Why does everyone recommend low flow through the sump?

[LobsterOfJustice]

I keep seeing people recommend 3-5x sump turnover. When I first got into the hobby a few years ago, I always saw 10x being recommended, but maybe I'm not remembering correctly.

Anyway, 3x is PAINFULLY SLOW. You realize your tank water is only being filtered three times an hour when it could be filtered much more? If you have a decent skimmer, your skimmer will clean up the sump water and then be looking for more, but the trickle of water coming from the overflow can't provide it fast enough. You dont need to worry about the sump flow being "too fast" for the skimmer... The water may have less contact time, but it is contacting more water, so it really doesn't matter, as long as you are bringing dirty water to the skimmer fast enough.

Also, faster return means more water going down the overflow, means more waste removed. If you have 3x turnover and turkey baste the rocks, 9/10 of that crap is going to settle back down somewhere else even if you have powerheads providing another 30x flow. With my 15x sump turnover, I can watch the cloudiness get sucked up to, and down the overflow tubes.

Lastly, you need the flow anyway, so why not use your return as a decent source of flow? With 2 Tunze 6000 in my 90, a 3x-5x sump return literally would go a few inches and diffuse, providing no curent to the tank. Now I have a return which is actually providing flow to my tank. You need the return pump anyway, and you need the flow anyway, so why not let the return pump be the source of some of your flow?

I just dont see any downsides to having a larger return pump. There is a limit of course on sump size. You can't push 2000GPH through a 10g sump.

I recommend the largest flow through the sump you can safely push down the overflows and get through your sump without microbubbles. I expect this to be somewhere in the 10X-20X range.

[psteeleb]

the lower flow is to allow for detritus to settle out and to minimize micro bubbles as you suggest. I also have read that it may be better for skimmer production but I would think there are pros and cons to this. A filter sock would deal with the detrius, but they need to be changed every couple days

[beanoil]

Here's a thread I posted a few weeks ago. It sparked an interesting debate.


http://www.reefcentral.com/forums/sh...readid=1391001

[Sk8r]

I thought it was sump/fuges that wanted slower flow.

On the other hand, my sump/fuge has a firehose style flow from a larger than 1" hose coming down a 15 foot gravity drop, powered by a (dialed back) iwaki 100. I use a big cheato ball and a breakwater of live rock rubble to break up the flow and it all works great, no shortage of copepods, no problem with cheato growth: cheato likes flow.

[Hop]

I'll admit that up until recently I was a 3-5Xs person. But now I'm on the "it depends" on the system wagon.

The slower flow is a good rule of thumb so that there is efficiency between the skimmer and the flow-through, as well as reducing the band-aids needed for some set ups in regard to micro bubbles.

But, with my most recent tank, I found that the 3-5 times rule was a horrible failure. I believe that this was due to the equipment, equipment size etc. I've now bumped up to about the 8X mark (this added about 800gph through my sump per hour, to about 1400gph) and the results were positive.

So ultimately I think it is still a good rule of thumb, but to set up the system so that you can fine tune it as you go. I really think that over 10 may still be overkill though. I can't see any benefit to bumping mine up any higher.

Thoughts?

[sjm817]

It depends on a lot of things. 3 - 5x is a good general rule, but systems are different. A 120G display with a 30g sump and one with a 75G sump shouldn't necessarily follow the same rules. Systems may have very different filtration setups in the sump such as different skimmer and refugium setups.

[Tang Salad]

Originally posted by LobsterOfJustice

Anyway, 3x is PAINFULLY SLOW.

Your making a lot of assumptions here. Would you say that a person's drip-drip-drip flow through a calcium reactor is also painfully slow? How about through a denitrator...is that super-slow flow also painfully ineffective? Not trying to be a wiseguy, just pointing out that different applications have different flow requirements, and it is not a given that skimmers are more effective if the sump has a higher turnover.

Originally posted by LobsterOfJustice
You realize your tank water is only being filtered three times an hour when it could be filtered much more?

Again, this may or may not be true. What if the skimmer removes 90% of DOCs when the sump gets 3X flow, but only removes 30% of DOCs when the sump gets 9X flow?

Also, if the return pump is much more powerful than the skimmer pump, then a lot of water is returning to the display without ever having even entered the skimmer. Surely that can't be more efficient.

Anyway, more than one way to skin a cat, just my .02 etc...

[LobsterOfJustice]

Originally posted by Tang Salad
Your making a lot of assumptions here. Would you say that a person's drip-drip-drip flow through a calcium reactor is also painfully slow? How about through a denitrator...is that super-slow flow also painfully ineffective? Not trying to be a wiseguy, just pointing out that different applications have different flow requirements, and it is not a given that skimmers are more effective if the sump has a higher turnover.

That's a completely different situation... those setups need low flow to work. Skimmers dont, or we would be drip feeding them as well.

Again, this may or may not be true. What if the skimmer removes 90% of DOCs when the sump gets 3X flow, but only removes 30% of DOCs when the sump gets 9X flow?

So it's removing 1/3 of the nutrients from a given volume of water, but it's seing 3x the amount of water. So it's the same for the skimmer. And it's letting any mechanical filtration you have work 3x better.

Also, if the return pump is much more powerful than the skimmer pump, then a lot of water is returning to the display without ever having even entered the skimmer. Surely that can't be more efficient.

It doesn't actually matter... its not like you skimmer is seeing any less water, or the amount of nutrients in the water are different.

I'm not arguing that skimmers are more efficient at higher flow rates. What I'm saying is that they AREN'T MORE efficient at LOWER flow rates.

[LobsterOfJustice]

Originally posted by sjm817
It depends on a lot of things. 3 - 5x is a good general rule, but systems are different. A 120G display with a 30g sump and one with a 75G sump shouldn't necessarily follow the same rules. Systems may have very different filtration setups in the sump such as different skimmer and refugium setups.

but what I'm trying to understand is... WHY are we saying 3-5 is a good general rule? Did someone just make that number up? Why do you believe it is a good general rule, because someone told you?

[LobsterOfJustice]

Think of it as a freshwater tank with a canister filter... you want the strongest filter to get the most waste filtered out of the tank. It's going to suck up more crap and take it out of the water!

You're not going to buy a canister filter for your 120g freshwater that's rated for 60g and then add powerheads to the tank... that's going to leave waste in the tank. And to further the freshwater analogy, lets use the nitrifying bacteria as a "skimmer". We want to have the most flow through the filter in order to remove the most particles, filter the most water, etc. But we dont worry about the bacteria not having enough time to do their thing, because just like the skimmer, the ratio of waste removed per unit of water to volume of water seen ensures the same total amount of waste is broken down and removed.

[Tang Salad]

Originally posted by LobsterOfJustice

I'm not arguing that skimmers are move efficient at higher flow rates.

It seemed from your first post that you were saying this. ("If you have a decent skimmer, your skimmer will clean up the sump water and then be looking for more, but the trickle of water coming from the overflow can't provide it fast enough. ")

Originally posted by LobsterOfJustice
So it's removing 1/3 of the nutrients from a given volume of water, but it's seing 3x the amount of water. So it's the same for the skimmer. And it's letting any mechanical filtration you have work 3x better.

Many of us don't use any mechanical filtration, so it's the same either way.

Slower flow through the sump has several important advantages. It's quieter, has fewer microbubble problems and greatly simplifies sump design. A smaller return is also more energy efficient and adds less heat to the tank.

I have just a plain old 20g for a sump; no baffles, no chambers, and no microbubbles. If I had a larger return pump I would be losing these advantages and gaining nothing.

[seapug]

There's also the issue of cost and power consumption. Powerful return pumps are expensive and consume a lot more electricity than powerheads and aren't as controllable in their flow pattern and direction. Also, if you use an in sump recirculating skimmer, slower flow through the sump allows the skimmer and any reactors to process water multiple times before it is sent back to the display. While the return rate may be slower, the water it sends back to the display may be "cleaner" than water that goes on a high speed loop from the sump to the display and back.

[LobsterOfJustice]

Originally posted by seapug
There's also the issue of cost and power consumption. Powerful return pumps are expensive and consume a lot more electricity than powerheads and aren't as controllable in their flow pattern and direction.

But now you need a return pump AND a powerhead, which is more electricity! Why not upgrade the return pump, run it through a sea-swirl, and eliminate a few powerheads since you need a return pump anyway?

Also, if you use an in sump recirculating skimmer, slower flow through the sump allows the skimmer and any reactors to process water multiple times before it is sent back to the display. While the return rate may be slower, the water it sends back to the display may be "cleaner" than water that goes on a high speed loop from the sump to the display and back.

So the trade-off is that it's seeing less water, but it's cleaning that water better. So overall, they equal out to the same amount of waste being removed, in theory. Now that I think about it, it actually could be more efficient at higher flow rates. It probably easier to remove the first 10%-30% of organics, but it takes longer to clean the next 10%-30%. (Think about it in terms of nitrate removal with water changes - lets say you have 100PPM. A 50% change removes 50PPM, but the next 50% change only removes 25PPM) So in this time frame, it would be better suited to be working on cleaning up dirtier water. It will skim out more in dirtier water, so I want to constantly be bringing tank water to it. I don't want it to have a chance to clean the water in the sump before more tank water is supplied.

[LobsterOfJustice]

Originally posted by Tang Salad
It seemed from your first post that you were saying this. ("If you have a decent skimmer, your skimmer will clean up the sump water and then be looking for more, but the trickle of water coming from the overflow can't provide it fast enough. ")


Many of us don't use any mechanical filtration, so it's the same either way.

Slower flow through the sump has several important advantages. It's quieter, has fewer microbubble problems and greatly simplifies sump design. A smaller return is also more energy efficient and adds less heat to the tank.

I have just a plain old 20g for a sump; no baffles, no chambers, and no microbubbles. If I had a larger return pump I would be losing these advantages and gaining nothing.

Well, I guess we have examples of both working - the only thing you've got on me is that I have baffles (and really, they're not much of an inconvenience ). I dont have heat issues from my pumps, don't have microbubbles, and my water is quiet.

Plus, I have one less powerhead in the tank because my return is actually putting out enough flow to count for something. Having one (or more, if you use it wisely) less pump is also energy efficient and adds less heat to the tank.

[Doahh]

Because upgrading from a mag7 to a mag 24 is over 120 watts more...
My Koralia 4 can make up for that difference in gallons per hour and it uses 11 watts

[Doahh]

EDIT: It's actually 200 watts more than my Mag 7

[just dave]

A few things to remember. People have been doing both ways for some time now with stellar results ,with horrible results ,and everything in-between. So the flow rate through your sump ( within reasonable boundaries) is not going to doom you to failure or make you an instant Tank Of The Month candidate. I assume we all have tanks with enjoyment being the primary reason for ownership. With that in mind , micro bubbles and horrible drain noises would definitely detract from that experience. So IMO the maximum flow rate through the sump should be determined by the sumps size and design and the drains capabilities. The minimum flow rate may need to have a number of variables considered based on equipment and how it's employed ( like heaters, chillers, UV units,...etc.)





I thought I already posted on this thread but I ended up posting on the thread beanoil linked.

[tkeracer619]

I run a dart as a return pump. Realistically I have a 30x turnover rate in the sump. I get plenty of settling on the floor of my sump/fuge and am glad I do. My skimmer pulls dark skimmate.

I promise I filter more then someone with the same setup but 5x sump turnover.

Cheato loves my sump.