Image copyright with photographer,please contact me before use Copyright © 2007

About Me

System Design

Tank Shots
Coral Stock
Fish Stock
Invertebrates Stock

DIY Projects
Online Magazines
Reef Chemistry Links
Reef Forum Links
Test Data
Useful Information
Sketchup Files
Site Map

Dive Trips:-
Sulawesi 2007

Contact me

If you wish to contact me, please email me at

Last Updated.
May 2007.


System Design (Page 1 of 2) Next >>>


As it has been quite a while since I setup my last reef tank, I was very out of touch with the current reef keeping methods. I found that visiting the forums and reviewing all the past threads very helpful. Any questions I had, I just posted on the forums and answer was forthcoming extremely quickly. Also there are a myriad of sites on the internet and a lot of information can be obtained from these.

My choice for filtration is to use the 'Berlin Method', combined with heavy protein skimming. This method uses the live rock in the tank itself as biological filter. The live rock is full of nooks and crannies, giving a massive surface area for hosting the bacteria. If you ensure that there is adequate water movement over the rock, the biological breakdown of ammonia, nitrites and nitrates will occur.

The heavy protein skimming removes wastes from the aquarium before it can mineralize, thus reducing the load on the de-nitrifying bacteria.

Additionally I am employing a Deep Sand Bed (DSB) in the sump. Water from the main tank is overflowed to the sump. The majority of the water will be sent to first chamber in the sump, where the protein skimmer will obtain it's water. However some of the unfiltered water will be diverted to DSB. The DSB will provide space to grow some caulerpa/chaeto outside the main tank. I am intending to light the DSB 24 hours a day. Hopefully the long lighting period will prevent the algae going asexual and polluting the tank.

I would love to take credit for the sump design. Sadly it wasn't my idea, I copied Simon Garratts design from his website. If you follow this link, he explains in great detail how the sump works and I can't see the point in regurgitating the information here. Also there is a very good section on DSB's.

Circulation is maintained in three areas:-

Overflow -> Sump -> Tank. Water in the sump is pumped back to the main tank by two Eheim 1262 (3400lph) pumps. They will need to pump water to a 7' foot head. Head pressure has a detrimental effect on the performance of the pump. So using the performance curves (that are supplied with the unit or are freely available on the web) , this should equate to approximately 2500lph each.

Tank-> Chiller -> Tank. Water is pumped from the main tank to chiller and returns back to tank. I decided to take the water directly from the tank. As I thought it would be easier to plumb in the saltwater reservoir which would be situated outside. I am using an Eheim 1260 pump (2400 lph). As this circuit is plumbed in a closed loop fashion, there is no effect of head pressure on the performance of the pump.

Closed loop in Tank . This will be powered by a Sequence 15000 pump (1500lph). The outlet of the pump will split into 5 branches; front of the tank, rear of the tank, left and right sides and underneath the live rock. Again as the pump is plumbed in a closed loop fashion, there is no effect of head pressure on the performance of the pump.

Tank Size

72" x 24" x 24". Many people have asked me, why do I want such a big tank. This simple answer is that within reef keeping the old adage 'Bigger is better' is certainly true. The larger the tank the more stable the environment. For example If you have snail die in a 10 gallon reef tank. the chemical breakdown of the animal after the death could upset the tank in such a small environment. Conversely if you have snail die in a 150 gallon aquarium, the death really isn't going to have much of an impact on the tank. Temperature swings are moderated in large tanks, if a heater malfunctions then the tank will take longer to change temperature. Also a larger tank is more forgiving in terms of overfeeding and overstocking. The tank was built to my specifications at Wharf Aquatics.

Tank Stand/Cabinet

Saltwater is heavy at approximately 8lb per gallon. So a heavily constructed base is needed. My intention was to build my own. First of all I built this in wood, however when placing the tank on the stand. I noticed a 2-3mm gap in the middle. I checked the wood for straightness prior to purchasing and bought dried wood. But after purchasing I made a big mistake, I left it in my garage for a week or so before taking it inside. Even though I had bought dried wood, the wood gathered moisture in the garage and after being my house for 2-3 weeks it warped! I hadn't noticed this warp and built it into the stand. the stand was built using screws and PVA glue, so there was no way I could dismantle it. This left me with only one alternative, which was to start again. Also when I placed the stand on my floor, I noticed that that the floor was far from level and needed a lot of packing to compensate..

  Image of old stand
Image of stand foot   Following a number of threads on the forums mentioning warping problems with wooden stands before and after you fill your tank. I decided that I would use a metal stand and then clad it in wood. The stand would be supported by eight adjustable feet. The feet have an articulated joint between the thread and the base, this will allow for any variations in floor level. I estimated (probably over) that the tank + sand + rockwork + equipment would weigh 1250kg. The maximum load of each foot is 700kg each. So as long as there is a minimum of 2 touching the ground, there is more than enough capacity.

Each foot has a 50mm metal base, 100mm M12 thread, with a load capacity 700kg. To estimate each downward load, the tank weight is divided by the number of feet. So each foot would take approximately 1250/8 = 156kg load. So the 700kg load capacity of the feet is more than enough. The reason why I am using heavier feet is that there was not a lot of price difference between a 150kg foot and a 700kg and I am erring on the side of caution. I am using eight feet to spread the load across the floor and to try to prevent my floor tiles from cracking. Nuts are welded into the base of the stand, to take the thread from the feet.

Leveling the tank was so simple. I started with with the four corner feet. The middle four feet were screwed up so they didn't touch the floor. Using a spirit level I then adjusted the level of the tank. Once the tank was level, I then screwed down the centre four feet so that they just took the weight. If you screw them down too much. The corner feet will lift. Overall the operation took less than 10 minutes to gain a perfect level tank on a very uneven floor.

  Image of new stand


The sump was also built to my specification at Wharf Aquatics.


Water is shifted from the tank to the sump via the overflow weirs. The overflows are designed to take up the minimal amount of space within the tank. This entailed getting the tank drilled and installing boxes with a comb filter to allow water to enter. Water is then overflowed down into the sump. I opted for 55mm hole to accept a 40mm bulkhead fitting. In case of blockage of the weir I decided that I would have 2 overflows, one at each end of the tank. The center of the overflow weirs are sited 2.5" below the water line. Pipework from the overflows down to the sump will be 40mm rigid.

Water is returned from the sump to main tank via two Eheim 1262 pumps. I do not believe that you need a great flow rate through the sump as only a small amount of water flow is needed for the DSB and the flow needed for the protein skimmer is small (1000lph). However I decided to use two, in case of a failure.

The return pumps (Eheim 1262) will be situated outside the sump to reduce the transfer of heat from the pumps to the water. These have a flow rate of 3400lph. The connectors are suction 25/34mm (1"), pressure 16/22mm (3/4"). 25mm rigid tubing will be used between the sump and the main tank. A 34mm hole will be drilled in the main tank to take a 25mm bulkhead fitting. As the return holes in the main tank is below the water level, an anti siphoning setup will be used to prevent floods in the case of pump failure.

I am also adding a closed loop system to the main tank. I opted for a closed loop system rather than the Tunze Streams because of the visual impact streams have on the tank. However with the closed loop system you still have a number of pipes in the aquarium. These can be hidden in the aquarium by passing the pipes through the bottom pane. But I am concerned that these will always create a weak spot. I think the method you choose is down to personnel preference.

The heart of the system will be a Sequence 15000 pump. the output of this pump will be divided into 3 outlets, which will situated front and back of the main tank, together one placed under the live rock. I have designated two inlets for this pump (55mm holes, 40mm fittings) in the tank. The reason why a chose two is simply is in case one gets block for whatever reason. Each outlet will be equipped with a number of nozzles, these are simply a 45° bend. At design time I am concerned that the water flow will be different from the first outlet when compared to the last. When the tank is setup I will test this and if necessary try to restrict the flow within the pipes or alternatively use reducers at the end of the nozzles.

There is further water circulation from tank to a cooler.

To calculate my total turnover in the tank, I added all the flow rates of the pumps:

(2 * 2500lph) (Sump Returns) + 2400lph (Chiller) + 15000lph (Closed loop) = 22400lph

Then I added the tank volume and the sump volume together, subtracting 15% for rock displacement, not filling to the top of the tank etc.

680 litres (Tank) + 250 Litres (sump) minus 15% = 790 Litres

So the total system turnover is 22400 / 790 = 28 times which just falls in the 20-40 current guidelines for keeping a SPS/LPS tank.


Next >>>