Why do we assume a large return pump is needed?

[Herbert T. Kornfeld]

I get flack sometimes when I suggest it to people planning their tanks, so I thougt I would open up this can in a seperate thread...

I see so many tanks with large overflows, bulkheads, and sump return pumps that can turn the tank over ten times or more per hour. Why? Is it just an idea that the industry side has implanted in our heads to sell larger pumps? Seems so.

Many european reefers, being more energy aware (utilities are more in most european countries), seem to use very small sump return pumps, and large, high-flow circulation pumps (TUNZE streams and Rio hyperflows for example), or nice closed loops (and flow eductors). In fact, we have had some TOTM examples as well...

http://reefkeeping.com/issues/2005-03/totm/index.htm

http://reefkeeping.com/issues/2004-12/totm/index.htm

Thats a large tank with only 100gph for the overflow! Seems enough. After all, you are just trying to keep the surface clear...thats where most of the oxygenation comes from (not the splashing in the overflow like some argue). Thats also where the proteins that skimmers remove builds up. Now, using a high-flow overflow might keep this clear as well, but because it will run so fast, it will also mix alot more of these proteins into the water...and since this water will run so fast through the sump, it will never get processed and be blended right back into the water column...to be processed another in another pass.

I have heard the argument that you NEED a large sump turnover because otherwise your skimmer will overskim or have its perfoemance hindered. Do you really think that a skimmer gets even half the crap out of your water in just one pass? If running a 1000gph skimmer in your sump with only 100gph passing through it is a waste, then all those deltec-style recirculating skimmers must be a waste as well.

Benefits to running a larger sump pump? Well, maybe dirt and mulm that settle on the bottom would get removed faster with a higher flow overflow, but a closed loop could easily do that as well with an inline cannister. Or a powerhead with a prefilter sponge. Or, with a good enough in-tank circulation, the stuff might never settle anyways. I think that when reefing started, it was the most logical thing to do...run the majority of the tanks water circulation through the sump. But now we have options like Tunze streams, Rio hyperflows, Wave timers, etc.

Benefits to a low flow overflow/ sump return pump? Well, less electricity. Less heat. Less noise from a large pump as well as the gurgling overflow (thats what drives me nuts!). My cousin runs his 125g with an eheim 1250 for the sump return...something like 200gph goes throught he overflow, but his tank is silent. And the circulation runs off of less than 100 watts at any given time thanks to his two 5000+gph Tunze streams on a wave controller. He spends his electrical bill on more lighting. His tank used to need a chiller. Not anymore. And he swears that his skimmer is actually pulling out more skimmate now.

So, next time you are planning a tank, think about better ways to spend your money than a large, high-pressure sump return pump. For a long time I was running my 60g with the larger pump in the sump, and the smaller one on the closed loop...recently I swapped them. The tank gets more flow. The temperature added to the system by the pumps is less. And its much quieter. Anyone have any other thoughts on this idea? Anything I am leaving out or overlooking? Please, if you feel your tank follows this 'low flow sump' idea, post a link or pic here, I am really interested. Any input is welcome. Its just my opinion, but I think the industry has planted the idea in our heads that we need big expensive pumps because it makes more money...thats it.

 

[dattack]

It's just the that people who have large tanks need to decide how they want their main tank circulation. Some people do not want additional powerheads in their tank and would prefer for example , a Sequence pump to run through an oceanmotions 4 way. If they were to run 500 gph on a 200 gallon tank with a small pump, then they would need to add too a couple of tunzes (which can be expensive) or a lot of maxijets.
In other words, it's sometimes more convenient and sometimes more efficent to run a reliable pump like a sequence to power most of the flow.

 

[gabe3d]

I run a sequence pump as a return and as the only means of circulaion for my entire tank. At first i was really debating the fact whether i should get some Tunzes but my tank being the size it is 125gallon, is a little bit small to have some real estate taken up by two colossal powerheads. You can try to hide it as much as possible but in the same you will be limited as to where you would want the flow directed.

Gabe

 

[Sheldon337]

Wouldn't you want your skimmer and return pump to operate at about the same speed so you are processing most efficiently. If your skimmer runs at 375per hour, why not get a return pump rated for the same?

 

[NOLACLS]

I am setting up a 280 and im thinking I will go with a seq barracuda (4500 pgh @ 0') on a OM 4way for a CL and another barracuda for the sump return. My thoughts on such a big pump for the return is I want to also run it through a chiller and the pump will be outside as well. It will be pretty much right behind the tank but I want to make sure I have more flow than I need...that way I can just cut the output with a gate valve if need be (that can be done safely right?). But I like the idea of being able to T off some more return nozzles if I want and just open the gate valve as needed. I figure with all the bends going up and running through a chiller I will see 2500 gph out of the pump but that is really just a guess. Is that to much overkill The tank will be a mix reef 72x36x25tall.

 

[gabe3d]

I am setting up a 280 and im thinking I will go with a seq barracuda (4500 pgh @ 0') on a OM 4way for a CL and another barracuda for the sump return. My thoughts on such a big pump for the return is I want to also run it through a chiller and the pump will be outside as well. It will be pretty much right behind the tank but I want to make sure I have more flow than I need...that way I can just cut the output with a gate valve if need be (that can be done safely right?). But I like the idea of being able to T off some more return nozzles if I want and just open the gate valve as needed. I figure with all the bends going up and running through a chiller I will see 2500 gph out of the pump but that is really just a guess. Is that to much overkill The tank will be a mix reef 72x36x25tall.

On the same line as what he mentioned .My pump is T'ed off, to feed my chiller which then feeds my skimmer. The other side is connected to two SCWd and one direct output that is Ted for surface skimming, for a total 10 directional outputs.

By buying a more powerful pump i was able to have good flow in my tank, feed my skimmer, feed my chiller, return water to tank, and surface skim, for the cost of 160watts.

Gabe

 

[dmirza]

I think the most important thing in deciding what return pump to use is the size of your skimmer and the size of your sump. My Euro Reef requires 2X - 3X tank volume turnover. That means I need at least 250 GPH for the skimer. I also have a seperate feed for a refugium, which would itself take about 250 GPH to get a 10X turnover in the refugium. Therefore I would need at least 500 GPH in my tank.

 

[TrojanScott]

This was a great post, and I agree! I felt like I didn't have enough during my initial set up, but it's just fine.

I have a 180, and my return is a Poseidon PS4, that is T'd off into two pre drilled returns. I have two SEIO 820's in each corner, (I have no problem with pumps in the tank, and when they're covered in coralline, is cool, to me anyways)
My skimmer is run by a Mag 7, and currently, I don't need a chiller. 35 gallon sump, and my tank is whisper quiet. If I were doing SPS corals, I may throw in another SEIO, or a TUNZE, but at the moment, everything is good. I get plenty of gunk out of my skimmer, and I'm not sure why I'd need any more force on my return.
Besides, I've always been a believer in keeping up with water changes. Do that, and most of your problems, (and equipment for that matter, ) are obsolete.

 

[NOLACLS]

Is it better to have the fuge fed by water from the tanks overflow or from the sump return pump?

 

[mikester]

Gabe, I'm doing the same as you with my setup, but I have been strongly leaning towards changing it. There are a few things I don't like about this setup.

1) To force water through the chiller, skimmer, calcium reactor, etc., I have to close off the unrestricted return line quite a bit. Otherwise, the water takes the path of least resistance and I don't get very much going through my manifold that feeds all the devices. I end up getting about 1200-1500 gph out of a 3000 gph pump this way, which seems like an inefficient way to do it.

2) I don't think this is the best setup to get the most out of my skimmer. I have a recirculating Deltec, and it only needs 150-200 gph for the feed pump. I notice that a lot of detritus settles out in the middle chamber of my sump, so it doesn't even make it to the skimmer when it's fed off the return pump. I think it would be more effective to put a small feed pump in the first chamber, where it gets the raw overflow water.

3) Single point of failure for so many devices. Even though I have a very reliable pump (Sequence 1000 series), I hate putting all my eggs in one basket.

Besides using the Sequence to feed the return, chiller, skimmer, and calcium reactor, I also have a pair of Tunze streams, and an Ampmaster 3000 on a closed loop.

What I'm strongly leaning towards doing is buying a Blueline BL1100 to use for my return pump. I had one of these on my 75g and it's a great pump - quiet, doesn't add any heat, very efficient. I would send it through the chiller and then straight to the tank return, and would end up with about 700-800 gph turnover and the pump only uses 70w. Then I would put the Sequence on my closed loop and get rid of the Ampmaster completely. My skimmer and calcium reactor could both be powered easily with a Maxijet 600 powerhead in the sump.

 

[Stasher]

Fantastic thread, thank-you very much for talking about something that I screwed up with when building my tank. I tried to put way to much flow through my sump and it didn't work out at all, I have since halved my sump return and built a killer closed loop.

 

[gabe3d]

Mikester,

I've also ran into some of those problems that you've mentioned. and tried to minimize the number of bends possible and have ball valves as close as possible to each other. These are bad pics but this is what it looks like. You cannot see the feed to the chiller but it is the tube that is pointing towards you.

Gabe

 

[Fursphere]

It's just the that people who have large tanks need to decide how they want their main tank circulation. Some people do not want additional powerheads in their tank and would prefer for example , a Sequence pump to run through an oceanmotions 4 way. If they were to run 500 gph on a 200 gallon tank with a small pump, then they would need to add too a couple of tunzes (which can be expensive) or a lot of maxijets.
In other words, it's sometimes more convenient and sometimes more efficent to run a reliable pump like a sequence to power most of the flow.

I'm glad somebody understands.

Ampmaster 3000 pumping through an Oceansmotions 4-Way right here. About 2400 GPH flow. I'm probably going to upgrade it to an eight way in the future to get this mixed up better.

So much power can be wasted on large pumps running 24/7. Proper planning can save you a TON on monthly electrical bills.

 

[Entropy]

I recommend you flow whatever rate your skimmer is flowing through your sump so your skimmer is working 100% and not just reskimming the same water. I don't like powerheads in the tank, so it is either tank to sump, or closed loop as the only options.

 

[polcat_4u]

Great thread. My skimmer and return pumps are the same model call it 1000gph on a 90g. The sump is 30g, basically the overflow goes directly to the skimmer and the skimmer output goes directly to the return pump chamber. I'm moving a 1000gph but not really "through" the sump, just through the skimmer. I'd say 95% of the the flow makes it through the skimmer on the first pass. I also have two eductors on the returns. So with two mak4 pumps I get a becket skimmer and 40X turnover, seems reasonable??

 

[Herbert T. Kornfeld]

"Wouldn't you want your skimmer and return pump to operate at about the same speed so you are processing most efficiently. If your skimmer runs at 375per hour, why not get a return pump rated for the same?" -Sheldon337

"I run a sequence pump as a return and as the only means of circulaion for my entire tank. At first i was really debating the fact whether i should get some Tunzes but my tank being the size it is 125gallon, is a little bit small to have some real estate taken up by two colossal powerheads. You can try to hide it as much as possible but in the same you will be limited as to where you would want the flow directed." -Gabe

And a few others like this regarding chillers and the like. I just want to point out that I am also considering closed loops (with higher flow / low head pumps that are more efficient) to be just as good if not better than the Tunze/powerhead methods due to the obvious advantage of being able to hide everything better. I like closed loops because it allows me to direce water in ways that powerheads and the like cant...like across the bottom of the tank, or behind the rocks. And for all those concerned about needing high flow for a chiller, why not run your chiller on a closed loop anyways? It would seem a preferable to running it on a sump return line (less heat from the pump adding heat). And as for the skimmer query as Sheldon asked...I think I covered that in my first post well enough.

"I recommend you flow whatever rate your skimmer is flowing through your sump so your skimmer is working 100% and not just reskimming the same water. I don't like powerheads in the tank, so it is either tank to sump, or closed loop as the only options." -Entropy

One question>>>WHY? Do you really think that a skimmer processes all it can in just one pass (what about deltecs)? Do you feel that reskimming the same water more than once or twice will not yeild results? I once also came from that camp, being a fish-head for almost 20 years now, everywhere around us we see 'bigger is better'. But why? To me it just means more waste, more noise....and FWIW, I think that because the skimmer on my cousin's 125 gets fed the water directly from the overflow at such a slow rate, it is able to do a better job getting the proteins out instead of a higher flow overflow that would bypass the skimmer and end up mixing the proteins back into the water. After all, the surface of your tank is where the skimmate collects naturally... why not take advantage of that? Running the overflow faster just draws more tank water through. Its like running a vent fan in your kitchen thats waaay too big for the kitchen...sure, you are keeping the kitchen air clean, but you are also drawing air from the rest of the house...not really doing any good, and at what expense?

 

[Herbert T. Kornfeld]

hey polcat_4u, what is the wattage of your water movement? How loud is it...or what room is it in?

 

[Entropy]

One question>>>WHY? Do you really think that a skimmer processes all it can in just one pass (what about deltecs)? Do you feel that reskimming the same water more than once or twice will not yeild results? I once also came from that camp, being a fish-head for almost 20 years now, everywhere around us we see 'bigger is better'. But why? To me it just means more waste, more noise....and FWIW, I think that because the skimmer on my cousin's 125 gets fed the water directly from the overflow at such a slow rate, it is able to do a better job getting the proteins out instead of a higher flow overflow that would bypass the skimmer and end up mixing the proteins back into the water. After all, the surface of your tank is where the skimmate collects naturally... why not take advantage of that? Running the overflow faster just draws more tank water through. Its like running a vent fan in your kitchen thats waaay too big for the kitchen...sure, you are keeping the kitchen air clean, but you are also drawing air from the rest of the house...not really doing any good, and at what expense?

I feel my skimmer is designed to flow about 700gph (according to the mfg). Now assuming my skimmer is doing a good job at that rate I feel it is best to have at least 700gph of water available to it to skim on a constant basis. Since this is also my only means of tank circulation it works out fine for me. I respect your opinion on the subject, I just don't happen to agree with it, and I don't think your kitchen vent analogy is very accurate because a kitchen vent does not clean the air, it vents it (some other method replaces the vented air like and open window or fresh air intake somewhere). If the kitchen vent was filtering the air for instance and then feeding it back into the house, then there would be nothing wrong with having it filter some of the rest of the house as well IMO.

 

[NOLACLS]

This is a very interesting thread! This makes me kinda rethink what my return pump should be for my tank(280gal). I was planning on 2 seq barracudas one on a OM 4way and the other going from the sump split to a chiller and then a return manifold along the top back of the tank. From that manifold I will tap in 3/4 loc line and also run the line from the chiller into the tank. Pumps will be outside on the same wall as the tank. Skimmer will have a pcx 55 that does around 1200gph...my return pump starts at 4500 and I am guessing ill get 2500 out of it.

After reading this...it kinda makes me think the sump return might be to much. What do yall think...before I order these things

 

[Herbert T. Kornfeld]

Just to clarify...your skimmer spec is that its pump should run at 700gph, or that it should be fed 700gph of tank water (Im assuming its the EV150 in your sig, in which case its just the pump rating)? Just because its main pump should be 700gph doesnt mean it should be taking in tank water at that speed, unless someone has invented a skimmer that can actually purify water in just one pass! Dont think so? If you plumb your skimmer so that it recirculates most of its water from its outlet, and maybe only passes an actual 100gph through, I bet you will see no reduction in skimmate. This is the same as if you were to run your overflow at 100gph with your 700gph skimmer. Since proteins naturally settle towards the top of the tank, and as long as you run the overflow fast enough to keep the surface clear, I see diminishing returns for runnning an overflow beyond that minimum (which feeds the skimmer). If we could actually see the proteins in the water, it would make more sense...as we need to keep in mind what we are skimming...proteins and raw waste, not water. So why not just target the main source...not the whole tank?

You are right about the kitchen vent, its not 100% accurate. Ill come up with a better one next time...