Part 4 The Filtration System

Part 4 The Filtration System

In this section I intend to present the filtration and circulation systems that maintains my show tank in a pristine clean condition for at least a month, without any intervention on my part. Information on how and why this system works will be presented later; for now I'm just going to specify the design of the system. The following list contains the key elements:

mechanical filter
purigen reactor
biological filter
tank water circulation
water exchange

Attention to detail is important in order to achieve the desired result. Thus, all of the elements in the list above must be accounted for and sized properly in order to handle the load. The load is ten adult discus in a 100 g tank. If the tank size or number of fish is different, the elements can be modified proportionally.

The picture below shows the mechanical filter, purigen reactor, biological filter, and circulation pump. For my application these items are housed in a basement closet located below the tank, thus eliminating noise and allowing easy access for servicing.

4-1 Filtration System:
Attachment 93077

Mechanical Filter-- Mechanical filtration is achieved with two 25 micron pleated cartridges. Each filter has 30 sq-ft of pleated material. Since there are two cartridges they provide 8460 sq-in of surface. This is normally a vary large filtration surface for a 100 gal tank, but all of it is necessary for the filtration system employed in my tank. Note that there is a pressure gauge mounted on the canister. This is very useful for determining when the filters need changing and cleaning. With my tank setup, the pressure gauge begins to indicate an increase about midway thru the fourth week after the last cleaning. The cartridges are then cleaned by placing them in a bucket of water using a weak solution of bleach, and they provide service for several years. The Nu-Clear canister is built like a tank (extremely rugged design), does not leak, and can be expanded with additional add-on modules.

A sump can be used to provide the same function providing it has sufficient means for trapping allot of material (equivalent to 60 sq ft or more). What I like about using a canister filter is that it is easy to service without adding a lot of debris to the aquarium, and the pressure gauge keeps tract of when it needs servicing.

Purigen Reactor-- My purigen reactor is a custom design and build. Here's a link that describes all of the details including how to service it:

http://forum.simplydiscus.com/showth...urigen-Reactor

This is one way of implementing a purigen reactor, but certainly not the only way. I have previously used a PhosBan 150 Reactor that I modified by adding two additional sponges and two mesh screens. Two of the sponges are placed at the top of the reactor. Sandwiched between the two sponges is a mesh screen that keeps the purigen out of the tank. A second set of sponges and mesh screen are placed at the bottom of the reactor. The two sponges hold the screen in place. This is a HOB type of setup and I use it for my QT tank. The screen material must be cut very carefully in order to keep the purigen beads out of the tank. PhosBan offers an optional pump for their reactor. However, I would recommend a pre-filter that operates in the 30 micron range in order to keep suspended material from clogging up the purigen reactor. A canister with a pleated filter and built in pump would work well for this pre-filter.

4-2 PhosBan Reactor:
Attachment 93078

There are other methods I have seen employed for a purigen reactor. Go to youtube.com and search "DIY Purigen Reactor" for lots of good ideas. As far as I know, no one markets a reactor specifically for purigen, and other reactors that are available will not work for purigen unless modified. Reference the link above for an explanation.

I use 250 ml of purigen, with a flow rate of 1.0 gal/min, for my 100 gal tank. A higher flow is not necessary and may cause the reactor to require servicing more often because it will eventually get plugged up. Purigen can be recharged using bleach and instructions are included with the product. Don't expect the 100 ml Media Bag to work. The bag material does not allow enough water contact to be effective.

Biological Filter-- The use of purigen greatly reduces the load placed on the biological filter and the reason for this will be explained in my next post. However, it is still absolutely necessary that a well established biological filter be present. I use a separate canister for this purpose because it allows me to pre-filter the water to 25 microns before entering the biological filter. That way I don't have to clean it often; in fact it can go at least 2 years without cleaning. I use 4 liters of SeaChen Matrix as the media.

Tank Water Circulation-- A large amount of water circulation is necessary to get good results; the reason for this will be explained later. Circulation is accomplished in two ways: the circulation pump, and the use of lift tubes. The picture below illustrates the how water circulates external to the tank.

4-3 External Circulation:
Attachment 93079

The Filter Pump transfers 1000 gal/hr. Since the tank is 100 gal, this means that the entire contents of the tank circulates thru the pleated canister filter 10 times each hour. Other features of the circulation system are the following:

1) New water is preheated by a separate 20 gal Pre-Heat tank.
2) The new water is mixed with the water entering the Filter Pump. This allows for good mixing.
3) Valves separately control the flow rate thru the Purigen Reactor and the Bio-Filter.
4) An overflow pipe assures that the automated water exchange system can never overfill the aquarium.

The following is a picture showing how circulation is controlled internal to the tank. Note the short PVC pipe with a cap on top and holes in it, located right where the water enters the aquarium. This pipe is located in the center of the tank and is hiden by the imitation tree stump. The stump is hollow and has been modified to allow water to exit around the bottom of it. Water is directed across the tank floor where the feces and uneaten food resides. The lift tube acts a water pump that draws water from the tank bottom and then transfers water to the top of the tank. The outlet of the lift tubes then direct the water towards the pipe that collects the feed water for the circulation pump. Do not assume that the water circulation pushes the feces and uneaten food into the water column, it does not work that way; this will be explained in my next post. Note that the following picture shows only the left hand side of the corner bow tank. There are three additional lift tubes on the right side.

4-4 Internal Circulation:
Attachment 93080

The filter pump and the lift tubes result in considerable circulation witin the tank, especially across the floor of the tank. This circulation is absolutely necessary to achieve the desired results for a pristine clean tank.

Circulation of water within the tank is also accomplished with the use of six bubble lift tubes. These are assembled from 1" thin wall tubbing and other components available on the internet.

4-5 Typical Bubble Lift Tube:
Attachment 93081

The lift tubes are driven by a Linear Piston Pump from Jehmeco, Model LPH45. This is by far the best air pump I have ever owned. The six lift tubes require allot of air and the Linear Piston Pump is very quiet and really does the job. Here's a link:

http://www.jehmco.com/html/central_air_pumps.html

Paul, what is the make/model of the yellow circulation pump in your system? Knowing its specifications may be useful to others trying to piece out a similar system.

There is always the ambiguity between the pumps rated GPH (or Liters/Hour, depending) and the achieved circulation rate in the system. You mention 1000 GPH but it isn't clear which of these two this refers to. Also, the circulation pumps like Pan World and Iwaki come in models that are good in high pressure situations, and other models good for high flow.

Choosing between them can be confusing. I'm guessing you chose high pressure, since the Nu-Clear filter requires it. I think they recommend filters rated for at least 14 feet of head.

Finally, I noticed that the Nu-Clear people recommend installations where some flow can bypass the pleated filter and go right back into the tank. They say this is to ensure that at least some circulation remains even if the filter is entirely clogged (clearly, at the expense of filtration efficiency when it is not actually clogged). I wonder if you'd considered or experimented with this option. With this layout, the pressure meter on the filter is useless. I know you like the pressure heater as an indicator of when maintenance needs to be done.

Matt,

Quote:

---

what is the make/model of the yellow circulation pump in your system? Knowing its specifications may be useful to others trying to piece out a similar system.

---

Pan World 50PX-X

Quote:

---

There is always the ambiguity between the pumps rated GPH (or Liters/Hour, depending) and the achieved circulation rate in the system. You mention 1000 GPH but it isn't clear which of these two this refers to. Also, the circulation pumps like Pan World and Iwaki come in models that are good in high pressure situations, and other models good for high flow.

---

The pump is rated at 1110 g/h at 0 head pressure. I just guessed that the flow rate was 1000 g/h when it was operational. Since you were interested in the actual flow rate, I took some pressure readings and determined that the actual flow rate turned out to be 960 g/h using the pump performance curve. I didn't know that some circulation pumps are good at high pressure while others were good at high flow. Makes sense when I think about it.

I believe that where the flow occurs in the tank is more important that what the actual total flow rate turns out to be. The purpose of the circulation is to get the feces and uneaten food to dissolve and/or disintegrate. Since they both end up at the bottom of the tank, that's where the circulation should be. Circulation does not have to be very much. If you can observe any slight movement at all, then the circulation is sufficient. Note that the feces and uneaten food are not swept away with high circulation. Rather they stay in place and shrink completely away. Those areas of the tank that have less circulation require more time.

Quote:

---

Choosing between them can be confusing. I'm guessing you chose high pressure, since the Nu-Clear filter requires it. I think they recommend filters rated for at least 14 feet of head.

---

I'm confused what you mean by this.

Quote:

---

Finally, I noticed that the Nu-Clear people recommend installations where some flow can bypass the pleated filter and go right back into the tank. They say this is to ensure that at least some circulation remains even if the filter is entirely clogged (clearly, at the expense of filtration efficiency when it is not actually clogged). I wonder if you'd considered or experimented with this option. With this layout, the pressure meter on the filter is useless. I know you like the pressure heater as an indicator of when maintenance needs to be done.

---

I seem to recall something about bypass. Never used this feature. I service the pleated filters once a month. By the end of the month the pressure gauge has increased about 1 psi. Thus I am changing the filters long before they have to be serviced.

Thanks for your questions. If you like more information, feel free to ask.

Paul

Thanks Paul. What is the range of PSI you observe on your filter?

Quote:

---

Originally Posted by afriend

I'm confused what you mean by this.

---

For the average person choosing a pump, the difference between versions with similar flow rates but different "maximum head" specifications may be confusing.

E.g. the Pan World 50PX-X specifies 1110 GPH with a max head of 13 Ft and consumes 90 watts, while the Pan World 150PS specifies 1100 GPH with a max head of 28 Ft and consumes 180 watts. The pumps have similar GPH ratings at zero head (and so zero PSI), but as head pressure climbs the second pump pushes more water at equivalent pressure (at the cost of electricity).

Matt,

Quote:

---

since the Nu-Clear filter requires it. I think they recommend filters rated for at least 14 feet of head.

---

I understand the part about the motor. It's the part above that confuses me.

The pressure difference between motor off and the motor on with 0 g/m flow is about 5 psi. I have some doubt about the accuracy of the pressure gauge.

Paul

I was going by this set of instructions for the Nu-Clear filter: http://www.inlandseasproducts.com/v/...structions.pdf

It says:

RECOMMENDED PUMP
Use with a pressure rated water pump rated at 7 to 15 PSI, or at least capable of handling 14 ft of head pressure with the desired flow rate. Suggested flow rates are a minimum of two times tank volume per hour for freshwater tanks and four times tank volume per for saltwater tanks. The pump should be able to supply the desired flow rate at a nominal head height of 12 ft. Pumps should always be installed in line before the filter(s).

Reading that again, it mentions both 14 ft of head pressure and 12 ft. head height. It also mentions 7-15 PSI, which is equivalent to 16-35 ft. of head (according to http://www.engineeringtoolbox.com/pu...ure-d_663.html). Quite confusing.

I suspect that all of this is just rough rule of thumb stuff. Technically, your pump is under-rated according to the Nu-Clear people. You'd like 1000 GPH, so they ask for a pump rated at 1000 GPH at 14 feet of head. Your pump is rated at zero GPH at 14 feet of head! But, I expect that their recommendations are very conservative to allow for a sizable safety margin for very clogged filters. If Nu-Clear became known as the filter that clogs easily, their sales would suffer.

This makes your empirical experience so valuable. We know that the Pan World 50PX-X is adequate in a system like yours, maintained as you've described.

Matt,

I believe that in my application the pump is working against a head height of 0 feet. Referring to figure 4-3 above, the tank water level is 9.7 ft above the pump which corresponds to a pressure of 4.3 psi. When the pump is turned off, the pressure at the inlet side and the outlet side are both at 4.3 psi. The the pump only circulates the water, it does not lift it from one level to another.

The pump does have to supply some pressure to over come pressure losses in the lines and a pressure drop across the filters. I estimate that this pressure loss is about 1 psi (about 2 ft head height). This corresponds to a flow rate of about 960 gph on the pump performance curve.

Look over my approach above and let me know if you understand it differently.

Paul

Paul, I think you're speaking sense.

I am simply surprised you see so little back pressure from your filter, given the manufacturer's recommendation to select a pump rated at 7-15 PSI. You've estimated 1 PSI, and that it the gauge increases by about 1 PSI as the filter clogs. Does the filter's PSI gauge really read between 1 and 2 of additional PSI (over and above the steady state pressure caused by gravity) through the course of month?

I know you have said you don't necessarily trust the filter's PSI gauge, so there may be no point to the question. :-)

Quote:

---

Originally Posted by MattArmstrong

Paul, I think you're speaking sense.

I am simply surprised you see so little back pressure from your filter, given the manufacturer's recommendation to select a pump rated at 7-15 PSI. You've estimated 1 PSI, and that it the gauge increases by about 1 PSI as the filter clogs. Does the filter's PSI gauge really read between 1 and 2 of additional PSI (over and above the steady state pressure caused by gravity) through the course of month?

I know you have said you don't necessarily trust the filter's PSI gauge, so there may be no point to the question. :-)

---

Matt,

Here's some additional data:

P1 = Pressure gauge indication when the canister is empty and sitting on my workbench: 2.6 psi (should be 0)

P2 = Pressure gauge indication when the canister is in place with no AC power applied : 6.9 psi

P3 = Pressure gauge indication when canister is in place with AC power applied with a clean filter : 8.0 psi

Explanation of the difference between P1 and P2: The pressure gauge is located 9.6 feet below the water surface of the tank. This produces 4.3 psi.
4.3 psi = (9.6 ft / 33ft) x 14.7 psi. Note that standard atmospheric pressure at sea level (14.7 psi) is equivalent to 33 ft of water.
6.9 psi = 2.6 psi + 4.3 psi

Explanation of the difference between P2 and P3: 8.0 psi - 6.9 psi = 1.1 psi. This is the effective pressure that the pump is working against. It increases to about 2 psi by the end of the month.

The expected performance curve for the pump is presented at this link:

http://www.panworldamericas.com/PX/60Hz/A3400466.pdf

The operating point for the pump is determined as follows:
From the performance curve, the no-load (zero head) flow rate is 70 L/M this corresponds to 1110 g/h.
The effective head pressure under load is 1.1 psi. This corresponds to 2.2 ft head pressure.
The operating point is located at a point on the load curve where the head pressure is 2.2 ft.
This corresponds to a flow rate of 60 L/M or 950 g/h.
At the end of the month the head pressure increases to 2 psi which corresponds to a flow rate of 800 g/h.

At the time that wrote the post about the 1000 g/h, I just guessed. Turns out that 950 g/h is a better estimate.

Hope this helps to explain it better.

Paul

Quote:

---

Originally Posted by afriend

P1 = Pressure gauge indication when the canister is empty and sitting on my workbench: 2.6 psi (should be 0)

---

Paul, thank you very much. This explanation is very clear. It is somewhat funny that the gauge is not zeroed when under zero pressure, but your other measurements and calculations show pretty clearly that the gauge is otherwise accurate.

Afreind, do you think that purigen my also absorb pheromones or hormones. The stuff that stunt discus? I wonder if i mix carbon and purigen would this help me get discus as large as possible. (as well as water changes of course