Manifold help

[Mihai]

Hi,

I'm trying to setup a manifold for my sump return (similar to the one
shown by Anthony Calfo here: http://www.wetwebmedia.com/pbh2oret.htm
or here http://www.advancedaquarist.com/issues/ ... /short.htm).

However, it seems that for a tank of my size (90 gal) I need a large
pump (that will be able to put about 1000 gal/h at 4' head). I notice
that all the pumps capable of doing that have 1'' fitting, presumably
because they need them.

On the other hand, the manifold cannot be made out of 1'' pipe (it
will simply not fit above the water level and/or would be very ugly if
you can see it from below); I already built a manifold of 1/2'' CPVC
that fits nicely in the aquarium.

You see my problem right there: 1'' return, but 1/2'' manifold.

To fix this problem, I'm thinking to split the return into two or
three 1'' pipes, and reduce them after that (see figure).


1. Is it a good idea to split the return? Any special plumbing
required or will a simple "cross" do it?

2. Should I get a wider pipe for the manifold (e.g., 3/4'' CPVC)?
(that implies that I have to re-build the manifold).

3. Should I use a smaller pump (100-200gph) as return from the sump (I
have a rather small sump - 15gal out of which 7 gal is meant as
algae scrubber and the rest is buffer for return) and use a closed
loop to get more circulation?

4. Should I get more/less outputs for the manifold (currently I have
the 6 ones shown in the figure, the tank is 4' long, 1.5' wide)?

5. How many gallons per hour do I need for the tank (again 90 gal, I
plan to get full reef eventually - in 1-2 years)?

6. Can you recommend any particular pump for the job (I know that it
should not be a Rio). I'm looking for a pump that is:
a) quiet (the tank is in the family room),
b) reliable,
c) reasonable priced,
d) that does not heat the aquarium too much (we have a pretty warm
house - 75F year round, I'll probably end up with 2 x 250W MH, and
I don't want to buy a chiller if I can help it)and
e) that fails in a way that does not take down the system (i.e., no oil
leaks or shocks all my fishes).
f) That are forgiving to running dry (if for watever reason my
sump runs dry)
I know that it's a lot to ask from a pump, but please send me your
favourites and their nice features (and the not-so-nice ones).

7. How can I connect the return to the manifold in a way that does not
leak and that can be undone for maintenance, cleaning, etc...
I heard about "quick connect" things, but I couldn't find any at
the local hardware store (Lowes); at least I couldn't find any made
out of plastic (metals are a no-no for marine systems, right?).

I pretty much read all the reefs.org on manifolds (search on
manifold), but I'm still confused...
Read the first post here if you want to see why
http://reefs.org/phpBB2/viewtopic.php?t ... t=manifold


Thanks a lot for the help,
Mihai

 

[taikonaut]

I will give general answer instead of address each of your issues.

A simple cross will do it, but 3 half inchers may not be a good idea.

Rule of thrumb is that 3/4 can handle twice as much flow as a 1/2 incher. An one incher can handle twice as much as a 3/4 incher's flow. Going from 1 inch to 1/2 inch you will need at least 4 of them unless you don't care about back pressure. Back pressure means more heat for the water, and you said you want to do without as much as possible.

For ease of maintenance, you may want multiple manifolds instead of one large one. Easier to take apart if one portion of it is clogged. Quick-connect is not reliable for large diameter setup unless you want to pay big bucks. Use PVC "union" instead.

 

[Len]

I'll copy this to the DIY forum for you Mihai.

 

[Mihai]

Len, thanks for placing it in the right forum, taikonaut thanks for the advice. You're right, what I want to minimize is backpressure, while maximizing flow.

Clogs... Hmm... That doesn't sound good. You're right, with that setup if it clogs I'm ****ed. Why would they clog anyway? I'll have a strainer at the pump...

Thanks,
Mihai

 

[Anonymous]

In answer to your questions.

1. Splitting the return in no big deal. The key is to kep the total of the brances cross section ar least as big ans the crossestion on the mail line. Bigger is better.

2. It would be better to be 3/4 but not necessary. Moving the center brance to the front would help to distrubute the flow through the headder. Make sute the sum of the exits cross section are approximatle equal to the main line.

3. I would have a return pump that turn the tank over about 10x an hour. You want it big enought to keep up with the skimmer also.

4. see answer to #2

5. softies and lps 10+ sps 20+

6. Iwiki, I think I spelled that right. Nice reliable external pump line.

7. from the pump put a union then a ball valve on both the inlet and outlet with the union being closer to the pump.

 

[Mihai]

I see, it makes sense, I'll start adding up stuff, and modify the central line to go in two places in front (as opposed as one in the back).

Thanks,
Mihai

 

[dow]

How about this, 1" tee-ing off to 3/4" with shutoff valves on each of the three legs, so that you could work on them individually.

 

[taikonaut]

Looks like we are getting there, aren't we?

 

[Mihai]

That looks really cool. It also increases the number of outputs such that I have both more flexibility and more flow. And everything will be 3/4'' from the cross up (as 3 x 3/4'' accomodates more than 1 x 1'').

Only one thing bothers me about this design though: Anthony Calfo said that it should be closed. I don't have strong engineering arguments for it, but I feel that if it's closed (i.e. no holes at the ends) it equalizes the presure better than a few open ones, i.e., it will have more equal flows among the outputs...

And once again, probably equal flows is not the really important thing here.

And talking about outputs, should I reduce the diameter to 1/2'' or less to increase the speed?

That was a great idea dow (Junior Member!??!), thanks a lot.


Less-than-junior member,
Mihai

 

[dow]

Only one thing bothers me about this design though: Anthony Calfo said that it should be closed. I don't have strong engineering arguments for it, but I feel that if it's closed (i.e. no holes at the ends) it equalizes the presure better than a few open ones, i.e., it will have more equal flows among the outputs...

Oops. Sorry, I should have given more information. Cap off the ends of the legs, or maybe add another nozzle there. Technically, the loop in Calfo's example isn't exactly "closed," since it has several openings for water to exit. A design with multiple risers coming from a single source and terminating at multiple nozzles should equalize nicely. You might want to put the same number of nozzles on each leg, however, just to make sure. As for reducing the diameter further to increase speed... that's a tough one. While the decreased diameter of the nozzle will make for greater speed, it will also increase back pressure. I'm not sure that it'll matter greatly, but it's something to consider. Also, by running your return water out through smaller outlets, you might be limiting the overall flow rate (gallons/hour) that you're putting through the system. Not a good thing with a reef tank if you slow it down too much.

Hmmm... Junior Member... I hadn't noticed. Been in the hobby for about two years. Just don't post much. Like many things, the interest waxes and wanes... This can make it hard on my tank during the low times, and hard on me (and my pocket book) during the high interest times, lol.

 

[Anonymous]

YEP add a cap or nozzel to the end or you will not get much pressure out of the nozzels. Water is like electricity - path of least resistance.

 

[Mihai]

Dow, you're quite a character
Thanks for the help. I think that I'll taper them down up to the point where they can still support all the flow that goes up through the original 1'' pipe. Thus I'll get some speed increase without a reduction in flow.

Thanks,
Mihai

 

[taikonaut]

I don't know if you really want the jet to shoot up like jacuzzi... it will generate lot of bubble and spray if you over do it. Besides, it is the flow rate that matters, not the speed... this is not a pressure washer or a water cutter, you know

Rather the setup is a loop "closed" does not matter that much. You can do a flow analysis on the different design (loop vs bar), but I doubt that you will see much different.

 

[Mihai]

Well, it will probably not bubble up anyway, as the nozzles will be under the water level anyway.

The idea with the speed is to create more water turbulence that will reach the great depths of my tank (90gal = 24''). If the water has no speed I will only get lots of action at the surface, but none or very little close to the bottom. I guess that I"ll need to experiment. Anyway, after the thing is constructed it will be quite easy to change the nozzles, so it's not a crucial decision at this point.

Regards,
Mihai

 

[taikonaut]

Maybe you want to bury the pipe in the sand instead..., but using water jet for your purpose is not a good idea. High velocity jet will make coral very uncomfortable, but it is fine for most good swimmers.

 

[dow]

Okay Mihai, obviously I have too much time on my hands this morning. That, and I'm using this as an excercise to work out the design for my next tank. Here's the same diagram, with the addition of a generic sump (add your own preferred skimmer, fuge, etc.) and a calfo style overflow with the shelf for you to display coral on or encrust polyps or whatever.

 

[taikonaut]

I will put one union with each of the ball valves of the manifold.

Ball valve is cheaper than gate valve, and it is what you need in this setup, not gate valve.

 

[dow]

Unions with the ball valves on the return legs would be a good addition.

You're right, ball valves are cheaper. From now on, "BALL" will be spelled "G-A-T-E." :lol:

Another addition that you could make, if you're an uncontrollable gearhead and just CAN'T leave well enough alone, is to replace the ball valves on the three return legs wiith electrical valves and then set them up on a timer so that you could simulate wave motion as well as increase turbulance by alternating which ones were on and which were off.

Once you start down that slippery slope, it just never ends, does it? :mrgreen:

 

[Mihai]

I see that you really have time

A few comments are in order:

1) I already have an overflow drilled at the bottom of the tank (right
in the middle, if you look in the original post, that's why the
manifold does strange things in the middle. Hence I don't
need/want/coudn't/etc want Calfo's overflow, right?

2) I saw that type of overflow, it's cool, but I think Calfo uses it
as an algae scrubber, not for coral shelf. As a coral shelf it will
be very shallow, and we'll only be able to get corals that
are <1'' high and love lots of laminar water flow.

3) I've been thinking on my way to work, and probably the cross should
be as high up as possible (to minimize total pipe length = loss of
flow to friction in pipes), probably close to the upper edge of the tank.

4) Taiko, I agree that it is not good to go to extremes, but I'm
worried that if I use 14 outputs like shown in the picture (I'll
probably go to 50 to have the flow balanced 5+5+5), there will be very
little speed (probably too little). All the flow from the 1'' pipe
(assuming that I'll end up with a pump with 1'' pipe, although I saw a
few with 1.5'' pipes) fits through 2 of those 3/4'' outputs, when you
divide this by 15, the speed will be very little. Regarding the corals
liking it, I'll probably not put them to flow into the corals, but
rather as Calfo suggested to intersect each other in the tank to
create turbulence.

5) Regarding the electrical valves that is SO COOL!!!! The coolest part
is that it can be upgraded at a later time. Don't get me started on
that now. I've been thinking that in a few years I'll build my own
system that will monitor and control the aquarium, based on an old PC
and a bunch of sensor and actuators. The main thing I was thinking
about was to prevent "accidents". The idea is to have checks at all
relevant points in the aquarium and the sump (I assume that there are
sensors that can tell you if the sensor is submerged or not), and have
the pumps under control (plugged on a strip controllable from the
computer). As soon as something goes wrong (to much water here, or to
little there) it will stop the pump and beep/email/page, etc. Now
there is something else for that dream controller to do: waves!

But I guess that you're right: the slope is very looong and slippery

6) Regarding the gate vs ball, it will certainly be ball: I'm having a
hard time finding all-plastic gate valves at the local hardware store.
And yes, That was the whole idea with three separate manifolds: to be
able to clean them independently, so there will be three union valves
after each gate (not before, like in the picture). This way
I'll be able to run two of them while cleaning the third... Am I
missing anything? If the unions are before the gate they'll not drip
as much when I take them off, right? But they should be easy to empty
if I stop the pump... OK, after the gates.

7) Any advice on the return pump - brand/place to purchase (see list of requirements in original post)?

8) Finally, why two different outputs for the Calfo's overflow?


All the best,
Mihai

 

[taikonaut]

The point of using a manifold with multiple outlets is to spread the flow over a wider area, instead of one output. If you want speed, you will have the overflow in one end of the tank, and the return on the opposite end. By using small nozzles, you defeat one of the main advantages of manifold, but I guess it is just a compromise, and there is nothing wrong with that.

I don't want to beat a dead horse, but I just want you to be aware that back pressure, decrease flow rate, and irration to livestock may make this otherwise great project less than optimal. But after all, this is your call. Just my 2 cents worth.