Return Line Design

[Forsaken77]

Wow learned something new! Thank you for sharing!

However I did notice the article stated that it is okay for cool water as restricting flow can cause additional heat. I wonder if this is going to harm the pump and shorten the life span.

You could always use 2 smaller pumps instead if splitting the lines. A lot of people use pvc on the drains and soft tubing on the returns to dampen noise as well. So if you had 2 of your dc pumps running with a dedicated line to each return, you wouldn't have a problem. And that way if a pump conks out, you still have the other one running.

I am in the process of setting up a dual return as well. At first I was going to split it, then decided to go with 2 smaller pumps.

 

[roberthu526]

You could always use 2 smaller pumps instead if splitting the lines. A lot of people use pvc on the drains and soft tubing on the returns to dampen noise as well. So if you had 2 of your dc pumps running with a dedicated line to each return, you wouldn't have a problem. And that way if a pump conks out, you still have the other one running.

I am in the process of setting up a dual return as well. At first I was going to split it, then decided to go with 2 smaller pumps.

It is a good solution but I just don't like extra equipments and maintenance associated with them. Also two puns will take a significant amount of room and create additional heat. I am just not sure if that's something I really want.

 

[Steve P]

If you want the flow to be evenly distributed between both return nozzles you will most likely need need to put a valve on the one closest to the pump. If you have a good
pump, you could build a 'manifold' type return, and have two or more returns. Manifolds are typically used to send flow to multiple reactors so that only one pump is used. The picture below is from a bulk reef supply tank build. They used one pump, (On the right). The single pump is used for both the returns and as a manifold to power reactors, but you need a good pump to do this. Hope this helps.

Off topic but does anyone know who makes the sump in that picture?

 

[Forsaken77]

It is a good solution but I just don't like extra equipments and maintenance associated with them. Also two puns will take a significant amount of room and create additional heat. I am just not sure if that's something I really want.

I considered that as well. But what else goes in the return chamber besides pumps? Nothing really. Splitting the line will mean running a bigger pump at a higher flow rate, compared to 2 smaller ones that aren't maxed out. I would guess that the bigger pump that's running harder would create more heat because it's doing a lot more work. But no one can say one way or the other.

I was looking into getting a Vectra L1 for my 180 to Tee the lines. Now I'm considering getting 2 Reef Octopus VarioS-8 pumps instead. They're a little oversized, which you have, so running them at 60% is less work. Plus I also read a lot of people complaining about the whining noise when the Ecotech pump is run above 60% and the RO pumps are dead silent and come with a slew of extras where Ecotech charges for. You're also splitting the maintenance on the pumps because not all of the water is going through each pump, only half. So you could probably get away with cleaning 1 pump every 6 months rotating.

Whichever way you go, good luck.

 

[Forsaken77]

Off topic but does anyone know who makes the sump in that picture?

It's a custom sump from Synergy Reef. Very expensive and a long waiting list.

 

[JoeIII]

http://www.cyclestopvalves.com/geninfo_2.html

Many of the return pumps we use are not centrifugal pumps referred to in the article.

Generally, a centrifugal pump will have a "snail shaped" volute, with the outlet offset to one side (most "big" external pumps, dc pumps like the jebao DCP, vectra, and similar)

Pumps like a quiet one or MAG drive with the outlet in the center are not centrifugal, and do not have the impeller design described in this article. these pumps have straight vanes on their impeller. This is common for less expensive AC pumps, as they can run in either direction and still pump water, so it does not matter what side of the sine wave the power is on when the pump starts.

This video from MrSaltwaterTank compared power consumption between a vectra L1 and a MAG18 - the vectra's output controlled by it's speed control and the MAG18 being restricted with a valve. As shown in their chart, the MAG does use less power as you restrict the flow but it is not a huge amount when compared to the amount the flow is restricted.

The video will skip to the chart, but feel free to watch the whole thing:

In my 125G system, I run two pumps. Currently I use one MAG9.5 (unrestricted) and one DCP 6500. Before the DCP (72% power) I had a quiet one 4000 as my second return and noticed two big differences when I made the change. 1) the NOISE - the quiet one was by far the loudest device in my system, now my wavemakers are. the DCP is near silent. 2) HEAT - I did not need a separate heater before - now I do. The DC pump uses less electricity to move more water because it converts less of it's used energy into heat.

My next system will hopefully have one larger DC centrifugal [return] pump. In my opinion the benefits from having one DC pump outweigh the drawbacks.

 

[siggy]

Someone mentioned the benefits of tubing, two equal lengths out of the tee should do the trick. or maybe place the tee in the center of the tank with equal pvc returns.
@saltyfilmfolks thanks again for the Goods.

 

[bif24701]

Really? Back pressure makes the pump more efficient? I have never heard of that.

Yes, because of cavitation is one main reason. If there isn't enough resistance or back pressure the impeller with have cavitation. This happens a lot with pressure rates pumps.

 

[saltyfilmfolks]

Many of the return pumps we use are not centrifugal pumps referred to in the article.

Generally, a centrifugal pump will have a "snail shaped" volute, with the outlet offset to one side (most "big" external pumps, dc pumps like the jebao DCP, vectra, and similar)

Pumps like a quiet one or MAG drive with the outlet in the center are not centrifugal, and do not have the impeller design described in this article. these pumps have straight vanes on their impeller. This is common for less expensive AC pumps, as they can run in either direction and still pump water, so it does not matter what side of the sine wave the power is on when the pump starts.

This video from MrSaltwaterTank compared power consumption between a vectra L1 and a MAG18 - the vectra's output controlled by it's speed control and the MAG18 being restricted with a valve. As shown in their chart, the MAG does use less power as you restrict the flow but it is not a huge amount when compared to the amount the flow is restricted.

The video will skip to the chart, but feel free to watch the whole thing:

In my 125G system, I run two pumps. Currently I use one MAG9.5 (unrestricted) and one DCP 6500. Before the DCP (72% power) I had a quiet one 4000 as my second return and noticed two big differences when I made the change. 1) the NOISE - the quiet one was by far the loudest device in my system, now my wavemakers are. the DCP is near silent. 2) HEAT - I did not need a separate heater before - now I do. The DC pump uses less electricity to move more water because it converts less of it's used energy into heat.

My next system will hopefully have one larger DC centrifugal [return] pump. In my opinion the benefits from having one DC pump outweigh the drawbacks.

Keep reading. We've had this debate before and the engineers won.

Yes, because of cavitation is one main reason. If there isn't enough resistance or back pressure the impeller with have cavitation. This happens a lot with pressure rates pumps.

 

[siggy]

Head pressure working as back pressure?

Yes, because of cavitation is one main reason. If there isn't enough resistance or back pressure the impeller with have cavitation. This happens a lot with pressure rates pumps.

Good article and another abstract to question logic
What the engineer left out of his criticism of designers using VFD'S on Centrifugal pump applications is that The Primary Benefit (VFD) is the SOFT START or RAMPING of large motors , with that function you no longer have the huge inrush of current or the need of conductors and fusing/protection for that level current.

 

[Reef of Fillory]

I just put ball valves on both of the return lines up towards the output on my 75g and it worked great for me.

I had mine set up the same way your drawing shows, with a T going straight up to one return line, and shooting over for the second return

 

[mcarroll]

The video will skip to the chart,

The vid started at the beginning – are you able to edit the post?

I totally get what you are saying but I am a little OCD. I want the tank symmetry by having the overflow in the center. I know it is not necessary but I can't help it...

It's fine if you want to! I was just saying that your diagram was fine too – if you didn't want to.

However I did notice the article stated that it is okay for cool water as restricting flow can cause additional heat.

You might find this post on my blog interesting (I am not the writer):
“AC and DC Pumps” – Internet Reefer’s Decoder Ring

There are no "bad guys", just the low-down about the pumps. Probably lots of new info on AC pumps and DC pumps in there! (Ever heard of a reluctance motor????)

 

[a-mused]

If you're concerned about heat, put a bulkhead on your sump and use an external DC pump. (though with a 125gal system + sump, a DC pump isn't going to add that much heat into the system.) Be sure to put ball valves on either side of the pump so it can be isolated easily for routine maintenance or replacement. As for valves to the returns, you're using them to balance the output, not restrict it. Otherwise your short & straight leg is going to be putting out a LOT more water than the more distant return off the leg of the tee. Why balance the returns? Because otherwise you may have so much flow returning from the short/straight leg that it's simply too much flow to put anything in front of it.

Pick a DC pump that has a speed controller - use that to control the amount of return rather than valves. Using your display volume of 125gal, plus at least 10' of head pressure (6' vertical, ~3' horizontal, two 90deg bends at the return nozzles, plus a tee (and unknown sump volume) - you'll want to spec a pump that flows a minimum of 625gph @ 10' of head pressure and 1250gph @ 10' of head pressure. You don't want this to be near the maximum points of a DC pump - which are near-linearly efficient when dialed down (and much quieter when run well below max speed). If you're looking at the EcoTech Vectra, the M1 will flow ~1100gph @ 10' of head pressure at max speed, so you're looking at around 660gph of return flow, running at 60% speed. Bonus points: it already has a 0.75" outlet , no reducers needed. Why don't we want reducers? Because you have a given volume of return water, regardless of pipe size. You're going to return X gph. If you reduce the size of the pipe, you still need to return X gph, so the speed of the flow will increase. 1.25" pipe has an area of 1.23 sq.in. 0.75" pipe has an area of 0.44". Since you're splitting into two 0.75" legs, you're looking at a total cross section area of 0.88", around a 40% reduction in cross section. That translates to a 40% increase in the flow rate through the 0.75" legs in order to meet your minimum target turnover of 625gph. With the Vectra, you don't have that flow rate increase issue .. if anything, you're slowing down the return rate since you're splitting it into two 0.75" legs. You're still getting the same turnover volume, but at a lower flow rate - yet another reason to put a gate valve in the short leg to balance the returns equally - or at least to give you the option to do so.

 

[KStatefan]

Many of the return pumps we use are not centrifugal pumps referred to in the article.

Generally, a centrifugal pump will have a "snail shaped" volute, with the outlet offset to one side (most "big" external pumps, dc pumps like the jebao DCP, vectra, and similar)

Pumps like a quiet one or MAG drive with the outlet in the center are not centrifugal, and do not have the impeller design described in this article. these pumps have straight vanes on their impeller. This is common for less expensive AC pumps, as they can run in either direction and still pump water, so it does not matter what side of the sine wave the power is on when the pump starts.

If they are not centrifugal what type of pump are they?

 

[Teddyboy16]

It's a custom sump from Synergy Reef. Very expensive and a long waiting list.

Aside from adding more John guest fittings or changing the size of the bulkheads. They don't do custom sumps anymore. Once your order is placed for one of thier standard sumps, it takes 2-3 weeks to make before shipping. Might take longer if they are really busy.

 

[mcarroll]

I see about 11' of plumbing length and about 4' of pumping height.

Assuming the addition of a 1.25" flow-through Tee to eliminate outlet valves.

Assuming the use of at least five 90º elbows.

5x flow would be about 600 GPH. On a friction loss calculator, I get +0.5' of head pressure:

Pressure Loss (psi): 0.22 Head Loss (ft): 0.5
Line Number: 1.25" @ 11' length @ 11 GPM
Date: 10/12/2017
Nominal Pipe Size: 1.25
Pipe Schedule: SCH 40
Flow Rate (gpm): 10
Viscosity (cP): 1
Specific Gravity (water=1): 1.025
Temperature (F): 79
Pipe Roughness (ft): 0.000016
Actual Pipe ID (in.): 1.38
Fluid Velocity (ft/sec): 2.15
Reynolds Number: 23490
Flow Region: Turbulent
Friction Factor: 0.025
Overall K: 7.05
Piping Length (ft): 11
Short Radius Elbows: 5
Tee Flow Through: 1
Pipe Exit : 2​

10x flow would be about 1200 GPH. I get +2.1' of head pressure on the calculator at this flow rate:

Pressure Loss (psi): 0.94 Head Loss (ft): 2.1
Line Number: 1.25" @ 11' length @ 21 GPM
Date: 10/12/2017
Nominal Pipe Size: 1.25
Pipe Schedule: SCH 40
Flow Rate (gpm): 21
Viscosity (cP): 1
Specific Gravity (water=1): 1.025
Temperature (F): 79
Pipe Roughness (ft): 0.000016
Actual Pipe ID (in.): 1.38
Fluid Velocity (ft/sec): 4.51
Reynolds Number: 49329
Flow Region: Turbulent
Friction Factor: 0.021
Overall K: 6.69
Piping Length (ft): 11
Short Radius Elbows: 5
Tee Flow Through: 1
Pipe Exit : 2​

Assuming the flow curve for the DC-10500 mentioned at the top:

This pump would be good for about 2600 GPH in this setup.

This pump is way, way overkill...even for a DC pump it'll be wasting power vs a "correctly sized" model.

Not to beat a dead horse, but I'd sell or repurpose these monster pumps if possible – they have to have some resale value don't they??? This is only a 125G tank, so even a DC-3500 would do the job.

 

[joro]

It's a custom sump from Synergy Reef. Very expensive and a long waiting list.

But worth the wait and every penny I had them make the same one for my 340g build and admire it daily!

 

[Dom]

I saw this on reefbuilders a while back. 1 pump, two returns. Even flow to both nozzles, not sure if its out yet.
https://reefbuilders.com/2017/08/16/maxspect-turbine-duo/#

Nice idea, but now, if the pump fails, you have to buy THAT pump again.

I like the idea of a centrally located return pump for even flow. The manifold is a nice idea too. By doing it this way, if/when your pump fails, you have flexibility to upgrade and make other changes.

I just feel that such a proprietary pump comes with restrictions.

 

[Dom]

Keeping the return simple is better than just about anything else.

Remember:

• Flow to the two returns in your plan do not have to be equal. That's not a rule.
• You don't really need more than one return.
• You need all of the flow to go into the tank. That's the only rule. ;Drowning
• Tank flow is via powerheads, not the return pump.
• The job here is just delivering sump water to the display in a timely and hopefully unobtrusive manner.
• We might debate whether 2X, 5X or 10X is "timely", but under normal circumstances there's no need to complicate the return beyond that.

And in my next build, I plan on having multiple returns because

1. I want to eliminate the clutter of power heads.
2. I will reduces the potential of stray voltage in the tank as my power heads are AC.
3. No power heads = less money; no purchase of the heads and less electricity.

 

[Dom]

I totally get what you are saying but I am a little OCD. I want the tank symmetry by having the overflow in the center. I know it is not necessary but I can't help it...

Yeah... I'm the same way... I dig symmetry. Thought that was only me.