Flow rate? Oh, no...not this again...

[jjohnwm]

I keep seeing flow rate being discussed and dismissed as a very simple straight-forward measure of the effectiveness of tank filtration. It is constantly being suggested that a tank needs 7x flow rate, 10x flow rate, etc. and the number of times that the tank's volume is turned over in a given hour seems to be regarded as a direct indicator of whether or not your filter is sufficient for your tank. I even see discussions of adding yet another dinky little canister filter to a tank, thus ostensibly increasing its flow rate.

Without getting into a discussion of the other relative merits of canisters vs. sumps, I think the idea of flow rate needs some examination. Let's say you have a tank of 100 gallons capacity, and it is filtered by one large canister that has X square inches of surface area of mechanical filtration and Y cubic inches of biological media. Let's say the filter pumps 500 gallons per hour; you have a flow rate of 5 turnovers per hour. Gee...common wisdom says that's too low; better buy another canister! So you get another identical filter, set it up and presto! You suddenly have twice the flow, all the way up to a blistering 10x! Congrats, you are filtering the crap out of that tank...literally...

But...your buddy has the same tank, except he has a 30-gallon sump for filtration. He is using a single return pump which has the exact same flow rate as your canisters, i.e. 500 gallons per hour. Pathetic, huh? Except...his sump has a mechanical filtration area 5x bigger than both of your canisters combined, and a biofiltration media volume that is 10x greater. Does anybody really believe that his filtration is inferior to yours? I know which one I would choose.

Does adding another canister really double efficiency of filtration? It has doubled the filtration area and volume, and doubled the gph, but the speed at which the water passes through the media is unchanged. How do we quantify this? It's reminiscent of power measurement in watts; you can operate a 10watt heater for 100 hours, or a 100watt heater for ten hours, and either way you have used the exact same amount of power: 1 kilowatt-hour...and that's how pay for power, by the kilowatt-hour. Very easy to calculate efficiency.

With the filters? Not so much. How do we measure the relative efficiency of these two set-ups conveniently?

Some tech-oriented aquarist will immediately pipe up "No problem, we'll just measure the nitrates in the water...or maybe the turbidity...or maybe both...or just look at the amount of waste visible on the gravel...or...or..." and at that point they will trail off because it becomes apparent there are just too many variables to quantify this. Different pumps will have different variances in gph caused by head pressure. Different mechanical media will have differing amounts of surface area per volume, both for mechanical filtration and for bacterial colonization. The canisters are pressurized and will force the water through the media, possibly allowing the use of different, "better" media...but they will begin dropping off in gph almost immediately as they gradually become clogged, while a sump would operate far longer before clogging would become an issue. Just too many variables to realistically state that one or the other is "better".

Is there an answer? Beats me. Personally, I would far rather have the vastly greater ease of cleaning a sump as opposed to the fiddly procedure required with sealed canisters, but plenty of folks would ooh and aah over a shiny sci-fi looking canister on display under their tanks and would put up with the increased hassles.

But I think it would serve us well to actually give some thought to these factors rather than blindly following the mantra of 7x to 10x flow rate as if it were the be-all and end-all of aquarium filtration. It is a simplified way to quickly compare one aspect (and only one aspect among many others) of filtration, and nothing more.

 

[esoxlucius]

I've never really been able to get my head round the huge turnover rates that are quoted for freshwater set ups. When I came back into the hobby and was researching everything just to try and catch up with developments, i must admit I felt a little paranoid that my turnovers were nothing like 10x. I even bought variable speed pumps which would give huge turnover should I require it. Those same pumps now run at half speed, and have done for a long time.

I have sumps, my parameters are great, water clarity is fantastic, fish are healthy. Will increasing my turnover improve all that? Nah.

 

[12 Volt Man]

the 10x turnover rate was a result of the unfortunately popularity of crappy cartridge HOB filters if I recall.

4x to 6x per hour is plenty with canisters and sumps etc with lots of media capacity.

 

[FJB]

I agree that an overemphasis on flow rate per se seems to be pervasive, and that it is at least partially mis-guided. Flow rate alone is nearly meaningless.
An important additional consideration is that a good reason for multiple filtration units is redundancy and safety, regardless of whether this also increases flow rate. If one has 5 empty large HOB filters along the back of an aquarium, flow rate would be high, but filtration would be nil (no media).
If one has a large sump for a large aquarium (great choice!), it would be said to be foolish to have a second large sump just in case something fails; instead, having a spare pump (or spare parts) would be the smart thing to do such a failure occur.
With canisters or HOB's, one should want to have at least 2 units (more if desired, depending on the size of the tank); should one fail; there would always be one already in place to handle the slack, at least temporarily. This also allows staggering maintenance of filters, which some people favor.
I believe a number of people have multiple filtration units for reasons such as this redundancy (definitely my situation) not for flow rate per se. And then, I am sure there are people who are hung up on the flow rate fallacy.

 

[duanes]

IMO nitrate is not useful as a tool in measuring filtration, because normal filtration does not reduce nitrate, it produces it.
Nitrate is a product of metabolism, so unless the aquarist uses a specific nitrate reduction system (such as a nitrate reactor, an anoxic plenum, or plethora of vegetation) water changes are your only optioning reducing it, not filtration.
Ammonia and nitrite might be better indicators, if the tank has either, after it is "really" cycled, this could tell you your flow rate is wimpy.

But I try to base flow rate in my tanks on the type flow rates fish have evolved to live in, in heir natural habitat.
If I were keeping discus, angels, or swamp fish like some anabantids I would use a low flow rate, just enough so water testing showed it was sufficient.

If I was keeping riverine species that come from moderate flow rivers, I might want to up it.
I use a pump rated at about 1500 gallons per hour in my 180 gal Andinoacara tank (the Rio Charges river where these fish are common, has an average flow rate of 2500 cubic feet per second). That is a serious moving wall of water, and the fish are barely phased by it.

If I was keeping rheophillic species like tigerish from the Congo, Retroculus from the Amazon basin, or Tomocichla, from Central America's fast flowing rivers, or even some barbs (like Denison) that come from fast flowing steams I'd probably double that 1500 gph rate in a similar size tank.
I find after keeping species of the genus Tomocichla, Rheoheros, and a few other rheophillic types, a lack of enough flow influences aggression, and lack of oxygen that comes with a high flow rate can be critical to their health.

If I were keeping species that live in the tidal wave zone like Eretmodus, or certain Goby's, instead of flow rate, I'd be considering a wave maker.

 

[neutrino]

Agree on flow rate and on most of this, except the idea that using canisters makes you more dependent on flow rate-- if that's what you intended. With a single exception (higher flow rate) my current tanks are running a theoretical 3.5-4.5 times water turnover per hour based on max filter specs-- mostly canisters with one Seachem Tidal. As most know, the reality of a running filter with media, tubing, etc. is nearly always less than advertised, so the real turnover number is maybe 50-60% lower.

All tanks are healthy, low nitrates, water is clear, and currently includes a couple of fish well over 15 years old. So high turnover rates aren't the end all, be all.

I've argued the obsession with flow rates in the past. At different times I've run the same tank on a small, fast power filter OR a larger, slower filter-- with the same results, excepting the fast, small filter needs frequent cleaning vs the larger filter. Either one can work. Sump, canister, power filter, moving bed, or whatever else, I find there's a sweet spot with tank size, stock, feeding, etc. where you're good where you are and more is just more.

I don't personally run multiple filters on each tank. Setups are different now and over the years, but between my tanks I'll have enough filters to move them around where I need them and I like to keep a retired filter or two as backups.

 

[TheWolfman]

Turnover is relevant to the quality of your water especially in a monster fish tank. We are looking to scrub particulate from the water column, and the more turnover the clearer the water. The nitrogen cycle does not need 10x turnover. You could probably get away with 3X turnover and still have a properly cycled aquarium. A lot of times when people are saying 10x turnover they are counting power heads and waves maker into that equation.

 

[jjohnwm]

...the 10x turnover rate was a result of the unfortunately popularity of crappy cartridge HOB filters if I recall...

i did not know that, but it sure makes sense. They were doing their best to sell us on the idea that a large tank would be best filtered by a contraption that forced all the water through a postage-stamp-sized cartridge, which would of course require expensive replacements on a regular basis. Higher flow = cleaner water = faster clogging = increased sales of cartridges.

...I agree that an overemphasis on flow rate per se seems to be pervasive, and that it is at least partially mis-guided. Flow rate alone is nearly meaningless...An important additional consideration is that a good reason for multiple filtration units is redundancy and safety, regardless of whether this also increases flow rate...I am sure there are people who are hung up on the flow rate fallacy....

I agree that flow rate...at least the way that term has come to be used in the hobby...is pretty meaningless. Without knowing the volume/surface area of the filtration media, that number is just a number.

Redundancy is always good. Two pumps in a sump is comforting; double the amount of water moving through the same already-large filter bed in the same unit of time, so moving twice as fast. Did adding that second pump double the flow rate? How about two canisters, both pumping the same gph. The total gph has been doubled, and so has the total filtration media size...so the speed of the water moving through the media is the same. Is that double the flow rate? Today's accepted use of the term flow rate could say that both statements are true...and that makes the definition of flow rate worthless.

...Agree on flow rate and on most of this, except the idea that using canisters makes you more dependent on flow rate-- if that's what you intended...I've run the same tank small, fast power filter OR a larger, slower filter-- with the same results, excepting the fast, small filter needs frequent cleaning vs the larger filter. Either one can work...

I meant that, as you state, the flow rate is only half the equation. The amount of filtration media through which the water is flowing at that rate is the other half. As you state, either extreme...big/slow or small/fast...can achieve the same result. The canister loses out only because its small size means that it should be cleaned more often, but its design usually causes that not to be the case.

...A lot of times when people are saying 10x turnover they are counting power heads and waves maker into that equation.

Yep...and that is yet another example of how the use of the term is so badly brutalized that it begins to mean nothing. Add a wavemaker or circulating pump and you might get somewhat clearer water or at least cleaner bottoms, simply because dead spots that allow waste to settle and accumulate can be avoided. But there is no reasonable way to say that this increases turnover rate, although there is more water movement...which is almost always a good thing if not carried to extremes.

I appreciate all the feedback comments; they are all based on experience, rather than just being a recitation of stuff found on the internet. Bravo!