Closed loop manifold help

[spieg]

Im in the design stages of a closed loop manifold and need some help with plumbing in the most efficient manner. I want to minimize friction loss to get the most flow possible.

The intake will be a 2" bulkhead that feeds 3' straight to the pump with no elbows at all. The return will go from the pump to a 1.5" closed loop manifold that sits on top of the tank and use either spaflex or (1) 45 degree elbow to get back to the manifold. The outlets will come from reducing tees on the manifold. Im considering both the super dart gold and the barracuda gold for pumps. The tank is a standard 125g and im trying to supply all of my flow needs without any powerheads and without running two closed loop pumps.

Now for the questions:
1. On the manifold, does it matter if I use (2) 45 degree elbows instead of (1) 90 degree elbow at each corner, or is the friction loss the same either way? How do long sweep elbows compare? Would (1) long sweep 90 be comprable/better than (2) standard 45's?

2. With regard to the outlets on the closed loop itself, does it matter what size the outlets are? Im sure 1/2" outlets would have higher velocity compared to 3/4" which may not be desirable. But having more outlets makes the system more flexible/adjustable/random. Would it be better to have (10-12) 1/2" outlets than (5-6) 3/4"? If you are moving the same amount of water, does it matter what size the outlets are? Where is the sweet spot on this type of setup to have enough outlets to provide random and adjustable flow and avoid dead spots yet not end up with a bunch of high velocity jets?

3. Where the output from the pump first meets the closed loop manifold, are there advantages/disadvantages to teeing into the loop vs. teeing off the loop, or does the pressure stabilize either way and not make any difference? (see pic)

4. Is this type of setup using a single pump on a 6ft tank with no supplemental flow from powerheads viable or am I wasting my time and money?

Any and all help or suggestions is appreciated! Also if anyone out there has a similar setup (1 external pump, 6ft tank, closed loop manifold) id love to hear what size your outlets are, how many you have, and how its working for you. Thanks in advance!

 

[stugray]

I run a setup exactly as you are describing, however my 2" intake goes over the back of my tank.

I use a variable speed Barracuda and cannot run it at full speed or I have WAY too much flow.

If you choose that pump, you dont need to worry much about friction losses as you might have to throttle it down anyway.

My manifold goes straight up the back to a TEE where it splits and goes to either side of the tank. There are three loc-line nozzles per side for a total of 6.

The first two locline nozzles in the line are smaller diameter than the rest to keep the pressure up and provide more flow to the ends.

You also might want to consider that the two pumps you are considering are NOT silent.

If you could put the CL pump in the basement below your tank, or on the other side of a wall, that would help.

Stu

 

[spieg]

stugray: Thanks man! Ive seen some pics of your setups in some other threads and was hoping youd reply! Since you have too much flow with the barracuda cranked up, and since im not sure im up to the task of rigging up that sweet variable speed stuff, do you think the super dart gold would work well for my setup? If I can can save on the 125w or so id like to. Am I correct in assuming you are using 1/2" for your first two nozzles and 3/4" for the other four? Have you tried the same setup with all six nozzles the same larger size and found that you were unhappy with the results? The tank will be set up in my finished basement, and the closed loop pump will be on the other side of the wall behind the tank mounted at the height of the 2" bulkhead so that I dont need to put any bends on the input of the pump. Thanks again!

 

[Perched Urchin]

Spieg and stugray,
I am in design stage of building a similar system for my new 60G cube (so a smaller scale than both of yours) and likewise hoping to use this as my sole source of flow (I can always augment if I have to down the road). My plan is not quite true closed loop, as I am using my return from my sump as the supply for the manifold. Also, my design is to come through a hole in the center top rear of the tank (I will have a siphon break built in just in case of pump stopage), elbow down to a tee/wye at the bottom of the tank, then bury the manifold under the substrate and hide the nozels in pieces of LR. I am using a Reeflo Dart as the return pump. Here are some ideas I have had, suggestions I have gotten from various folks as some questions that maybe one of you could answer.
1 - I have been advised to use 2 45's where ever possible versus 90's, as it is much easier on flow. Since I will have the lines under the sand, I plan to do this on my corners (unless I end up using "plan b" from below)
2 - it is possible to get true wyes (3 equal 120 degree angles), which I had considered versus a tee into the manifold as this would greatly improve flow there; however, I am struggling to find a corresponding 30 degree fitting to complete the 90 degree angle. This led me to consider "plan b" - using flexible PVC for the entire manifold loop, eliminating all the elbow fittings beyond the wye completely. What do you think?
3 - I had originally thought of using wyes instead of tees to connect the nozels to the loop; however, after more thought, I don't think this is necessary. Since this is a closed loop, once static pressure is reached, I think the tees are fine. I like stugray's idea of using a slightly smaller fitting on the first nozels on the loop to try to keep a more even flow throughout, as it seems to make sense - but I am no expert in fluid dynamics. Does anyone know if this is really necessary?
Thanks - very interested to read more on this.

 

[spieg]

Perched Urchin: Im glad im not the only one with closed loop interest/questions. Ive read a ton of old posts on it, but so much of it seems to be dated and/or untested, that im really hoping to get some more people to chime in with some real world experiences or some hard numbers at least.

As far as your plan goes, id be concerned moving that much water through your sump. Im not sure how big your sump is, and whether or not you plan to have a refugium in sump or anything like that, but that could pose an issue.

1 - Im not really sure on the fluid dynamics here either, so hopefully someone chimes in. Im going to say that either way, theres going to be similar friction loss, but in your case, youll have a lot more head loss because youre pumping the water up from your sump. When all is said and done, even if you only have to move the water 4ft up, youre probably going to have more like 8ft of head after your bends and everything else is added up. That reduces the output of the dart down to 2500gph or less pretty quick. A pressure biased pump might be more suited to your environment, or a separate return pump and a separate closed loop setup.

2 - True wyes are available, flexpvc.com sells them as do others. If the friction losses in a static loop are the same regardless of degree bends (and im not sure that is or isnt the case) then it shouldnt matter if its a true wye or a tee for any of the connections. The true wyes just make the plumbing job a little harder as youve mentioned.

3 - I think the tees are just fine for the outlets. As for whether or not the first outlets need to be smaller, im hoping to get some more feedback on that one, too!

 

[Perched Urchin]

Spieg,
Another unique aspect of my system is that my sump is going to be about 11 feet away from my DT in an unfinished part of my basement and you guessed about right - calculated head loss for the entire system ends up at about 8 feet. I actually discussed the entire set up with Chris at Reeflo just to make sure I wasn't going to lose all my flow to head pressure. His thoughts were that I would still get over 2000 gph all told, which would be about 36 times the actual DT volume. That should be pretty good.
The flow through the sump (40G breeder) will be brisk, but I have discussed the plan with a number of folks that have a good bit of experience and they don't think it should be a problem. Running through a couple sets of baffles and a fuge should give time for microbubbles to disperse. And a couple folks actually pointed out that Cheato likes flow too.

 

[uncleof6]

Although the friction loss of a 45 is half that of a 90, using two 45s to replace one 90 will equal the same friction loss. The advantage with 45s is replacing two 90s with two 45s, thus cutting the friction loss in half. (not always practical to do)

The friction loss in a tee is very significant, depending on how the flow enters the tee. Entering through the branch, for 1.5" the loss is 8' in straight pipe. Entering through the "run" it is only 2.7'.
Friction loss for 1.5" pipe (converted to vertical feet is) 2.3' for 10' of straight pipe.

the second drawing above, will show less friction loss-- entering through the run rather than the branch.

With the return manifold on top closed as in your drawings, terminating the loop through the branch of the tee will create enough back pressure to balance the flow through the outlets. This was Calfo's basic premise with this design.

HTH,

Jim

 

[spieg]

Jim,

Thanks, that does help! It sound like entering through the run of the tee is worth the extra distance and additional 2 turns to plumb it from the pump. Even with the extra length and bends, the overall loss should be lower.

Are there any major friction differences between a standard 90 and a long sweep 90?

Is stepping down the size of the first couple of outlets necessary or can they all be the same due to the backpressure created?

 

[uncleof6]

Jim,

Thanks, that does help! It sound like entering through the run of the tee is worth the extra distance and additional 2 turns to plumb it from the pump. Even with the extra length and bends, the overall loss should be lower.

Are there any major friction differences between a standard 90 and a long sweep 90?

Is stepping down the size of the first couple of outlets necessary or can they all be the same due to the backpressure created?

standard 90 hits a dead end and turns. A sweep keeps moving around the radius. The sweep would show less loss. The longer the radius, the lower the loss would reach towards straight pipe, but since it changes direction, it would remain above that of straight pipe. How significant it would be I leave to those that enjoy math more than I do.

The single greatest gain is accomplished with an increase in pipe size. Where 1.5" can flow 3600 gph with minimal loss, a 2" will handle 5400 gph with minimal loss. So flow 3600 through 2" will give quite an advantage.

I really don't think downsizing the first outlets is really all that necessary, with a looped system. With a dead ended line I can see the advantage.

Jim

 

[stugray]

Here's a pic of mine:

SInce I upgraded to the barracudda, I replaced those 90 degree turns with flex PVC.

Notice the Union below the TEE at the top.

This allows me to remove the manifold without breaking down the tank.
( although it IS a bit of a challenge to reach it from the front )

Stu

 

[spieg]

Thanks again Jim! I think ill plan on using long sweep 90s for my corners. I wish there was an easy way to just know what kind of output you end up with in a given situation. If there was enough difference between the 1.5" and 2" on the return side I might be inclined to try to find a way to hide 2" in my canopy. The 1.5" is already pretty obtrusive, though, and going to restrict the number of t5's I can make use of.

 

[spieg]

Thanks for the pic stugray. I just read through your VFD thread again and im almost thinking about trying it. We'll see. :thumbsup:

Heres pretty much what I have in mind for mine now:

I think I will go with about a dozen 3/4" outlets, and plan on capping half of them off. For curiosity's sake, would changing out the standard tee with a sanitary tee affect the dynamics of the flow within the loop? For the better or for the worse?

Im now leaning more toward the barracuda gold or else following stugray's lead. I might wait and see how Hookup's build goes and follow that as a guide.

 

[spieg]

Finally got to this stage of my build and put my closed loop together. Followed my design from above and ended up with this:

I capped off 4 ports and it flows "ok" through the 8 ports with locline. Ill probably end up capping off another port or two to get a little more velocity. The flow at all outlets is pretty even and it looks like its going to work pretty good! I went with the barracuda gold for now, with tentative plans on switching out the motor to a 3phase on a VFD for some more control down the road.