I've posted this update on woodcentral... pasting it here..

I took off the pulley-side arbor nut, loosened the pulley set screws,
and with a block of wood drove out the arbor from the bearings. The
blade-side bearing came out with the arbor. I looked it over, and it
seemed like the blade-side bearing might be a bit loose-- but very
small amount. I was going to get a new set of bearings and just
replace them. But since I couldn't find a local store with bearings
today, I took a look at what it was going to take to get that other
bearing out. With the top on, it was going to be a real bear to loosen
that spanner nut (then tap out the bearnig with a larger block of
wood. (And then I'd have to figure out a way to pull the other bearing
off nearly the full length of the arbor-- and get the new one back
on!) So I thought I'd just put it all back together as a practice run
to see how hard it was going to be to put back what I had taken apart.
I needed my son to hold the pulley and one spacer in-place, but we
mananged to get it back together. With a block of wood on the blade-
side arbor, using a hammer we tapped it mostly into place, and then
using the arbor nut we snugged it all together tight. Then I backed
off the arbor nut about 1/4 turn. It seems like it might be just a bit
stiffer on the rotation than it was before, but it does spin freely
and surprise, surprise -- all the play is gone! My guess is that maybe
all the inner runs -- arbor, arbor spacer, pulley, arbor spacer, and
arbor nut were not tight to begin with. When I spin it up, it sounds
fine. I does concern me a bit that I had to put quite a bit of
leverage on the arbor nut when I tighten it up -- to fully draw in the
blade-side bearing -- and that it seems to spin maybe just a bit less
freely. But time will tell I guess, and for now, it's solid like I
would expect it to be.

When I say quite a bit of leverage, I was using a 15/16" socket on a
1/2" drive ratchet with a handle that is 7" long -- while I was
sitting awkwardly below the saw. So not sure how much real leverage I
could apply. I guess that wavy washer/load space is in the arbor
assembly putting some pressure on the outer run of the bearing. The
leverage I put on the arbor nut should have just pulled the inner runs
(blade side arbor - bearing - spacer - pulley - spacer -- nut)
together, and the outer runs are sitting in the arbor housing. The
pulley-side bearing outer run is tight in the housing by virtue of the
spanner nut.