Closed loops tricks

[Dark]

Have tips or trick to avoid having micro bubbles with a system closed loop? Many people telled me this gernre problem with the closed loop.

And also if you cant tell me somethig about your experiments with the closed loops, they are welcome…

What would you make if you should remake your closed loop?
Also, I am better to take water by my overflow drilled in my tank, or to make an independent intake in plumbing?

I've planned using an ocean Motion Super Squirt as flow divider...

Thank you in advance...

Dark

[Icehockey_whino]

I just got my CL going today. The only thing I would change is when measuring and cementing together all the pieces... I would make myself remember that the tank has a cross brace and it is almost physically impossible to actually get it into the tank after it has been assembled. I tweaked and angered a few of my corals doing it, but finally did it.

Micro bubbles were a problem at first but i think they were just caught up in my ballvalve apparatus. Once I tweaked it, evrything was fine.

With all the research that I did on CLs, I would recommend doing an external pumped CL. That way you don't mess with your sump or need to worry abou how fast of a return your overflow has. With external what comes out ...goes in and what goes in ... comes out...etc...

[Jasen Hicks]

I think the issue you are having is not a big enough intake to the pump in your closed loop. If you have too little water getting to your pump it will cavitate, strip the gas out of the water, and blow it back into your tank. (like you have now)

The way to avoid this is to 1. Put a big suction line to the pump (the goes in part) and then put a valve on the outlet (the goes out) that way you can match your input and output using the valve.

Ideally you want the input to be greater than the output so you dont have to worry... most of the time though it can be difficult unless

1. you drill the bottom portion of the tank (back side) so you have the entire column of water FORCING water down the pipe to the pump.

2. Or you can put a 2" U-Tube into it and that would draw enough through the siphon but it would be ugly!

Its all about the net positive suction head!

[Dark]

Do you think that a "Y" at the intake with two 1" "U" pipes intakes will help the pump to not cavitate and create micro bubles?

Dark

[Dark]

Any other issues taht I'd better think in the desing of my Closed-Loop?

Is that necessary to have a dismontable union on each outlet of the Super Squirt, for further dismounting?

Are you using any filter/strainer at the intake of the pump, other thant a strainer in the tank, to filter debris?

Dark

[Jasen Hicks]

Any place you want to keep somthing accessible, i.e. the squirt id put unions. that way if it ever clogs or gets damaged you can easily remove it and not replace a lot of PVC.. the parts add up.

I dont filter my intake to my CL pump at all. I use a GEN-X 55HP for mine and as long as you keep your inlets screened to keep large stuff out the small stuff **shouldnt** hurt your pump.

If you do 2 pipes coming to the inlet join them into a pipe that is equal to thier diameters added together... i.e. 2 1" pipes into a 2" pipe. A little time in the pvc section at HD will help put it together, just be creative with reducers, fittings, etc.

When you do join them, make sure you get at least 6" of straight pipe to your pump inlet to help reduce turbulence at the impellar. And, ALWAYS put a valve before and after the pump so you can isolate the pump and replace the impellar if needed. I didnt do it on my stock tank and had to drain all 100 gallons to fix a pump. Wasted $70 in salt that endevor. But just make sure the valve on the inlet side is always 100% open, and yes it can count in the 6" of straight pipe since it will be wide open.

Jasen

[Dark]

Thanks Jasen! I've never thinked about that straight section of 6" before the inlet of the pump...

Does the turbulence is responsible of micro bubles sometimes?

Dark

[Jasen Hicks]

It can, think of the white caps on a wave as it crashes, same idea but smaller scale. The white is the foam or gasses being released from the turbulence.

Jasen

[BrainBandAid]

I would recommend looking at all your connections for leaks before you seriously worry about cavitation. From what I've read, the majority of bubble problems are from pinhole leaks or un-tight threads on the suction side. Especially on closed loops.

If the problem is suction cavitation, then the bubbles are usually reabsorbed back in the discharge side, due to the higher pressure there. You would throttle shut the discharge valve for suction cavitation, or increase the suction pipe diameter.

If it is discharge cavitation, then the pump is running too fast or building too much pressure in the volute, due to too high discharge head pressure. So increasing the size of the discharge pipe would be a better fix. You would have very low flow with discharge cavitation anyway.

And if you really have cavitation, you should be able to hear it very well. Sounds like little pebbles being bounced around in the volute. Or like knuckles cracking.

And if it's turbulence, wouldn't the water have to mix with air at some point? It shouldn't if it's really a closed loop.

Also, I read a long post a while back about a guy who's chemistry was way out of whack and it caused his water to be saturated with some kind of gas, like oxygen or something. Poor guy went crazy trying to get rid of the bubbles. Ended up being that some meter or test kit he had messed up his alkalinity or something...
I'll have to try and find that one.

Hello Jasen...
(Ex) MT2 from the Maryland Gold :wavehand:

[NoCarrier]

<a href=showthread.php?s=&postid=8490634#post8490634 target=_blank>Originally posted</a> by Dark

Also, I am better to take water by my overflow drilled in my tank, or to make an independent intake in plumbing?

Couldn't this be a problem? An overflow intake seems like a likely source for extra air entering the plumbing.

[Jasen Hicks]

He does need to use an independent intake.

As for the bubbles.... Your aquarium has a lot of gasses in it including oxygen, nitrogen, etc. If it didnt then the fish would die. The whole system is open to the atmosphere so there is a constant gas exchange. I'm not sure the exact amount of air, but its a lot more than most people give credit for!

Jasen

[samtheman]

"strip the gas out of the water"
"building too much pressure in the volute, due to too high discharge head pressure."

Neither of these describe cavitation. Cavitation is where the pump does not have enough flow into the intake. The pump puts the water into a low pressure area where some of it vaporizes, causing bubbles.

[Dark]

<a href=showthread.php?s=&postid=8501703#post8501703 target=_blank>Originally posted</a> by NoCarrier
Couldn't this be a problem? An overflow intake seems like a likely source for extra air entering the plumbing.

Sorry, my english is'nt very good...

I was talking about my "hole" for my actual overflow ( I have two in my 125G for overflows). Just wanted tu use one of them for plumbing intake of the losed loop, but I can easily make an intake for the CL without using it.

Dark

[newnano]

cavitation doesnt create microbubbles necessarily. you can get cavi without bubbles. so most likely you have air getting in somewhere. also, anytime you set up a new system it could be a cpl weeks before micros disappear and you get "slime" all over everything.

if your pump is cavitating then look at your plumbing. are you restricting the pump in some way? i did this once with a 90 degree elbow too close to the intake of the pump

[Dark]

We can easily stop cavitation with a ball valve on the output of the pump... To reduce flow?

What are your intake look like? A PVC pipe caped with many small drilled holes? Or normal cone stainers?

Dark

[ReefJerk]

<a href=showthread.php?s=&postid=8493401#post8493401 target=_blank>Originally posted</a> by Jasen Hicks
...If you do 2 pipes coming to the inlet join them into a pipe that is equal to thier diameters added together... i.e. 2 1" pipes into a 2" pipe... When you do join them, make sure you get at least 6" of straight pipe to your pump inlet to help reduce turbulence at the impellar.
Jasen

Would the two 1" pipes then be sufficient to feed a 2" inlet? I have a dilemma where I need to feed a 3" inlet with (2) 1 1/2" U pipes over the back of the tank. The manufacturer highly recommends against reducing the inlet. With these 2 feeds on the 1 1/2" inlet of a Gen-X 18000(4750gph), I could hardly feel any drastic suction from either side with my hand. I just want to make sure I don't cavitate, or kill, the pump as soon as I power it on.

Cliff

[Jasen Hicks]

If the inlet of the pump is 2" and you put about a 6" run or more of 2" pipe in front of the impellar you could get a 1" - 2" elbow and a 2"x1"x" T to tie it all together it should be fine.

somthing like this simple drawing:

| |
L ===T === PUMP
(keep the L and T as close together as possible)
(T to PUMP should be AT A MINIMUM 6")

| is 1" pipe
=== is 2" pipe

[Jasen Hicks]

Sorry, that TXT drawing sucks.. heres one i did PS:

http://reefcentral.com/gallery/data/500/57147CL.jpg

[samtheman]

<a href=showthread.php?s=&postid=8502436#post8502436 target=_blank>Originally posted</a> by newnano
cavitation doesnt create microbubbles necessarily. you can get cavi without bubbles. so most likely you have air getting in somewhere. also, anytime you set up a new system it could be a cpl weeks before micros disappear and you get "slime" all over everything.

if your pump is cavitating then look at your plumbing. are you restricting the pump in some way? i did this once with a 90 degree elbow too close to the intake of the pump

"Cavitation implies cavities or holes in the fluid we are pumping. These holes can also be described as bubbles, so cavitation is really about the formation of bubbles and their collapse. Bubbles form when ever liquid boils. Be careful not to associate boiling with hot to the touch. Liquid oxygen will boil and no one would ever call that hot."

Cavitation is the formation of bubbles as water vaporises. You can't have cavitation without bubbles. That is the defination.

[Jasen Hicks]

Ok, i agree it may or may not be cavitation. Let me explain a little bit better.

Pumps require NET POSITIVE SUCTION HEAD (abbreviated NPSH) at the inlet to function properly and not get damaged, cavitate or cause too much turbulence at the impellar. All NPSH is... amount of water weight at the inlet of the pumpthe VOLUME of water at the inlet of the pump (pressure into the inlet, because volume is proportionate to pressure)... . Put a pump at the bottom of a 55 gallon drum full it has a LOT of NPSH, put it very near the surface of the 55 gallon drum it has very little.

Equate that to the pipes feeding your pump... a 3/4" inlet line is a lot less than a 1" inlet line, or a 2" inlet line.

When you do not supply the proper amout of NPSH to the pump it will bring gasses out at the impellar. Why? A pump that is used in an aquarium is a centrifugal pump, it spins creating a low pressure part at the impellar and creating a high pressure part at the outlet or volute of the pump. Simple science, gasses stay in solution at high pressure and come out at low pressure. The impellar which drops pressure will bring this gas out of solution and send it to the tank as microbubbles if there is not enough NPSH.

[Jasen Hicks]

Please dont think of my post as showing off knowledge and trying to show how smart i am, I deal with this stuff every day as I am a Nuclear Engineering Officer on a Submarine, i have been specifically trained in all this crap and just want to share info, thats all!

Jasen

[Gooli]

i've done 2 closed loops so far, one on my 90 and one on my 180.
it's pretty basic really, you just want to make sure you have joints glued in as many spots as possible and minimize threaded connections. Make sure you got lots of teflon on those threads to avoid micro bubbles - that's it.

i know you might say to yourself, "i like threads better because i can take it apart and i can't do that with glued connections" - but if you plan in advance and do it well, u shouldnt need to take it apart.

hope this helps

[kgross]

I have to correct some errors here. First 2 one inch pipes does not equal the flow of 1 two inch pipe. The cross sectional area of the pipes is what is important not the diameter. Area is pi r^2, so with a 1 inch pipe the area is .785, so 2 of them gives you just a little over 1.5 square inches. A 2 inch pipe has 3.14 square inches of area, so to get the same flow through the 2 inch pipe you will need 4 1 inch pipes

Jason I am glad that you are a nuclear engineer, and not a plumber because you don't understand simple fluid dynamics and how friction loss in pipes work.

kim

[kgross]

To help with the original question. Normally with a closed loop you get micro bubbles from 3 places. You can have cavitation, which is pretty rare if you do your plumbing well, you can have leaks in the plumbing which is pretty common, or you can suck air bubbles into the closed loop, which is also pretty common.

To get max flow out of your pump make sure that all of your intake plumbing is at least one pipe size larger than the intake to the pump to reduce friction loss, if you have very little friction loss on the intake and your pump is mounted below the tank, you will not have cavitation in the pump. So if your pump has a 1 inch inlet, make sure your intake plumbing is at least 1 1/4, with 1 1/2 being better if you split it out to multiple inlets calculate the area of each inlet and make sure that they are larger (because the T's and Y's will create friction) then the main inlet pipe.

Do the same on the pump outlet, then if you need more velocity out of the outlets you can neck them down right at the outlet to increase the velocity without restricting the flow to much so the water shoots out of the outlet faster in a smaller area.

Kim

[Jasen Hicks]

Kim, thats pretty rude. I was just stating that it would work and be better than a single 1" pipe. not that the flow would be the same as a 2" pipe, it indeed would be less than a single 2" pipe...

The intent was to keep it so that there was no restriction from the 2 1" inlets to the 2" pipe.

Before you make mean and distateful remarks maybe you should read the actual words and not the in between things that arent there.

Jasen

[Jasen Hicks]

<a href=showthread.php?s=&postid=8504475#post8504475 target=_blank>Originally posted</a> by gooliver
i've done 2 closed loops so far, one on my 90 and one on my 180.
it's pretty basic really, you just want to make sure you have joints glued in as many spots as possible and minimize threaded connections. Make sure you got lots of teflon on those threads to avoid micro bubbles - that's it.

i know you might say to yourself, "i like threads better because i can take it apart and i can't do that with glued connections" - but if you plan in advance and do it well, u shouldnt need to take it apart.

hope this helps

I agree with this whole heartedly! Unions are a aquariust best freind, and can be glued onto regular pipes. The treads are ok, and useful in some spots but the teflon tape better be good! :D

On a seperate but relevant side note, a freind plubmed a 65 gallon tank up with regular slip fittings, and just put them in hand tight with no glue!

It worked for about 2 days.. fortunately he was putting in some rock when it failed and he avoided 100 gallons of total water on the floor! (It was only 5 in the end!)

[kgross]

Jason

The 2 1" inlets to the 2" pipe will restrict the flow no matter how you connect the plumbing together. You said that your way would not restrict the flow, which it will, and it could cause cavatation depending on the flow rate and static head pressure on the pump.

2 1 inch pipes will create twice as much friction loss as the 1 2 inch line and as you said cavatiion is determined by the net head pressure on the pump inlet.

If you want to split to 2 pipes from the 2 inch pipe you will need at least 3.14 square inches of area in the 2 pipes. so you need 1.6 square inches per pipe so take pi r^2=1.6 1.6/3.14 =r^2=.509 .509=r^2 r=0.7138, so the pipe would need to be at least 1.427 inches to not create more friction loss than the 2 inch pipe.

Now of course the fittings to connect the 1.5 inch pipes together and to the 2 inch pipe will create extra friction loss so it is still not a good idea.

For the poster with the 3 inch inlet pipe with 2 1.5 inch pipes. The area of the 3 inch pipe is 7.065 square inches, the 2 1.5 inch pipes are 3.53 square inches, so to get the same flow rate you will need at least 4 1.5 inch U pipes.

Just by checking the cross sectional area of the pipes you want to use you can get a very good idea of how well your idea will work, and the math to do this is very simple pi r squared. Also remeber that going up in one pipe size comes very close to doubling the cross sectional area, and the higher the cross sectional area, the lower the friction loss so the higher the flow rate. So if you want extra flow out of your pump, increase the size of the plumbing on the pump by at least 1 pipe size and if possible 2 pipe sizes to reduce the frictional loss to basicly zero so even if you have to put extra elbows or other fittings on the pipes it will not make a noticible difference in flow rate.

Kim

[Dark]

How big is you intake plumbing when using near 3000 GPH pumps? Becaus 2" pipe began to be really big... And not really cute in a reef. (What about 3"!!!)

My CL pump will be 4 feet under the tank in the fishroom, so there will be a head pressure on the inlet, but at the output too, so are they eliminates each other? (like at zero)

I'm begining to have a nice plan of my CL in my head... Thanks for your advice!

Dark

[kgross]

I have 2 2" pipes going into my tank to feed both of my closed loop pumps, yes they are large, but with my tank it is setup as a penisula so the end against the wall that has the overflow in it (calfo style on the end) has the 2 intakes going through the overflow into the tank. Doesn't look to bad is well hidden behind the rocks.

Yes the static head pressure on your pump will be at zero because you have pressure on the intake that is the same as the output, so the only flow loss you will have from the pump is frictional head losses, which are determined by the velocity of the water in the pipe and the slickness of the pipe. If you decrease the velocity, with a larger pipe you will reduce the friction loss, and so get more flow.

Kim

[ReefJerk]

So, for a 3" inlet, I need (4) 1.5" feeds, correct?

[ReefJerk]

And, by the way, there is no simple way for me to do (1) 3" feed. My hood was customized for the 1.5 and all the plumbing has been done.

[BrainBandAid]

<a href=showthread.php?s=&postid=8502266#post8502266 target=_blank>Originally posted</a> by samtheman
"strip the gas out of the water"
"building too much pressure in the volute, due to too high discharge head pressure."

Neither of these describe cavitation. Cavitation is where the pump does not have enough flow into the intake. The pump puts the water into a low pressure area where some of it vaporizes, causing bubbles.

I hate to get off topic, but I don't think you really understand cavitation. There are two kinds of cavitation, suction and discharge.

In suction cavitation, the suction, or vacuum, is so high that it causes the pressure around the eye of the impeller, which is where the water is sucked into the pump, to be low enough to "boil" the water, letting the water change to a gas. This bubble can cause damage to the impeller, usually around the inside of the impeller vanes close to the eye. It also sounds like rocks being chewed up in the pump. You probably won't see any bubbles with suction cavitation because as soon as the water moves to the higher pressure of the discharge side, the bubbles reform into water. That's where the noise comes from. Suction cavitation is fixed by reducing your suction head, most commonly by increasing the diameter of the suction piping. Throttling the discharge valve will increase the discharge pressure, and slow the pump down, decreasing the amount of suction. But you sacrifice flow.

Discharge cavitation occurs when the discharge head pressure is so high that very little water leaves the volute, so the water just spins in it. The very tip of the impeller vane creates a low pressure area behind it when it spins, much like a submarine propeller. When the impeller tip crosses the cutwater, which is right near the discharge tube, the pressure drops enough to cause cavitation. This can be fixed by increasing the diameter of the discharge piping, or some other method of reducing the discharge head.

Remember, there are two different sides to a pump. Suction and discharge, vacuum and pressure.

If you don't understand something, just try a quick google and learn.

http://en.wikipedia.org/wiki/Cavitation
http://www.pumpworld.com/troubleS2.htm
http://www.mcnallyinstitute.com/01-html/1-3.html
And my bread and butter, sewage pumps. (http://www.grpump.com/education~action~display~id~17~type~articles.asp)

The important part:
All these articles will say the same thing: when the pressure increases again, like when the water moves to a higher pressure area of the pump, the cavities or bubbles will implode or collapse. That's what causes the noise and damage. This doesn't mean there has to be "high" pressure, just high-er.
Like the difference between the inside of your pump and your aquarium.
What I'm getting at is that the chances of seeing bubbles in your aquarium that are caused by cavitation are almost impossible. It's like trying to boil water in a pot on the stove at room temp. Not gonna happen.

And throttling the suction valve can cause cavitation, too. It's kinda like a venturi.. the restriction causes low and high pressure areas after and before the restriction, respectively. This cavitation can eat away the valve pretty quickly.

[BrainBandAid]

And then onto the pipe diameter discussion:
The formula to calculate friction loss in pipe is:

hf = kf [L/D] Vsquared / 2g

where kf is a friction factor (PVC pipe, water only, at 80 degrees F) V is the water velocity in the pipe, and g is the acceleration of gravity (32.174 ft/sec squared). L and D are length and diameter. Since the flowrate is the product of the cross-sectional area of the pipe and the velocity, you can transform the equation to read:

hf (friction head) = kf [L/D] b Gsquared/Dfourth

where G is the flowrate in gph, L is feet, D is inches, and b = 7.2035 x 10 to -7th. (note: the L/D part must be in consistent units for the final equation)

The kf given for 2" pvc at 2000 gph is 0.0175. For 1" it's about 0.0147. So given the arbitrary length of, say, 8 feet of pipe (not counting fittings such as elbows which would be constant between the two anyway), the difference in friction losses would be:

8 feet of 1" pipe at 2000gph: 4.0678 psi
8 feet of 2" pipe at 2000gph: .3027 psi

difference: 3.7651psi
or, (1ft of head = 0.4321psi) 8.71 feet of head.

My brain really hurts now, and I can't remember if there was a reason I did all that. And yes, the math is from a book. (Not too shabby for an MT2)
But it shows there is a big difference between 1 and 2 inch pipe. However, that's 8 feet. The difference between 3 feet of 1 1/2 and 2 might not be worth the effort. Or, the difference between 2 and 3 inch pipe may not be worth it.
And I don't think the cross sectional area conversions you guys are doing will work with this formula. Maybe it will, I have no idea.

I would suggest you pipe your system with as big a pipe you can manage, figure out all your friction and other head losses, and choose a pump that will meet your desired flow requirements and be within the 40-60% range of the pumps head vs. flow chart. Then you can't fail. So if your TDH is 10 feet, and you want about 1500gph, find a pump with that curve.

What's easier, trying to design your system around a pump, or finding the right pump for your system?

[Dark]

Ouch... That's kinda really hard for my brain a this time of the day... But it help me unerstand a lot more about fluids.

I'm looking for two smaller pumps instead of one big. I don't really want 2" pipes in my tank, and 2 pumps allow more versability and security in case of a shut down/maintenance.

Dark

[kgross]

<a href=showthread.php?s=&postid=8505977#post8505977 target=_blank>Originally posted</a> by ReefJerk
So, for a 3" inlet, I need (4) 1.5" feeds, correct?

If you do not want to cause a a large restriction yes you will need 4 1.5 pipes to be able to carry the same flow as the 3 inch pipe. But it just depends on what flow rate you want out of the system.

Kim

[kgross]

<a href=showthread.php?s=&postid=8507482#post8507482 target=_blank>Originally posted</a> by BrainBandAid
And then onto the pipe diameter discussion:
T
And I don't think the cross sectional area conversions you guys are doing will work with this formula. Maybe it will, I have no idea.


I don't remember the exact equations but your formula does sound correct. The cross sectional area does work very well to compare flow rates from one pipe to another With the change in cross sectional area the velocity changes at the same rate since the velocity is equal to the flow rate divided by the area it is flowing through, so when you double the area the velocity is halved also.

Kim

[MayoBoy]

<a href=showthread.php?s=&postid=8505843#post8505843 target=_blank>Originally posted</a> by Dark
How big is you intake plumbing when using near 3000 GPH pumps? Becaus 2" pipe began to be really big... And not really cute in a reef. (What about 3"!!!)

My CL pump will be 4 feet under the tank in the fishroom, so there will be a head pressure on the inlet, but at the output too, so are they eliminates each other? (like at zero)

I'm begining to have a nice plan of my CL in my head... Thanks for your advice!

Dark


3000GPH = 50GPM so you can use 1-1/2" suction. You'd lose 13.5 feet of head over 100' (to friction) so 4' shouldn't be a problem. 2" would be better but 1-1/2" should work.

[Dark]

I think i should follow the intake diameter for intake pipes... It's a good clue!

By the way, Nice to know that 1 1/2" will do the job, I was afraid of that big piece of 2" pipe in my reef... Will look like a Manhole! :-s
Dark

[MayoBoy]

<a href=showthread.php?s=&postid=8512705#post8512705 target=_blank>Originally posted</a> by Dark
I think i should follow the intake diameter for intake pipes... It's a good clue!

Dark

Depends on the pump. A big 20 HP pump might only have a 5 or 6" flange on it while requiring 10 or 12" pipe. It just depends upon the application.