I'm on the 2-part dosing bandwagon again myself but I think the key to Dez's ease of use is the controller. That's the other thing I would use if I ever were go to a calcium reactor again (although I wouldn't
, but if I did, I'd want a pH controller)
One cautionary note about T'ing off the sump return like what has been described above: the flowrate should be controlled at the inlet side of the reactor (ie., like how Mark has shown, the valve must be between the sump return pipe and the reactor ... not on the output side of the reactor). The reason for this is that sump return pumps might be capable of producing enough pressure inside the reactor that the acrylic seams and the flange are under more stress. I actually blew up a reactor doing this (weak seam I guess, I don't think that's typical, but certainly the other issue I had was that the darn flange always had a little bit of wetness). It drove me crazy. That's why I switched to a powerhead feed - the flow control on the output side gave me a more steady effluent rate and a powerhead isn't capable of pressuring the vessel enough to cause any problems like that.