I'm only speculating because I haven't tried to model this, but if you look at any dual band C/Ku feed, one has some control of the C-band illumination pattern via scalar ring adjustment, but this is completely ineffectual for controlling the Ku-band pattern. For most intents and purposes there is no compensation for f/D for Ku, and these types of feeds seem to have a fixed f/D on Ku that is on the high side for a BUD. Hence the poor illumination that is often seen.
Furthermore, the Ku side on such a feed has no real scalar tuned for that band. Scalars provide more benefits that just pattern control, and this forum has ample demonstrations of the detriment of running a C-band feed without one. Recently I did a comparison of a Superdish 121 feed (no scalar) to a 105 feed (abbreviated scalar) to a Dish Pro feed (normal Ku scalar). The DP beat the 121 by more than 2 dB in CNR across the board. I would not be surprised if that was also happening here.
The above applies fairly similarly to any dual band feed. When putting any Ku feed on a BUD, it's important to realize that Ku is going to be about 3X more exacting than C-band on the quality of the dish surface. If one takes a 1/10 wavelength rule-of-thumb, we would want to keep the reflector surface within 2.5 mm for Ku-band, but only 7.5 mm for C-band. The latter is well within most BUD designs, but mesh dishes will have a hard time with the former. When one stretches mesh with ribs and considers local convexities or concavities plus the inevitable minor blemishes and damages, this will all exact a price in performance as the focal cloud will be large.