Multiple geometric tolerances at MMC

[AH_AK]

I am having a bit of confusion around how to interpret a feature that has both position and perpendicularity tolerances attached to it. I get that the position control will, by itself, restrict the orientation and my understanding was that an additional orientation control was only added if tighter control of the orientation was necessary for functionality.

The reason this is messing me up is because I was under the impression that cylindrical tolerance zones and VC's for position tolerances were located by basic dimensions (i.e. placed at the true position of the feature). Since orientation tolerances don't fix the position, I assumed their tolerance zones and VC's would be floating and that the UAME axis or feature surface just needed to fit in the appropriately sized cylindrical tolerance zone (or not intersect the VC), but that the cylindrical tolerance zone and VC could "slide" around to find a successful configuration (in the absence of positional tolerance).

This lead me to an interesting dilemma on how to apply the bonus tolerance when the UAME deviates from MMC.

Here is the example that got me thinking:

They seem to be analyzing only the orientation(perpendicularity) tolerance, which makes sense as the example is in the orientation tolerance section. They use a 50mm "orientation VC" this suggests to me that the orientation control is to be considered independently of the position tolerance. I can't tell from the pictures if the positions of the "orientation VC" and tolerance zones are fixed at the true location w.r.t. datum B or if they are "floating". Am I wrong? Are the positional VC and orientation VC both fixed at the true position?

I get the advantage of shifting the perpendicularity tolerance to the size tolerance and then allowing the bonus tolerance as the UAME departs from MMC to control the perpendicularity (and position), but I don't understand the correct approach to analyzing the combined effect of both the position and the perpendicularity tolerances.

Logically, I would think these two controls would be verified separately (along with the size tolerance). The positional VC (surface method) and tolerance zone cylinder (axis method) would be fixed at the true position, while the perpendicularity VC and tolerance zone cylinder would be "floating". At MMC, it is pretty straightforward as the tolerance zone diameter would be zero for the perpendicularity control and 0.3 for the positional control.

The problem is that as the UAME departs from MMC. Bonus tolerance is available, but how would it be "distributed" between the two controls? For instance, if the UAME had a diameter of 50.15, then there would be 0.15 bonus tolerance available. Would this be shared between the position and orientation controls, or, would each analysis be provided with the extra 0.15? The former makes more sense to me from a big picture assembly interference avoidance standpoint.

I feel like I am getting pretty deep in the weeds on this, but I really want to understand the appropriate methodology.

Can one of you sage GD&T folks help me understand the correct way to analyze this problem?

 

[3DDave]

Each geometric tolerance is separate. Nothing is distributed between them.

 

[Burunduk]

You are correct in your interpretation that the positional tolerance zone and VC are fixed to the true position axis by the basic dimensions (explicit, implied, or queried).
You were also correct that these are separate controls (even though the perpendicularity tolerance indeed refines the orientation error that the position tolerance is also able to control). Suppose that the axis method is implemented. The two tolerance zones are not bound to each other, but for the hole to be able to conform to both tolerance zones the perp tolerance zone would have to be at least partially included in the position tolerance zone, even after floating to a location that would minimize the actual perpendicularity error measurement, but nevertheless it does float. As the controls are separate, the bonus gained by departure of the UAME size from MMC would be fully available for each of the controls, so both tolerance zones would grow by the same amount from 0 and 0.3. Why should this be a problem?

 

[AH_AK]

Thank you both for the replies.

I was just thinking about the possible acceptable configurations incorrectly. I was assuming that the bonus added to both would push the hole inside the position VC surface while still remaining acceptable by the axis method. I have it in my head that as a designer the VC at MMC represents the worst cases scenario surface (not to be violated). The thing I was neglecting to realize is that since they are applied separately, this VC would remain unviolated. For instance if a larger hole was produced, it could be shifted by the entire 0.3 + full bonus, but the AME axis is going to be right on the edge of the tolerance zone. Any additional tilt will put it outside and it will no longer satisfy the position tolerance zone. In other words, if the actual UAME axis is in the worst case scenario, it must be perfectly perpendicular, otherwise it would violate the position tolerance zone. So yeah, no need to track or distribute bonus. Just form the separate tolerance zones and the produced hole either passes or fails. The position VC will not be violated. From a design intent standpoint (something needs to fit through here) this seems important. I was just thinking about it wrong. Thanks for the clarification and reassurance.

 

[Burunduk]

Another thing that is helpful to remember is that each control with an MMC modifier on the tolerance imposes a separate VC boundary. In the 9-14 example which you posted, the position tolerance produces a 49.7 mm VC boundary related to datums A and B, and the perpendicularity tolerance produces a 50 mm VC boundary related to datum A only (this one is a floating boundary since there are no constraints of translational DOF). These are different VCs of different sizes applying in different datum reference frames, so it's not useful to think of a single "VC" that guards some sort of collective effects variation...

A different point is that the perpendicularity tolerance acts as an orientation refinement of the position tolerance. So the one thing that could be problematic is if the perp. tol. was applied at MMC while the position tol. would be applied at RFS, in a way that the total perp. tolerance (including bonus) could exceed the position tolerance (impossible with the current tolerance values in the example). If done this way, it would be a mistake, because in such situation utilizing the full perp. tolerance would mean nonconformance to the smaller tolerance of position (the perp. tol. would no longer act as a refinement).

 

[AH_AK]

Excellent points.
I am trying to understand the implications of these tolerances w.r.t. mating parts (fit). I was just using the position VC to visualize the hypothetical mating boss. It would appear the size and positional tolerances are sufficient to ensure the fit of a mating component. The orientation tolerance further refines the orientation w.r.t. datum A without counterfeiting the fit.

There is a lot of nuance to this system beyond the rules. If you don't really understand the implications of the tolerances, I imagine it would be easy to follow the rules, but end up inadvertently compromising the functionality of the assembly. Practical experience seems to be key to fully realizing the advantages of the system.

Again, thank you for the thoughtful responses.