Sorry guys - I've been out of town, just back for a day, and out of town again tomorrow until next week. But I did receive the bit last week, and I put a micrometer on it this morning. Hopefully I'll have a chance to radius some wood this evening.
The bit looks well made; the carbide is sharp and the edge is smooth. The brazing to attach the carbide to the steel shaft is quite a bit cleaner than my other bits which show some small gaps.
The top bearing is a regular 1/2" bearing, the measured diameter is 0.512" which is standard spec for these bearings. Like my other bits, it can be removed and replaced if it ever wears out.
The diameter of the cutter just under the bearing is 0.494" so a little smaller than the bearing.
The depth of cut is a little over 2" so you can do a 4" wide board no problem.
The diameter of the bit at the 2" cut depth is 0.79" - with the measured top diameter, this measurement should be 0.846" (0.494" + 2*0.176") for a 12" radius. So the bit is a little flatter than 12" - when I CAD it out, my measurements suggest a radius of 14.3" - but when I hold the cutting edge up to 12" and 14" radii printed from CAD, it looks like the radius is just fine at 12". So it could be that the bottoms of the cutters themselves are ever so slightly tilted out of parallel toward the center of the shaft.
At the 2" cut depth, it doesn't take much to be off by a few hundredths and there is a big error magnification on the radius - in this example, if both of the cutters are off perpendicular by 0.028" that translates into a difference of 2.3" on the effective radius - an error magnification of almost 100. It is not an easy job - if you mill the radius on the carbide cutters before you attach them to the shaft, then you need a very accurate jig to weld them on straight and parallel to the shaft, and have both cutters the same distance from the center of the shaft (to within a few thousands of an inch). It is far more accurate to attach the cutters first, true them up, then mill the radius, bevel and sharpen.
My original bits are a bit better than this (±0.3" on the radius) but even still, they do leave the usual tool marks on the board that require some touch-up with a radius block - and it would not take much at that point to knock the radius down from 14" to 12". I'll try to mill some wood this evening and evaluate the bit "in action".