Defining the pulsation source in Puls software for centrifugal compressors ?

Hello Everybody,   

One of my company Customer want us to provide piping pulsation analysis for gas compressor station where theoretically pulsation sources are two centrifugal compressors. I know that Puls software is not designed for piping pulsation with centrifugal compressors. Centrifugal compressors particularly in these times should not be considered as pulsation pressure source, however it has been assumed and defined by the Customer that these pulsation sources should not exceed 10% of maximum pulsation value peak to peak based on API 618 in Puls software for given rpm of compressors? 

1.Based on requirement of 10% how exactly should I define the pulsation source to perform pressure and force simulation vs frequency in Puls? Would 10% of the pressure pulsation curve based on API 618 be my input, and having that I should calculate forces vs frequecy for each node then the harmonics would be imported to Autopipe (harmonics described by damping of ~2%, frequency for the 1-st, 2-nd, and 3-rd harmonics with phases and forces acting in different direction)?    Can I find any example in Bentley resources that could show me practical way of solving problem mentioned above?      

Given data:

Suction side – 4MPa

Discharge side – 6,5MPa

Methane gas flow – 140 thousand m3/h

             

2.   If the piping model in Puls will have less than 400 nodes for both suction and discharge lines it seems to me that option 1 of Puls software be good enough to make pressure and dynamic forces acoustic simulations (with two centrifugal compressors) ? Am I correct thinking so?  

Best Regards

Robert Augustyn       

 

  • 1. PULS allows for modeling pressure pulsation source, however it is necessary to know the frequency characteristics of this source. In this case it does not seem that you have any model of this kind. Also I am afraid API618 limits have nothing in common with centrifugal compressors. 

    If you could 'simulate' numerically the loading (i.e. pressure pulsation source, no matter how you gain the spectrum of this pulsation), you would be able to apply it in PULS. Then you could get the dynamic forces to be applied in AutoPIPE, and further calculate dynamic response.

    I believe you should return to you customer and discuss what are true goals of this analysis.

    2. Yes, I believe that 400 nodes is the model contents limit, however I am not sure about how many pressure pulsation sources are allowed for Option 1. This would be easy to verify - just build the sample model and insert predicted (for analysis) number of boundary conditions of this kind.

    Best regards,

    Maciej Rydlewicz, PhD Eng.

    maciej.rydlewicz@softdesk.pl

    +48 512206994

  • Thank you Maciej for your response. I may call you to try to discuss this issue when appropriate time will come.

    Best regards

    Robert Augustyn

  • Hello Robert, 

    To add a little more to what MaciejRyd mentioned above,  see WIKI here for limitations for nodes and BC in each edition of PULS.

    Regards,

    Mike Dattilio
    Bentley Systems Design Analyst
    Design Engineering Analysis group
    ===================================================

  • Thank you for your answer Maciej. Yes, you answered my questions except one issue: API 618 limits refer to reciprocating compressors that are included in PULS application; however, as I know there are no other standards that I could use for centrifugal compressors. What allowable limits - standards would you use in this case for centrifugal compressors?
    Regards 
    Robert Augustyn

     Wolny od wirusów. www.avg.com

    Hello,

    In document nr.1 at WIKI (PULS issues) it is stated: [cyt] While there are similarities, there are also basic differences between acoustic analysis of centrifugal and reciprocating compressor piping. The fundamental approach of avoiding coincidence between excitation and resonant response is common to both types of piping systems. Pulsation sources, however, are treated very differently. The role of the centrifugal compressor in acoustic analysis is also very different from the role of the reciprocating compressor. For example, centrifugal compressors are modelled as specialized elements that connect the suction and discharge piping systems.

    Pulsation sources in centrifugal compressor piping are flow generated and occur at all locations where the flow separates from the pipe wall. In practice, only tees and control valves generate significant pulsation. Flow generated sources occur simultaneously, however, for design stage analysis each source is modelled individually. Further, and in contrast to reciprocating compressor piping, effective solutions to flow generated pulsation deal directly with the pulsation source.

    The source modelling methods described in this paper have been presented as manual calculation procedures so they may be used by all designers who have access to an acoustic simulator. To simplify and reduce design effort, each of the methods described has been incorporated into PULS, the acoustic simulator developed by NHRC and AGTD.

    Finally, the influence of mechanical response on acoustic design, and vice versa, is very important. This is clearly demonstrated by the use of closely linked mechanical and acoustical guidelines. For example, the minimum mechanical natural frequency guideline and the acoustic shaking force guideline were developed in unison for design stage analysis based on designers' field and office experience.

    I’d like to refer to the sentences written above in bold and underlined and ask three questions?.

    1. To model centrifugal compressor what specialized elements should be used to connect discharge and suction piping system? The suction and discharge piping should be analyzed independently (separately) as for systems with reciprocating compressors in PULS (even some special elements connect suction and discharge side). I think you can confirm that.
    2. Should I model only all tees and control valve as a pulsation sources (that are flow generated) of the system (in Puls to simplify and reduce design analytical effort)? What about centrifugal compressors as a source of pulsation – it is not mentioned anything about compressor frequency characteristics. Centrifugal compressors are also the source of pulsation of course relatively with much smaller amplitude that the reciprocating compressors; however, it does not seems to be correct procedure to neglect centrifugal compressors as a source of pulsation. Do you agree with that statement.
    3. What are acoustic shaking forces guidelines for gas centrifugal compressors stations?

    Regards

    Robert Augustyn