CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   How to define humidity in CFX (http://www.cfd-online.com/Forums/cfx/126795-how-define-humidity-cfx.html)

 atulpat November 26, 2013 07:53

How to define humidity in CFX

Hello

I want to carry out thermal comfort analysis in a room using CFX. I also want to define humidity in the room. Dry bulb temperature is 25 deg C and RH is 55% but I don't know how to define such boundary condition in CFX? Please help me.

 sosat1012 November 26, 2013 08:18

If i'm not mistaken you could define your boundaries through mass fractions when it comes to humidity. Check this calculator and use the humidity ratio value (kg/kg)

http://www.sugartech.co.za/psychro/index.php

 atulpat November 29, 2013 06:45

Thanks Sosat
I am having psychometric chart but problem is with defining the boundary conditions.
I have considered the fluid as a mixture of Air Ideal gas and Water vapour at 25C. This fluid is then assigned in default domain by considering air ideal gas as constraint and water vapour as transport equation. Here kinematic diffusivity of water vapour is taken as
2.5E-5 (m2/s). Again air inlet temperature, velocity and mass fraction of water vapour are defined in inlet boundary condition. But I am getting higher value of relative humidity. Relative humidity is calculated as follows:
Insert→ Expression→ Pg
Definition = (0.61078 *exp ((17.269*(Temperature-273.15 [K]))/((Temperature-273.15 [K])+273.3[K])))*1000 [Pa]
Insert→ Expression→ Relative Humidity
Definition = (((Water Vapour at 25 C. Mass Fraction/Air Ideal Gas. Mass Fraction)*Absolute Pressure)/((0.622+ Water Vapour at 25 C. Mass Fraction/ Air Ideal Gas. Mass Fraction)*Pg))*100
Insert→ Variable→ RH and in expression set Relative Humidity

 sosat1012 November 29, 2013 08:41

If i understand what you say ( i am a bit confused) you cant define the initial temp and RH inside the room? Am I correct?
Well if this is the case you should use initialization. These are not boundaries but initial values of your domain. There you can define initial velocities inside the room and the concentrations of all the gases that are not the one defined as constraint (air for you). It will appear itself to have only water vapor and you can define there the initial RH as a mass fraction.

I hope i helped. If not we could go on brainstorming

 atulpat November 30, 2013 02:01

Actually I have carried 2 to 3 case studies on thermal comfort in room. But I am getting slightly different results.
Now I am just explaining one case study. A hospital room having one patient and two staff members and one lamp. Initial temp. of room is 20 C so I am considering it as dry bulb temp. and relative humidity is 68%. From psychometric chart humidity ratio becomes 0.01 kg/kg.
Boundary condition are defined as follows
New material: Mixture Air
Basic Setting: Option- Variable composition mixture
Material List- Air Ideal Gas + Water vapour at 25C+ CO2 (Here CO2 taken from import library data as Gas phase combustion )
Then this mixture air assigned in default domain.
Buoyancy model- Buoyant
Gravity- -9.81m/s2
Ref. density- 1.2 kg/m3
Turbulence model- SST
Air Ideal gas- Constraint
CO2- Transport Equation (1.2E-5 m2/s)
Water vapour at 25C- Transport Equation (2.5E-5)

Inlet:- Air temp.: 20C, Air velocity: 1 m/s, Co2 mass fraction: 0, Water vapour at 25C mass fraction: 0.01
Lamp face_wall:- Heat flux: 100 W/m2
Lamp side walls:- Heat Flux: 5 W/m2
Patient_wall:- Temperature: 34 C,
Source:Equation source:CO2.mf:1E-5(Kg/m2s)
Source:Equation source:Water vapour 25C.mf:2.5E-6(Kg/m2s)
Patient Back_wall: Temperature: 34 C
Staff_wall:- Temperature: 34 C,
Source:Equation source:CO2.mf: 0
Source:Equation source:Water vapour 25C.mf: 4E-6(Kg/m2s)

Here residual target kept as 1E-6 and after 1200 iterations I am getting converged results and areaAve RH for default domain getting as 65%(It is OK as per psychometric chart).
Now problem is that
In results, At inlet of domain RH becomes 81% and areaAve Default domain temperature increased to 24 C. Now If we goes for dry bulb temp. 24 C and 0.01 kg/kg humidity ratio then RH should be 53% as per psychometric chart.

Sorry for long post.

 atulpat December 3, 2013 00:55

Please somebody help me for this problem.

 A_Prakash December 3, 2013 03:50

1) You are initializing your room temperature at 20 deg.C. After that you are specifying your conditioned air to come into the room at 20 deg.C. That doesn't seem right. You say that Domain temperature has increased to 24 deg.C (which is logical and expected), then what is the point of supplying conditioned air at 20 deg.C?

2) Query the velocity at inlet (ave, areaAve in CFD-Post calculator). Is your BC of 1 m/s enforced?

3) What are room dimensions?

4) Is there backflow?

 atulpat December 3, 2013 06:42

4 Attachment(s)
Dear Sir,
Thanks for reply. Actually I have not initialize to the domain at 20 C. Domain walls are considered as adiabatic. and inlet temperature is given as 20 C.Now I just calculated the heat transfer rate Q using wall heat flux and it is 170.98 W for whole domain. Here mass flow rate is 0.17488 Kg/s (1 M/s) and Cp is 1013.53 J/KgK. Using Energy balance equation, 20.96 C outlet temp is obtained. But in CFD result I am getting 23 C outlet temperature and 23.5 C domain temperature. then is it possible?
Also, Here Domain temperature is increasing up to 23.5 C and water vapour mass fraction is 0.01 Kg/kg. So as per psychometric chart, we should get near to 53% relative humidity but in CFD I am getting it 68%. Also, at inlet portion it is showing 81% relative humidity. We can observe this results in attached images. Is it possible?
Here velocity of the air is 1 m/s.
There is no back flow and dimension are also provided in attached images.

 A_Prakash December 3, 2013 10:50

Ok. Now, it is clear why you are not happy with the result: You expect a 68% RH as per specified mass fraction at inlet, but model is throwing air at 81%RH.
I am no expert (yet)..but, we can troubleshoot...
First sanity check would be: How is your RH equation expression implemented? Is it giving you 68% RH when mass fraction is exactly 0.01?

Would you mind sharing the model?

 ghorrocks December 3, 2013 17:55

With streamlines going all over the place like that I find it very hard to believe that this flow is steady state. You are probably going to have to run it transient to get an accurate simulation as the simulation is bound to flap about.

 atulpat December 4, 2013 06:28

@A_Prakash:
with the help of psychometric chart I am calculating the RH. And It is difficult to post model due to large size. RH calculation in CFX-Post as follows:
Insert→ Expression→ Pg
Definition = (0.61078 *exp ((17.269*(Temperature-273.15 [K]))/((Temperature-273.15 [K])+273.3[K])))*1000 [Pa]
Insert→ Expression→ Relative Humidity
Definition = (((Water Vapour at 25 C.Mass Fraction/Air Ideal Gas.Mass Fraction)*Absolute Pressure)/((0.622+ Water Vapour at 25 C.Mass Fraction/Air Ideal Gas.Mass Fraction)*Pg))*100
Insert→ Variable→ RH and in expression set Relative Humidity

@Ghorrocks:

How the transient analysis will increase the accuracy of result? And why Energy balance equation is not satisfied? Previously I worked on HVAC of data center, in such cases energy balance eqn. was satisfied but not in this case.

 sosat1012 December 4, 2013 06:38

Hi all! As far as i know CFX does not include RH at results analysis. So you give the expression (as you mention and present above) in order to get RH contours. Right? I guess I am. Why not do a cross checking of the results?

If you try the Water Vapor mass fraction contours (provided by default in cfx results), do you get a result of 0.01 at inlet. If yes then there should be a mistake at the expression definition and then I am afraid the area averaged result could be also wrong (that now seems right).
Actually you should get the 0.01 value at inlet. I am guessing the expression is the problem. When you define a boundary value this value should be there. Like with temperature. You define 20C at inlet and you can see that at the contours you attached. If not then the problem is related to the inlet boundary input which would be strange because you only have to give a single number

 atulpat December 4, 2013 08:02

1 Attachment(s)
Thanks sosat...

I just did the cross checking of the results for water vapour mass fraction and it shows that mass fraction is 0.01 throughout the domain (refer image). So problem may be in the expression of RH. You knows any method (Expression) for defining the relative humidity in CFX Post? and my second question is that why Energy balance equation is not satisfied?

 sosat1012 December 4, 2013 10:18

Some quick calculations with assumptions led me to this type

RH = (Xwv/18)*((Xwv*18+Xair*29)/Pwv-sat)*Pmix*100%

Pwv-sat is the saturated air water vapour partial pressure which is equal to 2.338kPA (you can check it from the psychrometrics)
Pmix is the gas mixture pressure which can be considered almost equal to atmospheric pressure that is more or less 101kPA
Xwv = 0.01 in our case (water vapour mass fraction)
Xair = 1-Xwv = 0.99 constraint (air mass fraction)
Numbers 18 and 29 are Molecular weights for water vapor (ok water vapor's is 18.016) and air

with these values i get an RH=69.27% while psychrometrics result to 69.00%
I checked some other values for Xwv and they provide similar small error. If you need the contours for a presentation or such i guess this would do.

I have a question? Why do you need CO2 in the mixture. There seems to be no CO2 source. Initially it doesn't exist (it's concentration is 0 everywhere). On the other hand you are using air that kinda includes CO2. Anyway there is no reason to include staff you don't need (if i checked your data correctly if do not consider this comment).

As for the energy sorry my friend i am not your energy guy...hehe!

 atulpat December 5, 2013 00:37

Thanks Sosat...
I will definitely try to calculates RH as per yours RH calculations. Here CO2 is used for patient

 atulpat December 6, 2013 06:42

Thanks Sosat...
Except Temperature results now everything is OK (i.e. Relative humidity). Your expression is working in CFX P:)st. I have used following expression for RH.
Insert→ Expression→ Pg
Definition = exp((20.386-(5132 [K]/T)))*132.29 [Pa]

Insert→ Expression→ Relative Humidity
Definition = (Water Vapour at 25 C.Mass Fraction/18.016)*(((Water Vapour at 25 C.Mass Fraction*18.016)+((1-Water Vapour at 25 C.Mass Fraction)*29))/Pg)*Absolute Pressure*100

Insert→ Variable→ RH and in expression set Relative Humidity

Now, if any anybody knows reason of temperature plot problem then please help me. Here Radiation model is not activated and sometimes I am thinking that due to this only I am getting slightly higher temperature.

 All times are GMT -4. The time now is 07:20.