multi-phase joules-thompson effect problem

[ktgottfr]

Hi all,
I am currently using the NIST SUPERTRAPP database to determine the liquid/gas phase composition of a mixed fluid. The fluid is raw natural gas from a pipeline, and so contains c1, c2, etc plus some water.

Now, here is my problem. This fluid passes through a pressure regulator valve, where the pressure drops from 537psi to 200psi. I am trying to determine the final temperature of the mixture. When I use the NIST program, it gives me the option of doing a constant enthalpy expansion on the two phase mixture. However, I just noticed that the program is combining the enthalpies of both the liquid and gas phases to do this.

So, my question is this: to properly determine the amount of cooling due to the joules-thompson effect, do I need to do a constant enthalpy expansion keeping just the gas phase's enthalpy constant, or is it appropriate to do as the NIST program is doing and keep the total enthalpy for both gas and liquid constant?

At the moment, I am guessing that only the gas phase's enthalpy should be kept constant, as only the gas actually expands through the valve. But I am a mechanical engineer and so this is a bit outside my normal area...
Thanks everyone!
-ktgottfr

 

[zdas04]

I did a similar analysis this week on a CO2/CH4 blend and used REFPROP's "bulk properties" option and only got one enthalpy. The magnitude was consistent with the gas so I used it. There might be a way to do a similar thing in SUPERTRAPP.

David

 

[maddocks]

My approach would be to keep the entire stream enthalpy unchanged. Any differential enthalpy is represented as work extracted from the fluid and a JT valve simply can't do this.

 

[ktgottfr]

Ok maddocks, that sounds reasonable. I was just a bit unsure, as it was apparent that enthalpy from the liquid phase was being transferred to the gaseous phase if the overall enthalpy was kept constant. But I suppose this is probably due to some of the gas condensing into liquid?
-ktgottfr

 

[EngAbdelkader]

Hi ktgottfr, I suggest you before running any simulation to take a look at our old friendly thermodynamic tool, which is the Pressure/Enthalpy charts for your stream, you'll find a clear one in the GPSA handbook and probably in Perry's chemical engineering handbook. I know that the charts are for one fluid actually but you can get a good approximation by studying each fluid (gas) alone.

Keep in mind that the Joule Thomson effect concerns only gases and is an iso enthalpic process.

in addition having read that your stream contains water!! if it was a real stream you'll have a great Hydrate problem after passing the stream through a choke valve!! that's why in the gas treatment plant we process the gas stream through a dehydration column (to take rid of water!!)

Hoping it'll help!