5 replies to this topic

[rxnarang]

Our company, as most others, limits the PSV outlet lateral velocity to 0.7 mach. Of course, pressure drop is also a major criteria. Assuming the back pressure criteria is met, what is the justification for limiting velcoities in the outlet lateral? Can we increase this velocity to 0.8-0.9 mach? What are the considerations in specifying this maximum velocity?

Regards

[avsp]

rxnarang,

I am working on a LPG project, in which the client's standard specifies a maximum allowable velocity of 0.8 Mach and a momentum of ,

[rxnarang]

AVSP, Thanks for your inputs.

The question is more about individual PSV outlet. Normally meeting 0.7 mach is not a problem. I agree.

By the way what software are you using to size the outlet. We are migrating to Flarenet to size individual and all the queries are arising because of the discrepancies we observe bewteen our traditional line sizing software and Flarenet.

Regards

[avsp]

rxnarang,

You may find this tedious but this is how we did it for certain two phase relieving scenarios in several of our oil&gas upstream projects, but with fairly good results.

For the calculation, we used HYSYS and PIPESIM. We first used HYSYS to do a isentropic flash calculation, for the relieving fluid, to estimate the liquid, gas and mixed phase densities at various decrements of pressure starting from the relieving pressure to the required outlet pressure. These were inputted to an in-house EXCEL spreadsheet which calculated the mass flux at the various decrements of pressure. The corresponding decremented pressure, at which the spreadsheet calculated the maximum value of mass flux, was noted.

Then PIPESIM calulation was done for relief discharge line by defining the sink conditions and estimated the source conditions at an assumed pipe size. The source pressure would be compared with the above decremented pressure. Obviously, it may not match exactly, but if the deviation was within + or- 1% then the next step was to check the back pressure and the Mach number from the PIPESIM output file. If this was OK, then the pipesize was acceptable otherwise we either decrease or increase the pipesize and re-run PIPESIM till we get the desired results.

Anyways, all this was done prior to the company acquiring FLARENET (this was a recent mandatory requirement from the client) after which all the above time consuming exercise was stopped. Unfortuantely when this software had arrived, I could not use it extensively as I had left that company by then. However, before leaving, I did validate some of my earlier "tedious" calculations on this software and was glad to see that the results came quite close.

Cheers

[benoyjohn]

rxnarang,

You may find this tedious but this is how we did it for certain two phase relieving scenarios in several of our oil&gas upstream projects, but with fairly good results.

For the calculation, we used HYSYS and PIPESIM. We first used HYSYS to do a isentropic flash calculation, for the relieving fluid, to estimate the liquid, gas and mixed phase densities at various decrements of pressure starting from the relieving pressure to the required outlet pressure. These were inputted to an in-house EXCEL spreadsheet which calculated the mass flux at the various decrements of pressure. The corresponding decremented pressure, at which the spreadsheet calculated the maximum value of mass flux, was noted.

Then PIPESIM calulation was done for relief discharge line by defining the sink conditions and estimated the source conditions at an assumed pipe size. The source pressure would be compared with the above decremented pressure. Obviously, it may not match exactly, but if the deviation was within + or- 1% then the next step was to check the back pressure and the Mach number from the PIPESIM output file. If this was OK, then the pipesize was acceptable otherwise we either decrease or increase the pipesize and re-run PIPESIM till we get the desired results.

Anyways, all this was done prior to the company acquiring FLARENET (this was a recent mandatory requirement from the client) after which all the above time consuming exercise was stopped. Unfortuantely when this software had arrived, I could not use it extensively as I had left that company by then. However, before leaving, I did validate some of my earlier "tedious" calculations on this software and was glad to see that the results came quite close.

Cheers

Rajiv,

The only reason I could think was to keep a margin from choked flow. The tail pipe dimensions and fittings can often can be finalised only after the piping study has progressed substantialy in a project. The more closer you go to Mach during the design, the more chances of getting choked flow with minor changes in line equivalent lengths used for pressure drop calculations. Hence 0.7 or 0.8 Mach gives a reasonable margin for design in these cases.


Regards
Benoy

[JoeWong]

The only reason I could think was to keep a margin from choked flow. The tail pipe dimensions and fittings can often can be finalised only after the piping study has progressed substantialy in a project. The more closer you go to Mach during the design, the more chances of getting choked flow with minor changes in line equivalent lengths used for pressure drop calculations. Hence 0.7 or 0.8 Mach gives a reasonable margin for design in these cases.

Benoy,
I do agree with your statement.

We may design tail pipe to Choked flow conditions (Mach 1) and this approach has been adopted. You may refer Choked Flow and Sonic Velocity.

Design PSV outlet to Mach 1, there are many conditions and criteria binding to it. Some being tabulated in "Is PSV tail pipe & lateral at CHOKED (Mach no = 1) Acceptable ?".