HYSYS Enthalpy Values 1

[MatthewD]

I am curious as to why the values for enthalpy in HYSYS (Peng-Robinson with Lee Kesler Enthalpy Correction EOS) are negitive for typical natural gas. Does this mean that HYSYS compares the enthalpy to a refernce value because when I look up the enthalpy value in the API databook the value is always positive

 

[DickRussell]

Absolute enthalpy values have no meaning; they are always heat content with respect to a reference condition. Only enthalpy differences have meaning, as with duties in exchangers for example. When HYSYS and API TDB enthalpy values differ in sign, it just means that the reference point for ideal gas enthalpy is different. The older (~1980) API TDB used liquid at -200 F as the zero reference point, except for ten light gases, for which it was zero for the vapor at zero Rankine (I don't have the current issue in front of me; it may have changed). This avoids negative enthalpies in typical regions of engineering calculations. This has importance in computerized calculations, when in doing enthalpy balances one would home in on temperature until enthalpy error was a negligible fraction of inlet enthalpy. In calculating that fraction error, you don't want to be dividing by a number close to zero; you want the normalizing value to be comfortably removed from zero. Of course, using zero liquid enthalpy at -200 F means extrapolating latent heat well below the freezing point of most compounds and is purely hypothetical. A roughly equivalent zero reference enthalpy basis would be to use, say, 150 BTU/lb for the ideal gas at -200 F. Exactly what HYSYS has used probably is available in their documentation. It could well be zero for the ideal gas at the ice point. This doesn't mean that HYSYS will have enthalpy balance convergence problems. It just means that more care has been taken to address this; perhaps a different normalizing total enthalpy has been used for deciding when the total enthalpy error has become negligible.

One thing that must be considered when using enthalpy differences as duties is that the differences have meaning only when there is no overall composition change, as there would be across reaction, unless all components involved in the reaction have zero reference enthalpy on the same basis. With the 1980 API TDB system, any reaction involving any of the ten light compounds having different zero basis automatically would give a problem. Regardless of zero enthalpy basis used, net heat of reaction can be calculated by getting enthalpy change to bring all reactants to the standard state for which heats of formation are defined, getting the heat of reaction by changes in heats of formation, adding those two heat terms (observing sign), and using the sum to heat the effluent composition from the standard state to effluent conditions. Either the shortfall is heat added or the excess is heat removed from the system. Either way, one can't always take difference between feed and effluent enthalpies as being net heat of reaction; that assumes too much about zero enthalpy basis for the compounds involved.

HTH, although it may have been more than you wanted.

Dick Russell

 

[MatthewD]

So for comparitive purposes between sets of T and P as long as you use the same source (HYSYS or API) you should be good?

I guess I find the absoulte values easier to understand cause I can see and reproduce the calculations that lead to them. As for HYSYS I just can't see any calculations.

 

[DickRussell]

Oh, don't mix "absolute" enthalpies calculated by different simulators, because you don't know for sure what the zero basis for the component ideal gas enthalpies are. Mixture enthalpies are calculated using EOS methods to get departure from mixture ideal gas enthalpy down to the real fluid at its conditions. While the departures should be comparable between two simulators or calcs done by different people, it is the ideal gas enthalpy basis for the various components that gives rise to different real fluid enthalpy values for the same fluid & conditions. If you are trying to develop confidence in a simulator's enthalpy calculations, compare enthalpy differences across some piece of equipment, like a heat exchanger. I would be surprised if the enthalpy calculations in a major simulator had a serious error.

 

[MatthewD]

My faith in HYSYS has not faltered. I just get asked alot of questions about its validtiy (I work with alot of old guys that dont trust "That new fangled technology") and I find that I must provide the burden of proof for every new application we use it in.

Let this be a lesson to those who read this post that these simulators are not ment to spew forth thermodynamic properties but to simulate unit operations. Sometimes its just better to do it by hand