Discharge Header Velocity

[SAK9]

What is the ideal design velocity for a pump discharge header pipe that has multiple branches taking off from it?Is there any established guide line on header velocities?

I am looking at a large chilled water plant that has currently a flow rate of around 120 l/s.The header is 300mm dia.We need to add a new chiller which will take the flow to around 160 l/s.Will a high header velocity cause an imbalance of flow into the take off branches?

 

[BigInch]

You're probably fine with the new velocity of 2.3 m/s.
A high of up to around 5-6 m/s could be considered for short segments of pipe, provided that the liquid is clean and that overall pressure drop and waterhammer are not expected to cause trouble.

"If everything seems under control, you're just not moving fast enough."
- Mario Andretti- When asked about transient hydraulics

http://virtualpipeline.spaces.msn.com

 

[quark]

New velocity will be about 3.2m/s (for SCH40 and 3 for SCH10) but that shouldn't be a problem. Discharge velocity of 2-3m/s is quite OK.

Flow imbalance will be created by the pressure drop across the branches. No problem if you have flow setters for each chiller. Atleast, do the balancing by the pressure drop curve of the chiller.

 

[katmar]

Whether there will be a problem or not is more dependent on the pressure drop along the header relative to the overall pressure drop, than to the absolute value of the velocity in the header. Under your higher flow conditions you will have a pressure drop of around 15 kPa per 100m in the header, and unless you have unusually low pressure drops through your branches it should be fine.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[quark]

Ah...stupidity unlimited. When I checked the velocity, I couldn't believe that BigInch was wrong. Then I rechecked with the same mistake, before posting. The velocity will be 2.3m/s as suggested by BigInch.

 

[SAK9]

Thanks guys for your response.From the posts here it appears you do not really treat the header seperately from branch pipes.

I was trained to select headers at low velocity(around 1.5 m/s) and use higher velocities(2~2.5 m/s) for branches leaving the header.Why? I do not have an answer.May be to ensure water in the header had no knetic head so as to make the system self balancing

 

[katmar]

SAK9, yes - the kinetic head and the friction loss in the header should be low relative to the static pressure in the header so that each branch "sees" the same source pressure. Keeping the velocity low is the way to achieve this, but it is better to check the actual numbers than to rely on rules of thumb.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com