Norm D'Allura's Site

I built my calcium reactor following the design of Sanjay Joshi. Of course, being a consumate DIY'er, I had to make a few changes, some of which he suggested:

The circulation loop is built entirely on the reactor body. This makes it easy to remove the threaded cap to change media or do maintenance. I had to use teflon paste to get the cap to seal. By using unions, the loop and pump can be easily removed.
The reactor exit is a 1/4" bulhead union threaded into the threaded cap. Using a bulkhead union in the cap as the exit allows for unreacted, excess CO2 to escape and avoids buildup of CO2 in the reactor which will eventually cause the pump to suck the CO2 and not circulate water. Plugging of the exit is avoided since there is no restriction (e.g. valve).
I control the flow of tank water into the reactor via a clamp on the inlet tubing.
To supply water to the reactor, I modified a Rio90 pump by tapping the outlet and threading a 1/4" bulkhead fitting into it. This makes for a reliable water supply to the reactor.
To control the flow of CO2 into the reactor I use a flow meter (RMA-151-SSV, range 5-50 cc/min with stainless steel valve, $33) from from Dwyer Instruments. Unfortunately I have to use a range below the minimum range to get 10-15 bubbles/minute. I found that if I orient the CO2 tubing at a 45 degree angle up, water fills the tubing to the check valve and I can see bubbles leave the check valve and go through the tubing into the reactor. I do not use the check valve shown in the picture, but have since replaced the rubber flapper valve with a real check valve since the flapper valve type restricted CO2 flow too much.
I use a rubber flapper check valve on my water inlet.
I bought a CO2/argon regulator from Harbor Freight for $29.

I find that at 10-15 bubbles/min (less than 5cc/min) of CO2 and an eluent rate of 20cc/min I can maintain a eluent pH of 6.5 with a calcium level of 750ppm and alkalinity of 15meg/L. This seems to keep up with my tank demand. I aso use a Nilsen reactor which adds both Ca and alk (by reacting with excess CO2 from the calcium reactor and respiration). The Nilsen reactor keeps my pH between 7.8 and 8.2 and eliminates free CO2 and phosphates- I do not have any hair algae in my tank but I can grow calupera in my refugium.

I have worn out three Mag 2 impellers and one Mag 2 pump cover (the part that holds the impeller in the removable head of the pump). I think the high solids in the reactor eluent was just too much for the impeller and the disolved aragonite abraded the impeller magnet making a loose fit between it and the unworn creamic shaft. The high solids were due to the aragonite constantly churning in the reactor due to 1) a high flow and 2) poor dispersion of the circulating stream through the aragonite bed. This made for cloudy water above the aragonite bed which got very cloudy if the eluent rate decreased (which it seems to do at low flows).

I replaced the Mag 2 pump with an Eheim 1048 and I added a 1/2" piece of coarse filter foam on top of the screeen to better disperse the flow. The aragonite bed is churning, but there are no solids above the aragonite bed. The water above the aragonite bed is crystal clear, so the pump should last longer. I am steadily increasing my CO2 and elluent rates - right now it is less than 5 cc/min and my pH is 6.6 @ 15ml/min eluent rate with alk at 10meg/ml. I am much happier with this arrangement which is also much quieter. I stopped using my Nilsen reactor. It seems my grape calupera came back but the tangs keep it cropped back.

I started using a new media called Rowalith C+ (also available here) which is supposed to be better than regular aragonite and phosphate-free. We will see how it works. I also found a new media which is very coarse, phosphate-free and high in scrontium and I will try that then my Rowalith runs out. I again tried a Mag 2 and a Mag 3 pump to get more turbulence in the reactor like I saw with the DelTec model but after running a while they got noisy and the impeller was worn. The Mag pumps also run very hot compared to the Eheim. I want to have more of a fluidized bed for better water/solid exposure, so I am trying a Eheim 1250. This pump also has a larger opening where the impeller is held, so there should be less chance of particles accumulating and plugging the flow. With the heavier media, I should have fewer particles suspended and in the water column and they should stay out of the pump but just in case I put one of the circular foam filters that come with the Mag pumps inside the reactor over the exit of the tube. It just so happens I have about 1/2" of the exit tube extending into the reactor, so it holds the filter in place vertically while the length of the foam is perfect as it compresses slightly against the opposite wall of the reactor. With this set-up the top three inches or so of the media is very turbulent and the media is vigorously fluidized. I can actually hear the bubbles of CO2 strike the impeller.I also eliminated the problematic check valves as they were rendered non-functional after a few months. I put in quick-connect valves on the inlets to make removal of the reactor for maintenance easier. I am running about 25dKH or 9.1 meg/L with a pH of 6.6 and an output rate of 7 cc/min.

Norm's Web Page

Calcium Reactor