How much variance in rim diameter is normal?

[bzo]

I'm building some wheels and purchased a pair of alloy rims. One of the rims is noticeably smaller than the other one and the ERD measures 2mm less (578mm vs 580mm). As best as I can measure, the smaller one seems to have a bead seat diameter of 620mm.

I plan to run these tubeless so am a little concerned about the fit of the bead. When testing fitting tires with tubes, the larger one snaps audibly when tire is inflated and beat seats, whle the smaller one doesn't. The tire still seats on the smaller rim without a gap, but is not as tight as the other rim.

Is this kind of variance normal or should I be sending the smaller rim back as defective?

[Tenlegs]

Is the bead seat dia is outside ETRTO tolerances?, best I could find is Min 621.45mm Max 622.45mm
If your rim BSD is 620mm it seems it's far too small, the rim blank would have been about 6mm short to end up that small.
(all assuming you've accounted for the roundness of the rim)

Scroll down to section 7.1
http://engineerstalk.mavic.com/en/road- ... safe-fast/

More info

http://www.wheelfanatyk.com/blog/kgf-v-psi-part-2/

[bzo]

Thanks for the links Tenlegs. I wasn't able to find any info on the tolerances myself, so that was helpful. The WheelFanatyk blog post says that uniformity is usually tight, but the ones he measured seem to be more established brands with presumably better QC than this cheap Alex. I've also found some references to builders saying that a 2-3mm variance in ERD for alloy rims isn't uncommon.

I re-measured the bead seat again more carefully, and the small one is 621-621.5 depending on the spot, so I guess it's slightly below the published tolerances and something I can live with. I'll just put on an extra layer or two of tape. The larger one is 623-623.5, which makes the average right on, but me unlucky in getting the outliers.