Help me with my sump plan (pic)

[Carissa]

I'm in the process of designing a sump for a 33g fowlr tank. The sump is a 20g tank.

My conundrum is that in order to get good flow through my sump (~550gph) I have to have my bulkhead 3" below the surface of the water in the main tank. This means that in case of a power outage, I need to have the extra capacity in my sump to hold all the water that will siphon out, which comes out to be a bit under 9 gallons (12" x 35.5" x 4.5" (depth to bottom of bulkhead) = 1917 cubic inches x 0.004329 = 8.3 gallons). My return bulkhead will be higher, so that is not a concern.

Anyway, I'm left trying to figure out how to design my sump to provide at least 10" of water for my in sump skimmer (which I will have to cut off anyway - currently it needs 14").

My thought was that I could use a very wide PVC pipe filled with live rock rubble as the initial filtration mechanism, then through a sponge for particulates, then into the skimmer chamber, than the return pump chamber, which will be only 1/3 full. Here's a pic. See any issues or ways I could improve on this?


http://reefcentral.com/gallery/data/578/192325sump3-med.jpg

[crvz]

A couple thoughts. First, you're pushing 10x system volume through your sump per hour. Probably more, to be honest. You could easily cut that in half and be in good shape. It's not great practice to use the return pump as the main flow for the display, usually powerheads and closed loops are better suited. So I would seriously consider cutting it back (can be done simply by adding a valve to the return pump), as it's going to simplify a few things.

Second, that sponge is going to need frequent attention, but as long as you clean it out often that should be ok. Personally, I kind of get lazy with stuff like that, so I tend to avoid them.

Third, you have no bubble trap from the skimmer. Not sure what kind of microbubbles may be created there, but there's nothing protecting them from getting to the return pump. But that probably wouldnt matter anyways, because....

Fourth, the fall of water from the skimmer section to the return pump, particularly considering the huge flow rate, is going to be very violent. Microbubbles are going to be created here.

What would I do? Probably something along these lines;

- lower flow from return pump
- lower water level in middle two sections
- raise water level in return pump section to be even with the skimmer section at least (though it will fall as water evaporates of course)
- build a bubble trap between skimmer/return

[s ruppa]

an inline check valve will stop any water above it from draining back to the sump. i put mine after the last T in the return ( if you are using any T's) so if stubbed out PVC from your drain, placed a check valve and then continued to your sump the only water that would drain would be from the check valve output to your sump. i got one with a union also so it can be removed

steve

[Phyxius]

Dont solely rely on a inline check valve to take care of anything for you as it could result in problems. Slime builds up in the tubing and over time can cause the CV`s not to function properly. Just a thought.........

[s ruppa]

i agree with phyxius. good point. it is a solution but combining room for water to drain back and a check valve is a good idea

steve

[eznet2u]

My first impressions...

Lose the rubble pipe thingy. If your flow is too high, it will become a skimmer of sorts. (Mine did)

Put the skimmer in the same section as the drain.

Lose the sponge altogether.

Lower the water level in the skimmer section, this will make the skimmer not have to work as hard.
(The deeper the pump, the harder it is to pull air.)

Raise the water level in the return section. I run mine about 1/4" below the section next to it...
No waterfall, no bubbles.

Add an ATO.

[Carissa]

Ok. Some good comments. If I can reduce flow requirements, I can fill up the sump more, since I can raise the location of my bulkhead on the main tank (and therefore need less room for backsiphoning). I was told that I needed 10x water flow through the sump. If I don't need that much, it will simplify things a bit. Also I was already planning on having a T to regulate flow from the return pump, and also a valve of sorts to regulate flow coming FROM the tank. So if I don't like how fast it's going, I'm not stuck with it.

Those of you who commented on check valves must not quite understand what I'm doing. I'm not worried about water from the return line. I'm worried about water from the input line from the main tank. Obviously if I put a check valve on this, I will get no water from my main tank into my sump, unless I'm not understanding correctly.?


So going through those calculations:
At a 1.5" depth, a 1" bulkhead will give me about 415 gph max (still around 8 - 10x). If this is sufficient, I will need enough room in my sump to handle -
12" x 35.5" x 3" (depth from surface to bottom of bulkhead) = 1278 cubic inches x 0.004329 = 5.5 gallons of siphoning possible from the main tank.

Being that my sump's maximum volume is 17.7 gallons, I will need to ensure that it is not more than about 70% full at any given time on average, to handle the possible 5.5 gallons.

Since my sump is 14" tall, 70% of 14" is 9.8". So to be safe, I need no more than 9.5" of water depth in the sump at any given time.

My skimmer is the type with the pump at the top. The pump is supposed to be just covered with water. Currently, without modification, that means I need a depth of 14" to get to the minimum water level for operating this skimmer (although I am operating it in about 11" of water right now, and it's working, though intermittently, don't know if it's the skimmer's fault or the water level). But I can cut off the tube (thus making it less efficient as well, I think), down to 10" depth.

Ok so lets say I do that. I'm a bit concerned about having no particulate filtration - won't the water get cloudy? My skimmer is a cheap piece of garbage, thus I found out after I bought it, so I don't want to rely on that solely for mechanical filtration. And no I can't buy a new skimmer, unless someone is volunteering to send me theirs for free. :) Heck, I would even pay postage.

So, I'll lose the live rock, and stick with sponge for now. I can always just take it out if I want to, if I'm getting nitrate buildup or whatever. I'm thinking, maybe the sponge should be AFTER the skimmer, thus doubling as a bubble trap too. This skimmer creates zillions of bubbles, I'm finding out. So here's my new plan. Again, comments please!!

By the way, how does an ATO work?

http://reefcentral.com/gallery/data/585/192325sump4-med.jpg

[wrott]

"and also a valve of sorts to regulate flow coming FROM the tank. So if I don't like how fast it's going, I'm not stuck with it."

You need to control/adjust flow from pump w/ a ball valve on the output of the pump. Not on input, and no valve on drain to sump--this is asking for trouble--clog.
If you're going to spend the time and effort to build the sump, I would buy a new skimmer(somehow) and add a refugium(even a small one will help a lot w/ nutrient export and pods).

[Carissa]

For the line going from the tank to the sump, I was going to use a clamp that is normally used for clamping off radiator hoses when someone is working on an engine. It just squeezes the flexible hose gradually depending on how far you tighten it to either restrict flow or you just loosen it to allow flow. No moving parts inside the hose to get clogged up.

The pump will have a T that will either divert most of the water up to the tank, or allow some of it to go back into the sump. I was told that this will not burn up the pump, like trying to actually restrict flow.

I may add a refugium later on. What do you need to start a refugium?

[crvz]

A check valve allows water to flow in one direction, so you would put it on the return pump. It allows water to flow to the tank, but not back to the sump. Still I wouldnt trust it.

Also, I wouldnt ever but a valve on the drain lines to control flow (though sometimes their logistically great to have in place during maintenance). If you want less flow from the drain, you slow down your return line. It's what's determining how much water is flowing anyways, they're not independent.

Onto the new sump picture, you don't have anything in place to keep the water level of the skimmer section that high. It's just going to equalize with the other section, since water isnt flowing over the top of the first baffle. Place another baffle to the right of the one not touching the floor. Make this new baffle the same height as that one, but have it extend down to the floor. Now you have a good bubble trap (water flows over baffle 1, under baffle 2, and then over baffle 3) and a sealed section for the skimmer.

[crvz]

<a href=showthread.php?s=&postid=13168279#post13168279 target=_blank>Originally posted</a> by Carissa
For the line going from the tank to the sump, I was going to use a clamp that is normally used for clamping off radiator hoses when someone is working on an engine. It just squeezes the flexible hose gradually depending on how far you tighten it to either restrict flow or you just loosen it to allow flow. No moving parts inside the hose to get clogged up.

The pump will have a T that will either divert most of the water up to the tank, or allow some of it to go back into the sump. I was told that this will not burn up the pump, like trying to actually restrict flow.

I may add a refugium later on. What do you need to start a refugium?

Again, there's no need to stick a flow restricting valve or clamp or anything on the return lines, it's just asking for a serious flood.

Depending on the pump, restricting it in most cases actually reduces electrical draw (using less money) and creates less stress on the pump. I'm assuming you're using a mag drive or similar, and so restricting it actually slows down the propeller and decreases wear. It's not like a linkage in an engine or other motor, where putting a load on the moving end increases motor effort and work performed. These are free spinning propellers that are not mechanically joined to anything, so slower spinning only helps the pump.

What's my point? Put a valve on the return pump (a ball valve is good, a gate valve is much better) and don't worry about a T. It's overly complicated and doesnt help anything in this case.

[Carissa]

Ok, lets try this again. :)

Would this work? I don't have enough glass to do three baffles.

How is it that the return pump determines the flow? If the return pump stops or slows down, the flow won't stop or slow down, that's the whole reason why I need extra space in the sump, right? But now that I think about it I guess, after so much water drains down out of the tank without getting returned, the rate of water draining slows down too, since the depth of water over the bulkhead is decreasing. Hmm. I'm obviously retarded when it comes to water physics. Is this how it works? Will the rate of flow of the return pump determine the depth of water in that chamber, or is it solely based on how much water I put in? I guess it would be how much water I put in....

So what determines how full the tank itself is, then? I can't quite get my mind entirely around how all this stuff is interacting. Would it be possible for me to overflow my tank if I cranked up my return pump too high?

http://reefcentral.com/gallery/data/585/192325sump5-med.jpg

[crvz]

Yep, so when the return pump stops, a few seconds to a minute or so and the water stops returning from the tank. since the return pump is what's pushing water in the tank, when you stop or slow down the flow from the pump the water in the drain will equalize to match it. The whole input=output thing. ;)

[crvz]

I didnt really address your other questions. The water level in the tank is somewhat determined by how fast water is being pumped into it from the sump, as well as how quickly the drain in the overflow allows it to return into the sump. If your pump is pushing water very quickly, so fast that the return hole size can't handle it, then yes it is possible to over-fill the tank.

However, if you back off the flow from the return pump, such that the return drain easily handles the water flow (amount of flow a drain can handle is determined by the hole diameter), the water level in the tank will stay just high enough for water to flow over the overflow teeth.

Does that help?

[Carissa]

Straight bulkhead, no overflow teeth or box. Yes I think this is starting to make sense now.

So let me think this out...

If my bulkhead is 1.5" below the water surface, I will get 415 gph according to this chart

http://www.nano-reef.com/forums/index.php?showtopic=69372

So if my return pump is adjusted to provide exactly 415 gph, the water level will always remain exactly at 1.5" above the bulkhead since it's being returned at the exact same rate that it's leaving. It doesn't matter how much water I add to the sump or not as long as I'm not running the pump dry.

Lets say my return pump is now returning water faster than 415 gph. It stands to reason that the tank will now fill up, since the maximum of 415 gph for my bulkhead is being exceeded. But then, as it fills, the bulkhead ends up deeper. So once it's up to 2" deep, I would then start getting 478gph going into the sump. So if I adjust my pump to work at 478 gph, I will have a deeper tank as a result. If we take it to an extreme and crank up the pump, I can see what would happen would be either the main tank would overflow, or the sump would run dry.

Ok now if we go the other way - I'm getting 415 gph at 1.5" depth, and I don't return it fast enough to the main tank. So my sump has to start filling up, until the water level in the main tank is low enough that I'm getting the same gph as my return pump. Ok, so if it's too low, I will start sucking air instead of water through the bulkhead, and/or my sump will overflow if I don't leave enough space in it to handle all the excess gallons.

Alright, it makes sense now, I think. :)

So is my sump design ok like that?

[crvz]

your link didnt take me anywhere, just the forum index. So I'm not sure where youre getting your info.

But, your logic makes sense. I would, however, seriously reconsider running just a straight bulkhead for an overflow. why are you doing that? an overflow box is not a bad alternative, and may be a very good choice in this case.

[Carissa]

1. I don't have one and it will cost me money to build one
2. I don't want excess noise, the tank is in the bedroom
3. I want a lid on the tank that seals very well, thus preventing excess evaporation (I have to buy water, and the tank is housed in my closet, not where I want a lot of moisture condensing)

Is there a good reason why I need one?

[crvz]

1. Yup, it sure will!
2. An overflow box doesnt have to be noisy, mine makes almost no sound at all. Surely less than what the sump splashing sounds will be. In my opinion, it's a bit (too considerably) more reliable than just a hole in the back of the tank.
3. Another issue all together. ;) Evaporation is surely a result of air exposure, but if you seal it up too well I'd want to know how that affects my O2 concentration.

[eznet2u]

Try this...

http://img213.imageshack.us/img213/2319/192325sump5medly8.jpg

This gives you the refugium. The refugium will act as the sponge and keep micro bubbles out of your tank.

As for putting this thing in a closet...you are asking for mold. Evaporation in the way to remove excess heat. Without it your tank will overheat quickly.

[One Dumm Hikk]

<a href=showthread.php?s=&postid=13173420#post13173420 target=_blank>Originally posted</a> by Carissa
1. I don't have one and it will cost me money to build one
Is there a good reason why I need one?

Yes, there is a very good reason to use an overflow box instead of just the bulkhead. Surface skimming area. If you use a 1" bulkhead and no overflow box, then you aren't efficiently removing the top layer of water from the tank. The longer the edge of the overflow box, the better it is at removing the surface of the water so that the skimmer can get it out of the tank.

If you are building the sump, then adding an overflow box wouldn't be that much more expensive and you would possibly have a piece of scrap large enough from the sump to build the overflow box.

The longer the edge of the overflow box, the better. To build one the full length of your tank, you would only need two pieces of glass or acrylic that are 4" wide and 48" long (assuming you have the 48" long 33 gallon tank). That gives you 48" of linear surface for the water to run over. Then the water would drain out of the bulkhead into the sump. The full tank overflow boxes are generally referred to as "calfo style overflows".

[Carissa]

The return bulkhead will be very close to the surface (~1"), so I figured that would keep the surface stirred up anyway.

I don't think heat will be too much of an issue. I'm not going with very bright/hot lights, they won't even be sitting on the tank but will be suspended over it, and the temp here rarely gets above 20C in the summer outside. Normally it's much colder than that here.

The skimmer produces lots of microbubbles so I figured this would probably add some O2. Plus, the sump won't be all that sealed off, the skimmer will be sticking out, and hoses coming out of it.

Am I off base with this?

And it would cost me a fair bit, probably $50 (a lot of money to me right now) to build an overflow box. I live where I have to order almost everything via mail, and shipping one more piece of plexi will cost me a lot, especially if I have to order it from the US which is likely the case. Right now, I have to order in a valve for my return pump which I can't buy around here, which thus far I can't seem to get cheaper than $20 with shipping. Just for a 1" ball valve.

[cdness]

The overflow would be a great thing to add. Maybe some others in your town have some plexiglas or acrylic that they would part with cheap. All you need is a razor blade and straight edge to cut the stuff. Silicone the heck out of it and you are good to go.

Another option would be to use an elbow with a piece of pipe going up to the place you want your water level at. Cut teeth into the pipe and you just made yourself a small but surface skimming successful overflow. It's not ideal but would get you by until you could build an overflow.

Another option would ve to see if someone in another town could cut the pieces then ship it flat rate to you which would be pretty inexpensive. Then you just assemble it and put it into the tank...

Just my .02

[Radar1]

Why even have baffles at all or do you have way too many micro bubbles. I would recommend a blank sump, using a micron bag for the drain, skimmer then your return pump. I do use a foam filter on my return pump and rinse it out on a regular basis. For a tank your size you should not have overflow problems during a power outage so ditch the check valve idea. I would not recommend using live rock or any type of refugium in your sump. I have tried these for years and have now moved to a clean simple sump and have never seen my tank stay so clean. Keep it simple for a tank that size it will go a long way.

[Carissa]

The skimmer does produce an awful lot of microbubbles. Plus I would like to have the option of a sponge or even just a spot to put some other media if I wanted to, and the baffles would provide that. A micron bag is a good idea too though. I think my sump is too small to house a refugium. It's only 23" long.

[Radar1]

What skimmer do you have?

[Carissa]

Visijet. A really junky one, evidently. But it does in fact work, I've been using it on my curing rock and it's doing fine other than a problem when the venturi got clogged once. Lots of brown gunk.