Manifold outlet spacing

[Weaverofduart]

Ive read the threads where people have been talking about pressure loss across manifolds and ive got "Perry,s chemical engineers handbook" for all the formulas. In learning how to do the pressure loss calcs ive got the impression that the spacing of the outlets from a manifold effects the pressure loss. Perrys talks about distances between fittings effecting pressure loss and gives formulas for calculating it but dosn't do so for outlet spacing.

How do I do this or do I just need to ensure that the spacing is even? I am pretty sure that it is something to do with ratio of outlet diameter to manifold diameter but thats all ive got.

 

[ione]

As the distance between outlets increases, pressure drop due to friction increases as well. So in order to have an even flow distribution you have to minimize the pressure drop along the manifold length.

You can find a valid example of manifold (header) sizing at

Pag. 226-228 “Flow in a manifold” (example 7.3) Chapter 7 from “Chemical engineering Fluid Mechanics, second edition revised and expanded” by Ron Darby.

 

[Weaverofduart]

Yeah, ok.

So to minimise the pressure drop you can increase the manifold diameter and make the outlets very close together.

My fear is that it is not that simple. If it is then that is great but I think that the outlets have to have a certain seperation or the currents in the manifold into the outlets interact and cause a higher pressure drop.

 

[ione]

You do not have to take what above in absolutistic terms. The task of getting an even flow distribution is basically accomplished having a manifold that ensures a pressure drop considerably smaller than that in the branch pipes. I recommend once again the reference above, it could be enlightening.

 

[Weaverofduart]

I tried looking up that reference that you recommended but the only points I could find to download it from were rapid share which dosnt work very well on my connection.

However if it just tells me how to size the diameter of my manifold to ensure even distribution of flow then it won't help me because I already know how to do that. My concern is that the spacing of the outlets effects the pressure loss. For example if the friction loss for two 90 degree bends separated by a distance where r/d=0.5 (radius of turn dived by diameter of pipe) is four times that of two bends seperated by a distance where r/d=1 then it would follow that while you may reduce pressure loss due to the length of the manifold pipe in a manifold by making it as short as possible you could increase the pressure loss by having the outlets too close together.

 

[BigInch]

I think you worrying far too much about this. Have a look at the dimensions of a standard tee fitting. They've been known to be welded end to end on occasion. 2 outlets close together would tend to increase fluid velocity in the manifold, which reduces pressure in there, but as the velocity slows again as fluid runs down the branch, a lot of that pressure is recovered. Pressure drop from flow friction shouldn't amount to much over the length of a typical manifold.

 

[KernOily]

Your concern is even distribution across all nozzles (outlets) from the header. As long as the pressure drop across the header is << less than the pressure drop across any one nozzle, you are good. I can give you a number for << if you want, but I have always just made sure the drop across the nozzle was much much more than the drop across the header, and it worked great. A little oil patch horse sense.

 

[Weaverofduart]

The number I've been using is that the pressure in the outlets need to be greater by a factor of more than ten times for a maldistribution of less than 5%.

So if I have a very large pipe so that the loss along it s length is negligible for the required flow then what people seem to be saying is that the pressure loss between a design where I have my outlets as close together as is physically possible compared to one where the outlets are 5, 10 or more diameters apart is negligible, ie equal to the loss in the very large pipe.

That dosnt seem right to me. When water is entering an entrance water comes from straight in front, sides and all around. If outlets are close enough together are they not going to compete for water making the way that the water moves in the large pipe different to the way it would move if the outlets were further apart?

I know guys have said not to worry about it but I not only need to get this right but, I need to be confidant that it's right and I need to know why it's right.

 

[Weaverofduart]

Thinking about this some more if the distance of the outlet spacing doesn't matter just that the pressure differential be large then it would also follow that the eveness of the outlet spacing doesn't matter either?

 

[Weaverofduart]

Ok after being urged not to worry about this I am again.

Ive been told by my flow meter supplier that in order for them to work accurately I need 10 diameters of straight pipe before and 5 diameters after in order for there to be a stable velocity profile. This got me thinking that if I have the inlet to my manifold at one point and the outlets scattered to the left and right unevenly then it would be likely that the turblence of the water entering the manifold would effect the flow into the outlets.

People have said not to worry about this but it just doesn't seem right.

 

[Weaverofduart]

Here is a case that is neat in terms of pipe layout but seems like a bad idea.

M=manifolds 225mm
I=inlet 100mm
O=outlet 65mm

I
I
O O I O
O O I O
MMMMMMMMMMM
O O O
O O O

Hope that is clear.

 

[btrueblood]

Are you trying to balance the flow out of the varying branches to better than 5%? If yes, consider pressure-independent flow control valves, or similar high-precision control valves on each leg. If not, listen to the earlier posts and stop fretting. Trying to compute, a priori, the difference in flow or dP across the varying branches of a typical rectangular manifold is a subject for CFD lecture halls. Real engineers take a pretty good stab at it, and plan to tweak it in the test lab to get it right.

 

[osoloco]

The key to manifold design is to keep the velocity in the manifold low. Remember Bernoulli's Theorem... Think of the head at the exits, the change in height should be negligible, the change in static pressure should also be negligible if the velocity is low, leaving the velocity head for any differences. The portion of energy found in the velocity head applies only in the direction of flow!