Figure 4-16 in the '2009 version vs. figure 7-22, 7-23, and 7-24 in the '2018 version

[3DDave]

How fun. I hadn't looked until now but the figure that was 4-16 in the '2009 version and now 7-24 in the '2018 version no longer uses the virtual condition. I wonder why.

It still fails to mention that for the 3rd condition the maximum dimension only applies in one direction, but still, that's quite a change. Converting it to RFS means that there's less point in making that calculation as whatever the size the mating datum simulator has to collapse to meet it. Since D is referenced to arrest rotation the radial component of the RFS feature reference doesn't matter; the tangential one does and that is limited by the perpendicularity. Starting from a cylinder when only the width centered on datum feature B is important is not useful.

As I recall, the combined effects in the first example produced a dimension in a slightly diagonal dimension that was greater than the simple calculation showed so it is still wrong, but just in a different way. The addition of the straightness tolerance is a nice touch, but I don't understand why most values are changed. Change the text, change the test perhaps.


Edit: Title fix.

 

[pmarc]

4.5.3(d) in '94 was about RFS gages, so for 4-8 would not apply and a moveable pin would not ever be required.

The only difference between MMC and RFS gage pins in '94 was in the size of the pins (fixed vs. adjustable), not in their mutual distance relationship (basic vs. variable). Take a closer look at fig. 4-9 in '94 to see that the basic distance between the simulators/TGCs B and C is not fixed. Also, if you have access to Y14.43-2003 (that supports Y14.5M-1994) you may want to take a look at fig. B10(c) to see how the lack of the requirement for basic distance between pins B and C is realized. This changed with the release of Y14.5-2009, where para. 4.5.2(c) explicitly stated that the distance between the two simulators must be basic, unless translation modifier or movable datum target symbol was specified. Both cases are illustrated by figs. B10(c) and B11(c) in Y14.43-2011 (that supports Y14.5-2009).

How do you suppose a customized datum reference frame is able to shift constraints from a feature they ordinarily control to one later in the sequence?

As far as hard gaging is considered, you may want to take a look at fig. B24 in Y14.43-2011 for an example of how this can be achieved. This figure corresponds to fig. 4-46 in Y14.5-2009, except that the MMB modifiers were added in B24.

For '2009, "in 4-19 and 4-32(b) no DRF customization really takes place." That's true - both are MMC/MMB cases.

I am not sure why you are saying that these are MMC/MMB cases. Taking fig. 4-32 as an example, I am imagining the slot B produced at the LMC width and perfectly perpendicular to A. When mounted onto a gage simulating the |A|B[⊳]| DRF, the part will look the same as in fig. 4-32(b). But when mounted onto a gage simulating the |A|B(M)| DRF, the part will be allowed to look more like in fig. 4-32(a), except that the width of the simulator B will be fixed at 4.7 and the slot will be contacting the simulator B at only one wall as a result of rotational datum feature B shift.

FYI... My intention was not to sidetrack the thread. I just felt a comment was needed after you had suggested that the translation modifier could be simply replaced with the customized DRF concept. Can we now agree that we went too far down the rabbit hole?

 

[3DDave]

It's fine - otherwise it's just been comments saying that rules are rules and that's all there is to it.

Did I misunderstand "Side note: Although the gage pins B and C in fig. 4-8 in '94 would indeed be of fixed size, according to para. 4.5 3(d) the distance between them should be variable, and so the whole gage would not be that trivial anyway?"

That seemed to indicate the the MMC callout didn't require the distance to be fixed and therefore the gauge, as suggested, would not be trivial. Perhaps it was a short cut and was meant to be about the RFS version and that then 4.5.3(d) would apply - but that paragraph says nothing about translation.

Which is what was the division in opinion in the standard's participants that drove the translation modifier; however at the same time the customized datum reference frame was also being worked in, no doubt by a different committee and they all agreed "Why not both?"

Is there any case of a customized datum reference frame that can properly be gauged without translating and/or rotating gauge elements? If they can translate they could be doing the same thing as the translation modifier.

I don't have 14.43; Did they put a fixed round pin into the square hole and call it good? Because the customized datum reference frame callouts implies that there should be square pin that fits with a cylindrical base into a hole so it can turn to allow motion to let datum feature C participate. If they moved it to all MMC/MMB references that avoids putting the datum feature C simulator on a slide to vary the center distance. It becomes a much easier gage with one rotating part.

re: 4-32 - i'm dumb and got looking at the wrong figure instead of 4-32.

For 4-19 - similar problem as for the offset slot - why are they defined at MMC and then referred to at RMB? Why give bonus orientation/location to them based on the size and then say - naw, we're going to build the mating assembly to have expanding pins, and one of the pins is on a slide mechanism? I don't see an application that demands this - it seems contrived to support the translation modifier because the CMM guys hated looking for the center of the "expanding pin" because it requires generating geometry to see where it contacts and that just finding the center of the hole wherever it was was so much easier.

 

[Burunduk]

Is there any case of a customized datum reference frame that can properly be gauged without translating and/or rotating gauge elements? If they can translate they could be doing the same thing as the translation modifier

Yes, there is one in the standard, the one with the conical primary datum feature giving away the axial translation constraint to the secondary.

And what if the gage element needs to only rotate, like in another example, the one with the square hole? Would you then ask for addition of a rotation modifier?

On the other hand for cases with the translation modifier, often the explicitly stated degrees of freedom for a customized datum reference frame attempting to replace the modifier, would be the same as the ones implied by the types of datum references and the datum precedence order anyway, as the modifier is not intended to change the way how the constraints should take place in an ideal scenario like the costomized DRF does, but only to prevent a problem.

I take that you are still thinking through my last question from 8 May 22 05:04, or maybe you lost interest in the original topic of your thread? In that case you may as well want to change the title again.

 

[3DDave]

I thought I had already given an answer for that before you asked, but if you need it again - here it is:

If it is B only then it's too complicated as predicting the orientation of D relative to B is not trivial, but the pin feature will be able to occupy anywhere in some ring shape. A fixture to accept all locations/orientations of the pin will require a circular trough for that side of the part.

Oh - I realize you don't consider context - the context is that Only B preceding datum feature D, not that B is the only reference, which would make no sense due to the fact these all concern using D as part of the development of the datum reference frame. So, here, if you want it spelled out for you [B(m}||D(m)] is what I was referring to.

Why would I ask for a rotation modifier? There is already a customized datum reference frame control for rotation.

 

[3DDave]

Yes, there is one in the standard, the one with the conical primary datum feature giving away the axial translation constraint to the secondary.

That requires the fixture to move to engage the cone relative to the pin and flat face and to control the orientation of the entire part. It cannot be a single rigid gauge.

 

[Burunduk]

I asked you earlier:

"In the RMB case option (b), had the datum references been |B|D|, would you say the OB (Outer Boundary) for datum feature D is a 360° ring?"

To which your answer was:

"It would not be 360 because D is what gets you going 100 kph to align with D."

That's why I thought that for the "B only" option, you were referring to something else. Now you say that the "B only" option addresses what I asked about (which is by the way not "[B(m}||D(m)]" but never mind that). And if so, you do consider the OB or MMB of datum feature D to be a ring shaped boundary, according to: "but the pin feature will be able to occupy anywhere in some ring shape. A fixture to accept all locations/orientations of the pin will require a circular trough for that side of the part", and your initial "not be 360" statement is to be disregarded. Am I interpreting you correct this time?

That requires the fixture to move to engage the cone relative to the pin and flat face and to control the orientation of the entire part. It cannot be a single rigid gauge.

The same applies to certain applications without customization of the DRF, such as the case in fig. 4-30 (a) in the '09 version. Thus, you see that the customization does something different from what the translation modifier is intended for.

 

[3DDave]

If the frame of reference is [A|B] for the part then any feature using [A|B] will need a fixture that has 360 degree clearance for the pin to be located anywhere about the B-perpendicular-to-A axis.

If the frame of reference is [A|B|D] then any feature using [A|B|D] will have an oblong slot the for pin used as the D datum feature to fit into located from the B-perpendicular-to-A axis and not be 360.

The difference is that in the second case one has caught up with [D] and that has no dependency on the [C] feature - by which reference you caught up with C and saw the tangential movement allowed by the position tolerance. Catching up D means there is no longer tangential movement allowed - only the orientation remains. Since it is located from B, the radial position component remains - so it's a slot, but, because of the way they combine it's not a simple full rounded end slot.

In the example 4-30 the customized datum reference frame would be:
for RFS: [A[x,y,z,u,v]| B [w] ]
for MMC, since nothing moves there is no customize datum reference frame required.

 

[Burunduk]

If the frame of reference is [A|B|D] then any feature using [A|B|D] will have an oblong slot the for pin used as the D datum feature to fit into located from the B-perpendicular-to-A axis and not be 360.

I see that you avoid using the specific terms like I did in my question, but I conclude that you suggest that for an |A|B|D| referencing control, the outer boundary calculated in fig. 7-24 of the '18 standard, and likewise the RMB datum feature simulator for the external cylinder designated as datum feature D, should be an internal oblong. This so that the tangential direction of the outer boundary accounts only for the variation in size, form, and orientation of datum feature D relative to datum A, while the radial direction accounts for the position relative to A and B (and also includes the other variations).

If you are consistent with that approach, then if you were requested to consult a manufacturer designing a gage for a part he produces which is similar to the one from fig. 7-24 in the '18 standard (an RMB case), you would instruct him to build a custom datum feature simulator for datum feature D, that would be a contracting device with an oblong slot of adjustable size.

Correspondingly for the MMB case, it would be the fixed size oblong slot as you already showed in your calculation back on the day. By that you would ensure that the measurement method doesn't provide as much datum shift as allowed by the cylindrical datum feature simulator for a cylindrical feature as described in the standard.

Is that so?
If it is then I'm sorry, but I'm afraid you could get fired for such consultation, the way you suggested firing that vendor in a recent discussion.

In the example 4-30 the customized datum reference frame would be:
for RFS: [A[x,y,z,u,v]| B [w] ]

I don't know how you imagine that the cylindrical datum feature A can constrain 3 translational degrees of freedom, at least as long as everything applies in free state and the diameter is not chucked with force. Is that "z" addition the reason for the customized DRF? If not, what is the purpose of customization when everything else works exactly as it would by default?
The point with this example was that the customized datum reference frame from fig. 4-45, the one with the conical primary datum feature, does not work by implying translation of the datum feature simulators relative to each other in any way that differs from the behavior imposed by the default datum simulation rules for comparable cases. Yet, default degrees of freedom constraints are overridden.

This is significantly different from what is achieved by the translation modifier, which works by imposing an exception to the datum simulation rules that would otherwise apply, and doesn't necessarily override any default constraints. On the contrary, in the examples it is shown ensuring them.

 

[3DDave]

Oh, you caught me. I was tired and added the z. Redo your analysis ignoring "z"

Is there any case of a customized datum reference frame that can properly be gauged without translating and/or rotating gauge elements? If they can translate they could be doing the same thing as the translation modifier

Yes, there is one in the standard, the one with the conical primary datum feature giving away the axial translation constraint to the secondary.

The point with this example was that the customized datum reference frame from fig. 4-45, the one with the conical primary datum feature, does not work by implying translation of the datum feature simulators relative to each other in any way that differs from the behavior imposed by the default datum simulation rules for comparable cases.

First, how honestly amazing you think that they created an example of a customized datum reference that had an identical interpretation as if they had not done anything at all.

Second, the "default" is for orientation and not location. There is a basic dimension required (not shown) to place that flat surface relative to the location where the apex of the cone should be and another from there to the hole. The hole true position axis is not parallel to the flat surface - it is perpendicular to the axis of the cone. The figure 4-30 rule doesn't apply.

---

How does the fixture feature for the pin affect the fixture feature for the hole for the MMB case - it wasn't part of the 4-16(b) calculation; why would it be added?

By that you would ensure that the measurement method doesn't provide as much datum shift as allowed by the cylindrical datum feature simulator for a cylindrical feature as described in the standard.

Yes - that's the point - what is shown in the standard doesn't reflect what the restrictions placed on the features produces. I worry more about those who see a set of rules and then don't see an obvious problem with a propose interpretation that gives the wrong answer.

Still haven't asked a CAD guy to model it up for you have you?

 

[pmarc]

Did I misunderstand "Side note: Although the gage pins B and C in fig. 4-8 in '94 would indeed be of fixed size, according to para. 4.5 3(d) the distance between them should be variable, and so the whole gage would not be that trivial anyway?"

That seemed to indicate the the MMC callout didn't require the distance to be fixed and therefore the gauge, as suggested, would not be trivial. Perhaps it was a short cut and was meant to be about the RFS version and that then 4.5.3(d) would apply - but that paragraph says nothing about translation.

The paragraph says that the TGC cylinders must only be oriented to each other, not oriented and located. This applies to both cases, RMB and MMB (RFS and MMC using the '94 terminology) and fig. 4-9 shows that.

I don't have 14.43; Did they put a fixed round pin into the square hole and call it good? Because the customized datum reference frame callouts implies that there should be square pin that fits with a cylindrical base into a hole so it can turn to allow motion to let datum feature C participate. If they moved it to all MMC/MMB references that avoids putting the datum feature C simulator on a slide to vary the center distance. It becomes a much easier gage with one rotating part.

The simulator B pin is indeed a square pin with a cylindrical pilot sticked into a hole in the base so that the pin can rotate around the axis of the pilot/hole. The simulator pin C is at fixed distance from the pilot/hole axis. However, the distance is fixed not because C has been referenced at MMB, but because in Y14.5-2009 they changed the default rule for the distance between the secondary and tertiary simulator from variable to basic. In other words, if C was referenced RMB (not MMB) and in absence of the translation modifier, the distance between B and C would also be fixed.

 

[3DDave]

What I find interesting is that in the diagram in '1994 Figure 4-8 (b) is the indication of how the datum feature could be depicted to make clear that only the tangential control is being applied.

The interpretation that the distance isn't fixed I think was added in 1994, but I don't have access to the 1982 or 1973 versions to confirm. My recollection is that since requiring a fixed-boundary gauge to have moving parts is not what is reasonable except to CMM software sellers that this resulted in the 2009 reversal of that change and the adoption of the translation modifier. The customize datum reference construction is a superset of what translation does - there doesn't need to be both, though the translation modifier is a convenient short-hand.

If you write software you'll recognize that

i=i+1 gets tedious to write a few thousand times vs.
i++ or ++i, which also allow in-stream incrementing.

The translation modifier is a shortcut that does the same thing as i=i+1, but as i++. Yeah it's even more convenient, but, unlike i++, it isn't explicit as to what the effect is and then the reader has to go digging through the rest of the references to see what might be the case unless it's a trivial example.

And still, I'm left with the question - what designs demand rigid separation of features and have flexible/mobile mating parts that won't contact anything but the width (for example)?

It all seems a gift to CMM operators/software developers to not have to search for the optimum solution that fits the constraints and instead just removes the real problems of force-fits and interference.

 

[Burunduk]

I was tired and added the z. Redo your analysis ignoring "z"

What is there to redo? As I mentioned apart from the "z" your suggestion for customization of fig 4-30 (a), :"[A[x,y,z,u,v]| B [w] ]" didn't customize anything. The degrees of freedom are constrained by the primary and secondary datum references, exactly like in the original figure.

First, how honestly amazing you think that they created an example of a customized datum reference that had an identical interpretation as if they had not done anything at all

How is that an identical interpretation?
Unlike your customization of fig 4-30 (a) the customized DRF of 4-45 overrides the default constraints. If the DRF wasn't customized, A would constrain the z translation.

You are incorrect that the same rules that apply to fig. 4-30 (a) datum feature B do not apply to 4-45 datum feature B:

(g) Secondary and Tertiary Surface RMB. Where the datum feature (secondary or tertiary) is a surface, RMB applied to the datum feature requires the datum feature simulator to expand, contract, or progress normal to the true profile of the feature from its MMB to its LMB until the datum
feature simulator makes maximum possible contact with the extremities of the datum feature while respecting the
higher precedence datum(s).

This is not a rule for rotation constraining datum references only.

How does the fixture feature for the pin affect the fixture feature for the hole for the MMB case - it wasn't part of the 4-16(b) calculation; why would it be added?

Please clarify. I was addressing your explanation of the (c) case (" If the frame of reference is [A|B|D]...") first RMB (2018 7-24) then MMB (2009 4-16, 2018 7-22), what effect of pin fixture to hole fixture did I suggest adding?

I worry more about those who see a set of rules and then don't see an obvious problem with a propose interpretation that gives the wrong answer

I worry more about those that reference a standard in a drawing and then expect that the rules are ignored at interpretation. Do you reference Y14.5 in your drawings? Are you aware of what shape a true geometric counterpart of a datum feature is supposed to have? The math in the standard is not wrong as long as it follows its own rules. You could suggest that the rules are wrong, but then good luck promoting your own "workable solutions" of making users derive oblongs from cylinders.

 

[3DDave]

Should "allowed to have" be the same as "will be able to have"?

You ignored the geometry and have reverted to "the rules are the rules are the rules" again. That's not what the mathematics shows; you'll continue to be disappointed until you get your CAD guy to show you what is happening.

In 4-45 there is no boundary given that can move - more to the point, if that boundary were allowed to change, the location of the hole would not be constrained to the surface itself. One could say that anywhere along the shaft the hole was drilled that the surface could expand to reach that 24 mm dimension.

Anyway, "what size/shape is it" questions need to have a nice diagram - your CAD guy can help. I don't care to spend more time trying to figure out what you mean. I've shown you what I mean for this case; that's done.

Specifically, Correspondingly for the MMB case, it would be the fixed size oblong slot as you already showed in your calculation back on the day. By that you would ensure that the measurement method doesn't provide as much datum shift as allowed by the cylindrical datum feature simulator for a cylindrical feature as described in the standard.

There would only be one cylindrical feature on the fixture - at the hole. So, make a nice diagram that shows the feature in some acceptable orientation and location that complies with the limitations allowed for that feature and how that diameter you like so much is appropriate in meeting the requirements in a different datum reference frame.

 

[Burunduk]

You ignored the geometry and have reverted to "the rules are the rules are the rules" again

You reverted to "the math is the math is the math" again, but ignored what any math is based on: definitions.
And the math shown in the figures is based on different definitions than you base your calculations on. You miss the context of the shown calculations and that is setting the sizes related to the datum feature simulator, which is defined by certain rules. It's not intended as a math exercise to describe some virtual limit of vector-directed translation variations, the way you seem to think.

If the analysis had to be such that it accounts for specific vector-directed constraints like radial and tangential, as opposed to the standard's consistent equal treatment of displacement at any direction on the XY plane as long as there are a YZ and XZ planes established in the DRF, then if there was a position control applied to the pin located at 29 basic from B axis referencing only the planar datum feature A as primary and hole B as secondary, then the diameter symbol for the tolerance of position would be inappropriate. The tolerance zone would need to be ring shaped, with the width of the ring as the tolerance value. But the tolerance zone is generally cylindrical for a pin, regardless of the constraints that apply, with some very specific exception cases described separately.

"what size/shape is it"
questions need to have a nice diagram - your CAD guy can help

Or, it could be covered by the definitions that you haven't learned. Your copy of the standard can help. But until you get to that, the word "counterpart" in the term "true geometric counterpart" may give you a clue. By the way, I am my own "CAD guy". I hope that not your entire understanding of tolerancing comes from making CAD simulations, as useful as they are.

Your approach seems to be just as practical as that custom-made datum feature simulator for the pin to be made with the uniformly contracting oblong slot to contain the cylindrical pin in the RMB case.
And If you were that consultant hired to aid the gage design for the MMB case, you would cause the part manufacturer to build a fixture rejecting parts which are acceptable by the designer's intent, and could be approved by the inspection team of any other vendor that may get the same job.

In 4-45 there is no boundary given that can move - more to the point, if that boundary were allowed to change, the location of the hole would not be constrained to the surface itself. One could say that anywhere along the shaft the hole was drilled that the surface could expand to reach that 24 mm dimension.

I don't know why you say all that. The "boundary" can't move, but the datum feature simulator is allowed some limited level of adjustment around the basic location from the apex, per the profile tolerance. And as you could have seen in the "means this" portion, the datum reference frame contains two planes derived from the primary cone and one plane from the planar shoulder face. The datum simulator B needs to engage with the actual face, and it does that according to the requirement to progress from MMB as described in the text which I quoted for you to learn from in my previous post. Simulated datum B can't travel away from the actual face to offset the hole's tolerance zone that is basically located from it.

 

[3DDave]

Anyway, "what size/shape is it" questions need to have a nice diagram. I didn't see a diagram and you didn't ask the CAD guy. If you are a CAD guy who cannot make a diagram you need to take some classes. Why do you feel a wall of text is an answer to a geometry problem?

The diagram in the standard also didn't show a diagram of how they determined that number - same failure.

Finally - the boundary cannot move in the fixture without parts that translate. So the customized datum reference frame supports translation.

 

[Burunduk]

3D,
I agree that a CAD diagram with a bit of explanation is a good idea. I will soon post it in a separate dedicated thread.

 

[Burunduk]

Finally - the boundary cannot move in the fixture without parts that translate. So the customized datum reference frame supports translation.

That's the opposite conclusion to the one you should have made.
You've been clearly shown that the translation in the needed range for a secondary or tertiary planar datum feature simulator that has a location relationship to preceding datums, is perfectly within the default requirements, and not necessarily related to a customized datum reference OR the translation modifier.

And since the translation modifier is not intended to override default constraints of degrees of freedom, it would make no sense to try and replace it by using the customized datum reference frame for its applications.

 

[3DDave]

Oh, not intended. Well, that settles it.

The translation modifier explicitly applies extra constraint(s) a datum feature simulator places on the motion of the part. The translation modifier is a special case of a customized datum reference frame and a customized datum reference frame makes explicit what a "default" requirement does by implication.

 

[Burunduk]

If you wanted to specify the datum feature simulator behavior shown in fig. 4-19 in the '09 standard by the customized datum reference frame, what would the customized datum reference frame tell the user that the regular and unmodified one doesn't?

 

[3DDave]

Were I to write a computer program that read customized datum reference frames I could immediately parse the exact control required and would not depend on having to do an analysis of the relationship between portions of the part surfaces to figure out which directions might be allowed to control the motion.

You do understand the difference between "explicit" and "implied"?