Manifold design

[evilchild]

I've been reading a lot of threads of manifold design but it is still confusing to me. Here is my problem:

I've a 4 inch inlet pipe which will bring in about 300-500 gpm of water from a pump. I need to design a manifold of about 7 ft length. We are going to drill holes into the 7 ft pipe to create orifices. I have to figure out the spacing and size of these orifices to get uniform flow rate from all the orifices. The pipe thickness would be 4 inches and the manifold will be used in vertical direction. Any help is greatly appreciate. Thank you

 

[Latexman]

Please provide a well labelled sketch. A 4" inlet pipe. A 7' manifold. 4" thick pipe walls. I'm confused!

Good luck,
Latexman

Need help writing a question or understanding a reply? forum1529

 

[evilchild]

Length of the manifold: 6-7 ft
Diameter of the pipe: 4 inches
Inlet flowrate: 300-500 gpm
Output flowrate: Uniform flow through all the orifices

The idea is that we have pump that is going to pump water into a tank. We want to design a manifold so that the water is distributed uniformly along the depth into the tank. The tank depth is about 7 ft. Hope this makes sense?

 

[LittleInch]

Still very odd, but I would just drill equal size holes at equal intervals on the basis that the pipe comes in from the bottom and goes up. Why your pipe is so thick doesn't make sense, but a drawing or Fuller explanation always helps.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[evilchild]

@LittleInch: Just to clarify that the pipe goes into the tank from the top. Would equal size holes at equal intervals give uniform flow rate at all orifices? My fluid mechanics is a little rusty. I will try to come up with some schematic diagram if it helps. The pipe is thick because we want to reduce the head loss as much as possible. As you can see the flow rates are very high.

 

[evilchild]

Quickly made this in powerpoint. Hopefully this will help you understand the problem better. Thanks!

 

[zdas04]

First, flow is never uniform no matter what you do. A microbar difference in resistance can completely change the flow. If you are looking for "kind of" uniform then it is possible, but not in your configuration. The pressure at the lowest hole is significantly higher than the pressure in the highest hole due to hydrostatic gradient. You could try to fake it by making the upper holes slightly larger, but that would give you some pretty unstable fluid dynamics.

If the manifold just has to be vertical, then I would enter it from the bottom and gradate the hole size (so that the first hole the water sees is the smallest hole at the highest pressure, and so forth until the top is the biggest hole at the lowest pressure).

You could do equal hole sizes if you could turn your manifold horizontal, but you would still get more flow from the hole(s) closest to the source.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
The plural of anecdote is not "data"

 

[cvg]

and to add on to zdas comment, if the fluid level in the tank changes, than all the work you did calculating the flow and headloss to each orifice will be wrong and need to be re-calculated with a different discharge pressure...

best to make the manifold pipe larger (than 4 inches) and make the orifices larger and then the headloss in the manifold will be reduced to negligble.

 

[tkall]

What a fun little problem. To simplify the process of explaining I would first assume no water inthe tank to remove the differential pressure on the orifices. That can be dealt with later.
1) select an orifice size at the bottom of the tank.
2) select an arbitrary pressure on this orifice.
3) calculate the flow from the orifice
4) calculate the pressure loss in the segment of pipe between the bottom orifice and the next orifice.
5) the pressure the next orifice "sees" is the sum of the pressure loss in the segment of pipe and pressure selected in step 2.
6) repeat 3-6
Do this for several values of pressure from step 2 and draw a system curve.

 

[evilchild]

Thank you all for your suggestions.

@tkall: I was doing something similar and I thought there must be something much more easier. Also, it should be assumed that the tank is full of water. The height of water will probably up to the top of the tank at all times.

@zdas04: Kind of uniform is acceptable. We want to reduce unwanted eddies due to the difference in velocities when submerged in water. Can I vary the spacing instead of the diameter? It would be easier to use one size drill hole.

 

[Latexman]

My company design manual specifically recommends to NOT use distributors for tank inlets because there is usually no need for good distribution in a tank. Why does this tank require uniformity? The eddies will mix the tank. Why not let them?

Have you looked into "jet mixing"? That will be easier, IMO.

Use Search (undet the thread title, between Forum and FAQs) on distributor or maldistribution and you will find some useful threads.



Good luck,
Latexman

Need help writing a question or understanding a reply? forum1529

 

[evilchild]

@latexman: these are the requirements of the project. I'm not in a position to make changes to these requirements, unfortunately.

Other option is: can I move the inlet into the center for a better pressure distribution in the manifold? I guess this means more bends and more minor losses.

 

[Artisi]

pumping into the centre of the manifold is a good start in equaling the flow distribution, change to 6" diameter pipe, this will reduce the friction loss between orifices to next to nothing you then only have to consider the differential pressure at the orifice.
However, can't see how this is such a critical problem, the flow rate isn't even nailed down as yet, 300-500 gpm is quite a range and the pipe is about 7ft, staying with this same near enough spec. "X" number of drilled orifices should give approx the same flow rate.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[LittleInch]

Have you though about a short stubby manifold with identical lengths of flexible tubing / hose with the first one coiled up but all attached to a long bar so they are all pointing horizontally. If the length of the manifold is short compared to the length of the small diam of the tubes (7 foot each), then flow should be even between each hoops as the extra head for the lower one will be balanced by the head of water in the tank.

For a variable flow a fixed orifice system will provide variable distribution but for a equal resistance system it should stay more or less the same.

Go for a diam and number of hoses which gives a similar total internal area to your incoming pipe.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[Artisi]

simple solution is go to your local plumbing supplier and ask to see the water restrictors use in bath room shower heads. These are a rubber diaphragm which restricts flow to a set volume by deforming the orifice based on the pressure. cheap, easy to fit and have a fairly tight flow range irrespective of the driving pressure.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[Latexman]

The design guidelines I've seen for drill hole distributors uses the area ratio (total perforation or orifice area to pipe cross sectional area) and distributor length to diameter ratio (L/D) versus the % maldistribution. % maldistribution = 100 - % distribution. Maldistribution is affected strongly by area ratio (Ar) and more weakly by distributor L/D. Maldistribution of < 5% is present for Ar < 0.5, no matter the L/D, so this gives a simple design guideline with wide application. Keep Ar < 0.5 and the distribution will be acceptable (95-100% uniformity).

Good luck,
Latexman

Need help writing a question or understanding a reply? forum1529

 

[BigInch]

Drill equal holes at equal spacing. Outflow from each will be almost the same. Considering static pressure increase as velocity slows as fluid is removed at one outlet and less and less fluid moves to the next outlet, static pressures actually increase in the downstream direction within the manifold, except for those extremely long manifolds where the distances between outlets are so far apart the friction from flow between outlet to outlet begins to amount to something, or fluid is extremely viscous and friction loss/unit length is very high. Frictional resistance between outlets must be much more than the pressure increase due to slowing flow in order to get any difference in outflow between similar outlets. It's not going to happen in a short 7 ft long manifold.

What controls the flow from a short manifold is the resistance to flow through each outlet, which is primarily controlled by the square root of the differential pressure across each outlet, Q = K dH^2. The square root of the difference between all manifold outlet differential pressures in a 7 foot length, even if vertical and immersed in water (3 psi) will not make a hill of beans difference to the flow across each orifice, as long as the pressure inside the manifold is somewhat high relative to the 3 psi, all flows will be nearly the same, high being only 20 psi. That should produce outflows that vary by no more than +/- 15%. A 10 psig manifold pressure would have around a +/- 25% rate of equality. at 50 psig, +/- 5%

Independent events are seldomly independent.

 

[davefitz]

One needs to know how much pressure drop is available for flow distribution. Also, how "equal " is neccesary- will +/- 10% be acceptable?

If you size the orifices for an orifice pressure drop equal to approx 3 times the manifold's inlet velocity head, then you would likely meet the +/- 10% variation at the design flowrate. Alternately, if the total flowrate is relatively constant , then you can drill each of the 4 orifices differently , to account for the varying velocity head in the manifold as the gross fluid flowrate decreasses from 4 to 3 to 2 to 1 downstream flowpaths. Or, you can reduce the manifold diameter followign each orifice so that the velocity head upstream of each orifice is a constant value.

"Nobody expects the Spanish Inquisition! "

 

[BigInch]

Adjustable nozzles would be so much easier, if you need that kind of fine tuning.

Independent events are seldomly independent.

 

[katmar]

Am I missing the point entirely here? This seems like a trivially easy problem. The fact that the manifold is vertical should not make any difference because as you go deeper down the manifold you are also going deeper in the tank, and the net effect of static pressure on the pressure drop across each orifice is zero.

If the manifold is designed for a velocity of 6 ft/s at a flow of 500 gpm then you get a 6" pipe and the friction pressure drop over 7 ft would be less than 0.1 psi. If the pump can deliver (say) 15 psi at the top of the manifold then the friction loss is negligible and the static pressure changes offset each other. The result is that an orifice at the top of the manifold "sees" exactly the same available pressure drop as the bottom orifice. Even if you continue the 4" pipe as the manifold the friction drop will be less than 0.4 psi and is probably still negligible.

You can safely space the orifices evenly over the length of the manifold. If they are spaced every 4 inches you will have about 20 orifices. Size each orifice for 25 gpm (= 500/20) at a pressure drop of 15 psi and you will have a simple manifold with 20 identical orifices that gives a very close to uniform flow over its length.

Katmar Software - AioFlo Pipe Hydraulics

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"