Process header design 1

[JeroG]

For pressure relief headers, it is recommended as per API RP 521 to have lateral entry for the connecting pipes that's going to be joined to the header.

However, what about for process headers? What if the process header is also going to have many connecting pipes that will be joined to it and that these connecting pipes are located close to one another on the header? Will there be a difference in having them coming in laterally (45 deg) or directly (90 deg)? Is there a minimum distance between each connecting pipe that should be observed?

Appreciate if anyone can share their experience on this.

Thanks.

 

[pennpiper]

"For pressure relief headers, it is recommended as per API RP 521 to have lateral entry for the connecting pipes that's going to be joined to the header."
This is true as far as it goes BUT! It does not consider the variables.
What is the size of the Flare Header?
What is the size of the incoming branch line?
What is the Pressure, Velocity and Phase (all gas, all liquid or mixed phase flow) of the release?
What is the Client's Project Design Standards dictate?
The answers to these questions allow the designer to male the most cost effective manner for the connection. Some may be

"However, what about for process headers? What if the process header is also going to have many connecting pipes that will be joined to it and that these connecting pipes are located close to one another on the header? Will there be a difference in having them coming in laterally (45 deg) or directly (90 deg)? Is there a minimum distance between each connecting pipe that should be observed?"
All Process and Utility Branch-to-Header connections should be the simplest, most cost effective method. This will almost always result in a 90 deg connection. This would be a connection using a Tee, Weld-O-Let or a Stub-In (un-reingforced or reinforced).

A 45 deg. connection of a large branch to large header will be more expensive, subject to greater Stress Problems and cause a magnitude of clearance problems in Pipe Rack layout. It is not necessary and not worth the cost and effort to make it work.

Sometimes its possible to do all the right things and still get bad results

 

[LittleInch]

Will there be a difference - yes, your header will be bigger and longer and more complex for little impact as pennpiper says.

Minimum distance - basically physical issues - i.e. enough room for flanges if you're using them or access to weld the connecting pipes, but normally limited to tee to tee fittings if you've got big branches compared to the header or I would go for min 50mm between edge of welds from weldolets.

You need to be careful when putting multiple branches on a header in the same plane as it can easily become a banana shaped header due to the weld heat input which then cools.

The latrolet on reliefs is mainly to reduce losses on the basis that the velocity could be very high and hence you want to reduce wear on the far side of the header and reduce entry losses.

You need to look at velocities of the gas or liquid entering the header and high velocities, especially if there is any hard material can quite easily drill a nice neat hole in the header opposite the entry point, especially if this is either at the start of the header or comprises a majority of the flow of the header.

Kind of a vague generic question so gets a vague and generic answer. If you have a specific problem, please provide some more details (no of connections, size of branches and header, flow rate / velocities etc


My motto: Learn something new every day

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

 

[MJCronin]

jeroG...

Please understand that the question that you are asking deals with some of the deeper subtleties of piping design.... there are no codes/standards that govern this matter.

PennPiper and Littleinch, as always, give good advice. !

Realize that your term "process systems" covers a lot of ground. What works well in one case may not apply in another

My two cents:

With regard to boiler condensate collecting systems, especially those with a high pressure drop..... watch out ! As noted by Littleinch, erosion can occur in the header, directly opposite the branch connection. Erosion resistant materials and increased header wall thickness can help in these circumstances.

http://www.eng-tips.com/viewthread.cfm?qid=45487

Beware of high velocity, two phase flow !



MJCronin
Sr. Process Engineer