Return plumbing and pump questions

[Reef monkeys]

This is my first set up with a sump

When I first fired it up I had pretty good flow using a Sicce 4.0, thinking I should have enough left over to run a carbon reactor. About 3 week from firing it up (reactor not set up or fuge running yet)I came home to a c2c with barely any water in it the BA overflow sucking air, forcing me to to dial the system way back at the gate valve. I took the pump out gave it a check over to make sure nothing was stuck or blocking flow and couldn't find anything, I'm at a loss. Should I redo my plumbing to a single return outlet and a bigger diameter than 3/4 or is the pump gone defective? The height from pump to top is 4', any advise would be greatly appreciated.

 

[woodnaquanut]

The 't's you have in the return are flow killers. Taking them out and going with one return that is larger will help. The real mystery is why it was working like that and is now not.

Did you take the pump completely apart? How about a vinegar bath? Does the pump by itself seem to have good flow?

Could there be something stuck in the return plumbing?

 

[mmn]

Not related, but those metal pipe clamps are probably gonna give you trouble down the road. Plastic ones are real cheap and don't corrode.

Regards
Michael

 

[reefer freak]

I did not disassemble the whole pump just down to the impeller and I have only had it out of the box for three weeks on a the new setup, I would hope I would not have to give a vinegar bath that often I did not check the flow before I I hooked it up so I really have nothing to go by as far as before and after flow. So the question is if I go to 1" plumbing and single return I will still need to T off for the fuge and reactor which will add a T's back into the equation, or is there another way?

 

[reefer freak]

Ha Thats funny I just noticed I have a different account on my phone for tapatalk than on the computer

 

[Pife]

I would be inclined to think something happened with your pump.

 

[4fishman]

The Sicce 4.0 puts out about 950 GPH - 15.8 GPM at a total loss of 12.5 ft.
To estimate the total pipe loss in an open system, you add the height in feet to the loss through the valves, elbows and other fitting. You said you already have a height of 4'. The straight pipe loss alone is about 56.4 ft of head per 100' of pipe, the TEE adds about 2 equivalent feet an elbow adds about 1 equivalent foot.
Total loss for about 12 feet of straight pipe (6.8'), fittings (1.2') and the lift of 4'gives you a total of 12' of head loss. You can see how you have very little pump pressure left. If you change your pipe size and fitting to 1" the 56.4 ft of head per 100' reduce to 16.5. This should give you some relief.
The quantity of returns should make no difference as the pump sees them both as the same pressure, not additive.
You could also remove that little return to the sump with the line-loc fittings, not sure what that's for, but the fluid wants to take the least resistive path, which that appears to be.
Hope this helps

 

[reefer freak]

The t and loc line with ball valve are feeding the fuge, it is not open at this point but will be in the near future. So what you are saying is the pump is insufficient? Ad far a the quantity of returns if pipe feet come into play if it was cut down to one return that would decrease and help some if I'm understanding what you are saying

 

[reefer freak]

Not related, but those metal pipe clamps are probably gonna give you trouble down the road. Plastic ones are real cheap and don't corrode.

Regards
Michael

Yea im in a small town, the hardware store here does not have the biggest selection or choices

 

[uncleof6]

The Sicce 4.0 puts out about 950 GPH - 15.8 GPM at a total loss of 12.5 ft.
To estimate the total pipe loss in an open system, you add the height in feet to the loss through the valves, elbows and other fitting. You said you already have a height of 4'. The straight pipe loss alone is about 56.4 ft of head per 100' of pipe, the TEE adds about 2 equivalent feet an elbow adds about 1 equivalent foot.
Total loss for about 12 feet of straight pipe (6.8'), fittings (1.2') and the lift of 4'gives you a total of 12' of head loss. You can see how you have very little pump pressure left. If you change your pipe size and fitting to 1" the 56.4 ft of head per 100' reduce to 16.5. This should give you some relief.
The quantity of returns should make no difference as the pump sees them both as the same pressure, not additive.
You could also remove that little return to the sump with the line-loc fittings, not sure what that's for, but the fluid wants to take the least resistive path, which that appears to be.
Hope this helps

The running in the branch in a tee, in a dual return, is like running the water into a brick wall.

The path of least resistance through a tee is run -> run, not run -> branch. To get significant flow out of the branch, the output side of the run would have to be restricted.

~12' dynamic pressure head, assuming your numbers are correct, which I don't think they are: because the Sicce 4.0 output is 952gph @ 0 foot of pressure head, and the actual flow rate is unknown (friction loss depends on the actual flow rate, through a given pipe diameter, friction coefficient, etc.) To figure out the actual pressure head, the actual flow rate needs to be known, or set as a constant, the pumps rating is irrelevant; the relationship is not linear. (The simplest method) Also, a 3/4" tee flowing through the branch is 5' in equivalent feet of straight pipe. 1" being 6' equivalent feet of straight pipe; 3/4" elbows are 4.5' in equivalent feet of straight pipe, and 1" is 5.3'. (Engineering Toolbox.)

The 54.6' per 100' of straight pipe looks about right, by interpolation, @ ~15gpm, however as soon as there is 1' of pipe on the pump, it no longer is flowing 15gpm, and friction loss drops. Put simply, ignoring friction loss, the pump flows 952gph at the outlet, if the water goes up 1', it is no longer 952gph, but less due to static lift (no pipe.)

The right way to select pumps and design plumbing systems is to draw it on paper, set a target flow rate, (constant) then run friction loss calculations to determine what the pump has to do, (gpm @ x' total dynamic head,) before you waste money on plumbing parts and a pump that won't do the job. After calculating it out, then buy a pump that will do the job, plus a bit extra. The way it is approached in the hobby (for the most part) is purely guess work, and often you actually need a larger pump, than the pump you have in hand, to run the plumbing system you built. That way you can have your dual returns, the gyres, and other flow robbers, without having to scratch your head and ask "why don't I have any flow?":hammer::headwally:

 

[woodnaquanut]

I did not disassemble the whole pump just down to the impeller and I have only had it out of the box for three weeks on a the new setup, I would hope I would not have to give a vinegar bath that often I did not check the flow before I I hooked it up so I really have nothing to go by as far as before and after flow. So the question is if I go to 1" plumbing and single return I will still need to T off for the fuge and reactor which will add a T's back into the equation, or is there another way?

The reason I asked...
Perhaps something got down in the impeller. Check the plumbing too. I'd assume the pump is not bad but everything fails. Run the pump in a 5 g. bucket. There should be lots of flow. If it did fail, at least it should still be under warranty!

Do you have life in the tank? Snails? Them little buggers get in everything. Also, until there is stability, the likelihood of Calcium precipitation is high. Wouldn't hurt to give the pump a vinegar bath.

All the above is related to the 'it did work now it doesn't' issue and figuring out what happened.

Changing the plumbing is more of a 'how can it be improved' issue. There is little functional reason for two returns. Getting that one 't' out will help. So will upsizing the pipes. You might not have enough pump to run both the fuge and reactors. Adding a small pump like a MJ/Cobalt 1200 to handle the reactors is not a big deal.

 

[reefer freak]

Thanks woodenaquanut I will give these things a shot, I do have the pump that came with the reactor just was trying not to use it. as far as snails go they seem to be something that dont last long for me so I quit buying them as part as a cleanup crew and rely on other critters to help which seems to be working thus far. The two returns seemed like a good idea as to get more even in tank flow but I also have 2 Jebao WP-40 in there so maybe thats enough, Im still pretty new to this hobby and still learning but this tank is a far cry better that the 55 I had, I have a 125 freshwater that might end up being my next saltwater build lol.

 

[reefer freak]

The running in the branch in a tee, in a dual return, is like running the water into a brick wall.

The path of least resistance through a tee is run -> run, not run -> branch. To get significant flow out of the branch, the output side of the run would have to be restricted.

~12' dynamic pressure head, assuming your numbers are correct, which I don't think they are: because the Sicce 4.0 output is 952gph @ 0 foot of pressure head, and the actual flow rate is unknown (friction loss depends on the actual flow rate, through a given pipe diameter, friction coefficient, etc.) To figure out the actual pressure head, the actual flow rate needs to be known, or set as a constant, the pumps rating is irrelevant; the relationship is not linear. (The simplest method) Also, a 3/4" tee flowing through the branch is 5' in equivalent feet of straight pipe. 1" being 6' equivalent feet of straight pipe; 3/4" elbows are 4.5' in equivalent feet of straight pipe, and 1" is 5.3'. (Engineering Toolbox.)

The 54.6' per 100' of straight pipe looks about right, by interpolation, @ ~15gpm, however as soon as there is 1' of pipe on the pump, it no longer is flowing 15gpm, and friction loss drops. Put simply, ignoring friction loss, the pump flows 952gph at the outlet, if the water goes up 1', it is no longer 952gph, but less due to static lift (no pipe.)

The right way to select pumps and design plumbing systems is to draw it on paper, set a target flow rate, (constant) then run friction loss calculations to determine what the pump has to do, (gpm @ x' total dynamic head,) before you waste money on plumbing parts and a pump that won't do the job. After calculating it out, then buy a pump that will do the job, plus a bit extra. The way it is approached in the hobby (for the most part) is purely guess work, and often you actually need a larger pump, than the pump you have in hand, to run the plumbing system you built. That way you can have your dual returns, the gyres, and other flow robbers, without having to scratch your head and ask "why don't I have any flow?":hammer::headwally:

Are you saying take the T out and make it a single line return here? Eliminating the "brick wall" And only have one T that would run to a regulated manifold for the reactor and fuge?

 

[uncleof6]

Are you saying take the T out and make it a single line return here? Eliminating the "brick wall" And only have one T that would run to a regulated manifold for the reactor and fuge?

Correct, along with larger pipe. As well as going on about how pump fed system design should be approached instead of the guesswork hearsay method that is currently popular.

 

[reefer freak]

Thanks uncle! My hydro engineering skills are are pretty much non existant.. Where would I find the information on how a pump fed system should be approached? Or is there someone that might be able to offer the correct information For the system I have here ?

 

[uncleof6]

Thanks uncle! My hydro engineering skills are are pretty much non existant.. Where would I find the information on how a pump fed system should be approached? Or is there someone that might be able to offer the correct information For the system I have here ?

You would have to ask now, after having a good step by step up earlier, that I now cannot find period. The detailed link I had several years ago, is kaput.

Draw your plumbing system on paper. Include all fittings, etc. in your drawing. Pick a flow rate: for instance for a 60 gallon tank 600gph (10gpm.) I suggest you use a round number in 10gpm increments. Once you get used to the process, you can hunt up more detailed friction loss charts. (these charts do the hard math for you.) Indicate exact lengths of straight pipe used in the system, and indicate the static lift. (Distance from the sump water lever, to the water level in the DT.) Then you jump onto Engineering toolbox, and start figuring out the friction loss at your picked flow rate, for all the straight pipe in the system. Then you figure out the straight pipe equivalent for all the fittings (also listed on EB) add it all together, and multiply the result by the friction loss per 100' divided by 100, (move the decimal point two digits to the left; e.g. 16.26/100 = .1626) to get the friction loss per foot. Add the result to the static lift, and you are done. Now go buy a pump that will do the job required (600gh @ x' total dynamic head) plus a bit more. Generally since pump curves don't look at friction losses, you usually wind up with the next closest pump anyway, which is larger. If you go smaller, you will end up scratching your head...

If you don't like the results, use larger pipe and/or less fittings.

http://www.engineeringtoolbox.com/pressure-loss-plastic-pipes-d_404.html

 

[laverda]

I would eliminate as many T's and elbows as possible by using supeflex tubing and feeding your sump and reactor from the overflow drains. That saves electricity as well. Super flex can be ordered from Marine Depot and others.

 

[uncleof6]

I would eliminate as many T's and elbows as possible by using supeflex tubing and feeding your sump and reactor from the overflow drains. That saves electricity as well. Super flex can be ordered from Marine Depot and others.

Feeding the sump from the drain is how it should be. I suspect you mean the "fuge" and reactor, rather than the sump. However, this is a poor idea as it turns the "fuge" into a garbage dump, and the reactors mess with the physics of the drains. As far as in sump "fuges" go, they are not nearly as necessary or effective as the popular would suggest, and most high end systems do not bother with them.

This is not about saving electricity. This is a luxury hobby, and requires almost a luxury budget. It is about designing a system, and making a smart decision by selecting a pump that will do the job required of it, without the guess work and hearsay. Part of that line of thinking, is while optimizing pump performance, we don't do things that are counter productive to the system as a whole, such as that mentioned above. Sometimes in the process, you discover that you actually can use a smaller pump, and save a few pennies that way... pumps I recommend for average systems, draw less than a real light bulb, but you lose that by pumping up the lighting higher than it really needs to be, or use under developed technology for the lighting, which really is a wasting electricity.

 

[laverda]

I did mean to say refugium or if you prefer your DSB area of the sump. I disagree with you about saving electricity. Not being energy efficient is just wasteful! Attitudes like yours are why many people find the hobby too expensive and quit or never get started. I have 5 reef tanks running including a 240 gallon and my electric bill is less then many people in the same area with just a single much smaller tank.
I agree with a lot of what you say, but not everything.
Their are many ways to accomplish a successful reef tank, just because someone does not do it your way does not make it wrong.

 

[uncleof6]

I did mean to say refugium or if you prefer your DSB area of the sump. I disagree with you about saving electricity. Not being energy efficient is just wasteful! Attitudes like yours are why many people find the hobby too expensive and quit or never get started. I have 5 reef tanks running including a 240 gallon and my electric bill is less then many people in the same area with just a single much smaller tank.
I agree with a lot of what you say, but not everything.
Their are many ways to accomplish a successful reef tank, just because someone does not do it your way does not make it wrong.

I don't run DSBs in the sump. Most of them are in-tank and several in buckets.

Actually trying to assign an "attitude" to my comments is rather presumptuous. My comment was a statement of fact, as your response shows: it is too expensive, so they get out of it, or don't bother getting into it. Aquariums are simply not energy efficient systems. The harder you try to make them energy efficient, the more likely you are to run into problems.

My discussion involves the designing and implementing of pump fed plumbing systems, not how many ways there are to accomplish a successful reef tank. It involves the physics to accomplish a specific goal. It involves Centrifugal Pumps 101, and there is one right way to approach it, and a whole lot of wrong ways that are merely guesswork. Often that may mean a larger pump, and a higher energy bill. It is just the way it is. Considering you can run 3000gph for the cost of a 150watt light bulb, the whole debate is over nothing, and it is done simply for the sake of arguing over it, and hence it will never end. Then there are those that don't want a higher electric bill, but they want to pump up out of the basement to a 40 gallon tank... kinda makes ya think doesn't it? Then they add dosing pumps, auto water changers, (pump driven of course) controllers, ad infinitum, put a mega watt of lights over their tanks, or use lighting technologies to don't provide enough bang for the buck, and then say "I can't afford to run a good pump..." that kinda should make one think too. :idea:

I am especially beginning to like the comment "there are many ways to..." because they seldom say how that is or why, and I look at a system with an ATS with say 45 watts of CFLs, (won't go into what those things are really worth) and then they have a fuge with Chaeto with another use of wattage for lighting all wasted electricity, because to pump a couple hundred gph through a bucket DSB will use less watts, (try 28watts for 1000gph using a second pump...) if you buy a pump the "smart" way rather than the commonly recommended garbage. Yeah, there are many ways to do things, but they don't get into what you go through to get there. Again, it simply keeps the debates alive.

As far as methodologies go, what I discuss is not the only way, but what I discuss is a heck of a lot simpler, (save for the discussion on why and how these things work) and does the same, if not a better job of it. The nice part about it is I can lead the horse to water, but it is up to him whether or not he takes a drink. :deadhorse1:

The other one I like is "I do that, and my parameters are fine, there many ways..." but that myth has not come up yet, in this thread.