PRejto had the same sort of halo problem.
http://www.iceinspace.com.au/forum/s...d.php?t=126274
You will normally get larger FWHM in blue in most seeing conditions - the seeing varies with lambda^(-0.2) so there is a broadening at short wavelengths. My impression is that it might be a bit worse than the theory suggests - it certainly is for "lucky" planetary imaging.