Can we add MMC modifier on a datum feature estabilished by datum target point?

[SeasonLee]

Hello All

A simple question---Can we add MMC modifier on a datum feature established by datum target points? If yes, what is the VC size and how to build the gage? Please see the attached for details. Thanks

Season

 

[pmarc]

Even if datum feature B had an orientation specified wrt A? Say a 0@MMC perpendicularity was applied - it wouldn't be acceptable to say the 2x targets had a directly toleranced location (radius) of 0.8985-0.8670 from the center of B with a basic included angle of 90deg - aka the horizontal distance between them would be between 1.2707-1.2261 ? Therefore an MMB tolerance would fix this radial distance at 0.8985 ?

The way I interpret both feature control frames on SeasonLee's drawing is that the datum B is derived from two datum targets pins that must be located at the MMB diameter, but in absence of the basic linear dimension(s) between the targets there is no requirement for the datum target pins axes to cross at the center of the MMB diameter.

 

[chez311]

3DDave,

I wasn't trying to distract from the OP, I was trying to just pull an example of a datum target configuration which would be more conducive to establishing MMB so I could understand your point and apply it to the case at hand.

Lets imagine the OP print has a third datum target point opposed from the 2x already shown which would "capture" the 1.797-1.734 dia. If I understand you correctly, you are saying that because the datum target points only touch the feature at a single place along its height instead of acting over the entire height like a standard FOS simulator would the effects of perpendicularity error are negligible. Is that nearly correct? I think I see what you're saying but I'm not sure I agree that "perpendicularity is meaningless" - perhaps the effect could be very small or negligible depending on the amount of error allowed, but it is not zero. If the feature is produced at MMC of 1.797 and has orientation error which causes it to be oriented at a 85deg angle relative to A then the resulting elliptical shape taken at a cross section parallel to A would have a circumscribed diameter of 1.804 - now of course since this is being simulated by 3 discrete points the effect might be a little nebulous as the two points on the major axis of the elliptical section which create this 1.804 circumscribed circle can slip between the 3x datum target simulators depending on their orientation.

So this in mind, what would you say the MMB is in this hypothetical case where the feature has three datum target points specified - is it still 1.797 ?

 

[chez311]

in absence of the basic linear dimension(s) between the targets there is no requirement for the datum target pins axes to cross at the center of the MMB diameter.

I guess I don't understand why the two 45deg basic angles which intersect at an implied zero basic distance from the center of the diameter shown isn't sufficient to dictate that requirement.

 

[Burunduk]

chez311, I was thinking about this too. Maybe it is because it isn't implied that the center of the feature (shown) is also the center of the MMB envelope. The feature can shift relative to the MMB envelope.

 

[3DDave]

It isn't zero. It also isn't equal to the perpendicularity tolerance.

 

[Burunduk]

Thinking of it I doubt my interpretation from 26 Nov 19 20:42 is correct. pmarc probably meant something else in his post from 26 Nov 19 19:44. MMB boundary is never fully defined in space relative to the feature - it is defined relative to the DRF. So I still wonder why the phantom lines through the targets intersecting at the center don't help to fully define the spacing between the gage pins tips. Let us assume datum feature B is specified zero perpendicularity error at MMC, MMB of 1.797 would be fully defined. Why the datum targets location still isn't?

 

[pmarc]

The spacing of the targets is defined - I was wrong.

 

[3DDave]

Looking at '2009, 4.24.10 and 4.24.12 and the interpretation of MMB wrt to target points I can only conclude the committee member in charge of those paragraphs lost their mind. Of what value are datum target points that do not contact the part? If MMB is OK then surely an LMB interpretation can also be made. Even though the targets would be inside the solid material the part compliance with the specification would be determined the same as any other LMB case.

I am left with the question about fig 4-53. Absent an FCF governing the relationship between the cylindrical surface and the datum feature A, what exactly is the MMB of that cylinder in the A frame of reference? Is it allowed for by the default angle tolerance for the part? Could it be 85 degrees off the nominal axis? Is this just the usual 'things are left out for confusion purposes?'

As I mentioned on similar changes, I'm sure this use of allowing gaps with the targets is the result of carefully documented industrial practices that the user community was desperate to have standardized and not just blindly filling out a checklist of "did this case get covered?" If those who recall ratifying this interpretation would be so kind as to publish the request and the supporting documentation it would be of great interest. It should explain why MMB was added, but not LMB.

In '2009, PRINCIPAL CHANGES AND IMPROVEMENTS, there doesn't seem to be any notation of this significant change from the '1994 version and all previous versions.

 

[Burunduk]

Of what value are datum target points that do not contact the part?

Same value as of the portions of a cylindrical MMB datum feature simulator that do not contact the part - allowing shift. That's the whole (target) point.

 

[chez311]

Of what value are datum target points that do not contact the part? If MMB is OK then surely an LMB interpretation can also be made. Even though the targets would be inside the solid material the part compliance with the specification would be determined the same as any other LMB case.

Now that you mention it, I think I agree - it seems to me that datum targets are utilized typically because contact with the entire surface may not be desirable or would create instability (surface irregularities, casting flash, etc..). If the part will not be required to contact the simulators in 99.9% of cases unless it comes in at MMC size then I would question why they are even specified. LMB specification I think would also be dubious**.

That said I think per 2009 in OP's case, if we assume an MMB diameter is determinable, since there are only 2x target points and the diameter is not fully "captured"/surrounded (ie: translation away from the simulators is not physically limited) then I think contact is actually required even though the pins are fixed at MMB. As I said in (25 Nov 19 15:45) it seems to me it has more in common with 4-31(c) than most other examples. In 2018 with the change to fig 7-36 the part would no longer be required to contact the simulator.

It should explain why MMB was added, but not LMB.

In regards to MMB vs LMB I interpreted it to say that both are allowed even though only MMB is mentioned in the subsections 4.24.10 and 4.24.12. Per Y14.5-2009 section 4.24 "Where targets are applied to a feature of size, the appropriate material boundary modifier is specified or implied" and per Y14.5-2018 section 7.24 this has been slightly modified to "Where datum feature reference is made to datum targets applied on a feature of size, RMB is applicable unless the datum feature reference is otherwise modified."

**Edit: by dubious I mean the reasons behind specifying LMB, not whether or not it is allowed. As you can see from my response I agree it should be allowed.

 

[chez311]

pmarc/3DDave,

So do you guys think we can make the following statements about the OP case? I know you might not agree with all these, especially (1) and (2) so I don't mean to make them sound "absolute" or definitive - just trying to pull together a concrete summary. Please feel free to dissect as necessary.

1) An MMB diameter can be determined of 1.797 and two target points can be fixed in location coincident to this diameter and fixed included basic angle.

2) Even though specified at MMB and the feature which drives it is a FOS, per Y14.5-2009 contact is still required as it is only being simulated by 2x targets and translation away from the simulator is not physically limited. Per Y14.5-2018 this may not be the case.

3) The effects of orientation on a target point specified at MMB is not significant, however is nonzero and is significantly less than the orientation tolerance unless the diameter/height ratio is very large. The standard does not provide a roadmap to deal with these calculations as they vary depending on the specified target (influenced by type ie: point/line/area as well as size/location relative to the related feature) and feature geometry. Its probably best to consider this on a case-by-case basis.

4) Taking into account (1) - (3), its probably best to avoid a specification as shown in the OP case for several reasons. First - even if we agree an MMB size is determinable so the location of the simulators (target points) are determinable, MMB applied to a FOS which is simulated by a nonFOS (or targets which do not fully "capture"/contain the FOS) is confusing, as shown by the correspondence on this post. It would be better to specify at RMB with basic location of the targets. Second - in situations where the simulator might not be required to touch the part (as in Y14.5-2018 or MMB simulators which completely "capture"/contain the feature) the use of datum targets may be questionable.

 

[3DDave]

"Same value as of the portions of a cylindrical MMB datum feature simulator that do not contact the part - allowing shift. That's the whole (target) point"

That's a control on the entire surface, all elements of which is constrained, not potentially microscopic points that can allow the remainder of the surface to extrude between the target points in ways that exceed explicit limitations.

 

[Burunduk]

Datum target points are not meant to control the surface of the datum feature.

 

[3DDave]

Really - the surface is allowed to be outside the datum feature simulator? Or are you acting as if I said that it controlled the surface characteristic? That's not part of this conversation.

 

[Burunduk]

As long as the datum points simulator pin tips contact the feature, the feature is not outside the simulator.

*Edit: not all of them must be in contact in the MMB scenario.

 

[3DDave]

Take a jar lid. Draw three points that are on a larger diameter than the lid to simulate an MMB - move the lid to contact two of the points and compare to that larger diameter. You will see that the points do not constrain the lid to that larger diameter. The mobility of the lid will exceed that MMB and it gets worse as the item to be constrained departs to LMC.

 

[Burunduk]

That's why the examples in '2009 that show a primary datum axis derived by target points display 2 sets of three points. It is also specified that the 2 sets should be spaced as far apart as practical.
The secondary datum axis example deals with a much more stable procedure as some of the degrees of freedom which are involved in the misbehavior of the jar lid are already constrained.

 

[3DDave]

I should have been more clear - the face of the lid is constrained to the planar surface upon which the three points are marked and it still allows the lid to exceed the diameter on which those points are located - exactly like the secondary datum axis example.

 

[Burunduk]

Why should datum target points simulator prevent the datum feature of exceeding any diameter? There are the size and rule#1 controls for this job.

 

[3DDave]

The MMB diameter not the feature MMC diameter.