Are these gages getting a lot of use and subject to wear? My remondation would be to toss them, get a new set that is certified and then just replace the set every few years. It will be more cost effective in the long run that having them calibrated and then buying individual replacement for the worn ones.
Wipe each gage clean, check for nicks or burrs on the gaging surfaces and dress the edges if necessary, using a hard Arkansas stone. Visually check to make sure there are no flats or interruptions in the gaging surfaces. Lay each gage on a surface plate or other plane level surface to make sure the gage is not warped or bent.
Size and conformity of the radial contour of each gage may be verified using an optical comparator which is equipped with a radi chart
Record any discrepancies or adjustments made on the Gage Calibration Record
The radius gages are acceptable if they exhibit no flats or deep imperfections in the contour surfaces are within 0.005” of the size marked on the gage.
I've seen two methods. A visual (preferable) or touch probe CMM with a lot of hits, or comparison to calibrated pin gages using an indicator for outside radii. A CMM with a touch probe doesn"t work well on the smaller gages. Optical comparators generally aren't accurate enough but help to locate flats.
When you consider the cost of buying new, calibrated gages, you'll say ouch.
I'm not a radius gage expert, but from a calibration management perspective, if you discard a gage rather than recalibrate, as recommended above, don't you lose the opportunity to confirm the gage was in-tolerance, and the chance to recall/retest affected products if it was not?
AJ's procedure is right on the money. We have been calibrating them this way for 10+ years. Under normal use these kind of gages last for decades so repurchasing is not necessary. Al the "Tolerance" AJ referanced was .005 not .0005.