When we began keeping acroporid corals in 1985, we believed it would be necessary to provide significant turbulence in the water to simulate the corals' natural environment. Recent research has indicated that turbulent water is indeed vital to the health of these corals as it greatly facilitates the diffusion of nutrients and waste in and out of the corals. The problem for us was how to develop a device that provided a strong but periodic surge that was inexpensive, and most of all would be maintenance free. 

Dump buckets are notoriously troublesome, requiring constant adjustment and repairs. Powerheads are generally too small for big aquariums and generally unsuitable for use outdoors. Larger pumps provided the output, but a continuous flow of water is not good for the corals, and did not simulate the natural conditions of periodic surge. 

The best design to meet our criteria was an automatic siphon device which has proven to be reliable and effective and is easy build and to operate. Delbeek and Sprung (1994) published a diagram of the surge device that I have received many requests for more details, hence the publication of this report. 

Before proceeding, I must acknowledge the advice of David Powell at the Monterey Bay Aquarium. When we first considered building one of these devices in 1985, Dave gave me a verbal description of a similar system that he installed at the Steinhart Aquarium. Many public aquariums use Surge tanks for a variety of applications but we were probably the first to employ one with living corals. 

How It Works The accompanying diagram illustrates a 150 gallon aquarium (#4). A Little Giant 3MD pump (#6) draws water from the aquarium through a 3/4" PVC pipe (#5) and delivers it to a 35 gallon barrel which is the surge tank (#1). When the surge tank is full, the 2" diameter PVC siphon pipe (#2) is also completely filled with water and automatically begins to siphon at a rate faster than the pump is filling the tank. When the surge tank is empty, the siphon sucks air and stops. The pump continues delivering water into the surge tank and the cycle repeats itself.

Much smaller devices can be constructed on the same basic plan. Although the pump and pipe sizes have to be scaled up or down as the dimensions change, particularly the vertical dimensions. At this time all I can suggest is trial and error for sizing, although some details for larger systems are presented below. 

Trouble-shooting Sometimes when the surge tank is full the siphon never really gets going. Instead, the water flows out the siphon pipe as fast as the pump is pumping the water in. To remedy this, we have placed the discharge end of the siphon pipe so that it is underwater at all times ( position #9 ). This creates a bit of back pressure in the pipe, allowing in the upper surge tank to rise a bit higher and thus start the siphon. Also a smooth bend at the top of the siphon, rather than a right-angle bends as shown in the diagram, will facilitate siphon start-up. At other times the siphon wont start no matter how high the water level is in the upper reservoir. We reason that this is due to air trapped in the discharge siphon which has nowhere to go because the lower end of the pipe is always under water. To overcome this problem, we installed a 1” PVC air-vent pipe (#3) at the point on the siphon pipe just above the highest water level in the aquarium ( at position #10 ). But this air vent may also prevent the siphon from starting and therefore a mechanism is requires to close it at the crucial moment when the siphon is to start. To avoid using a valve, which increases maintenance, we simply ran the air vent pipe back up into the reservoir tank with the opening of the air vent set just at the upper water level in the tank ( #8 ). The rising water acts like a valve to close off the opening of the pipe thus allowing the siphon to begin - presto, no moving parts ! 

Note: an air vent pipe smaller than 1” in diameter may work, but we find they clog too easily with salt and therefore require more maintenance. Unfortunately, once the siphon starts the air vent quickly reopens and sucks in air like a venturi. The discharge into the tank may be disturbing to aquarists who don’t want bubbles in their aquariums. This can be resolved by installing a simple flapper valve, or check valve on the air vent, but there will always be some air discharged into the aquarium. 

Sometimes the siphon wont stop. A 45 degree angle cut at the intake end of the siphon pipe, as shown in the diagram, can help resolve this by allowing more air to be sucked in when the siphon reaches the bottom of the surge tank. 

Drawbacks Depending on the size of the surge tank, the fluctuation in water level in the aquarium can be several inches ( 3” inches in our set-up ). We have not figured out a simple way to resolve this problem to monitor evaporation in a closed system. One possible solution would be to install a float switch in the aquarium or a separate reservoir and set it for the lowest “normal” water level. When the water level drops below that mark because of evaporation, the float switch will activate the flow of new replacement water. Of course, this will be periodically interrupted when the surge device dumps water back into the aquarium but none the less should work. 

Besides space requirements, the other annoying part of these surge devices is noise. The gurgling sound when the siphon starts and sucking sound when it stops can be very disturbing - definitely not recommended for the bedroom if you want a good night’s sleep. 

Applications We have a 200 gallon surge tank fitted with a 6” diameter PVC constantly filled by a 1.5 hp pump that delivers a powerful surge of water to our outdoor 7,000 gallon “Edge of the Reef” exhibit. This device has been in continuous operation since 1986 and has required virtually no maintenance except for pump repairs.

All of our outdoor coral culture tanks are fitted with 35 gallon surge tanks, while our indoor 390 gallon exhibit tank is operated by a ½ hp pump which lifts water over 6’ to a 55 gallon surge tank that delivers a powerful jet of water every 3 minutes via a 3” PVC siphon pipe. Wilfred Fong of the San Francisco Aquarium Society recently built one of these devices for his home aquarium after visiting the Waikiki Aquarium. His surge tank is much smaller ( about 10 gallons ) but still produces a good surge of water to his reef aquarium. 

Our newest exhibit will have two surge tanks, both operated from a single pump and each delivering surge to different points in the aquarium. One surge tank will be 55 gallons and the other 35 gallons so that the periodicity of the output of the two tanks will vary. Occasionally they will both discharge simultaneously creating different churning of the water in the exhibit. To handle the combined output of the two tanks we will add a 150 gallon sump at the same level as the exhibit aquarium connected by a 19” long, by 2” high overflow slot to allow excess water to flow from the exhibit tank to an adjacent sump where it will be pumped to the surge tanks ( similar to Delbeek and Sprung, page 170 ). 

Once you build one of these tanks and learn its idiosyncrasies you will find it a very useful devise for a variety of situations requiring a strong but periodic surge of water. Best of all, once you get it started it will run forever.