolerance zone as in fig. 2-21), the smaller the representative size of the envelope will be.

 

[pylfrm]

Even if the dimension origin symbol is applied to the side of the dimension originating at the flat face, would you agree that there is still no requirement that the boundaries be coaxial?

The figure essentially says the boundaries are coaxial and I don't think adding the dimension origin symbol would change that. If you're asking about the "more natural meaning" I proposed, I don't see any reason that adding the dimension origin symbol should change whether the boundaries are required to be coaxial.

What if we were to consider the inverse case, say if the dimensions applied were a basic 10 axial location and directly toleranced 23+/0.1 diameter.

I think the easiest interpretation to justify based on the standard is that the meaning would be identical (if you had said 23+/-0.03) to that described for Fig. 2-21.

If we're to make some pretty big assumptions, perhaps one could say theres an "implied datum feature" (I know, just saying the words "implied datum" makes me uneasy) and that the basic 10 should originate from the tangent plane contacting the flat face - I don't think I'm comfortable making such assumptions though.

I think making that assumption (or something very similar) is the only way Fig. 8-18 makes any sense.

Perhaps the size dimension in 8-17 means something less well defined - I'm not even sure where to begin if thats the case. What do you think?

The fact that an edge is dimensioned does seem to open up some other possible interpretations. Tolerances involving edges don't get much coverage in the standard.


pylfrm

 

[pylfrm]

First, how can size be "relative to" something?

Size in what I'd consider the usual sense isn't relative to something else, but the tolerance zone illustrated in Fig. 2-21 doesn't control size of the conical surface in that sense. It controls where the conical surface is allowed to be relative to the large flat surface. I'm not particularly opposed to referring to that as "size", but it's important to recognize that it's a different meaning of the term.

It is always a characteristic of only one feature.

That doesn't seem like a particularly meaningful statement because that one feature can be defined to include whatever you want.

Suppose that an external cone is considered. My suggestion is 1. Establish a theoretical envelope of the specified basic included angle. 2. Mate the envelope with the actual produced cone so that it contacts the conical feature on the highest points. If there are multiple solutions for this, refer to the default stabilization algorithm per the version of Y14.5 you are using. 3. You have now established the "the smallest size envelope" for the actual produced feature.

It sounds like you're saying that the relevant sort of size is not an intrinsic property of the envelope geometry, but rather something that depends only on its location relative to the actual feature. That seems rather strange, and is completely different from how it works for cylinders and many other shapes.

Wouldn't this reasoning also allow you to say that a plane at a particular location is "the smallest size envelope" for a nominally planar feature?


pylfrm

 

[Burunduk]

It sounds like you're saying that the relevant sort of size is not an intrinsic property of the envelope geometry, but rather something that depends only on its location relative to the actual feature. That seems rather strange, and is completely different from how it works for cylinders and many other shapes.

When the standard explicitly discusses "limits of size" of conical features (para. 8.4.2 Y14.5-2009) I see no other interpretation but a location-dependent size being relevant to this type of features.

Wouldn't this reasoning also allow you to say that a plane at a particular location is "the smallest size envelope" for a nominally planar feature?

I can't imagine how this is possible.
What would be the equivalent of directly toleranced diameters?

 

[pylfrm]

When the standard explicitly discusses "limits of size" of conical features (para. 8.4.2 Y14.5-2009) I see no other interpretation but a location-dependent size being relevant to this type of features.

I did not mean relevant to the feature. I meant relevant to your proposed UAME.

I can't imagine how this is possible.
What would be the equivalent of directly toleranced diameters?

Here's what you said, except with four words changed:

1. Establish a theoretical envelope of the specified basic geometry. 2. Mate the envelope with the actual produced feature so that it contacts the feature on the highest points. If there are multiple solutions for this, refer to the default stabilization algorithm per the version of Y14.5 you are using. 3. You have now established the "the smallest size envelope" for the actual produced feature.​

Seems to work for planar features now. Directly toleranced diameters weren't mentioned.


pylfrm

 

[Burunduk]

pylfrm, directly after the wording in your modified quote I wrote in the original post from 23 Jan 20 17:09: "This smallest size envelope can be indicated by the diameters along the portion of the simulator which contains the feature, these will vary from one actual feature to another.", and further there is a description of how these diameters can be evaluated and compared for different produced features, or for the same actual feature for checking the repeatability of the simulation. How would the equivalent evaluation of "smallest size" for a planar feature work?

 

[pylfrm]

Burunduk,

The part about diameters probably wouldn't work very well. I skipped all that because it didn't seem to be involved in the actual procedure you suggested.


pylfrm

 

[Burunduk]

pylfrm, it is not involved in the actual procedure I suggested because the procedure is not actively sized, it is automatically self-sized (I know you won't like this argument and I am aware it doesn't sound right in any aspect but perhaps this may be more effective anyway than me trying to find the exact wording to say that thing in a long post).