Jason,
     You might want to try Shipley SPR220, which can get a coat that
thick with one coat - I have personally gotten 3um resolution out of
that material with an average thickness of 3.5um, so I believe it
would work well in your case.
     That being said, double coating S1827 is difficult.  For certain
you do not want to put one coat down, then cure it, then recoat.  The
thermal differential that creates causes the second layer to not bond
well to the first one.  If you do that, you will have to soft bake and
then let it cool, which I also do not recommend unless you can do so
under an N2 ambient to avoid both reflow and particulate contamination
of the surface.  (I've tried both of the above methods, and they don't
really work).
    Your best bet for that thickness is try spinning it at around
1500rpm for 30 seconds with a ramp of probably around 850rpm.  I would
then put down another layer.  The key to doing this successfully is to
cover the ENTIRE surface of the wafer with resist in both cases before
spinning it.  Doing so does use a lot of resist, but if you do not do
that, the coat will be severely lopsided and will be unusable for
effective lithography.

-Eric