Sr GDTP Y14.5-2009 Exam Review Ch-4 Part-II Sep2021

[metrologic]

Sr GDTP Y14.5-2009 Exam Review Ch-4 Part-II Sep2021

Study review Part-II of III for Section 4, Datum Reference Frames. This thread covers subsections 4.12 thru 4.20.

Q1. 4.12.1 Simulation of a Single Datum Plane. Section 4.12 discusses multiple datum feature references, but I'm not concerned about that here. Fig 4-22 and Fig 4-23 illustrate some unopposed flat planes being used as primary datum features. These flat datum features are controlled with a simple profile tolerance. At least it seems simple. But if I'm not mistaken, this is the first instance where such primary datum features are being controlled by profile, instead of flatness. In these examples, does the flat datum feature simulator surface progress from an MMB toward an LMB? And since we are talking about moving a plane, what is the movement relative to? That is to say, does the datum feature simulator move relative to the location of the geometric tolerance zones in Fig 4-22 and Fig 4-23? Is the datum reference frame (DRF) established relative to the basic location of the datum feature simulator? Or does the DRF progress with the datum feature simulator. Please compare and contrast this with the example presented back in Fig 4-2, where simple planes controlled by flatness or perpendicularity establish the DRF.

Q2. 4.12.4 Pattern of Features of Size at RMB. "The datum feature simulators shall expand or contract simultaneously from their MMB to their LMB until the datum feature simulators make maximum possible contact with the extremities of the datum features(s). See Fig. 4-25." What does that mean? Are we simultaneously iteratively offsetting an equal distance for each simulator? Or are we scaling everything from a datum point or axis? I recall a lot of different adjectives being used to describe this process, but there never seems to be a precise definition developed. (And what does maximum contact mean, and does that always occur at minimum separation? Maybe it's covered in Y14.5.1M.)

Q3. 4.12.3 Pattern of Features of Size at MMB. "The origin of the datum reference frame may be established at the center of the pattern...or at any location defined with basic dimensions...." What about all those prior examples showing where to put the DRF in relation to this and that? Now we can just put it anywhere?

Q4. 4.12.4 Pattern of Features of Size at RMB. It seems like patterns referenced at RMB would suffer from instability frequently. I mean, it just takes one feature that is smaller or bigger than everything else in the pattern to prevent the datum feature simulator set from continuing its advance from MMB to LMB.

Q5. 4.14 Multiple Datum Reference Frames. "Where more than one datum reference frame is used and it is necessary to determine the relationships and calculate the boundaries between the reference frames, the relationship between the reference frames shall be specified." I guess I kinda understand what is being discussed here, but this could really use an illustrative example. I've never seem something like this before.

Q6. 4.16.6 Offset Planar Datum Feature Set at Basic Constraining a Rotational Degree of Freedom. Look at Fig 4-31. The figure has three parts (a), (b), and (c). On the left side they show a "This on the drawing" and on the right side they show a "Means this". In Fig 4-31 (a), the offset secondary planar datum feature B is referenced RMB. And the "Means this" illustration says "No translation or rotation of datum feature is allowed" -with a leader pointing to datum feature B. In Fig 4-31 (b), the offset secondary planar datum feature B is referenced as B[BSC]. And the "Means this" illustration says "No translation or rotation of datum feature is allowed" -with a leader again pointing to datum feature B. In Fig 4-31 (c), the offset secondary planar datum feature B is referenced MMB. And the "Means this" illustration says "Datum feature B must remain in contact at a minimum of one point". Why is the "Means this" description different for datum feature B referenced at [BSC] versus MMB? Is there a subtle difference between B[BSC] and B modified with MMB that I am not seeing?

Q7. 4.16.7 Offset Planar Datum Feature Set at MMB Constraining a Rotational Degree of Freedom. As mentioned earlier, in Fig 4-31 (c), the offset secondary planar datum feature B is referenced MMB. And the "Means this" illustration says "Datum feature B must remain in contact at a minimum of one point". It also states in this subsection: "Where the datum feature simulator and the higher precedence datum axis do not limit rotation in both directions about the datum axis, the datum feature must always contact the datum feature simulator." And back in 4.11.9 it said, "The datum feature shift/displacement shall always be limited or constrained by the datum feature simulator. If the datum feature simulator geometry is such that it does not fully limit or constrain the feature...beyond the established boundary limits...then the feature must remain in contact with the datum feature simulator, and datum shift or displacement is not allowed." Is one point of contact enough in all cases? Or does it depend on the circumstances such as order of precedence and degrees of freedom under consideration, etc.? I mean, if only one point of contact were necessary, wouldn't that allow for some significant additional freedom of movement in many scenarios? That sounds like datum shift. How exactly is datum shift disallowed for non-compliant datum feature simulator combinations?

Q8. 4.17 Application of MMB, LMB, and RMB to Irregular Features of Size. Back in subsection 2.7 I read: "Unless otherwise specified, the limits of size of a feature prescribe the extent within which variations of geometric form, as well as size, are allowed. This control applies solely to individual regular features of size...." Now here we are in subsection 4.17 and I read: "MMB, LMB, and RMB may be applied to irregular features of size when they are selected as datum features." If irregular features of size can have RMB, MMB, and LMB modifiers applied when used as datum features, why can't they have perfect form boundaries at MMC like regular features of size? Why were irregular features of size excluded from subsection 2.7's form control dictum?

Q9. Fig 4-35. This figure shows four different ways three pins could function as irregular features of size for datum features. How would these examples typically be toleranced? Are the pins themselves toleranced? Or would there be a tolerance associated with the irregular diameter or width?

Q10. 4.20 Restrained Condition. "In a restrained application, it is permissible to use as many datum targets as necessary to establish the datum features." It's not permissible to use as many as we want for free-state applications????

 

[3DDave]

Too tired to deal with all of them now - low hanging fruit:

A10: At some point in the free state, all possible degrees of freedom have been controlled making more targets than that either redundant or conflicting. Recall that the standard is written to address reasonable uses and not to prevent users from malicious or nonsense uses*. In contrast, restraint adds as many degrees of freedom as there are targets being restrained against.

*very frustrating over the years of "The standard doesn't say I cannot do this stupid or ridiculous thing so why can't I?"

 

[greenimi]

metrologic,

I said it before: Do you think you are efficient and effective having posts with that much information?
Normally, you are expecting answers on all your questions, right? Otherwise, you wouldn't write them here, correct?
Do you realistically, think someone will get each and every question, dissect it and try to answer it?

Again, multiple posts will / maybe, be a solution....

 

[metrologic]

@greenimi
I think I've probably reached peak questions per thread. See if you can help me out with Q1.

@3DDave
Regarding A10: Are there redundant points when the entirety of a feature is used as a datum feature? The points of contact may fluctuate from sample to sample due to manufacturing variation.

 

[3DDave]

In the ideal condition trillions of points are in contact - ideally every atom on the part surface of a datum feature is in contact. There is only one requirement, that elements on that surface are in contact so none of those contacts is redundant.

Each target point is a separate requirement and when there are more than required to establish a constraint then the extra requirements are redundant.

 

[metrologic]

@3DDave
I can't see the difference you are describing between full datum features and target sets. Some targets might not make contact, some will, but on the next sample, things might be reversed. The whole concept seems set up to function in a manner identical to multiple datum references, or hole patterns, etc.

 

[3DDave]

Not sure how else to say it; final try: If one is to require contact with all the targets and the part is warped and can only contact a few in the free state that is not the same as requiring contact with one datum feature which is warped only allowing contact with that datum feature - oh, wait - it is both required to and does make contact with that one datum feature.

Contact with a datum feature or a datum target is generally not optional - perhaps you believe it is?

 

[pmarc]

metrologic,
I don't think the standard says that one shall not use as many datum targets as needed in a free state. See para. 4.24.8, where words like "at least" and "usually" are used.

I am visualizing a traditional four leg table. While the contact with the floor will most likely be at 3 legs only in free state, I don't see why I should not be able to define all four legs as datum targets or datum features to establish a common datum plane from which I would measure the height of the thing and/or parallelism of the table top in a free state.

 

[3DDave]

If the four surfaces at the ends of the four table legs are given a single profile tolerance and the four surfaces collectively identified as a single datum feature resolving to a single plane then it's allowed/expected the table will rock - a condition already accepted as the case if it was a single surface with the same span and covered in the math standard.

If they are identified as individual datum targets there is the expectation that they will each have to contact their respective datum simulators/true geometric counterparts. If the targets are overconstraining the part in the free state the standard has no explanation of what to do for that case.

However the committee never really says they have to not overconstrain - maybe they just don't expect people to want to make poor choices, or maybe they don't care if they do.

Their reasoning on restraints relative to datum targets tells what they do expect.

 

[pmarc]

Perhaps I am reading your reply incorrectly, but I don't think that in the case when someone decided to take the datum target approach, like A1-A4, (and still use profile to control relationship between the faces) it would be any different than in the scenario where the four faces were just labeled as a single common datum feature.

 

[3DDave]

The standard is silent on that isn't it? Do what you want.

 

[3DDave]

Examining the sentence:

In a restrained application, it is permissible to use as many datum targets as necessary to establish the datum features.

What is the other case, for the unrestrained application? If the previous statement is only true for the restrained application then the clause should be negated for the unrestrained application:

In an unrestrained application, it is not permissible to use as many datum targets as necessary to establish the datum features.

Therefore, using targets without restraining the part must never be enough to establish the datum features.

Is that a reasonable statement? If not, then what is the defined case in the standard for the number of datum targets necessary to establish datum features for the unconstrained case?

 

[pmarc]

If you really want to play this game, one way to refute your argument would be to say that the sentence you quoted merely describes what is allowed in a restrained condition and says nothing about the free state. And since para. 4.23.8 uses words like "at least" or "usually" without making an exception by mentioning a restrained state, this opens the door for using more datum targets than needed.

But putting this aside, and going back to the four leg table example, I don't really see why someone shouldn't be allowed to use four datum targets without having to use a restraining note. Functionally, it may make no sense to restrain the table against the floor to measure its height or parallelism of its top to the floor.

 

[3DDave]

Why waste time specifying targets that are optional?

 

[pmarc]

I am pretty sure there are tables or table-like parts where not every portion of the bottom surfaces is supposed to touch the base.

I have also seen aircraft engine parts where 4 datum target areas were defined at 4 corners of a single surface without restraining anything just because there was no chance to arrange the targets in the triangular pattern as shown in fig. 4-47 and 4-48 in 2009.

 

[3DDave]

I've seen some really bad applications of Y14.5 myself. Usually it was because QC didn't have the right tools and needed a fictional way to get around some lack of ability. Not once was there a matching tolerance analysis that showed a functional effect on delivered items. Half the time it tossed usable variation out the window and half the time it would accept unusable parts but QA assured engineering that "they won't make them that way." Same QA turned a blind eye to the use of grinders at final assembly for those parts.

 

[pmarc]

Believe me, none of that happened in the case I brought up.

 

[3DDave]

Of those items tied up in "none of that happened" is "functional effect" a part?

What was likely done was a geometric fit analysis alone, but not any stress analysis, nor any performance analysis.

How do you know that usable variation wasn't excluded without a related stress or performance analysis? Or that the choice of targets and allowing a rocking condition wasn't a work-around? Or that some combination of acceptable variation would not lead to unacceptable performance?

The fact that these analyses are not part of any generalized dimensioning and tolerancing document I've come across, when they should be a top topic, tells me that it is so rarely done as to be not considered. Likewise missing are "This is the performance requirement, what is the matching geometric requirement to get there?" The closest I've seen is how much interference on a single interface is required to produce some amount of friction - a one-dimensional analysis.

 

[pmarc]

Of those items tied up in "none of that happened" is "functional effect" a part?

What I was trying to say was that all necessary aspects had been analyzed before the four datum targets found their way to the drawing.

 

[3DDave]

I'm sure they were; the state of the art for such analysis is well understood as the volumes of similar analyses clearly demonstrate. The ANSYS GD&T [sic] Module is a top seller. /s