I use this setup currently on my frag tank (Kalk with 2 part). To keep levels stable, I use a timer to dose a fixed amount of kalk through a peristaltic pump over a 24 hour period and use my normal ATO to make up the shortfall between evaporation and the amount of kalk being added. As long as your added kalk remains lower than the evaporation amount, this works fairly well.
The other factor in keeping alk stable when using kalk is making sure that the kalk you are adding is fully saturated. I have found that my Avast K1 reactor starts to lose potency after a couple of days. At first I was thinking it was the standing water at the top of the reactor that was absorbing CO2 (the reactor is not pressurized), but another reefer alerted me to the fact that the RODI water also contains CO2, so it reduces the saturation of the kalk over time. I really started to zero in on the saturation level when I switched to using a EC probe rather than PH for measuring saturation. Fully saturated kalk is about 10.3 ms/s according to Randy.
I had planned on experimenting with a sealed non stirred Nilsen style reactor using mason jars and a couple of 1/4" bulkheads, but when I found out that the CO2 carried with the RODI also impacts the saturation, I am no longer as eager to go to this trouble. I still may try this with some smaller quarantine tanks. I found that the HM Digital TDS meters could be used to monitor the saturation of the kalk exiting the reactor. Fully saturated Kalk reads ~ 6400 ppm (about 640 on the display along with 10x flag).
Dennis