Composite Position Tolerance Simultaneous Patterns 2

[jeballes]

I've got a cylindrical part with two patterns of notches around both ends. I'd like to position both of them as one pattern, but have the two patterns controlled by separate feature-related tolerances. If I tolerance them as shown in the attached drawing, will the PLTZF for both patterns be set at the same time by the simultaneous requirements rule? Or does that only apply to the datum reference frame?

 

[powerhound]

If B was not in the FRTZF then the centerplanes of the cutouts could be out of parallel to datum axis B by .015. With reference to B the error is limited to .010. What am I missing?

John Acosta, GDTP Senior Level
Manufacturing Engineering Tech

 

[jeballes]

Datum axis B would be perpendicular to datum plane A, and the FRTZF would already be perpendicular to A, so their parallelism to B is already controlled, right?

 

[powerhound]

Yep, I was too hung up on B only, to see the big picture. Sorry for the distraction.

John Acosta, GDTP Senior Level
Manufacturing Engineering Tech

 

[greenimi]

Datum axis B would be perpendicular to datum plane A, and the FRTZF would already be perpendicular to A, so their parallelism to B is already controlled, right?

Yep, I was too hung up on B only, to see the big picture. Sorry for the distraction.

So, do we agree that B as secondary in FRTZF does nothing?
No degreees of freedom stopped and no valuable addition in the FCF (again the secondary B in FRTZF)

 

[axym]

Hi All,

Interesting thread, sorry that I'm late joining in.

If you want a non-intuitive callout with subtle and hidden rules, nothing fits the bill like composite position ;^). Here are some comments:

I agree that the B reference does nothing in this example. B cannot constrain the one rotational DOF that is not already taken by A. But if the FRTZF referenced A|B|C, then the B reference would have an effect and would be different than A|C.

The analysis of the "relative clocking" of slots in the two patterns brings up some subtleties of composite FCF's and simultaneous requirements. Even some gray areas, I think. Here are some issues that come to mind:

1. Each FRTZF can rotate independendently of the other since a clocking datum feature was not referenced.

2. Each FRTZF can translate independently of the other, and independently of the PLTZF. This complicates the relative clocking, because each FRTZF can be rotated around a different center.

3. By default, FRTZF's are not subject to simultaneous requirements. To override this default, Y14.5 allows SIM REQT annotations to specified beside each FRTZF. But I don't think that the effect of SIM REQTS is completely straighforward in this case. Would SIM REQTS control the relative orientation and location of the two FRTZFs (so that the two zone patterns share a common center), or just control the relative orientation of the two FRTZFs (so that they don't share a common center) ? I would say that if you asked several Y14.5 experts, the results would not be unanimous.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[jeballes]

If we give the tolerances MMB modifiers and used a functional gage to inspect the SIM REQT FRTZF, wouldn't the notches simulating the virtual conditions of the slots be fixed in orientation to each other? So if any two slots were clocked to opposite sides of the PLTZF, the part would pass PLTZF inspection, but fail the FRTZF inspection.

 

[chez311]

3. By default, FRTZF's are not subject to simultaneous requirements. To override this default, Y14.5 allows SIM REQT annotations to specified beside each FRTZF. But I don't think that the effect of SIM REQTS is completely straighforward in this case. Would SIM REQTS control the relative orientation and location of the two FRTZFs (so that the two zone patterns share a common center), or just control the relative orientation of the two FRTZFs (so that they don't share a common center) ? I would say that if you asked several Y14.5 experts, the results would not be unanimous.

Evan - that took me a second to wrap my head around, but I think I understand what you are saying. It seems to me that even adding datum C to the FRTZF would only constrain rotation of both FRTZF's around datum plane C and doesn't solve the ambiguity you have pointed out about whether the FRTZF's would be share a common center/be constrained in location relative to one another.

If that is the case, what would be your recommendation to create a tolerance that is unambiguous? My only thought would be to either switch to Multiple Single Segment (which would of course change the nature of how the lower segments could float relative to the datums) or make them all part of a single pattern (which would require either tightening/loosening of the FRTZF tolerance on one set of slots).

 

[axym]

jeballes,

The tolerance zones (or virtual condition boundaries) for each FRTZF are fully constrained in orientation and location relative to each other. The two FRTZF's are also constrained in orientation relative to each other. But I believe that it is possible for two slots clocked to opposite sides of the PLTZF to pass the FRTZF inspection. It would have to be one slot from the 4X pattern and one slot from the 6X pattern.

This is because the two FRTZF's could translate relative to each other even with SIM REQTS (in my opinion, anyway), and this would have an effect similar to clocking error. Some of the slots would get displaced "clockwise" and others "counterclockwise". If the two FRTZF's translated in opposite directions, then two of the slots could be at opposite extremes of their PLTZF's.

chez311,

I think you're right that even adding a reference to C in the FRTZF would still allow the translation effect described above. If relative clocking is the main functional requirement for the lower segments, then it may be that multiple single-segment FCF's would be a better choice than composite FCF's. Perhaps this:

6X 1.000
POS|dia .015|A|B|C
POS|dia .005|A

4X 1.000
POS|dia .015|A|B|C
POS|dia .010|A

This would give you the relative clocking of slots in the two patterns. The 6X pattern of zones and the 4X pattern of zones would be constrained in orientation and location relative to each other, because of simultaneous requirements.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[jeballes]

If the FRTZF could translate relative to each other even with SIM REQT applied, then what does SIM REQT do? The patterns should be constrained to the same datum reference frame, even if that reference frame is allowed to rotate or translate. Figures 4-40 and 4-41 in 2009 describing SEP REQT illustrates this well.

 

[chez311]

jeballes,

I initially had the same thought, but I think it has more to do with how the FRTZF behaves in relation to the PLTZF for composite tolerancing. And I think you could argue both ways, as Evan noted it is not clear cut.

Imagine if there were a clocking datum plane in the figures you mentioned 4-40/4-41. Then both keyseats would be constrained both in location and orientation to the center axis (datum A in the figure) and imaginary datum B (clocking datum - not shown because i made it up). The same is not so for a composite tolerance - even in your original example if A|B|C were called out in the FRTZF it only constrains rotation/orientation of the pattern - it is still free to translate. I could see it argued that if it is not constrained in translation/location to the datum reference frame then why should they be constrained in relation to each other.

I'm not advocating for one over the other, I'm just saying I can see how it may be ambiguous and if it is a critical feature - relying on simultaneous requirements with a composite datum may not be the best way to go.

Edit: just realized that you are the OP jeballes - I was referencing "OP" when I was referring to you!