Hardware
Flow within the tank is currently as follows:
Return Pumps: 2 of Eheim 1262 (600Gph)
Power Heads: 2 of Tunze 6205 (1,320 to 5,811 USgal./h), 2 of Ecotech VorTech MP60w ES (3500 to 7500 gph)
Wavebox: Seems to be a lot.
Total Flow (Maximum):
Return pumps: 1,200 gph
Tunze 6205s: 11,622
VorTech MP60s : 15,000
Wavebox: Lots
Total: 27,822 + Wavebox
Turnover: 27,822G (Flow) / 300G (Tank) = 92.74X + Wavebox
Note: Because of the variable nature of the powerheads, the actual turnover is difficult to calculate.
Closed loops vs. powerheads
In this tank, all of the flow will be provided by powerheads.
By far the best resource for closed loop systems is
Oceans Motions. I used a closed loop on my previous 90G tank. It was based on two seperate loops (left and right) using an independant Oceans Motion Super Squirt 4-way on each side. Each loop was driven by a Poseidon PS4 pump. The flow patterns were very dynamic and the system was silent. Overall, it worked perfectly but I ultimately pulled it out and replaced it with Tunze powerheads.
There was really only one thing that I liked about the closed loop when compared to powerheads. When viewing the tank through the front pannel, it was barely visible. Aesthetically a closed loop wins every time, but that seems to be where the benefits end.
I pulled the closed loop out of my 90G for several reasons. The first being the heat generated by the two Poseidon pumps. The second being the electrical consumption and the third being the maintenance of the pumps and plumbing.
Note: I realize that Poseidon pumps are known to have unusually high heat transfer charecteristics, but they are also unusually silent. Another pump might have solved my heat issue, but I would then have a noise issue.
Granted there are some cool things that you can do with a closed loop which aren't possible with powerheads, but the reverse is also true. Consider the various modes that a Vortech pump can operate in.
I also like the fact that I can move the powerheads as the tank matures and corals grow. I can ramp up and down between 30% and 100% power. There are no holes drilled in my tank and cleaning is as easy as a vinigar soak.
In my opinion, simple is always better. Although I completely agree that powerheads are ugly, there's nothing simpler than a powerhead.
Surface Skimming - Powerhead orientation
It's necessary to skim the entire tank surface. The most effecient way to accomplish this is with a coast-to-coast overflow, but a coast-to-coast overflow doesn't work well with waves. It surprising how poluted the water surface becomes if it is not properly skimmed. The consequence of bad surface water is reduced light penetration, poor gas exchange and a visible scum build up.
Successfully skimming the surface of the tank really has everything to do with the proper placement of return plumbing, powerheads and close loop intake/outlets. It took a lot of experimenting to get this perfect on my tank.
With only the return pumps running, the surface water in my tank remains almost completely flat. Almost like a sheet of glass.
In this first image, the dark blue is the surface area that is skimmed most effeciently. The light blue area is also skimmed, but not nearly as rapidly. The easiest way to observe this is with floating fish food. If you drop it in the far front corner, it is eventually skimmed but tends to circle around a bit first.
When I first positioned the powerheads in the tank, the orientation was based on flow within the tank and did not give consideration to the surface water. This resulted in a bulge roughly centered in the tank. This is the logical consequence of two opposing powerheads aimed towards the center area of the tank.
The problem here is that the surface water at the edges is not skimmed at all and the scum quickly builds up. In the image shown above, the black lines indicate areas where no surface skimming occours. Protiens on the surface tend to travel down hill. With the center bulge, they were trapped at the edges.
To really achieve proper surface skimming with a center overflow, some of the flow needs to be oriented at the outside edges. This creates a slop inwards and allows the entire surface to be skimmed.
A smaller bulge in the center is still acceptable, but the outer edges should be within the confines of the overflow.
Connecting the Powerheads to the Apex Controller
This is pretty straight forward. The Apex controller has 4 variable speed ports (0-10V). Physically, there are two RJ45 ports, so a special cable is required. The cable splits the port and has two male Tunze connectors at the other end.
Important: Do not allow your Tunze powerheads to run at a power level between 1% and 29%. The Tunze controllers will not allow you to do this, but the Apex will. You can ruin your powerheads by doing this. I would assume everyone already knows this, but just in case...
Connecting the Tunze Wavebox to the Apex Controller
It would be great if the Apex could control the wavebox directly, but for most people it is not possible to set the frequency of oscillation to match the resonant frequency of the tank. This is because the smallest increment of time on the Apex is .1 seconds. It is possible that the resonant frequency of your tank will be an exact multiple of .1s, but this is somewhat unlikely. A much smaller increment such as .01 would likely be required.
There are a number of benefits of controlling the wavebox directly from the Apex.
- Include the wavemaxer in feed timers.
- Night mode that isn't dependant on a photocell.
- Single point of control for all pumps.
There's a trick to make this work.
I used the wavebox controller (6091) to set the wave and used the Apex controller to enable/disable the 6091 wave function.
The wiring diagram looks like this:
The Y adapter isn't strictly necessary but it makes things much easier because it serves as a gender changer for the cables. Both the wavebox controller (6091) and Apex have male ends whereas the Y adapter is female.
The second wavebox isn't necessary either. However, if two waveboxes are deployed in this configuration, they have to be on the same end of the tank running in synchronous mode.
The jumper in the 6091 controller needs to be set to slave mode to enable control from the Apex. You have to open the 6091 controller to change the jumper. This is explained in the user manual.
I haven't spent a lot of time experimenting with different voltages, but in the simplest of terms:
10V from Apex = 6091 stops creating a wave. The pump shuts off.
0V from Apex = 6091 operates in wave mode.
This seems counter intuitive, but it does make sense on one front. If the 6091 controller is disconnected from the Apex it operates independently and generates a wave. Therefore, it is not dependant on the Apex for operation; rather it is dependent on the Apex to know when to shut off.
Feeding
<rough notes>
I've found that shutting off the return pumps during feeding can be problematic. On my 90G this would cause the water level in the sump to rise, which meant that I had to shut off the ATOs. It also allows water to back siphon from the display which is supposed to be prevented by the return line check valves, but in practice it really only partially works.
I have programmed a feed mode for the Tunze 6205 powerheads and wave maker. It basically stops the pumps completely and then pulses each pump simultaneously for 1 second once every 10 seconds at 40% power.
The VorTech MP60s also have a built in feed mode.
Night Mode
I believe in reduced flow at night.
<<<This post is not yet finished. I am using the edit feature to add content. Check back later.>>>