CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   domain imbalance for enrgy equation (http://www.cfd-online.com/Forums/cfx/106677-domain-imbalance-enrgy-equation.html)

 happy September 5, 2012 07:05

domain imbalance for enrgy equation

Hi all,,
the below is the last iteration of my simulation for buoyancy-driven flow. I can see the domain imbalance for the N-S equation and mass equation is nearly zero; however, for Heat equation is -0.0418 %.
Do you this is quite enough?
I can found that imbalnce of energy equations at interair and top boundaries is higher than other boundaries. I tried to refine the mesh over there, but still have high values. Do you think that can cause problem within my results? so to solve the problem how can I treat with this problem, Please?
Any comments will be appreciated.

Regards
CFX user
******************
================================================== ====================
OUTER LOOP ITERATION = 1529 CPU SECONDS = 1.932E+05
----------------------------------------------------------------------
| Equation | Rate | RMS Res | Max Res | Linear Solution |
+----------------------+------+---------+---------+------------------+
| U-Mom | 1.00 | 8.8E-06 | 4.9E-04 | 4.7E-04 OK|
| V-Mom | 1.00 | 1.0E-05 | 4.9E-04 | 3.2E-04 OK|
| W-Mom | 1.00 | 9.1E-06 | 8.9E-04 | 4.7E-04 OK|
| P-Mass | 0.73 | 1.8E-09 | 8.1E-07 | 4.8 9.4E-02 OK|
+----------------------+------+---------+---------+------------------+
| H-Energy | 0.42 | 2.7E-08 | 1.2E-05 | 5.6 3.9E-04 OK|
+----------------------+------+---------+---------+------------------+
| K-TurbKE | 1.00 | 5.1E-07 | 3.1E-05 | 5.6 1.2E-05 OK|
| O-TurbFreq | 0.65 | 6.8E-09 | 8.3E-07 | 7.8 1.9E-05 OK|
+----------------------+------+---------+---------+------------------+

CFD Solver finished: Tue Sep 4 04:01:52 2012
CFD Solver wall clock seconds: 5.3389E+04
================================================== ====================
Termination and Interrupt Condition Summary
================================================== ====================

CFD Solver: All target criteria reached
(Equation residuals AND global imbalances)
================================================== ====================
Boundary Flow and Total Source Term Summary
================================================== ====================
+--------------------------------------------------------------------+
| U-Mom |
+--------------------------------------------------------------------+
Boundary : back -5.8039E+03
Boundary : bottom -1.8993E-06
Boundary : chimney walls -1.1411E-03
Boundary : front 5.8039E+03
Boundary : inletair -1.0861E-07
Boundary : left 7.7466E-04
Boundary : right -4.4811E-05
Boundary : top 8.2928E-04
-----------
Domain Imbalance : 7.7249E-04
Domain Imbalance, in %: 0.0000 %
+--------------------------------------------------------------------+
| V-Mom |
+--------------------------------------------------------------------+
Boundary : back 1.9882E-02
Boundary : bottom -2.5480E-02
Boundary : chimney walls -2.3605E-01
Boundary : front 1.9764E-02
Boundary : inletair 1.2138E-03
Boundary : left 2.0296E-02
Boundary : right 2.0737E-02
Boundary : top 1.1605E+04
Domain Src (Neg) : room -1.1605E+04
-----------
Domain Imbalance : 1.2969E-02
Domain Imbalance, in %: 0.0001 %
+--------------------------------------------------------------------+
| W-Mom |
+--------------------------------------------------------------------+
Boundary : back -8.3537E-04
Boundary : bottom 2.9684E-06
Boundary : chimney walls 1.2648E-03
Boundary : front 6.9932E-04
Boundary : inletair 1.7835E-07
Boundary : left 5.8039E+03
Boundary : right -5.8039E+03
Boundary : top -1.5814E-03
-----------
Domain Imbalance : -5.8874E-04
Domain Imbalance, in %: 0.0000 %
+--------------------------------------------------------------------+
| P-Mass |
+--------------------------------------------------------------------+
Boundary : back 1.2765E+00
Boundary : front 1.2474E+00
Boundary : inletair 6.2579E-04
Boundary : left 1.2757E+00
Boundary : right 1.2716E+00
Boundary : top -5.0718E+00
-----------
Domain Imbalance : -1.9602E-06
Domain Imbalance, in %: 0.0000 %
+--------------------------------------------------------------------+
| H-Energy |
+--------------------------------------------------------------------+
Boundary : back -1.2937E-02
Boundary : front -1.2638E-02
Boundary : inletair 8.3896E+02
Boundary : left -1.2930E-02
Boundary : right -1.2885E-02
Boundary : top -8.3926E+02
Domain Src (Pos) : room 1.6441E-03
-----------
Domain Imbalance : -3.5057E-01
Domain Imbalance, in %: -0.0418 %
**********************

 Felggv September 5, 2012 14:26

Have you double checked your boundary conditions? Any source terms? Is this a transient or steady state case?

Give us more information of what you're doing and how are you modeling it.

 ghorrocks September 5, 2012 19:36

For most applications you have achieved a tight convergence. But to check you are OK in your case you should do a sensitivity analysis. Do a simulation with tighter and looser convergence and see the difference in parameters of importance to you. When it converges to an accuracy you are happy with then you know you have adequately converged.

 happy September 5, 2012 20:26

Quote:
 Originally Posted by Felggv (Post 380413) Have you double checked your boundary conditions? Any source terms? Is this a transient or steady state case? Give us more information of what you're doing and how are you modeling it.
Hi,,
I did not have any source for heat or momentum. I'm sure.
my simulation is steadt state.
Tin at inetrair=1300C and the top is opening with atmospheric pressure.
left, right,front, and back B.Cs is opening with atmospheric pressure.
my modeling is about flow driven by buoyancy.
so what you think, is it enough?
I'm asking because when I did mesh independence study, my intersted variable in some importnat region can not reach constant value or at least
the variation is less than 1%.
This has killed me, I can get any progress through my research. so that I'm asking enrgy imbalnec may cause this problem.
regards
CFD user

 happy September 5, 2012 20:31

Hi again
does the model can cause the problem?
By the way, I modeled the fluid region only and I considered the walls as adiabatic as well.
any help will be great.

 evcelica September 5, 2012 20:50

You say you do not have a source term but it states you do in your energy imbalance???

fluctuations of 1% are pretty small for a buoyant flow problem as they sometimes do not have steady state solutions, 0.04% imbalance would be pretty damn good, but you should have set the reference enthalpy closer to your inlet conditions, or average conditions. You can see from looking at the numbers, your inlet and outlet enthalpies are very high compared to the rest of the boundaries, so this energy imbalance is divided by a very inflated number due to the enthalpy being so high.

 evcelica September 5, 2012 20:53

I just reread you original post, those are not the energy imbalance at the boundaries, that is the total energy going in/out of those boundaries. It is very high because your reference enthaly is probably at 20 deg C, not at 1300. Those being really high does not mean it is is imbalanced there.

 happy September 5, 2012 21:58

Quote:
 Originally Posted by evcelica (Post 380456) I just reread you original post, those are not the energy imbalance at the boundaries, that is the total energy going in/out of those boundaries. It is very high because your reference enthaly is probably at 20 deg C, not at 1300. Those being really high does not mean it is is imbalanced there.
Hi and thanks for your input
my fluid is air and my ave temp is 662.5 C.
air=O2+N2. As well I went threough the CFX help, it report that:
" he reference specific enthalpy is the enthalpy of formation at the specified Reference Pressure and Reference Temperature (often 1 atm, 25°C). The reference specific entropy is also evaluated at the specified reference pressure and temperature."
the heat of formation of air is arount 162 cal/g
entropy=162/1573=0.103 cal/ gm.K
Regards

 happy September 5, 2012 22:43

Quote:
 Originally Posted by evcelica (Post 380456) I just reread you original post, those are not the energy imbalance at the boundaries, that is the total energy going in/out of those boundaries. It is very high because your reference enthaly is probably at 20 deg C, not at 1300. Those being really high does not mean it is is imbalanced there.
Hi EriK
Do you expect that my results will change too much when I change myrefernce temp and enthalpy?
Regards
CFX user

 evcelica September 5, 2012 23:23

No the results should not change, just pointing out why they are high, and telling you those values are not imbalances, just the amount of total enthalpy traveling through those boundaries, which is based off the enthalpy at your reference enthalpy point plus the specific heat *difference in temperature from that reference point.

If you change this reference enthalpy now and try to continue iterating from your current solution, it will probably diverge, so I would not recommend changing it now.

 happy September 5, 2012 23:30

no heat source

Quote:
 Originally Posted by evcelica (Post 380455) You say you do not have a source term but it states you do in your energy imbalance???
Hi Erik,
could U plz tell me where it states you do in your energy imbalance.

Regards
CFX user

 evcelica September 5, 2012 23:36

Not sure on this, I actually thought source terms didn't show here, but in you script it says:
Domain Src (Pos) : room 1.6441E-03

 happy September 5, 2012 23:50

Hi Erik,,
I used T ref=25C and reference enthalpy=0 J/Kg& ref. enthropy=0 J/Kg.K; but now my Tref=662.5C, ref. enthalpy=681.78 kJ/kg & ref. enthropy= 728.67 J/kg. K. is it correct? or I should remain ref. enthalpy &enthropy=0 because it is reference point?
I start calculating from the beginning. my residual is going to convergence.
I should mention that i have inlet; however, I did not specify outlet where I set the top as opening with entrenment option for opening pressure.
the problem was also at the top of my domain.

Any comment could be helpful.

regards
CFX

 evcelica September 6, 2012 02:13

You are confusing me now.....
I was just trying to explain that it is not a problem. It is not an imbalance. It doesn't matter what the value is, its arbitrarily based off the reference point and reference enthalpy. Whatever you set it to shouldn't matter other than giving you different values in your enthalpy boundaries and changing you energy imbalance percentage. It shouldn't affect results. The large values at those two boundaries are not a problem.

 happy September 6, 2012 02:54

Quote:
 Originally Posted by evcelica (Post 380478) You are confusing me now..... I was just trying to explain that it is not a problem. It is not an imbalance. It doesn't matter what the value is, its arbitrarily based off the reference point and reference enthalpy. Whatever you set it to shouldn't matter other than giving you different values in your enthalpy boundaries and changing you energy imbalance percentage. It shouldn't affect results. The large values at those two boundaries are not a problem.
thanks a lot
you are right.
Regards

 All times are GMT -4. The time now is 14:35.