The Plumbing and Water flow Workshop

[mojoreef]

Ok folks Since we are in between workshops I thought I would put one together on the above mentioned topic.

Water flow is one of the most under rated aspects of keeping a reef aquarium, and to that some general misunderstandings (Overkill, one way flow and so on) and it seems to be a good topic to cover.
Their are many different types of tank set ups and each need to have particular plumbing flow plans in order to accommodate the type of filtration being used. Add to this the vast amount of different equipment types and it can be confusing. So lets start with an overall concept behind good flow and patterns and then work our way into breaking down the systems. When someone is setting up a reef tank they must try to not only recreate the flow that a particular corals require, but they must keep in mind the confines and drawbacks of it being a closed tank system.

When designing a system that is going to be lagoonal, keeping soft corals, lps and simular and uses a dsb they have some consideration to keep in mind. Soft corals, clams and lps are big eaters and must be able to capture food and absorb nutrients in order to survive and meet their energy budgets. So we want to makes sure as much food and detritus stays active (moving around) in the tank as possible. Once the food/detritus hits the sand it becomes the food for it and not the corals. Now add to the equation that one of the limitation of the sand system is on how much food it can process, so this makes it make sense to keep as much available to the corals and reduce the over all load put on the bed. Now we just have to do this and not blow the sand all over the place, lol.
From my experience I find we don't have to put alot on the size of overflow as we want to keep it in the tank for the most part, so normal overflows will be fine. We start the overall plan by taking care surface water. This can be done easily by using the return water from the sump. So just come over the top with at least two returns, they can go directly into the tank pointing across the surface about two inches deep or by using seaswirls if desired. Keep it simple here and use the sides of the tank to bonce flow off of with your ultimate destination being the overflow
for normal outputs

For the main part of the flow you can use either a closed loop system or powerheads to create the overall flow for the mid section and lower section of the tank. The idea here is to try to interlace the flow sources so that they help each other accomplish the overall plan.
So here is an end view of a tank, it shows some good flow coming from the back of the tank and going from a point just above the sand bed and pointing slightly higher towards the front. Then Their is a second level of flow that takes the flow from that point and pushes it higher to the upper flow which will take it out of the tank.

Remember its not so much as "how much flow you have" as it is "how well placed it is". Ok now for a top down flow. What we are looking for is an over all coverage and the ability to keep as much of the food/waste/detritus in the water column. So having more outlets with less flow per is going to make that easier to accomplish then to have just a few outlets with a ton of flow.
So in this pic it gives you a general concept looking from the top down as what kind of coverage we are looking for

Ok so that is a basic idea of what we are trying to do with flow in the tank, so the next thing would be how to accomplish that using pumps, pvc and other neat toys. Feel free to toss in what you want for questions and ideas.

MIke

mike

 

[NaH2O]

Mike - the second diagram down. Is that representing if you have the return coming in the front of the tank? or is it just showing what the flow does as it bounces off the front?

 

[CarlaW]

Don't forget to talk about flood control, say if the power goes out, ie., siphon holes and such. I think alot of people forget that what goes in has a way coming back out and making an awful mess!!
I think this is gonna be an awesome thread, there are so many different ways to create the flow that you want.

 

[reedman]

The other factor that I think most people forget about is rock work. All the diagrams in the world aren't going to do any good unless you account for the rock work (and coral growth). Flow in an empty tank is easy. Flow around a rock structure that is visually pleasing isn't.

Also when placing outlets low in the tank you have to remember that there will be less flow out of those than other outlets higher up. Simple fluid dynamics. And finally, spray bars work great when they have enough power behind them....I doubt many reefkeepers have a pump big enough to make a spraybar worth while.

Great workshop topic Mike.

 

[mojoreef]

Mike - the second diagram down. Is that representing if you have the return coming in the front of the tank?

Yes that is another forum of return, say if you used sea swirls or simular.

CHarlie/reed good posts, we will try to cover s much as possible. What I am trying to show is that folks should design the system to work with the filtration concepts they use, and to not worry about a large amount of flow as compared to multiple amounts of lower flow.

Mike

 

[NaH2O]

Would the flow patterns change at all if a tank didn't have overflows and only a hang on skimmer/filtration?

 

[mojoreef]

Ok lets tackle the closed loop concept first. A closed loop is a flow system that basically draws water from the tank and blows in back in with a pump inline. So in never seeing air it is closed with in the system.
When we put a closed loop on a tank we have to keep some math in mind, so lets get passed this first. Lets assume we are using a ampmaster or sequence pump that is blowing 3000 gph with an 1 1/2 input and output. For location of the input of the pump we need to have two things in mind. One is we need to put it in a location where we can access it from time to time for cleaning, the second is its always best to double the input if possible so we dont have so much suction coming from one hole that it sucks in a fish (its also good for redundency). Ok one more to, lol look for a spot that will hide the plumbing when you have your rock work in place to. So in the case of this 3000gph we are going to do 2 x 1 1/2 inch inputs.
Now for the outputs of this closed loop we are going to go with 3/4 inch. The reasoning is that locline is 3/4 and will screw right in. Now to figure out how many you can use and maintain descent pressure, I always use the 3 to 1 rule. Example: it takes 2 times 3/4 openings to make one 1 1/2. So times that by 3 and you should be fine, so in this case a max of 6, 5 would be better. We will use 5.
Ok so we have the basice design in mind and we know how many holes to drill so lets get busy


Mike

 

[mojoreef]

Nikki a little, but the basice premise would be the same. Instead of the top most flow pointing at the overflows they would point to the filtration devices, all else should remain viable.

Mike

 

[bc_slc]

great thread. Taggin along.

 

[reedman]

Example: it takes 2 times 3/4 openings to make one 1 1/2

Not to go all engineer on you Mike, but it's actually 4 outlets @ 3/4" each gets you the cross sectional area of a 1 1/2" pipe.

3.14*radius squared = cross sectional area

so 3.14*(.75)^2 = 1.76 for 1.5" pipe
3.14*(.375)^2 = 0.44 for .75" pipe

4*.44 = 1.76

sorry...engineer in me.

 

[mojoreef]

Not to go all engineer on you Mike, but it's actually 4 outlets @ 3/4" each gets you the cross sectional area of a 1 1/2" pipe.

Excellent Reed, that why I just event mystical formulas like

I always use the 3 to 1 rule.

Ok so lets go after the closed loop. The first thing I do after having a general idea of the plan is to set up the plumbing peices above the tank so I can have a good idea of how its going to look, and to get more detail on the measurements

So this is my tank, but the concept and proceedure should be the same, just ignore the motorised ball valve. So for this closed loop I like to use 1 1/2 inch pipe as the manifold and then reduce it to 3/4 inch bulkheads to come in the tank wall. The above picture basically shows the Tee's layed out to where they will sit on the back of the tank.
Ok once that is done and I have all the peices layed out sitting on top, I make marks on the back of the tank wall right where the centers of all these Tee's are sitting. If you look at the Tee it will have a center mark from the mold when it is being made. To make a mark I usually just use a razor blade, but if you have a clear back or you dont want the marks you can just run peices of tape and then use a pen to mark it.
On my tank I have a coast to coast overflow and standard overflows, so I mark those out on the back also so I know where everything is (better then drilling through your overflow :idea: ) . So once they are layed out you can now take the Tee location measures and transfer then down to where they are going to be in the tank. In my case they are going to site just below the coast to coast overflow.

Mike

 

[mojoreef]

Ok now comes the part where we take a perfectly sound aquarium and drill it full of holes!!
To do the drilling I use a hole saw like this one

From here we just drill the holes, when drilling becareful when the pilot hole goes through and then the hole saw hits the acrylic, so dont push to hard. Also when drilling you dont want to stop until the hole is all the way through and it helps to make small circluar motons when drilling this way the hole is not to tight for the saw. the end product is

Ok so now its time to plug up those holes. The first thing I install is the Bulkheads, in this case 3/4 inch bulkheads which are thread by thread

I always use silicone gooped between the gasket and the BH and then gooped between the gasket and the tank wall, this keeps water from leaking through the threads. Into the bulkhead I use a 1 inch closed threaded nipple. It looks like this one but only an inch long and all threads

Again I silicone these peices together, the result should look like this

Mike

 

[mojoreef]

Ok moving on, from here I make the transition from the 3/4 to the 1 1/2 manifold. So to do that I want to attach a 3/4 female threaded by 1 1/2 slip male bushing to the nipples. Again it will look like this one but in the measurements given

the result is this

Ok that was the easy part, lol. Now we need to make the manifold and then attach it. I dont have pictures, but here is the process. I take one of the tees and slide it onto the bushing about half way, then go to the next bushing and slide another tee on. At this point I take a measure of the 1 1/2 pipe needed to span the two. I make the cut and then glue it together, IMPORTANT!! While the glue is still wet between the two tee's I slide them onto the bushing again, this makes sure that they will fit perfectly when we glue the manifold to the bushings. You do NOT want to glue the tees to the manifold at this point. I follow this procedure for each of the out put bushing until I have a manifold assembled, in this case it will consist of five outputs. From here you can glues all five bushing to the maniflod and slide them on. It should look like this

Ok just to add a few things here. In my case I have a MBV so I have 5 to 6 outputs on either side of the valve. Now if you arent using the MBV you can just use another tee facing down towards your pump. Or ome in from one side, so thier are options.

Mike

 

[mojoreef]

Ok so now we finish the output manifold. You can cap either side of the manifold with a 1 1/2 cap or you could have used 90 on the ends instead of the end tee's. From here you complete the plumbing to the output of the pump.
From here we do the inputs of the closed loop in the same manner as the outputs, but we dont reduce the pvc as we are going to keep all the plumbing 1 1/2 including the bulkheads.
Once both the input and output manifolds are inplace we cut and glue in the balance of the closed loop. Make sure you use unions and valves on the balance of the plumbing so you can remove the pump for maintence down the road with out having to drain the tank. In my case I have unions built into the pump and just added valves. your finished product should look like this sorta, lol

Ok lets hold on here and see if thier are any questions


Mike

 

[reedman]

That last picture illustrates a great use of location to minimize head loss. You have a straight out from the output to the T fitting. Also, you have the pump elevated above the floor to optimize the pump flow. Many people underestimate how much flow is lost to pressure head. Also, reducing at the last part before the locline reduces losses due to friction in the plumbing. This all adds up to more flow, less back pressure and more bang for the buck.

 

[jlehigh]

Your observations are right on Reed. This plumbing design influenced mine by far more than any other tank out there.

I went ahead and diagrammed mine... I wish it was more straight forward to show but with 14 inlets it get's tough. The longer the line the longer the draw distance in that relevant perspective. If I extended each line it's full flow path it would look like one massive mess.. The purple inlet is the sump return line. 4 different opposing inlets are being fed at all times rotating via an 8-way Oceans Motions.

There are several weaknesses in my design.

- I rely heavily on inlets placed high in the tank to blow flow to the lower areas instead of having inlets lower in the tank (with exception of the spray-bar)

- Though flawlessly doing it's job the spray bar uses up what could have been 4 additional inlets

- Having inlets placed lower in the tank could increas the efficiency in getting detritus into the overflows

 

[mojoreef]

Also, you have the pump elevated above the floor to optimize the pump flow.

Reed if the pump was on the floor and the inputs and out puts were in the same locations it wouldnt matter. As in it wouldnt take a head loss. Now I will bow to the engineer in you, lol, but from my understanding. if the pump were dropped the pressure from the water coming down into the pump would cancel out the extra head pressure for it having to go up higher. So basically thier is no head loss from the point of the input in comparison to the output line??

Jleigh thanks for the pic, it will give us more diagrams for folks to look at. One thing to remember is that flow patterns will never stay static for the life of the tank. As corals grow or rock gets moved you will always be adjusting the directional flow. that is the good thing about using lockline on the interior.

Mike

 

[NaH2O]

Here comes the female with the crazy female questions . How would the drilling differ if it was a glass aquarium? Is there a formula or anything to help you with hole placement so you don't break the tank? I'm also assuming the hole saw might be different?

Reed - I have my closed loop pump sitting on the floor behind the tank, and the flow seems to be great! I don't know how to determine how much loss there is, but wouldn't it be just a minimal amount? Also, Mike indicated in the first post

Remember its not so much as "how much flow you have" as it is "how well placed it is".

So, based on this, the amount of loss you might see by having your pump 3 feet lower (in my case) doesn't matter that much.

 

[reedman]

Yep, I'll back off the head pressure loss... a bit. Here's a blurb off the web on calculating head loss in an aquarium that indicates there are some losses due to head pressure, but not large ones.

[font=Arial,Verdana,Sans-serif]First, determine the lift or height the pump will need[/font] [font=Arial,Verdana,Sans-serif]to move water. If the pump draws water from a reservoir below the aquarium, measure (in feet) from the pump to the point water returns to the tank. This will be your initial head pressure in feet.[/font]

[font=Arial,Verdana,Sans-serif]If the pump draws water from and returns water to the aquarium[/font] [font=Arial,Verdana,Sans-serif]through a filter, divide this figure in half as the pump is getting assistance from gravity on the intake side.[/font]

[font=Arial,Verdana,Sans-serif]Keep in mind that every turn the water takes[/font] [font=Arial,Verdana,Sans-serif]as it returns to your aquarium increases head and reduces flow. Keep tight turns to a minimum. A good rule of thumb if using 1/2" to 1" I.D. tubing is to add one to two feet of head for every 90 degree elbow or tee on the return side of the pump. The smaller the diameter of the tubing, the MORE head you should add.[/font]

[font=Arial,Verdana,Sans-serif]If you reduce the diameter of the return line at any point,[/font] [font=Arial,Verdana,Sans-serif]add a number of head-feet to your calculation equal to the percentage of reduction. If, for example, you've calculated a head pressure of ten feet in a 1" tube and you then decrease the diameter of the tube to 3/4", add 25% or 2.5 feet to your head pressure.
If you are using mechanical or chemical filter modules, UV sterilizers, or other inline filters or devices as part of your filtration scheme, add two to ten feet of head per device to your calculation. The finer the material the water must pass through, the more head you should add. Keep in mind that as filter material clogs, head pressure increases. Err on the safe side here.​

[/font]

My pump sits on the floor too Nikki. Not practical for me to have it any higher. The big kicker for me was experiencing first hand how much you flow can lose when you have hard plumbed bends in the plumbing (especially on the output side of the pump). ​

 

[Skipper J]

Here comes the female with the crazy female questions . How would the drilling differ if it was a glass aquarium? Is there a formula or anything to help you with hole placement so you don't break the tank? I'm also assuming the hole saw might be different?

Good Question Nikki!I was beginning to feel that AGA is getting obsolete because of new flow system :lol:

Very Helpful thread Mike! Apart fron the other threads you posted of course!
I'm listening!!!Thanks!