DIY Multi-Canister Filter

[Olly613]

So I woke up this morning having had an inspiration for a diy filter in my sleep (happens more often than it should I think). Tell me what you think, keeping in mind it's in very early stages without exact dimensions nor does it account for the exact parts I'm going to have available to me (if I build it).

So as the title says, it's supposed to be a multi-canister 'canister filter'. In my experience, canister filters have been quite good (esp. for mechanical filtration), but can be annoying to to start and when you clean them, and you tend to lose much of your BB in doing so. So...

I thought about making a filter with several canisters that can be shut off independently - and where you cycle through cleaning 1 different one per week (for example), leaving all the good stuff in the remaining canisters. That's the premise, and I drew a diagram (hopefully I attach it correctly).


The intended use for this would be on a 240g tank I'm setting up in May. The diagram shows 2 canisters, but I would likely have 4 instead (see top right for the routing change, where the circles are ball valves). That being said, it could very well be expanded to as many as you like, so long as it's relatively symmetrical to keep an even flow running. With that in mind, my plan for dimensions would be approximately as follows:
-2" PVC for connecting pipes
-6" PVC for Canisters and 2-3 ft. High (11-17 US gallons each)
-~1000GPH pump (?)

I'm not terribly concerned with how it looks/size because it will be hidden behind a wall, mainly interested in how well it will work for mechanical and bio filtration, while trying to keep it as easy as possible to maintain.

It's my first fully DIY filter, and I haven't seen too many similar designs to build off of, so if you have any suggestions or recommendations on how to improve the design, or perhaps why it may fail disastrously, I'm all ears! Also any pump or filter media ideas would be super. If this gets approval from you guys (being more experienced DIYers), I'll build it and then post the results. Thanks!

(Quick questions too: why do most diy projects like these typically use PVC over ABS pipe?)

 

[pjsmetana]

I've been working on a similar design, single canister though. The problem I'm running into is the reducer and, more so, caging the media so the water cannot go around the filtering material. One issue with your design is the chicken wire for a media cage... which means your in the same boat as me. Welcome Aboard! You need a quick disconnect to remove the canisters and pump from the plumbing without making a mess. I really like the idea of multiple smaller canisters, vise 1 large, because you could probably hang them from the back of the tank and get less head pressure, thus more pump efficiency. I like your pump location, as post filtration is much better.

People typically use PVC over ABS pipe because PVC is rated for human drinking water usage, and ABS typically is not.

 

[ruddybop]

My only concern is that canister filters need all the air removed or they are noisy... Having the water come in on the top and come out at the bottom you will get air pockets up were the lids are....You need the output/return somewhere at the top most part of the canister to evacuate all the air......OR you will hear gurgling sounds within the canister. Here is a pic I found for a better understanding.....

 

[pjsmetana]

Looks like a Rena Filstar xp2 there! Good little canister, that is.

 

[Olly613]

One issue with your design is the chicken wire for a media cage... which means your in the same boat as me. Welcome Aboard! You need a quick disconnect to remove the canisters and pump from the plumbing without making a mess.

People typically use PVC over ABS pipe because PVC is rated for human drinking water usage, and ABS typically is not.

First, thank you, now...

I think you mean being able to pull off the whole cannister for quick cleaning as a unit right? I think that would be super, but you'd need twice as many valves and as you said, some type of quick disconnect. It's quite clever, and I'm sure it's possible, but if I'm interpreting correctly that makes everything more complicated.

What about keeping the canisters stationary, but using something more akin to a filter sock to contain the media. Not necessarily a commercial one, perhaps just some form of cheese cloth, or large polyester stockings - either of which are not rigid and should stretch a bit. Most imortant here though is that it's a material that lets water/etc. pass through with ease. Heck, even plastic window screen could work. That way the soaked media 'should' expand to fill the space (with some form of chicken wire or rigid plastic grate and the very bottom still). This has the advantage of providing extra filtration too. Perhaps like either of:



These could be built very cheaply, and as I said the second one you could potentially machine wash if even only a few times before complete replacement.

Keep in mind in most commercial canisters at least some water will escape around the cages. But understandably, this should be minimized to increase filtration.

 

[Dan F]

Looks plausible, but I think that you will be surprised at how much the 6" fittings will cost you. I would bet each canister unit would end up costing $100+.

One thing you could change that would save you a bit is using 1" or 1.25" pipe to connect the canisters, 2" is way bigger than necessary for 1000 GPH.

 

[Chaz88]

My only concern is that canister filters need all the air removed or they are noisy... Having the water come in on the top and come out at the bottom you will get air pockets up were the lids are....You need the output/return somewhere at the top most part of the canister to evacuate all the air......OR you will hear gurgling sounds within the canister. Here is a pic I found for a better understanding.....

Put a small vent valve in the top to bleed the air out. That is what my large inline canisters have, they flow in the top and out the bottom.

It might not be a concern but one problem I see with the design is that the filters are not going to have equal flow when run in parallel.

 

[pjsmetana]

... the filters are not going to have equal flow when run in parallel.

I saw the same thing, but really, that doesn't matter at all unless one clogs up.

 

[Olly613]

Sorry I was making a reply, I didn't see you message there ruddybop.

On that: that a really good point. Hm? Could you link to these vent valves Chaz88? That might be the answer I need. And yeah, based on how soiled each cannister is, they won't necessarily run perfectly even - but they should auto-regulate to some extent shouldn't they? As in, if one fills up really bad, then the other three will start running a bit faster and thus soil at a slightly faster rate. But there's part of the advantage to multiple cannisters right? If one clogs, the whole filtration doesn't fail!

Gonna have to have a think/sleep on that - keep it coming though, this is really helpful (though very distracting from studying for my exam tomorrow, ha!)

 

[Olly613]

Oh, and Dan F, just look into the 6" pipes, not only are they expensive (not being all that standard) but they're also tougher to get. So I might revise the dimensions to being 4" pvc with a length more like 2.5-3 foot. Also, 1" pipe does sound more manageable for the other plumbing. Thanks!

Furthermore, the top of each canister may benefit more from using a 4" 45deg 'Y' fitting and a reducer rather than connecting the 1" pipe into the side. Also adding (gluing) some chunk of plastic (or anything) to the cap might remove any space for an air pocket to form. Yay/Nay?

And would there be any merit to me essentially running this in reverse instead? I mean in terms of air build up and what have you?