New 360Gallon sump & circulation design

Nathan - ++LOL "the coming back from vacation disaster".

This is why I had to exit reef keeping for 7 years. Came back from our honeymoon to a half empty tank (salinity 50% towards fresh water) and 100 gallons of water on the floor... It's taken nearly a decade for the memory to start to fade.

There is a balance between convenience (automation) and the risk of a disaster. I am hoping that probes (salinity, water on floor, etc...) and the on-line controller (DIY) can send me alerts to avoid this.

I live in Dallas and it's hot and reasonably dry most of the year. I also travel for 10 days at a time. The evaporation rate when I used metal halides was obscene! I had to keep an automated RODI topoff or I would have to water the tank daily.

Is there a really robust topoff technique that isn't prone to these issues? Maybe using 3 probes as redundancies in case one gets stuck?
 
Here's the latest iteration based on some of the feedback.

I removed the wet/dry and added a forced drain outlet from the skimmer pump with a valve (speed up water changes). Few other changes too.

<a href="http://s1062.beta.photobucket.com/user/karimwassef/library/" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/2cf15381.jpg" border="0" alt="Photobucket"/></a>
 
...I personally don't like in-tank pumps. I find them very expensive for their performance and cosmetically unattaractive. I want the tank to look as devoid of technology as possible. So the surges are really the only way I know to move sufficient volume for a healthy reef.

So you are thinking closed loop for main tank circulation? Or will the surge tanks and return be carrying the flow alone?

I'll agree with you (partially anyway) that controllable internal pumps are expensive for what they are, and can be an eyesore, but completely disagree on performance grounds. The cost of operation is a pittance comparatively speaking. That I can move them at will or remove them without compromising my tank, priceless.
 
There is a main tank circulation loop that is driven by one large external pump. Even without the surge devices, the closed loop keeps the tank flow and filter going.

The surge devices would augment to this by taking water from the sump and alternate in their release into the main tank to create random wave motion. I would like to set them up on slightly different frequencies, allowing them to reach sync in phase and then 180degrees out of phase over time.

It all goes back into the large overflow (6" x 96") and down two 2" drains to the sump. There is a third emergency 2" drain as well.

As far as in-tank flow vs. plumbed flow... that's a matter of preference and access - to each their own. When I was in my college apartment, I had no chance to build a surge device, so back then powerheads did the job.
 
updated with detail of the calcium reactor to account for the additional equipment there

<a href="http://s1062.beta.photobucket.com/user/karimwassef/library/" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/e49a7416.jpg" border="0" alt="Photobucket"/></a>

no denitrator coils
no UV

anyone strongly supporting either?
 
For the DSB, I have it plumbed in the main flow, so the surface water in the refugium section will be high. I'm assuming that this will be sufficient to create sand-level flow of 200gph. I can see that the live rock can impact this flow, but why would the presense of macroalgae or mangroves interfere with the DSB?

The case against algae: Along with macro algae comes the need for light, The case for neither is compelling.

Sand bed infauna is more active during periods of darkness, than during periods of light, so a permanently dark sand bed will have a higher degree of constant activity, than a bed with macro present requiring that the bed be lit.

Rooting algae, such as Caulerpa (yes Caulerpa is still a very beneficial macro to use, though a couple of species behave rather badly) will inhibit movement in the sand bed. Other algae, such as Chaetomorpha, while not directly damaging the sand bed, slow the flow over the bed, and trap particulate matter within their structure. This limits settlement to the sand bed. In extreme cases, the particulates are filtered to the extent that the sand bed infauna begin to starve. Sometimes this occurs to the extent that the infauna move up into the algae mass looking for food--and your sand bed goes south for the winter.

The case for a bucket DSB, is a little easier to make. There is no room for algae, and to eliminate any possibility of problems, no other life in it.

The case against in-sump, in-line DSB's: Most significant is that it is separate from the main food chain (which is in the DT,) so the DSB will require additional attention to insure that an adequate supply of food is present.

Many are tempted to combine methods along with the DSB--see above.

What do you do if the bed goes south on you for whatever reason? The whole system has to go down, to A) Fix the problem, or B) remove the sand bed completely--which in many cases is exactly what is done.

In all cases, I see it as best that the DSB, if not in the main tank, be run truely remotely--not in-line with the main flow. Due to its separation from the main food chain, that it be strictly a "chemical filter" the sole purpose being the processing of ammonia to nitrogen gas, and to prevent the possibility of problems with it, it be strictly bacteria based. Also that it be plumbed into the system in such manner that a simple turn of a valve removes it from the system.



For the overflow, I have a 6" x 96" section of the main tank to handle the overflow and 2 x 2" drain lines (durso top) to handle the flow to the sump. I have a third emergency 2" overflow in case the quiet durso's fail.

Quiet Durso is an oxy moron, unless the flow rate is far lower, than the flow rate is going to be in this system, or they are very near siphon, in which case, they will not be stable, and a single 2" emergency back up will not be much of a back up. Back up needs to exceed the flow rate capacity of the main drain(s).

I will say this one time: You will want to run a high performance siphon system on this tank--for silence and absolute reliability--after you find out the durso's aren't going to get it done--start with the best, and save yourself the problems.

Another thing that is going to affect your drain performance is exactly how you plan on going through the wall, to the sump in the garage. Horizontal runs in drain lines cause starting problems (air locking: if it is a siphon system) and additional turbulence (read noise, bubbles, unreliability) in "durso" type drain systems.

In your design I also see flow issues with the balance pipe between the 1st and 2nd sump, and the gravity drain from the second sump to the return pump. This is going to a a pretty significant balancing problem, and I doubt that it will be able to keep up with the return pump--unless: both sumps are essentially the same body of water--e.g. very large pipe; or double overflow scheme (overflow baffle in, overflow baffle out with the second sump) Again, you would be better off with a single tank sump. The double sumps work out sometimes, with lower flow rates.
 
uncle - thanks for all the feedback.

DSB & Caulerpa:
Interesting that you mention Caulerpa since that's what I was thinking of planting. My most succesful sump evolved into a canopy of Caulerpa that was overflowing with cocepods that kept my fish and corals well fed. So, the plan is to achieve shadow and avoid digging roots with a suspended eggrate canopy. This also helps with the trimming of the overgrowth. The idea is to put the refugium on an off-cycle lighting as well. I also agree with the idea of keeping it clear of any other animals that can feed on the bed (love Ron Shimek’s article).

Durso v. BeanAnimal
I've looked at both and I'm not sure I see the vast difference in the physical implementation. Both consist of three overflows, two of which are bent down and the third is an emergency that points upwards. In the BA version, all three have water valves. In my version of Durso, the emergency is intentionally open. The two bent down overflows in both designs have caps. Durso has air valves at the top of both and BA has an air valve on only one. Having both an air valve and water valve on both siphons allows for individual tuning of air and water flow. The BA approach essentially provides an optimized solution where one air valve (middle) is completely closed and the other air valve is completely open. Then the middle water valve is used to ensure a full siphon through that drain. The drain with the open air valve is intended to carry very little water flow to avoid gurgling and bubbles. Essentially, the physical implementation is the same. BA is a special case of the Durso with a proposed optimal state. I do like it and it makes good sense.

<a href="http://s1062.beta.photobucket.com/user/karimwassef/library/" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/13a8b838.jpg" border="0" alt="Photobucket"/></a>

<a href="http://s1062.beta.photobucket.com/user/karimwassef/library/" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/d4a44cff.jpg" border="0" alt="Photobucket"/></a>

<a href="http://s1062.beta.photobucket.com/user/karimwassef/library/" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/844d0a4b.jpg" border="0" alt="Photobucket"/></a>

Overflow sizing
I have three 2” drains. One of them is an emergency which should not be activated unless something goes wrong. I’m not sure I understand this statement “a single 2" emergency backup will not be much of a backup.”… how large does an emergency drain need to be? 2” should be able to handle a very high flow of water. I don’t know the limitation of a 2” pipe and would appreciate any external reference to it (couldn’t find one myself).

Horizontal drain
This is a legitimate concern and my plan is to keep it very short. The total drain drop will be about 3ft (water in overflow to water in sump). The through wall distance is about 1ft. I don’t like this ratio, but it’s unavoidable. I plan on using a 45degree connection under the tank and another at the sump to avoid going really horizontal… 1ft drop down + 1ft down and across + 1ft down. This will force the hole in the wall to be larger… I don’t see an easy fix for this.

Sump to Sump imbalance
This is a very good point and is actually intentional. The connecting pipe is a 2” diameter pipe and the travel is about 2ft. I want to keep the sand bed sump always submerged. To do this, the connection there is well above the sand bed ensuring that it can never siphon back lower than the top of the pipe. The first sump is really for holding a large volume of water. It allows me to perform water changes of varying sizes without impacting the living system. I can close the skimmer inlet and the valve between the sumps and then open the main drain valve. This would evacuate the first sump without changing the sand bed sump or the main tank (return pumps off, of course). I can then open the valve to the water change tank and refill the first sump. Open the valves, turn on the pumps and I’m back in business in a few minutes.

Thanks for all the ideas and feedback. I really appreciate it.

I just realized I don’t have one way valves on the main returns! Easily fixed on paper (LOL).
 
<a href="http://s1062.beta.photobucket.com/user/karimwassef/library/" target="_blank"><img src="http://i1062.photobucket.com/albums/t496/karimwassef/1e80fe42.jpg" border="0" alt="Photobucket"/></a>

added missing one way and ball valves
 
uncle - thanks for all the feedback.

DSB & Caulerpa:
Interesting that you mention Caulerpa since that's what I was thinking of planting. My most succesful sump evolved into a canopy of Caulerpa that was overflowing with cocepods that kept my fish and corals well fed. So, the plan is to achieve shadow and avoid digging roots with a suspended eggrate canopy. This also helps with the trimming of the overgrowth. The idea is to put the refugium on an off-cycle lighting as well. I also agree with the idea of keeping it clear of any other animals that can feed on the bed (love Ron Shimek’s article).

Good luck with that, hope it works out for you. The number 1 reason for DSB's going bad is improper setup and improper management.


Durso v. BeanAnimal
I've looked at both and I'm not sure I see the vast difference in the physical implementation.
The appearance of the two systems is similar, however the operational characteristics are very different.

Both consist of three overflows, two of which are bent down and the third is an emergency that points upwards.
For clarity, there is only ONE overflow; either in or coming out of that single overflow there are three standpipes/drains.

In the BA version, all three have water valves.
Only one valve is required--in the siphon drain line. The other two were a convenience, that Bean felt would be advantageous in terms of maintenance. The only valve used, is the valve on the siphon.

In my version of Durso, the emergency is intentionally open.
Durso, is the name given to a particular design of standpipe, invented by Richard Durso. The standard capped tee, with a down turned elbow, and air hole to allow air to enter the standpipe. This design solved a particular problem for him, however, he did not understand or know anything that governs the operation of air assisted drains (of any style, Stockman, Glass-Hole style, what have you,) by his own admission.

The two bent down overflows in both designs have caps. Durso has air valves at the top of both and BA has an air valve on only one. Having both an air valve and water valve on both siphons allows for individual tuning of air and water flow.
Again, the valve in the open channel, is not used--it is left open. A partially closed valve in an air assisted drain line, creates turbulence (air and water mixing) that causes the noise, bubbles and instability associated with ALL air/water standpipe drain systems. Water flow tuning is accomplished with a valve on the pump output, not by restricting a drain line. It presents a plug risk, which is why BA's system uses a fail safe and a redundant fail safe, that each individually--have a higher flow capacity than the main siphon.


The BA approach essentially provides an optimized solution where one air valve (middle) is completely closed and the other air valve is completely open.
No, it is an implementation of a siphon system, using two redundant fail safe's, that self adjusts. It is not an optimized "durso" setup.

Then the middle water valve is used to ensure a full siphon through that drain.
No, the valve on the siphon line is used to raise the water level in the overflow, after the siphon has started, so a portion of the flow leaving the tank, flows through the open channel (durso, air/water mix, whatever you wish to call it.)

The drain with the open air valve is intended to carry very little water flow to avoid gurgling and bubbles.
Yes it is indeed intended to carry a small portion of the total flow. It is low, because if the pipe is more than 1/4 full of water, it will start making noise, bubbles, etc. The stand pipe is vented to prevent it from becoming a siphon, and is used as a "buffer" to allow the system to handle variations in flow rate--without the need to make adjustments to the flow to keep the siphon from breaking, or keep the tank from overflowing--or in this case, keep the dry emergency dry, unless there is a plug in the siphon line, or other such emergency. The tubing from the open channel is a trigger, that when submerged, causes the open channel to become a siphon--the second fail safe.

Essentially, the physical implementation is the same. BA is a special case of the Durso with a proposed optimal state. I do like it and it makes good sense.
They may appear the same, however the implementation is different, the operating conditions are different and the capacity is way higher than a air assisted drain system. In the case of a 1" durso, the problem free flow rate is around 50 gph, (and such are not recommended for practical use) for 1.5" pipe it is around 350 gph. I can't say what it may be for a 2" "durso," I have never run a 2" durso, I converted exclusively (no small task for 100 systems) to siphon systems a long time ago--however once the pipe is over 1/4 full you will have problems with it. A 1" siphon will handle ~2000 gph with a 36" drop, and a 1.5" siphon will handle ~4500 gph with a 36" drop, 2" ~ 8000 gph. This is idealized to a point, and assumes no restrictions in the line. (smaller bulkheads, valves etc.

I think you need to study the BeanAnimal system further, and understand the physics that govern the operation of siphons and durso's. The difference between the BA system and a dual durso system is as the difference between night and day...

No offense, but it seems a bit silly to me to use an inferior drain system, when you are planning the setup to accommodate three drain lines, and the only modifications needed to your plans are eliminating 1 air vent, and two valves. The headaches you will save yourself: priceless.

Overflow sizing

I have three 2” drains. One of them is an emergency which should not be activated unless something goes wrong. I’m not sure I understand this statement “a single 2" emergency backup will not be much of a backup.”… how large does an emergency drain need to be? 2” should be able to handle a very high flow of water. I don’t know the limitation of a 2” pipe and would appreciate any external reference to it (couldn’t find one myself).
I am looking at this strictly in terms of capacity of the drains--2 - 2" drains vs a single 2" emergency.... indeed with the system you are planning, you will not be able to flow enough water through it without problems, to overpower the 2" emergency.



Horizontal drain
This is a legitimate concern and my plan is to keep it very short. The total drain drop will be about 3ft (water in overflow to water in sump). The through wall distance is about 1ft. I don’t like this ratio, but it’s unavoidable. I plan on using a 45degree connection under the tank and another at the sump to avoid going really horizontal… 1ft drop down + 1ft down and across + 1ft down. This will force the hole in the wall to be larger… I don’t see an easy fix for this.
I pretty much guarantee the horizontal portion will cause you problems, especially with air assisted drains. With a siphon system it may or may not cause a problem. It is a known issue with both styles of drain systems.

Sump to Sump imbalance
This is a very good point and is actually intentional. The connecting pipe is a 2” diameter pipe and the travel is about 2ft. I want to keep the sand bed sump always submerged. To do this, the connection there is well above the sand bed ensuring that it can never siphon back lower than the top of the pipe. The first sump is really for holding a large volume of water. It allows me to perform water changes of varying sizes without impacting the living system. I can close the skimmer inlet and the valve between the sumps and then open the main drain valve. This would evacuate the first sump without changing the sand bed sump or the main tank (return pumps off, of course). I can then open the valve to the water change tank and refill the first sump. Open the valves, turn on the pumps and I’m back in business in a few minutes.
Good luck with this also. That balance pipe is not going to handle the flow rate through this system--unless it is rather low. The flow rates I mentioned above, do not apply to your 2" balance pipe, nor the drain out of your second sump--you are going to have to have it sucking in air in order to get the surface water out--gas exchange issue. Systems like this seldom work as planned--there are far more reliable ways to move water from one area to another--which is why sumps use baffles rather than pipe.

You really should do some more planning and come up with a solution for water changes on the fly.

Your surge system is not going to supply adequate flow, to shut the system down--which brings up the question of a black out.......but both are another 2 page topic.......... ;)

Thanks for all the ideas and feedback. I really appreciate it.

I just realized I don’t have one way valves on the main returns! Easily fixed on paper (LOL).
Check valves can, do and will fail. It is a 100% probability--and you won't be around to stop the flood. Sound system design demands that the need for such devices is eliminated. A passive fail safe is required: sufficient empty volume in the sump to contain all pump off drain down. If it cannot, redesign it. Good system design 101.
 
Last edited:
Nice looking system karim, I think you will enjoy it immensely. Any idea on when you will be able to start? Hope you can document it with pictures.

I am going to use 3 10" solar tubes on my 4x8 basement build. Have given any thought to solar tubes? Will your location allow? I would be interested in your ideas.

By the way, you mentioned you created the schematic with excel. Do you have a template you could let me have?
 
(1) As others have said, ditch the surge tanks. Too complicated, too many things to go wrong, etc. Internal pumps with multicontrollers have come a LONG way and can give you flow control that you will not achieve with surge tanks.

(2) Figure out your lighting. It will be one of the most (if not THE most) expensive parts of your build. If you go with LED's, you can toss your chiller.

(3) Glad to see you tossed the wet/dry :)

(4) Don't know how you are planning your overflow - whether it is a box or along the entire back of the tank. I am a big fan of dual overflow boxes in the tank if you are going with interior boxes - redundancy is good. If you are going external, consider redundancy in your design; imagine a big anemone sucking up against your overflow screen - can your system handle it?

(5) I may get flamed for saying this, but I'm not a fan of DSB, mangroves, or macroalgae. I prefer to run nutrient lean systems and do water changes for nutrient export. It is just one more system to crash/go bad - particularly when I am away from home,

(6) Consider a battery-powered solution for water movement in the event of power outage. Plan for it NOW so when it happens you aren't running around trying to figure out how to keep your DT from crashing.
 
Lavoisier,

I travel a lot and I'll be in Europe and China most of September and October, so I will be in the planning stage for the next 8 weeks or so. In the meantime, my plan is to work up a complete bill of materials ($) so I can budget the work. I believe that good planning is the key to success without excessive overcosts.

I do plan to take pictures of the whole process. I'm working with a local supplier for the tank. The plan is to go with a PVC base and glass eurobrace and Starphire front on an 8' x 3' x 2'.

The tank is on the first floor of a two story so I don't think I'll have the pleasure of a solar tube. I would love to... but no easy (or cost effective) way to do it.

I've attached my excel file - use it as you like.
 

Attachments

  • reef5.xlsx
    57.2 KB · Views: 0
Bonsai,

I can only have the main tank inside the house (nothing below it even). Everything else has to go in the garage. I plan on having a 10 foot protein skimmer and 10 foot calcium reactor since I have the room. My overall allocated garage water management area is 4 foot wide x 14 feet long x 10 foot high... makes a sump kind of necessary.

I have MH (from my old school days) for now, but I will migrate to LEDs once I get further along. It looks like a great DIY project. The chiller is also a function of location - Texas can get very hot. If the temperature is under 75F, the chiller will stay off.

My overflow is a straight bar across the complete length (8 ft) back of the tank. The overflow section is 6" wide and 2 foot deep (drains on the bottom). The idea is to give myself the greatest degree of control and the least flow speed at any point against the tank.

I've had great success with sump DSB with Caulerpa so I'm going to try and replicate that experience (watch for future posts on crashes and failures to give everyone a chance to say "I told you so"... LOL).

I totally agree that the tank needs to survive at 2 "disasters" without intervention. The first is multiple power outages and restarts. The second is an extended outage (24 hours).
 
uncle - not sure why a 2" diameter pipe would be unable to handle the flow. I still can't find a reference that points to this being the case.

I agree on the check valve problem, but the sumps are quite large and should be able to manage...

I apologize that I didn't explain the dimensions of the tank and sump sections up front. The tank is 8' x 3' x 2' (~360 gallons). It'll hold less water since the overflow section is 6" wide (60 gallons)... even less actually since the water will be 3" from the lip. Each sump is 85 gallons (total 170 gallons). So the the sump tanks (170 gallons) are 60% of the viewable tank volume (300 gallons).

Given that the total volume of both surge buckets (10 gallons each max) is under 20 gallons... the overflow section at 60 gallons and the sump (170 gallons) should be ok. The return lines will be 2" below water level (25 gallons to the siphon break opening)... again, the sump and overflow should be able to deal reasonably well.
 
uncle - not sure why a 2" diameter pipe would be unable to handle the flow. I still can't find a reference that points to this being the case.

An open 2" bulkhead, with 1" of head above it, will flow around 1300 gph. Why? Head pressure and gravity. The same bulkhead with 2" of "head pressure" will flow around 1900 gph. Why? Head pressure and gravity. This is for a bulkhead dumping into air rather than a pipe, however, through a submerged outlet pipe, it is the same.

What do you expect the water level difference between your two sumps to be? (head pressure) And what do you expect would flow through the bulkhead without the acceleration due to gravity? Now take from the above flow figures, the gravity, and add friction loss (drag going through the pipe.)

The flow through the bulkhead, with head pressure, and gravity, is a simple application of Bernoulli's Equation. I don't want to go to the trouble to figure out an exact figure for your balance pipe, but I already know it is not going to be enough if you are above 2000 gph coming down the drains. Even at your desired flow rate of 1800 gph for the 1 tank, let alone the flow from the other tanks, running on the same pump--it doesn't appear that the physics are in your favor.

The way to get this to work as you want, is to raise the first sump up and gain some acceleration due to gravity, and for the sake of keeping the oil slick off the surface of that first area--pull water from the top of the tank, rather than the bottom.


I agree on the check valve problem, but the sumps are quite large and should be able to manage...

I apologize that I didn't explain the dimensions of the tank and sump sections up front. The tank is 8' x 3' x 2' (~360 gallons). It'll hold less water since the overflow section is 6" wide (60 gallons)... even less actually since the water will be 3" from the lip. Each sump is 85 gallons (total 170 gallons). So the the sump tanks (170 gallons) are 60% of the viewable tank volume (300 gallons).

Given that the total volume of both surge buckets (10 gallons each max) is under 20 gallons... the overflow section at 60 gallons and the sump (170 gallons) should be ok. The return lines will be 2" below water level (25 gallons to the siphon break opening)... again, the sump and overflow should be able to deal reasonably well.
Then why on earth would you want to put check valves in the return lines? They accomplish nothing but restrict the flow........I could see it, if the volume in the pipes was that of a basement drop, and you did not want to fill the tank with bubbles from a dry return line.....
 
Last edited:
uncle - good point on keeping the two sumps at different levels to help the gravity flow. My original idea was to stack them but was concerned with the height of the first sump being too close to the bottom of the show tank. Maybe an intermediate solution with the first sump about a foot higher than the second is a reasonable compromise. That would yield about 4800 gph.

Using check valves is a habit I picked up a long time ago. I don't like the idea of dropping the top 2" of water in my main tank if I turn off main return pump. I would sometimes stack live rock and sps close enough to the surface that dropping 2" would be a problem. Not full exposure, but would only give me 1" of water over the corals with MH beating down.
 
so on another point.. what is a reasonable cycle flow rate for a 360 gallon show tank? I was thinking of something like 7000 gph?

360g - overflow (60g) - sand & rocks (100g?) = 200g

200g x 35turns/hr = 7000gph

Can't find my old Julian Sprung book, so working off of memory again...
 
Last edited:
Back
Top