tentative header size 1

[DEEPBLUE76]

hi all
i wanted to know that how do i assume the main header size if i have say 6 tappings. does this have any criteria or a specification. please help me out

 

[MJCronin]

Deep,

Use an equivalent internal area technique....

If, for example you have a header to distribute liquid to six seperate taps..... and they are all 4 inch NPS (sch 40....lets say)

Total internal pipe area required is: (6)(12.73 sq. inch) = 76.38 sq. inch

Looking up internal areas for sched 40 pipe, we find that a 10 inch pipe meets the requirements (~78 sq inches)....Therefore, a ten inch header is probably suitable for six 4 inch NPS branches..... Flow velocity should be checked in all branches of a system to ensure that it is "reasonable" (5 to 12 ft/second)

My opinions only......

Anyone else out there !!??

MJC

 

[quark]

MJC!

Cheers to you!!! That is the same method I apply when the flow requirement is not known specifically. (and fortunately it worked).

Regards,

Truth: Even the hardest of the problems will have atleast one simple solution. Mine may not be one.

 

[Flareman]

I read a study many years ago which indicated that you need to keep the equivalent area of the "header" to twice (2x) the area of the outlets in order to avoid a maldistribution of more than 10% between the first and last outlets. I've always used that parameter as a starting point and it seems to have worked for me. If I can turn up the original work in my archives, I'll post the reference.

 

[blasterman]

The 2X branch areas is the better approach and does minimize maldistribution. 1-1/2X header is the minimum. References in TEMA, API etc provide an initial guidance in this regard.

 

[Mechy]

I agree with MJCRONIN,your calculations are good, but sometimes theory and practice are diffeent. SO i suggest that you should consider one size higher pipe size for header than is calculated. This is always safer then to midfy the header.

mechy