CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   Impact of Humidity on CFD Results (

Jerry January 22, 2004 12:36

Impact of Humidity on CFD Results
Hello all. This is my first post here. I'm definitely a newbie to CFD. Could anyone answer this question:

-What is the impact/effect of moisture upon CFD results with everything being equal?

What I mean by this is if I were to run an analysis problem using a heating element and airflow, what would happen if I compared this to the same problem with the addition of moisture? We make Infant Incubators. We add humidity to our heated airflow to keep the infant's skin moist. This prevents drying out of the skin, and helps the infant develop by preventing heat loss through skin moisture evaporation. The only thing I can not determine is how moisture impacts heated airflows. I'm new at the company, and I've been charged with finding a CFD solution that can perform this type of analysis. I searched the internet, and have yet found any info on the impact of moisture. Could you lead me in the direction on literature that discusses moistures impact? I put in a call to Fluent to discuss this with an engineer.


Dean January 26, 2004 13:06

Re: Impact of Humidity on CFD Results
At room temperature, water vapor can make up anywhere from 0 to about 8 percent mole fraction of the air. This results in a variation of a few percent in the mean molecular weight, transport coefficients, specific heat, ...

In a warm incubator, the amount of water could be a bit higher - the vapor pressure of water increases rapidly with temperature.

Jerry January 26, 2004 15:39

Re: Impact of Humidity on CFD Results
Our humidity levels within the incubator environment could reach as high as 95% (even higher I believe). In the Japanese market they run their incubators at this level. You can hardly see inside the incubator because of condensate on the inside of the polycarbonate enclosure.

I realize that our CFD requirements are on the extreme side with flow, heat (convective, conductive and even radiant), moisture (0% to 95%), and condensation. We would also like to change certain geometry of the system effectively changing the boundary conditions (ie. Opening the side panels for access exposing previous results to ambient air). I would like a CFD system that allows for these types of iterative designs by changing geometry and variables (ie. moisture from 0 to 95%, heat source temperature, and fan speed) Here's the dilemma -> Money. We are a relatively small company. We cannot afford spending the 10's of thousands per year that is required to purchase a higher end solution (ie. Fluent, StarCD, et al.).

The reason I asked my initial question was to see whether our company could go with a lower-end CFD solution (CFDesign or FloWorks) to run what-if analysis. I believe this would give us great insight upon thermal conditions and flow patterns of our system (sans moisture) given varying degrees of geometry changes (internal changes as opposed to external). I asked my initial question concerning moisture to gage its impact upon the above scenario. If it took an additional 10 minutes to get to a given set point for a given humidity level than we could run our analysis without moisture, and then estimate the additional time it would take with the addition of moisture. We would verify the best model from our initial CFD analysis runs, and then perform empirical testing upon the physical model. Again this is without considering the impact of condensation. I was also curious as to the impact of moisture laden air to both thermal and flow results. If the results would be impacted nonlinearly then it's not so simple.

CFDesign claims to "do" moisture. On their website it states it can perform Two-Phase Fluid Flow. Is this "moisture"? Has anyone exposure to either of the two CFD programs I listed above? What is their suitability to our application (besides not having moisture and condensation)? Any other solutions out there (I've heard of StarWorks that is offered by Adapco)? What is the argument between Finite Volume Method and Finite Element Method (it seems everyone out there is using FVM, CFDesign uses FEM. They claim to have a superior method)? What about meshing (FloWorks uses Adaptive Hex, CFDesign uses all Tet)?

Sorry for the additional questions :)


All times are GMT -4. The time now is 12:38.