I spent some time experimenting this morning and I think I may have figured
out what's going on.

First, I agree with Larry, as long as the blade is held parallel to the
platter surface, the amount or speed of the abrasive passing under the blade
should be of no practical consequence. Consider a worst case scenario; no
abrasive at the center of the disk tapering to unused abrasive at the
circumference. Even in this worst case, the difference in the amount of
metal removed between the inside and outside edges of the blade, would be
limited to the size of the abrasive particle. In the case of the 100 micron
disk, this would be hard to measure. It certainly wouldn't account for the
error that I and others on this group have observed. On the other hand,
there is no question that Mark's observations are correct. The outside edge
of the blade definitely gets ground down more than the inside edge, and by a
rather significant amount.

As I gave this some thought, it occurred to me that the thickness of the
platters may not be consistent. Sure enough, when I measured the thickness
across the radius of the platters, I found that the outside edge is
approximately .004 inches thicker than the remainder of the disk, probably
as a result of a manufacturing operation. I verified this on three disks.
I'm sure this is one of the reasons why Veritas recommends that you stay
clear of the outside edge. My second thought was that the turntable that
the platters rest on may not be perfectly flat. I checked this by
installing a new (without abrasives) 4mm disk and then laying a precision
straight edge across it. After accounting for the slight increase in
platter thickness at the circumference, everything was dead flat, or at
least within .001 inches. As I indicated in my original post, I was able to
measure a slight bit of wobble at the outside edge as I rotated the platter,
but this was consistent between platters at around .003 inches and shouldn't
affect the bevel.

I next installed a new 100 micron disk on the 4mm platter and a new 9 micron
disk on the 3mm platter. I used a spare 1 5/8” plane blade that I checked
to make sure was both flat (no warp) and had parallel sides. I also ground
the edge square to the sides and put an edge on it using conventional
methods (water stones). I installed the blade in the holder and verified
that it was absolutely square. I also checked to make sure that the edge
remained flat after it was tightened into the holder.

I mounted the 100 micron disk and proceed to sharpen the blade, moving the
holder back and forth along the guide bar, until a slight burr formed across
the entire edge. When moving the holder, I remained approximately ½” from
the center and outside edges. With the care I used setting everything up
and the new media, I expected that I would initially see an edge that was
pretty much square. It wasn't!

At this point I didn't know what to think, so out of frustration I once
again put my straight edge across the platter, this time with the abrasive
disks applied. I was really surprised to see a depression toward the center
hold down. When I measured the disk without media, it was dead flat. I
think what's happening, is that the rather thick cloth backing on the 80
grit paper is being compressed by the brass hold down and pulling the center
of the platter with it. I measured the depression at around .004” which
combined with the slight rise at the circumference, is more than enough to
account for the error we're seeing. It also accounts for the micro-bevel
not being parallel, because the thinner media on the 3mm platter doesn't
result in the same problem, in other words, the 3mm platter remains flat.

Your thoughts?