Slot GD&T

[GravyBagel]

I read this article by Bill Tandler and thought it was very interesting, but I wasn't satisfied with the solution.

https://go.gale.com/ps/i.do?p=AONE&...|A328422062&v=2.1&it=r&sid=AONE&asid=fb82bcc4

With a linear slot that is aligned with the datum reference frame, would it be possible to create a multiple single segment position callout for the slot where there is a position tolerance referencing the AB direction (A being flat, B bottom edge, and C side edge) that dictates form error, then a position tolerance referencing datum C that dictates the length tolerance perpendicular to datum C.

Would this work or am I missing something? Has something been updated in the 2018 standard that allows for loosening up of length on a slot without loosening up the width in proportion?

 

[3DDave]

Typically there is a separate segment associated with each orthogonal direction of the slot. This has been in the standard for a long time; no need to update.

---
I like the hand-waving Tandler does about the computer controlled mill.

I have pointed out that such machines are what often makes this building of sequential references nearly moot. A part milled from a plate will get the datum features cut in the same setup as the rest of the boundary and every other feature in that plate.

The main stackups will be from errors in the size of the cutter, deflection of the tool, normal errors in the positioning of the machine.

At that I am underwhelmed by the argument for creating +/- tolerances on basic dimensions.

---

The biggest (and still insignificant problem) is that mathematically the displacement perpendicular to the narrow width of the slot creates an overlap at the round ends of the slot in terms of its virtual condition, creating a pinch-in that isn't a smooth arc. It looks in part that this is the effect Tandler is dealing with, but obliquely, by trying to force a uniform boundary around that part of the slot. If the slot was rectangular there would be no questions about this.

Draw a picture of the virtual condition if the narrow-way is allowed to have half the width of the slot as a location tolerance - say a .500 slot with a .250 location tolerance on the midplane.

His other suggestions are terrible. If the feature is created in one operation as suggested then there won't be two (or 3) separate surfaces so the depth of the counter-feature (15 nominal) is over defined.

 

[GravyBagel]

Thank you 3DDave. This makes good sense. For the slots I as making, I ended up using a max material condition callout with different size tolerances on the width and length and this seemed an appropriate path forward. Seemingly, the bonus tolerance would only get applied in the direction of the dimension.

Hopefully I'm not off base here.