5 replies to this topic

[samrat]

Process Description:

The flooded screw compressor has the following gas composition at the inlet:
Hydrogen: 64.96% by mol
Methane: 11.33% by mol
Ethane: 13.49% by mol
Propyline: 10.21% by mol
The suction pressure: 0.533kg/cm2g
The discharge pressure: 5.03kg/cm2g
The gas is known as a tail gas.
The gas is compressed in a synthetic oil flooded screw compressor using POA based synthetic oil. After the compression the gas is passed through the OIL TANK SEPARATOR (OTS) and then though the Secondary oil filter  coalescer type  to make the gas devoid of syn oil. The gas is then cooled to around 60Deg C.

Question:
1. Which equation of state do I have to consider as the gas is hydrogen rich.
2. How do we model the Secondary oil filter where the entrained foil from OTS is filtered to about 5ppm level.
3. Has anybody done the simulation, if so can they help?
Request answers to my above question.

[PaoloPemi]

you can select a EOS with some specific parameters for hydrogen (see for example the alpha function proposed by Grabosky and Daubert to replace std. SRK alpha for H2) or some other thermodynamic model,

take care that H2 will be well above critical point, an area where std. and extended alpha functions may generate unaccurate results.

I do not know which simulator you have access and I cannot provide specific comments,

with PRODE PROPERTIES for mixtures with H2 I usually select Peng Robinson with TWU alpha function, as alternative I have the Grabosky and Daubert alpha for H2 (probably your software has something equivalent).

For the other questions (how to simulate the screw compressor, secondary filter etc.) again, that depends from the features available in your simulator,

for my work I utilize mainly Excel so I can create my own models for filter etc. while I can solve compressor unit with a specific operation.

[samrat]

Thanks Mr. Paolo Pemi. Appreciate your answers.

We use Honeywell UniSim Design R410 process simulator. Could you throw some light, having known the specific process Simulator? I do not know whether this software will have the specific EOS for the composition stated in my earlier communication.

Regards,

samrat

[PaoloPemi]

I would recommend the SRK modified by by Grabosky and Daubert for hydrogen but I am not sure that your software has this feature (you may ask technical support),

another possibility would be the Zudkevitch-Joffee model altough I have not direct experience, (as said I have different options in my software)

hoping this helps,

Paolo

Edited by PaoloPemi, 28 June 2013 - 02:06 AM.

[samrat]

Thanks and appreciate. Will check with Honeywell Tech Support.

samrat

[RockDock]

An SRK model should work well. Just be sure to define your reformers and shift reactors well. I have a completed ProMax model of it, but I don't have access to Unisim.