Modular Filtration

[Kefner]

Seems this would be a great proejct for one to DIY. Seems like the heater module is already designed. Has anyone done a modular filter? I have some 3" PVC and fittings I was going to give it a shot. Just seeing if anyone else as done it yet.

Kefner

30L Medium Light
29H Low Light
5hex Med-High Light

60 - Coming soon....

[iointerrupt]

For the past couple months I've been working on one, although with 2" PVC.

All four "canisters" are identical. 2" union at top for access to the contents, one foot of clear PVC, and then a reducing T at the bottom with a 1/2" union "accessory" port. From left to right in the picture above: Mechanical filtration with sponges, biological filtration with Ehfisubstrat, CO2 reactor with inline check valve and diffuser, and finally a heater module. When I installed it, I reversed the order:

I started out with a 265gph Hydor pump, but it was loud and I wanted more flow so I switched it for a 300gph Eheim.

I ran it like that for awhile, but the CO2 reactor would fill up with air because it still didn't have enough flow, and was not nearly tall enough. So I did a redesign where I combined the mechanical and bio modules into one and replaced the CO2 module with a Mazzei injector. That left me with just two modules, no CO2 buildup problem and great flow.

One problem I ran into (and never really solved) was priming. Getting all the air out of this filter could take a good half hour of fiddling around. At one point I had an air purge valve installed, and that helped a bit. Now with just two modules it is easier to prime then with four.

Another thing I struggled with was the interconnects between modules. I used 1/2" fittings and tubing, but really wanted something less restrictive. The next common size up is 3/4", and that would have been overkill. 5/8" is just right, but the variety of fittings available in this far less then 1/2" or 3/4".

Don't start this project thinking you will use up those "spare" PVC parts, as you will end up with more spares then you started with by time you are finished. I know I did, and now I'm fiddling around with the idea of turning those spares into a fluidized bed filter...

Overall the thing cost probably two if not three times that of a commercial filter, and isn't anywhere near as user friendly. That said, if you are into this sort of DIY, it's a fun project, and people are impressed when you say you designed your own filter.

[gmccreedy]

Seems like a pretty cool project if you have the stuff laying around, but I have to agree...the cost seems to far exceed any standard canister filter by the time your done just becaue of the pump...

But if your looking for some special filtration methods, this looks like something that is very doable.

How about an article on something like this guys so that others could streamline the cost of materials without buying to much extras. I think it could be done within a resonable budget if a parts list was available and certainly a part by part installation manual.

As fas as priming...i think if you force flow through the intake it should force the air out the outflow to fill the system. Like maybe a powerhead...just to get the water in. You have alot of bends and twists, so a siphon may not start as easily...I would also suggest that the CO2 not be turned on until the filter is running (since this is just going to introduce moe air pockets that need to be overcome during the intial priming.

i also notice that you have the pump below the filter...if you can get it above, this would cut down on the head loss and the only real lift that the pump needs to provide is above the outlet of the last module. there are still minor losses through the modues themselves, but the static head is cut by at least a foot or more from what i can see.

But nice little project guys...I like it. It also looks cool and is definetly a "bragging rights" to others.

Edit: I think you can ignore my comment about losses through the modules if you have the pump sucking the water through the filter as opposed to pushing it through. either way though, I would get the pump above the modules. this may also allow you to get more modules inline

[swylie]

The height of the pump isn't important, and won't affect flow. Raising or lowering the pump will change the water pressure inside the pump, but the water pressure on both the inlet and outlet side of the pump will be the same. As long as the filter inlet and outlet are at the same height, there will be no head loss (from height) anywhere in the system. If you ignore the pump for a moment, think of it like a siphon. If you have the ends of a siphon hose at the same height, there's no flow 'cause there's no difference in pressure. Of course, there'll be head loss from the restrictiveness of the filter...

[Kefner]

I started to design out what I was going to do. I'm using the Rainbow as a inspiration to the design that I'm mapping out. I'm thinking three parts or cans. I found I'm able to get the parts I need only in a 2" design (elbows, tees, screw caps, etc) which are readily available at my LHS. I figure using a 2" can with a 1 or 3/4" inter-tube will help keep the pressures high enough.
Mechanical Can (Sponge Filter), Biological Can (Bio-Ball or Ceramic Disc) w/ the CO2, and Heater can. With a Quiet One Pump.
I too started looking at a venturi adaptor for my co2. But most of the sites I found they are in the $30 range. Since I'm already going to have a biological can might as well plumb the co2 right in the center of this.
I figure by using the inter pipe design this should help eliminate alot of the air going into the system. I'll have to fashion a purge valve of sorts.
Of course the goal is to keep the cost of everything down. Once I get a working system, I'll post my build on it.
I guess is should really be titled "Modular Filtration DIY Journal"

[gmccreedy]

Originally Posted by swylie

The height of the pump isn't important, and won't affect flow. Raising or lowering the pump will change the water pressure inside the pump, but the water pressure on both the inlet and outlet side of the pump will be the same. As long as the filter inlet and outlet are at the same height, there will be no head loss (from height) anywhere in the system. If you ignore the pump for a moment, think of it like a siphon. If you have the ends of a siphon hose at the same height, there's no flow 'cause there's no difference in pressure. Of course, there'll be head loss from the restrictiveness of the filter...

edited...i am not getting into this.

Your wrong.

[Wasserpest]

Nice project indeed. Couple of thoughts, not meant to be discouraging...

Comparing this to a canister filter, the canister would offer a vastly larger filter area/volume. With a 2" diameter the mechanical filter area is very small.

The canister would be easier to clean too... And use up much less space.

I'd like a heater module (inline heater), and CO2 module (external reactor), but for filtration, I'd prefer a canister filter and its capable pump.

Like you said io... this is a fun project. Thanks for sharing.

[iointerrupt]

I won't pretend to know enough about fluid dynamics to say where the pump should placed. I couldn't even find a consensus if the pump should be on the inlet or outlet side of the filter. I went with outlet so that debris wouldn't pass through the impeller.

I will say though that one downside of mounting it higher, is it makes vibration dampening a bit more tricky. I spent a lot of time playing around with trying to get my pump quieter, since it sits exposed in the room.

Kefner: If I was doing it again, and was trying to save money, I might suggest not even building the canister modules yourself. You can buy pre-made ones with 3/4" FPT ports on them for half the price of building your own. Here is one from Aquatic Eco for $22. That is the price of a 2" union and 1ft of clear PVC, let alone the bushings and other fittings you need. You can probably find similar ones even cheaper at a home improvement store.