DIY Canister Filter, new idea

[nc0gnet0]

They don't create much vacuum at all, of the canister pumps I know of use an open empeller design, or in the case of the fx5, a semi-open impeller. In order to acheive any significant vacuum you would need a closed impeller design.

The majority of the alleviation of internal pressure is do the following:

internal canister pressure is caused by the wieght of the water in the tube(s)

When in a powered down state where syphon has been achieved this effect is x2 due to the fact both the intake and outflow tubes are full of water and exerting downward pressure

When the pump is running gravity in the outflow tube is overcome so its no longer is exerting pressure on the canister, this is why when you turn the pump on it appears to cause a vacuum.

I used to have a chart that showed the volume of water a given diameter pipe would flow in gph in a siphon. This was used in designing overflow systems for saltwater aquariums using a sump. I remember that a 1.5" schedule 40 PVC can flow approx 2100gph (gravity flow). Add a few bends etc and it will drop a bit but you get the idea. I beleive that a 1" pipe would do approximately 1500 gph (that is what the fx5 uses).

You start to see the picture, it is going to take a pretty decent pump to overcome what the siphon can put back into the canister.

 

[Cluunox]

I base my opinions on a setup I have that is simular, in the diagram the "pump" is a little 300 gph submerable pump designed to run a feature in a pond (spitter), you can pick them up for 17 bucks on ebay. Is it pretty? Nope... Does it work? Yup......and I do beleive that was the question. It's been running strong for 6 month now, and yes the filter does occasionally clog and water slows to a trickle, replace the filter and back to normal. Does the pump heat up when the filter gets clogged? Probably, but not eough to effect the tank temperature.

View attachment 118252

The main problem that would bother me on this design is that if there is a leak that stops the water from returning to the tank the pump will empty the tank down to the lvl of the pump intake.

In a design with the pump in the canister and just intake/return tubes in the tank a siphon break hole can be created below the water line to limit the amount of water lost from the system regardless of the pumps ability to return water to the tank or not.

Would guess its up to the user to decide the risk lvl on the specific design features of the setup they are using.

 

[jones57742]

Would it be possible to use a powerhead to pump water directly from the tank, into the canister (at the bottom) forcing the water up, through the filter media, and out the top - back into the tank.

Dw:

The reason you do not see this configuration in the real world is that the impellers in power heads are low head but produce high velocities.

They are not designed for pumping against the pressure generated by the media in the cannister.

This would work as a "wet/wet" closed system but the cannister (ie. sump) would need to be approximately 1/3 the volume of the tank volume.

Only a portion of the sump would contain media and the water would pass through rather than be forced through the media.

The one thing I see on your drawing you may want to reconsider is using siphon gravity to bring the water to the canister and have the pump move water out of the canister back to the tank, as it will save energy and power. Otherwise the pump has to work twice as hard.

CW:

I do not believe this to be true.

As depicted the power head is pumping against virtually zero static head.

TR

 

[nc0gnet0]

Jones,

As I said earlier, it depends on the powerhead and which one he uses. You can see the term "powerhead" affixed to many different pumps nowadays.

In a design with the pump in the canister and just intake/return tubes in the tank a siphon break hole can be created below the water line to limit the amount of water lost from the system regardless of the pumps ability to return water to the tank or not.

This is something that is done on a wet dry HOB overflow, but I have never seen anybody do it to an external canister, I wold be concnerned with air entering the system in such a setup.