8 replies to this topic

[bponcet]

for a certain critical condition I wish to evaluate the speed of sound in hydrocarbon's mixture (mainly C1) at about 50 Bar (pipeline) with gas value fraction of about 0.8 (gas + condensate)
from literature I have estimated the speed in gas and liquid fractions,
however I have read that speed of sound could be very low in mixed phase,
how should I proceed ?

[Bobby Strain]

Cautiously.

Bobby

[marchem]

it is true that HEM model predicts a lower speed of sound in mixed phase (compared to gas and liquid phase)
honestly I do not know if that is the correct value to consider in your case (depending from process, model etc.).
However to give some results I calculated the speed of sound in gas, liquid and mixed phase with the HEM model available in Prode Properties by entering the macro =StrMSS(stream) in a cell of Excel

for the mixture Methane 0.8 Ethane 0.06 Propane 0.1 n-Butane 0.02 Co2 0.02 at 50 Bar.a with model Peng Robinson
the values calculated for the speed of sound at different temperatures (and pressure 50 Bar.a) are

Vapor Fraction 1.0 (dew point) 275.4 K speed 325.42 m/s
Vapor Fraction 0.8 240.2 K speed 236.9 m/s
Vapor Fraction 0.5 215.8 K speed 166.4 m/s
Vapor Fraction 0.2 209.1 K speed 119.8 m/s
Vapor Fraction 0.0 (bubble point) 207 K speed 534 m/s

as you see the calculated values (with HEM model) are lower for mixed phase,
as said I do not know if that is true (or not) in your case

[kkala]

Several information can be retrieved through googling, e.g. http://www.physicsforums.com/showthread.php?t=287816, http://www.mathworks.com/matlabcentral/fileexchange/36920-two-phase-medium-sound-speed-calculator, http://www.sintef.no...s/lund_an10.pdf, http://www.potto.org/gasDynamics/node74.php, etc. This probably needs assessment by someone knowledgeable and interested in the subject. First two web references mention bubble flow, without saying that other flow regimes are excluded. Correlations indicate lower sound velocity in the two phase mixture, compared to its components. Do these have validity for all flow regimes? Advice would be welcomed.
Imagine gradual cooling of a high velocity vapor (even water steam), passing through an orifice at critical (sound) velocity. Will there be decrease in volumetric flow rate, as soon as condensate droplets are formed? Or the resulting fluid (mist flow, constant upstream pressure) will still have same volumetric flow as before?

Edited by kkala, 05 October 2012 - 08:29 AM.

[PaoloPemi]

it is not easy to calculate the speed of sound in mixtures keeping in account the mass transfer between phases and the effect of capillary pressure,
you need a EOS to model both energy and phase equilibria, see the paper "Two-Phase Isentropic Compressibility and Two-Phase Sonic Velocity for Multicomponent-Hydrocarbon Mixtures" by Abbas Firoozabadi and Huanquan Pan.
If the HEM model is acceptable (that depends from the process which you wish to model) you can solve (rigorously) with a software as Prode Properties (see the above values given by Marchem), differently you need to create your own model, not an easy task...

[bponcet]

thanks,
HEM (homogeneous equilibrium model) means that velocity, temperature and pressure between the different phases are equal,
right?

[PaoloPemi]

that's correct,
you do not provide the details and it is difficult for us to judge if HEM is suitable (or not),
an advantage is that HEM is less complex to solve than other approaches,
presuming there is constant composition you can calculate the speed of sound (with Prode Properties or another tool) in a range of temperatures and pressures, then perhaps you can define a simplified model and use that in your analysis.

[bponcet]

PaoloPemi,
I am evaluating a critical condition in a pipeline (hydrocarbons transport),
under expected conditions HEM is acceptable,
I have been in contact with Prode (I have requested assistance to solve several cases) and the software seems ok for this project.

[marchem]

as said by others for a pipeline you may probably consider also different models,
two phase flow is a complex condition which you cannot simulate easily,
in most cases a decent estimate is a good result.