4 replies to this topic

[Hafiz Zakaria]

Greetings,

I'm a fresh graduate engineer and currently i'm given a task to do a flare network study of a certain oil terminal plant. The software that i use is Aspen Flare system analyzer. During the job, a few questions lingered in my mind:

1) Why rated flow is used for tailpipe calculation? Is it a matter of conservative or any engineering code recommending that? If yes, what code applied?

2) Why is mach number matters in flare header calculation? I know the constraint, 0.3 for main header and 0.5 for tailpipes. But what will actually happen when those constraint are breached? Can someone explain this to me please?

Thanks a lot in advance

[fallah]

Greetings,

I'm a fresh graduate engineer and currently i'm given a task to do a flare network study of a certain oil terminal plant. The software that i use is Aspen Flare system analyzer. During the job, a few questions lingered in my mind:

1) Why rated flow is used for tailpipe calculation? Is it a matter of conservative or any engineering code recommending that? If yes, what code applied?
API 521 recommends rated flow for tail pipe sizing except about modulating pilot operated PSV.
2)
Why is mach number matters in flare header calculation? I know the constraint, 0.3 for main header and 0.5 for tailpipes. But what will actually happen when those constraint are breached? Can someone explain this to me please?
You can take 0.7 for tail pipe and 0.5 for main header.Consequences of passing these limitations are having (at least) more vibration,noise and stronger supports of piping.
Thanks a lot in advance

Edited by fallah, 28 January 2011 - 02:25 AM.

[Hafiz Zakaria]

Thanks for the answer

[Lowflo]

Rated flowrate is used for pop-acting PSV tailpipes because those valves are designed to reach their rated capacity fairly quickly. It would be non-conservative to use the required flowrate. Only modulating valve flow at the capacity demanded by the system (required flowrate).

API 521 says to use the required flowrates for modeling headers (everything other than the tailpipes). When there are a large number of sources, it's reasonable to assume that the flowrate in the main headers will approximate the sum of the required flowrates. That's because all of the PSV are not open at the same time. The higher the number of sources, the truer that assumption becomes. Understand that API 521 is general guidance. That is, you always have to look at your specific system to see if that guidance fits. There are some subheaders for which it makes sense to use the sum of the rated flowrates rather than the sum of the required flowrate. That's true for subheaders that have only a few sources of flow, from vessels of comparible size, and the sources are in the same Flarenet scenario. Judgment is always required.

The default mach number constraints are perhaps OK for a new design but they are very conservative for an existing flare header system. Personally, when modeling an existing system, I turn them off because they create a clutter of warning messages. For an existing system, the bottom line is will it work or not work. A mach number of 0.8, or even higher, is often acceptable for a worst case load scenario on an existing flare header. When you have high mach numbers, you need to be more sure of your input data. There's less margin for error, so you need to be sure your flowrates and compositions are conservative. In such cases it's also a good idea to run some sensitivity cases by varying your composition (often there is a range of possible compositions from a given PSV source). Mach number if composition dependent, and you'll sometimes discover that a mach of 0.8 can be mach 1 if the composition varies.

[Hafiz Zakaria]

Rated flowrate is used for pop-acting PSV tailpipes because those valves are designed to reach their rated capacity fairly quickly. It would be non-conservative to use the required flowrate. Only modulating valve flow at the capacity demanded by the system (required flowrate).

API 521 says to use the required flowrates for modeling headers (everything other than the tailpipes). When there are a large number of sources, it's reasonable to assume that the flowrate in the main headers will approximate the sum of the required flowrates. That's because all of the PSV are not open at the same time. The higher the number of sources, the truer that assumption becomes. Understand that API 521 is general guidance. That is, you always have to look at your specific system to see if that guidance fits. There are some subheaders for which it makes sense to use the sum of the rated flowrates rather than the sum of the required flowrate. That's true for subheaders that have only a few sources of flow, from vessels of comparible size, and the sources are in the same Flarenet scenario. Judgment is always required.

The default mach number constraints are perhaps OK for a new design but they are very conservative for an existing flare header system. Personally, when modeling an existing system, I turn them off because they create a clutter of warning messages. For an existing system, the bottom line is will it work or not work. A mach number of 0.8, or even higher, is often acceptable for a worst case load scenario on an existing flare header. When you have high mach numbers, you need to be more sure of your input data. There's less margin for error, so you need to be sure your flowrates and compositions are conservative. In such cases it's also a good idea to run some sensitivity cases by varying your composition (often there is a range of possible compositions from a given PSV source). Mach number if composition dependent, and you'll sometimes discover that a mach of 0.8 can be mach 1 if the composition varies.

Thanks lowflo..really appreciate the answer..now i understand