Ordinate dimension precision(s)

[mamartin1323]

In my role as CAD Manager from time to time i get a question that seems to only have one answer, one which i hate but i'm hoping you smart folks might be able to help me not to not use, that is "because I said so..." We are currently working with sheet metal parts where the mounting holes are the critical dimension(s) the part overall dims are not as critical i.e. holes are 3 place dim's and overall part maybe 2 places dim's. in the past when doing ordinate dimensions I have always kept all the precisions in the set the same, when 0.000" is origin then the last will be 3.312" in order to lessen the tolerance to the final i would add a +/- tolerance not change the precision to that particular dimension. Is there a CAD standard (other than ASME Y14.5, checked there first) that speaks to this or have I just got this rule stuck in my head? Thanks for your comments and insights
Ps I realize I could just have them change to linear dim's but ordinate are cleaner on this particular part.

[tedg]

Not really my particular discipline, but have a little experience in mechanical design...
Are you saying the hole locations are to the thousandths but the overall sheet metal dimensions are to the hundredths?

I don't know what the "normal cad standard" would be, but I would simpy have a different dimstyle for each of those tolerances.
It would be as easy as dimensioning your part with one dimstyle, and locating your holes with other dimstyle.

And maybe have a note on your title block explaining the required tolerances (+/-) etc.

Hope that helps, but I'm sure there's some AUGI members with more experience with this.

[mamartin1323]

that is correct, overall (highlighted) could be hundredths, hole/ feature loc's thousandths, the title block states 3X = +/-.005", 2X = +/-.01"
as I stated I could have him change them to linear dim's or tolerance the overall dim but for some reason changing the precision on that dim in a set doesn't seem like good drafting practices to me... which is what he did, i redlined it and sparked this discussion... I could have an instructor from 25 years ago stuck in m head telling me so but was hoping there was a standard other than Y14.5 that may speak to this...
ex.JPG

[jaberwok]

I believe that ordinate dimensioning is no different from linear dimensioning from a common point - except, of course, that, for consistency, the dimension from the corner to the first hole would also be to two places so the origin would actually be the first hole centre.
Since the titleblock states tolerances against decimal places this works.