If you’re familiar with the different GD&T modifiers, then you probably know that the circled P creates a “projected tolerance zone.” This is often used on threaded holes to keep any fastener that protrudes beyond the threaded hole from causing interference with a mating part:
Without the “P” modifier, the tolerance zone exists only within the depth of the threaded hole itself. The result is that the threaded hole could tilt, and be passed for position tolerance, yet cause interference:
So “P” is a good thing. However, when this concept is presented in our GD&T classes, someone will occasionally ask if we could — as an alternative to “P” — simply tighten the position tolerance number instead. The dialog might go like this:
“Couldn’t we just change the 0.3 to 0.2 (or 0.1) and achieve the same effect of preventing too much tilt?”
“Yes, that would be legal,” I answer. “But using the P allows us to keep a larger tolerance, while preventing interference.”
“But it has the same effect of tightening the position tolerance anyway,” the student might reply.
This is where we have to be careful. It’s true that projecting the tolerance zone has the effect of tightening the perpendicularity aspect of a position tolerance (because it’s extended higher), but it still permits the threaded hole to use the 0.3 for lateral position. In other words, if the threaded hole doesn’t tilt, then the 0.3 is still allowing the axis to drift left or right within that zone. By eliminating the “P” and dialing the tolerance down to 0.2, we’d be robbing ourselves of some tolerance which might still be helpful in terms of x-y location.
So a projected tolerance zone is not the same as merely tightening the tolerance number. That said, there are times when a projected tolerance zone doesn’t make sense, even on a tapped hole. An example might be for a hole whose fastener doesn’t protrude very far up — such as a bolt that engages the threads for 16 mm but then clears through an adjacent plate of only 4 mm thickness.
But in general, a designer should at least consider the “P” option when imposing GD&T on threaded or press-fit holes.
This gets tricky when trying to inspect/verify that the threaded hole axis is indeed within the specified tolerance. With most threads, there is the play between the mating threads which can make measurement difficult.
Does one use a threaded gage pin that will bottom out on the reference surface to eliminate that play, or is there a better way? In my application, the threaded hole is shallow and a relatively large diameter (1.0 x 32 TPI).
Yes, I would suggest a threaded gage pin that goes into the full depth of the hole, but then protrudes upward (in this case, 14 mm). Then gage it as if it were a pin being positioned. There’s not really a bottoming out; as long as the two faces (datum features A) are in full contact.
Your question brings up a side note; on the threaded hole there is an MMC modifier — which is the “ticket” to using a fixed-size gage.
In the GD&T community there is often a debate about using the MMC modifier on screw threads because MMC typically implies that there is a bonus tolerance. I think your question kind of touches on that: even threads are subject to a little looseness (until the part reaches final torque).
If we didn’t show the MMC modifier, then it would be RFS and you’d be obligated to use an expanding-thread gage, but… good luck with that. So the play that you speak of can be used to your advantage to verify the position tolerance.
Dear Mr. Belanger,
I heard about the debate of MMC versus RFS for threaded holes, but I never quite understood the “ticket to using a fixed size gage” argumentation by the camp that sides with MMC. Maybe you can clarify this a bit; when the MMC modifier is used in a position callout for a simple hole (not threaded), a fixed size gage simulates the virtual condition boundary not to be violated by the surface of the hole, and it is fixed not just in size, but also in location and orientation to the referenced datums. Contrary, for a threaded hole, any threaded gage assembled into the hole can’t be fixed in location and orientation to the datums, from what I think are obvious reasons. In addition to that, its purpose is different from setting a limit for where the surface of the threaded hole may be. So aren’t VC and thread gages have too little in common to have both implied by the same modifier?
Yes, that’s a somewhat subtle (and admittedly debatable) idea. Here’s the key… while it might not get a slip-fit type of functional gage, threads do still have tolerance on their pitch diameters, which can be seen on all the thread tables that are easily accessible. Thus, there is still some (admittedly small) clearance between internal and external threads, such that their pitch diameters might not be coaxial. What follows from that idea is that the any pitch diameter that deviates from the MMC based on the charted tolerance could have greater coaxial error.
However, while I stand by the fixed-gage idea, it would be imprudent to try and make a bonus tolerance chart like we often see for non-threaded clearance holes modified with MMC. The coaxial offset that I mentioned above is mitigated by other factors, such as the fact that the form of the threads makes them want to self-center.
So I think there is such a thing as VC on a threaded feature with a position tolerance. But there probably won’t be any bonus or “jiggle” sensed, because the threads are assembled together with the proper torque to hold things tightly.