bimr,
I understand your point; however, the solubility of CaSO4-2H2O (not SO42-) is about 2,000-2,200 mg/L. Also, RO can be applied to concentrate streams like this in a zero-liquid discharge application. You seem to have missed the implication that Ca must be removed prior to RO or, as you say, it will not be very effective concentrating, because it will likely salt out. Yeah...ok... Remove the Ca hardness using NaCO3, essentially dropping the Ca as CaCO3 up front using NaOH, to elevate pH. This leaves you with a Na2SO4 salt going to RO that is effectively concentrated very will up to about 6% by weight. In a ZLD facility I was involved in, it is true that we evaporate in a forced circulation crystallizer to drop the salt. However, if one was to recycle the concentrated NaSO4 solution and ppt with lime, it is something that, in my opinion, should at least be explored. It may be conceivable to get creative and at least minimize the load on the downstream crystallizer.
Don't be so quick to dismiss ideas. That is the point of the thread. And with all due respect, you seem to have missed the chemistry.