[xiaoyue1990]

Hi CFX Users,

I'm doing a transient analysis for a vacuum chamber:


basic setting are presented as follows:
reference pressure: 0Pa
inlet mass flow rate: 0.0208 g/s
outlet Pressure: 60Pa

Int. condition
relative pressure 10Pa

time step
total time 3s
time step 0.05s

the model is simple and all the settings are based on the results of experiment but the solver report error 1 after 2 coefficient loops. However the solver runs okay when I improve the reference pressure to 1 Bar but it's obviously not the actual situation. So what should I do to get a result under the actual boundary conditions?

[xiaoyue1990]

+--------------------------------------------------------------------+
| Job Information |
+--------------------------------------------------------------------+

Run mode: parallel run (MPI)

Host computer: 3P1EFAABB0VR3YJ (PID:6988)
Par. Process: Master running on mesh partition: 1
Solver Build: Sat Oct 2 02:26:59 GMTDT 2010
Attributes: single-int32-32bit-novc8-noifort-novc6-optimised-su...
Job started: Mon May 21 19:57:02 2012

Host computer: 3P1EFAABB0VR3YJ (PID:9156)
Par. Process: Slave running on mesh partition: 2
Solver Build: Sat Oct 2 02:26:59 GMTDT 2010
Attributes: single-int32-32bit-novc8-noifort-novc6-optimised-su...
Job started: Mon May 21 19:57:02 2012

Host computer: 3P1EFAABB0VR3YJ (PID:12340)
Par. Process: Slave running on mesh partition: 3
Solver Build: Sat Oct 2 02:26:59 GMTDT 2010
Attributes: single-int32-32bit-novc8-noifort-novc6-optimised-su...
Job started: Mon May 21 19:57:02 2012

Host computer: 3P1EFAABB0VR3YJ (PID:12768)
Par. Process: Slave running on mesh partition: 4
Solver Build: Sat Oct 2 02:26:59 GMTDT 2010
Attributes: single-int32-32bit-novc8-noifort-novc6-optimised-su...
Job started: Mon May 21 19:57:02 2012


+--------------------------------------------------------------------+
| Memory Allocated for Run (Actual usage may be less) |
+--------------------------------------------------------------------+

Allocated storage in: Kwords
Words/Node
Words/Elem
Kbytes
Bytes/Node

Partition | Real | Integer | Character| Logical | Double
----------+------------+------------+----------+----------+----------
1 | 20005.4 | 2095.5 | 3540.4 | 80.0 | 318.0
| 3345.95 | 350.47 | 592.14 | 13.38 | 53.19
| 3245.00 | 339.90 | 574.27 | 12.98 | 51.58
| 78146.2 | 8185.4 | 3457.4 | 78.1 | 2484.3
| 13383.79 | 1401.88 | 592.14 | 13.38 | 425.48
----------+------------+------------+----------+----------+----------
2 | 20052.9 | 2123.7 | 3540.4 | 80.0 | 318.0
| 3235.39 | 342.64 | 571.22 | 12.91 | 51.31
| 2989.41 | 316.59 | 527.79 | 11.93 | 47.41
| 78331.8 | 8295.7 | 3457.4 | 78.1 | 2484.3
| 12941.55 | 1370.57 | 571.22 | 12.91 | 410.45
----------+------------+------------+----------+----------+----------
3 | 20072.5 | 2131.6 | 3540.4 | 80.0 | 318.0
| 3244.31 | 344.53 | 572.23 | 12.93 | 51.40
| 2879.02 | 305.74 | 507.80 | 11.47 | 45.61
| 78408.4 | 8326.6 | 3457.4 | 78.1 | 2484.3
| 12977.24 | 1378.13 | 572.23 | 12.93 | 411.18
----------+------------+------------+----------+----------+----------
4 | 20073.6 | 2133.6 | 3540.4 | 80.0 | 318.0
| 3201.03 | 340.24 | 564.57 | 12.76 | 50.71
| 2913.02 | 309.63 | 513.77 | 11.61 | 46.15
| 78412.6 | 8334.6 | 3457.4 | 78.1 | 2484.3
| 12804.10 | 1360.96 | 564.57 | 12.76 | 405.67
----------+------------+------------+----------+----------+----------
Total | 80204.5 | 8484.4 | 14161.6 | 320.0 | 1272.0
| 3499.17 | 370.16 | 617.84 | 13.96 | 55.49
| 3104.73 | 328.43 | 548.20 | 12.39 | 49.24
| 313299.0 | 33142.3 | 13829.7 | 312.5 | 9937.4
| 13996.69 | 1480.64 | 617.84 | 13.96 | 443.95
----------+------------+------------+----------+----------+----------

+--------------------------------------------------------------------+
| ****** Notice ****** |
| |
| No reference value for the specific enthalpy was specified for |
| material "N2 Ideal Gas". In this case, the solver will use a value |
| of zero. Please note that this is different from previous releases,|
| where the reference value was set to H_ref = T_ref*Cp. |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| ****** Notice ****** |
| |
| A zero Reference Pressure is set for an entropy recipe. |
| Material : N2 Ideal Gas |
| Entropy recipe : S=CP*LN_T-R*LN_P |
| |
| The solver avoids a floating point exception by clipping to a |
| small number. In some situations robustness problems may |
| result; eg, static-total pressure conversions for MCF fluids. |
| Consider setting the Reference Pressure to a positive value. |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| Mesh Statistics |
+--------------------------------------------------------------------+
| Domain Name | Orthog. Angle | Exp. Factor | Aspect Ratio |
+----------------------+---------------+--------------+--------------+
| | Minimum [deg] | Maximum | Maximum |
+----------------------+---------------+--------------+--------------+
| Default Domain | 9.9 ! | 699 ! | 60 OK |
+----------------------+---------------+--------------+--------------+
| | %! %ok %OK | %! %ok %OK | %! %ok %OK |
+----------------------+---------------+--------------+--------------+
| Default Domain | <1 6 94 | 6 21 73 | 0 0 100 |
+----------------------+---------------+--------------+--------------+

Domain Name : Default Domain

Total Number of Nodes = 22921

Total Number of Elements = 25833
Total Number of Tetrahedrons = 3631
Total Number of Prisms = 504
Total Number of Hexahedrons = 14242
Total Number of Pyramids = 7456

Total Number of Faces = 11218

+--------------------------------------------------------------------+
| Average Scale Information |
+--------------------------------------------------------------------+

Domain Name : Default Domain
Global Length = 6.7680E+01
Minimum Extent = 2.0000E+01
Maximum Extent = 2.2000E+02
Density = 9.8174E-14
Dynamic Viscosity = 1.7700E-08
Velocity = 0.0000E+00
Speed of Sound = 3.7755E+05
Mach Number = 0.0000E+00

+--------------------------------------------------------------------+
| Checking for Isolated Fluid Regions |
+--------------------------------------------------------------------+

No isolated fluid regions were found.

+--------------------------------------------------------------------+
| The Equations Solved in This Calculation |
+--------------------------------------------------------------------+

Subsystem : Momentum and Mass

U-Mom
V-Mom
W-Mom
P-Mass

Subsystem : TurbKE and Diss.K

K-TurbKE
E-Diss.K

CFD Solver started: Mon May 21 19:57:05 2012


+--------------------------------------------------------------------+
| Convergence History |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| Writing transient file 0_full.trn |
| Name : Transient Results 1 |
| Type : Standard |
| Option : Every Timestep |
+--------------------------------------------------------------------+


================================================== ====================
| Timestepping Information |
----------------------------------------------------------------------
| Timestep | RMS Courant Number | Max Courant Number |
+----------------------+----------------------+----------------------+
| 5.0000E-02 | 0.00 | 0.00 |
----------------------------------------------------------------------

================================================== ====================
TIME STEP = 1 SIMULATION TIME = 5.0000E-02 CPU SECONDS = 5.772E+00
----------------------------------------------------------------------
COEFFICIENT LOOP ITERATION = 1 CPU SECONDS = 5.772E+00
----------------------------------------------------------------------
| Equation | Rate | RMS Res | Max Res | Linear Solution |
+----------------------+------+---------+---------+------------------+
| U-Mom | 0.00 | 4.4E-11 | 2.8E-10 | 1.4E+06 ok|
| V-Mom | 0.00 | 2.0E-02 | 4.7E-01 | 5.0E-02 OK|
| W-Mom | 0.00 | 4.4E-11 | 3.1E-10 | 1.6E+06 ok|
| P-Mass | 0.00 | 1.0E-02 | 2.3E-01 | 9.1 6.8E-02 OK|
+----------------------+------+---------+---------+------------------+
| K-TurbKE | 0.00 | 7.4E-03 | 1.4E-01 | 15.4 3.0E-02 OK|
| E-Diss.K | 0.00 | 2.5E-02 | 6.9E-01 | 12.3 1.1E-02 OK|
+----------------------+------+---------+---------+------------------+
----------------------------------------------------------------------
COEFFICIENT LOOP ITERATION = 2 CPU SECONDS = 8.642E+00
----------------------------------------------------------------------
| Equation | Rate | RMS Res | Max Res | Linear Solution |
+----------------------+------+---------+---------+------------------+

+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| The ANSYS CFX solver exited with return code 1. No results file |
| has been created. |
+--------------------------------------------------------------------+

End of solution stage.

[ghorrocks]

The reason for this is because with a reference pressure of 1 bar your 50Pa pressure difference makes a gentle breeze. Nice and easy. With 0 reference pressure you have a pressure ratio of 6:1, which means supersonic flow. No wonder it is hard to converge.

So put the reference pressure at 10Pa, the inlet at 50Pa and the outlet at 0Pa (like reference pressures are meant to be set) and configure this as for a high speed compressible simulation. Expect to see supersonic flow, shock waves and all that sort of good stuff.

[xiaoyue1990]

thank you very much! But the solver still can not run normally.

[ghorrocks]

Do the tutorial examples on compressible flow. I think there is an example of compressible flow over a wing at supersonic speed.

[xiaoyue1990]

Originally Posted by ghorrocks

Do the tutorial examples on compressible flow. I think there is an example of compressible flow over a wing at supersonic speed.

But actually the compressible gas cannot applied in this case. I think It would be better to use the ideal gas (in this case, specifically, N2 or Water) to conduct my analysis since the environment pressure is rather low. So is there a way to conduct transient analysis with Ideal gas in a low pressure environment? I've tried to set the reference pressure to 60Pa and it didn't help.

Thank you

[ghorrocks]

An ideal gas is a compressible gas. In fact ideal gas is what most people use for compressible gas flow

I've tried to set the reference pressure to 60Pa and it didn't help.

I never said it would fix it, I just said it was bad practise to not use an appropriate reference pressure.

[xiaoyue1990]

Thank you for replying so soon.

Errr...isn't the definition of ideal gas is INcompressible gas?
and I've tried again, when I use air@25c, ref pressure=0Pa, it solves normally. A tutorial example of circular vent also set so and I remember in the wing tutorial it also use the air@25. So I guess the ideal gas is the problem.

Is there a way for ideal gas to solve under a low ref pressure?

[ghorrocks]

Errr...isn't the definition of ideal gas is INcompressible gas?

Ummm - no. I think you need to read a fluid mechanics textbook. P=rho.R.T, so rho = P/(RT), therefore density is variable hence it is compressible.

air@25C is an incompressibel gas with properties taken at 25C. You are looking for "air ideal gas", and use total energy for the thermal equation. Again, have you done the compressible flow tutorial examples?

Ideal gas works fine at low pressure. But it might not be too accurate, you might need to run a modified EOS. But get ideal gas working first before worrying about that.

[xiaoyue1990]

I've done the wing tutorial and it's a steady state case with the ref pressure of 1 atm. I think this example cannot applied on my case(transient, low pressure)...

[ghorrocks]

But it has showed you how to set up a compressible flow simulation. That is why I referred you to it. Of course there is no tutorial example for exactly what you are modelling.