For a manufacturing tool, we're designing what is essentially a die set using off-the-shelf shaft, bushing, & roller element components and pressing those into in-house designed plates. Two shafts are press fit into the base and are quite long. The two mating bushings are press fit into a plate. A cage with captive ball bearings fits between the OD of the shaft and ID of the bushing, providing a rolling contact. Die sets like this are commonly available off-the-shelf (e.g., here: xxxxx, we’re just mimicking with custom-shaped plates.
I've been tasked with assigning the shaft & bushing holes' positional tolerances & perpendicularities. The dilemma arises since the die bushing vendor indicates that the shaft-ball-bushing fit is preloaded, ie a “rolling press fit”. As I understan it, there’s actually miniscule interference between each of the dozens of ball bearings and the shaft & bushing. So this is a very rigid fit with no play (clearance). This poses a theoretical problem for pairs of these bushings in terms of shaft-to-shaft spacing & perpendicularity. You’re basically trying to tolerance corresponding press fits, so fixed fastener type approaches are out the window (zero clearance in the “clearance hole”).
A vendor competitor’s design guide recommends shaft-to-shaft spacing tol of +/- .0005” which I’m guessing they plucked out of the air as a tight but achievable machining tolerance. Mating two pairs of zero clearance interfaces needs perfect location and perfect orientation though. Like I said, these sorts of assemblies are available off-the-shelf lots of places so this is achievable. Commercial die sets don’t use slotted components or flanges w clearance holes – they’re exactly as we’re attempting. Thoughts on how to approach tolerancing the location & orientation of these holes?
You can use gauge tolerances for dies or ANSI 4.1.
Tell me and I forget. Teach me and I remember. Involve me and I learn.