Dear All,

I am starting my first slug analysis, on an 8" and 4" blow downline (2-phase flow,liquid + gas), which has displaced (jumped) on site.

I have been through the CAESAR II manual, COADE Newsletters and this forum and they were very help full, giving me a jump start to dynamic analysis.

There was a post with the topic "slug analysis", which recommends a spectrum analysis for slug flow problems due to the many unknowns in slug flow. Thereto I am adopting the spectrum analysis method for the slug analysis (conservative).

I have obtained data regarding the flow from the process department and was able to calculate the forces acting at the elbows of the system, along with assigning the associated directions.

I realize that for each elbow, there will be a separate load case, as we are assuming a single one time occurance of the impulse.

The next step of how to generate the spectrum, is unclear to me. I can use an assumed slug length and the liquid velocity in the stream to calculate the time taken to round the bend, and that this force will be applied in a trapezoidal form. An example of generating the spectrum is given in the application guide but how can it be applied to slug flow case.

In summary my doubts:

1. How to estimate slug length
2. How to generate the DLF/Spectrum for slug flow to take care of the total system response.

I would gratefull for any help.

Regards

1) Slug length: no one knows. Your best bet is to estimate three slugs - the minimum it could be, the most likely, and the maximum it could be.

2) The DLF is generated from your trapezoidal pulse. You'll need a different DLF for each such trapezoidal pulse you have.

Once you have the force (defined in the "Force Set" tab), and the DLFs, you associate them (combine them) together together on the "Load Case" tab.

Thanks for the quick reply,

Slug length
1) Estimating three slug lengths, what if the maximum length of the slug is taken to be the length of the low point leg x vapour fraction %
which in my case comes out to be 4370 mm, is this a good enough approximation for the maximum slug length, if at all it is correct?
What is the practice is estimating the minimum and most likely lengths.
Am I going about estimating the length in a wrong way?

2) For two phase flows do we have to estimate the slug lengths for each length (elbow to elbow) of piping?

3) My concept of slug flow in the line is that the slug is existing in the piping and is accelerated when some equipment discharges its contents to the blow down line. Is this the correct concept.

Regards

1) No one can answer this one. The answer is based on experience in this sort of analysis and your specific system.

2) "Two phase flow" - nothing is constant. Ordinarily I'd say "no" to this question, the slug length is the slug length and is not affected by the system geometry. The distance between elbows affects your (trapezoidal) pulse, not your slug size.

3) Yes, this is the basic assumption.