Are our sump turnover rates missing a factor?

[UncommonSense]

Ah, if it is for cost savings on a smaller pump that makes sense.

It could also realistically save on the sizing of other equipment/dry goods!

Smaller equipment, less rock, less power hungry pumps/refugium light(s), etc!

It’s funny how practically applying this math isn’t really discussed… almost like vendors like upselling the average reefer!

 

[BeanAnimal]

"Dwell time" would be a function of throughput—it does not change based on dimension.
"Contact probability" would be a function of dimension—the filter is in the path of the molecule or not.
"Contact time" would be the speed of the molecule passing over or through a filter. There is a distinction that needs to be made: over or through.


Longer dwell time and non-linear current may increase contact probability. Slower current may mean a better chance of reacting for contact-type reactions. For pass-through (mechanical), faster current means faster processing.

There is no ideal "turnover." It depends on what you are trying to accomplish.

If you are trying to mechanically filter as much as you can, then a narrow or constrained path through a mechanical filter, as fast as you can manage, is the most efficient.

If you are trying to chemically or biologically filter, then slower flow would increase contact time. Larger dimensions and/or lower turnover both work toward this goal.

Early rules of thumb were likely based on the return pump also being the primary flow source for the display. That conflated sump processing needs with display flow requirements, which are now typically handled by dedicated powerheads.

 

[Aaron Stone]

How would it effect less rock and refugium lighting? In mind those are variables governed by the filtration needs of the tank. The turnover rate will have a small part to play as a variable in that equation, but is a very minor variable. Though I could be persuaded otherwise.

 

[UncommonSense]

"Dwell time" would be a function of throughput—it does not change based on dimension.
"Contact probability" would be a function of dimension—the filter is in the path of the molecule or not.
"Contact time" would be the speed of the molecule passing over or through a filter. There is a distinction that needs to be made: over or through.


Longer dwell time and non-linear current may increase contact probability. Slower current may mean a better chance of reacting for contact-type reactions. For pass-through (mechanical), faster current means faster processing.

There is no ideal "turnover." It depends on what you are trying to accomplish.

If you are trying to mechanically filter as much as you can, then a narrow or constrained path through a mechanical filter, as fast as you can manage, is the most efficient.

If you are trying to chemically or biologically filter, then slower flow would increase contact time. Larger dimensions and/or lower turnover both work toward this goal.

Early rules of thumb were likely based on the return pump also being the primary flow source for the display. That conflated sump processing needs with display flow requirements, which are now typically handled by dedicated powerheads.

Aptly put, good sir!!

Thank you for putting this “on paper”, with such specific detail!

I guess the follow-up question would be:

As a community of passionate hobbyists; should we start taking these factors into more consideration when designing our own systems, or making suggestions to others?

In my eyes, it seems like best practice would be trying to write up a modernized “Q&A” for sump type filtration design, and best practices… is this overcomplicating things?

 

[Aaron Stone]

If you are trying to chemically or biologically filter, then slower flow would increase contact time. Larger dimensions and/or lower turnover both work toward this goal.

I think for both purposes, assuming the chemical treatment is not side streamed, contact time has little impact for filtration purposes outside the extremes.

Mechanical filtration would have an increased efficiency with greater turnover.

 

[UncommonSense]

How would it affect less rock and refugium lighting? In mind those are variables governed by the filtration needs of the tank. The turnover rate will have a small part to play as a variable in that equation, but is a very minor variable. Though I could be persuaded otherwise.

physical dimensions of the refugium can affect rock and light needs; increased chamber width, and reduced water height for example! (Re: what @BeanAnimal said!)

 

[X-37B]

Honestly I don't think flow through the sump matters all that much.
I run 2-3times tank volume through all my systems and have for years.

Aio's are higher due to a fixed pump speed with no control.

 

[Aaron Stone]

physical dimensions of the refugium can affect rock and light needs; increased chamber width, and reduced water height for example! (Re: what @BeanAnimal said!)

I guess I lost the thread of the conversation. I thought we were talking about water turnover instead of sump design.

 

[BeanAnimal]

As a community of passionate hobbyists; should we start taking these factors into more consideration when designing our own systems, or making suggestions to others?

I don't honestly think it matters that much as long as flow is reasonable.

I have seen wildly successful systems with extremely low flow sumps and those with extremely high flow sump, and everything in between.

My suggestion is moderate flow that doesn't carry bubble through is good enough, and that would depend on the size and configuration of the sump more than the display turnover rate. Sure we can do some complicated math for velocity and air entrainment distance, but to what end?

My system
75 G display
50 G sump

Initially i ran 1500 gph return and a 1500 gph closed loop. The closed loop was a maintenance PITA, so I removed it. I did not want powerheads at the time, so I upped the return pump to close to 3000 gph.

I recently added two MP10s and migrated the ReeFlo pump to a Red Dragon 3, so am down in the 2000 gph range. Still insane turnover. Would I be better off with 1/3 of that and more MP10s, maybe... but that means more expense and more maintenance.

So whole 30x and 40x sound like huge numbers, remember that the discharge are providing a very significant portion of the flow required for SPS.

My skimmer feeds from the sump and returns to the sump. There is no other filtration, just a pile (maybe 50 pounds) of rock in the sump.

 

[Johnd651]

This was addressed in the message above… also, consider installing an anti-siphon device if you’re having difficulties maintaining sufficient reserve sump volume during return pump shutdown or power outage!

I have seen them fail. They do not always seal will if there is something growing or stuck on them.

 

[UncommonSense]

I guess I lost the thread of the conversation. I thought we were talking about water turnover instead of sump design.

I guess it’s a bit of both? We’re getting into the weeds of getting the maximum efficiency out of every part of your filtration system…

As several have said (and I agree!), success can be had with a plethora of different turnover rates, display tank mixing flow rates, etc!

the finer points of system design are primarily focused on not using more of anything than is realistically necessary! (Power, equipment, dry goods, etc!)

 

[UncommonSense]

I have seen them fail. They do not always seal will if there is something growing or stuck on them.

Check valves, absolutely! There are multiple ways to prevent excessive return plumbing siphoning from your DT into your sump though. There are quite a few articles on “anti siphon device(s)” around here, in fact!

 

[BeanAnimal]

Check valves, absolutely! There are multiple ways to prevent excessive return plumbing siphoning from your DT into your sump though. There are quite a few articles on “anti siphon device(s)” around here, in fact!

The primary reason that we design the sump to hold the entire system back flow, is because check valves fail -- even if they are well maintained.

While there is no law, or rule, one would be better suited having a sump with ample extra space. It is almost certainly a case of when, not if a check valve will fail.

 

[Reefering1]

I know, I know… Sump turnover rates are a topic that’s been discussed to death…

But, I’ve been thinking…

I believe the general consensus of sizing your sump return pump to turnover DT water 3-5X max per-hour might be too generalized…

Before you start sharpening your pitchforks… let me explain!

This 3-5X turnovers per-hour benchmark takes no account for sump dimensions; specifically water height, and chamber dimensions…

To clarify: we’re shooting for a target amount of contact time, or dwell time of water in each stage of filtration within the sump. That we can all agree upon!

But, our generalized turnover numbers don’t assume tank dimensions vs. sump dimensions! — that is to say; 1000gph moving through a sump chamber with a 12”X8” water column cross-section happens twice as fast as a 24”X8” sump chamber water column cross-section…

This means a 18” deep sump at 12” water height has 2.25X more filtration contact time than a 12” deep sump at 8” water height; both using the exact same flow rate.

Thoughts? Questions? Interesting ideas?

 

[MnFish1]

I know, I know… Sump turnover rates are a topic that’s been discussed to death…

But, I’ve been thinking…

I believe the general consensus of sizing your sump return pump to turnover DT water 3-5X max per-hour might be too generalized…

Before you start sharpening your pitchforks… let me explain!

This 3-5X turnovers per-hour benchmark takes no account for sump dimensions; specifically water height, and chamber dimensions…

To clarify: we’re shooting for a target amount of contact time, or dwell time of water in each stage of filtration within the sump. That we can all agree upon!

But, our generalized turnover numbers don’t assume tank dimensions vs. sump dimensions! — that is to say; 1000gph moving through a sump chamber with a 12”X8” water column cross-section happens twice as fast as a 24”X8” sump chamber water column cross-section…

This means a 18” deep sump at 12” water height has 2.25X more filtration contact time than a 12” deep sump at 8” water height; both using the exact same flow rate.

Thoughts? Questions? Interesting ideas?

I have never heard - nor followed - this 'benchmark' - nor can I imagine how it would ever matter. I don't think you can really completely control the rates in each section anyway? I mean - take your skimmer - you will set that at whatever - if your water if flowing too quickly through your sump, you will still be skimming the same amount of 'skimmate' - since the skimmer pump is the limiting factor. etc etc etc.

 

[MnFish1]

I think it primarily matters in return pump sizing, and by proxy, price… secondarily for power consumption!

I don't want to get off topic - have you really heard this is a 'benchmark' that people follow? It reminds me of the statement that you can only have x inches of fish per gallon of water. I do not see how filtration through the sump would be affected except for huge extremes in flow - certainly not a difference of 3-5x? Additionally - using a larger return pump - means you can use (potentially) less powerhead - so power consumption is not an issue either. In other words I don't know that a return should somehow be limited to 5x.

 

[MnFish1]

Aptly put, good sir!!

Thank you for putting this “on paper”, with such specific detail!

I guess the follow-up question would be:

As a community of passionate hobbyists; should we start taking these factors into more consideration when designing our own systems, or making suggestions to others?

In my eyes, it seems like best practice would be trying to write up a modernized “Q&A” for sump type filtration design, and best practices… is this overcomplicating things?

Its way overcomplicating things IMHO

 

[Randy Holmes-Farley Verified Community Expert]

I think a better way to think of it is whether anything happens in the display water in an hour that makes it appreciably worse. Since in most cases, it does not, then that drives far lower needed turnover rates. 1-3x may be plenty.

Only caveat I’ve heard is some folks think higher flow takes more detritus into the sump system. If that matters to you, it might be a reason to up flow.

 

[BeanAnimal]

I think a better way to think of it is whether anything happens in the display water in an hour that makes it appreciably worse. Since in most cases, it does not, then that drives far lower needed turnover rates. 1-3x may be plenty.

Only caveat I’ve heard is some folks think higher flow takes more detritus into the sump system. If that matters to you, it might be a reason to up flow.

I would agree with both points. I don't see anything wrong with a low turnover sump, other than it maybe not being ideal for mechanical filtration efficiency. However, that begs the question -- does one need mechanical filtration efficiency to begin with?

I may be apt to argue many of the modern problems that we see are due to roller filters being more efficient than even filter socks were.

It is amazing for how high the flow into my sump is, how much settles out. That was not by design. However, I had drawn up some designs that were aimed at settling and easy draining of detritus -- even going as far as contemplating and pricing conical settling tanks as an option to build a sump around. The idea was to put a shelf for the rock, pump and skimmer at the top of the neck -- then plumb the bottom to the floor drain. Water changes would drain the detritus. Still an interesting concept if one has a large system and enough space.

 

[UncommonSense]

I would agree with both points. I don't see anything wrong with a low turnover sump, other than it maybe not being ideal for mechanical filtration efficiency. However, that begs the question -- does one need mechanical filtration efficiency to begin with?

I may be apt to argue many of the modern problems that we see are due to roller filters being more efficient than even filter socks were.

It is amazing for how high the flow into my sump is, how much settles out. That was not by design. However, I had drawn up some designs that were aimed at settling and easy draining of detritus -- even going as far as contemplating and pricing conical settling tanks as an option to build a sump around. The idea was to put a shelf for the rock, pump and skimmer at the top of the neck -- then plumb the bottom to the floor drain. Water changes would drain the detritus. Still an interesting concept if one has a large system and enough space.

Ah yes… these so-called “swirl filters” are quite popular in larger aquaponic systems! There are even some off the shelf options, though many farmers make their own!