Minimum size for datum feature
Minimum size for datum feature
(OP)
It often seems to happen in my part designs that only a small percentage of the overall size and mass constitute the critical interface with mating parts. I want to use these features as datums and position other critical features to these but I wonder if they are too small. For example I have in work some linkage rods where only the two tabs at the end (which are at skewed angles to each other) mate with connecting parts. The bulk of the rod (which could make a sizable datum feature) has no interface and no need to be of precise size or shape. I could not find any info on this is 14.5 1994. Is there a general rule for minimum size of a primary datum feature?





RE: Minimum size for datum feature
There's no established threshold for how long something must be. It just has to be suitably long/large enough to repeatably establish the real datum, which is a theoretical plane, axis, or point.
John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems
http://www.gdtseminars.com
RE: Minimum size for datum feature
One idea I have: If a feature can be fully verified to be within it's specified form or position tolerance then it's size and location to the remainder of the part can be established; therefore, it can be used as a primary datum. While that sounds true in theory there must be some practical limit, and as John-Paul stated I think the key is "repeatability".
The work around I see is to make the non-critical bulk of the part the primary datum and in the case of the linkage rods I mentioned the two tabs are located to it. The sucky part is that to have the tabs at the tolerance I need wrt each other I can only allow half the tolerance wrt the primary datum.
Anybody from the inspection world who can shed some light and provide guidance on this?