How to resolve cfx error 123

Onig · February 24, 2015, 00:27

Hi
I getting the message
Ansys cfx solver exited with return code 123

Pl See the attachment...n reply your sggestions

Opaque · February 24, 2015, 07:56

In your case, the absolute pressure is out of bounds, i.e. less than 0.

You should review your boundary conditions setup, and reference pressure for consistency. Otherwise, the solver is not able to maintain the absolute pressure within bounds based while trying to converge the solution.

Is your setup a steady state analysis ? Perhaps reducing the timestep ? or is the mesh resolution enough, or too coarse ?

Onig · February 25, 2015, 05:20

Thank you for your reply
Its a steady state analysis..mesh is fine...Why cant i use mass flow rate as inlet conditions and supersonic at outlet...

Onig · February 25, 2015, 05:34

Hi
I am doing a conjugate heat transfer..Flow along with solid..two fow are seperated by solid...In one fluid domain reaction takes place and in other only coolant

How to define coolant as it is not taking part in reactions..whether coolant has to be mentioned as seperate domain...Say it is H2-O2 reaction and coolant is Liquid H2...

Hoping for your suggestions
Regards

ghorrocks · February 25, 2015, 05:56

As opaque said, the error in because the simulation resulted in negative absolute pressure. So you need to find what caused the negative absolute pressure to occur. This is often boundary condition choice. What boundary conditions have you imposed?

Onig · February 26, 2015, 01:25

I am doing heat transfer in C-D nozzle..Conditions: Inlet mass flow and outlet supersonic...And am giving rotational periodicity(in domain interfaces) for each sides as i have modelled only a sector of 60deg..

Onig · February 26, 2015, 01:37

Also if i am using a rotational periodicity for the sides in Domain interfaces...At inlet whether i have to give total flow rate or flow rate corresponding to that sector only...

ghorrocks · February 26, 2015, 04:06

You specify the flow rate for the segment you have modelled only.

Are you sure your boundary conditions are valid? If your outlet is pulling harder than your inlet can supply gas then you will get a negative absolute pressure.

Your simulation has progressed quite a way before crashing. So rerun it and output a backup file just before it crashes. Then you should be able to see where the region of low pressure is and that should help debug it.

Onig · February 26, 2015, 06:12

Thanks for you peoples suggestions..I have set pr as inlet condition and supersonic at outlet..now works fine...

Now two flows are seperated by solid...In one reaction takes place(fuel n oxi) and in other only coolant

How to define coolant as it is not taking part in reactions..whether coolant has to be mentioned as seperate domain...Say it is (gaseous H2- O2) reaction and coolant is Liquid H2...

Pls look in to my prob

ghorrocks · February 26, 2015, 16:47

If you disable the "constant domain physics" option then you should be able to have one domain a single phase single component flow and the other domain multiphase/multicomponent.

Onig · March 2, 2015, 01:38

Thank you..i could do different domains with that..

I have modelled only sector of 60 deg of the C-D nozzle...If i am giving both the sides 'rotational periodicity' in 'domain interfaces' OR if i am giving 'symmetry' in 'boundary conditions' for the sides...
1. What is the physical difference and which one is correct for my problem...Rotational periodicity or symmetry
2. Since am using 60 deg..whether symmetry conditions on sides is valid
3. Pls explain the "symmetry" and "rotational periodicity" difference?
Pls comment on my problem

ghorrocks · March 2, 2015, 04:34

You should read the documentation and a CFD textbook on this. You really need to understand the basics to set up a good CFD simulation.

Symmetry implies no flow normal to the boundary and zero normal gradient for all variables.

Periodicity implies that the values at one side of the interface get mapped to the equivalent location on the other side of the interface.

Onig · March 11, 2015, 00:49

Hi
Got a warning when i opened my result file...Did nt understood..Pls help
WARNING
Object(s) 'Domain 1 Default' referenced in object 'EXPORT/Location List' either do not exist or are currently unavailable for use. During state load this could be because the referenced object(s) have yet to be fully initialized.

Onig · March 11, 2015, 00:51

Where can i find the documentation

ghorrocks · March 11, 2015, 05:24

Look under help.

Onig · March 12, 2015, 02:06

Got it...When i do conjugate heat transfer (fluid domain where combustion occurs and solid domain of 1.2 mm thick simulating chamber) the results doesnt shows any temp variation in the solid outer wall..??
Conditions given were
1. fluid -solid domain interface - general connection
2. Heat transfer- Thermal energy for solid domain
3. No boundary conditions on solid domain..only initial temp ~ ambient 300K

Is there anything i am missing....Conduction has to takeplace right?

Gone thru documentation...but as you can see very little they mentioned abt this

Suggestions pls...(My actual objective is to know the coolant temp variation along the CD nozzle ...that is two fluid domains and in between the solid domain)

ghorrocks · March 12, 2015, 04:00

Please post your CCL and an image of what you are modelling.

Onig · March 17, 2015, 01:34

The fuel and oxi enters thru different paths n comes in contact only in chamber
Inlet - pressure and flow normal to boundary, Outlet - supersonic
sides- symmetry

Fluid - reacting mixture
reaction -h2 -o2 single step, eddy dissipation and finite rate chemistry
stoitiometry - H2-1 and O2 -0.5

SOLVER
upwind
RMS - E-4

I have put the model n the result shows low temp in chamber?? why??

ghorrocks · March 17, 2015, 03:50

You have not provided much information so I cannot say anything for certain.

Your minimum temperature is 88K. That's pretty cold. Are you sure that is correct? If the starting temperature is way too cold that might explain why the flame never gets too hot.

Onig · March 17, 2015, 08:39

What else inf is needed..can u specify..
Lower temp is 88K ...But the comb of H2-O2 should provide higher temp rt?

February 25, 2015, 05:34	Defining Domains in CFX	#4
Onig New Member GPJ Join Date: Feb 2015 Posts: 27 Rep Power: 11	Hi I am doing a conjugate heat transfer..Flow along with solid..two fow are seperated by solid...In one fluid domain reaction takes place and in other only coolant How to define coolant as it is not taking part in reactions..whether coolant has to be mentioned as seperate domain...Say it is H2-O2 reaction and coolant is Liquid H2... Hoping for your suggestions Regards

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
CFX Treatment of Laminar and Turbulent Flows	Jade M	CFX	18	September 15, 2022 07:08
CFX pressure in Simulations problem	nasdak	CFX	1	April 14, 2010 13:22
Different flow pattern between OpenFOAM and CFX	AirS	OpenFOAM	0	January 12, 2010 07:08
PhD using CFX	Rui	CFX	9	May 28, 2007 05:59
FSI using CFX and ANSYS	Bi Chang	CFX	2	May 10, 2005 04:47

February 24, 2015, 07:56		#2
Opaque Senior Member Join Date: Jun 2009 Posts: 1,804 Rep Power: 32	In your case, the absolute pressure is out of bounds, i.e. less than 0. You should review your boundary conditions setup, and reference pressure for consistency. Otherwise, the solver is not able to maintain the absolute pressure within bounds based while trying to converge the solution. Is your setup a steady state analysis ? Perhaps reducing the timestep ? or is the mesh resolution enough, or too coarse ?

February 25, 2015, 05:20		#3
Onig New Member GPJ Join Date: Feb 2015 Posts: 27 Rep Power: 11	Thank you for your reply Its a steady state analysis..mesh is fine...Why cant i use mass flow rate as inlet conditions and supersonic at outlet...

February 25, 2015, 05:56		#5
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	As opaque said, the error in because the simulation resulted in negative absolute pressure. So you need to find what caused the negative absolute pressure to occur. This is often boundary condition choice. What boundary conditions have you imposed?

February 26, 2015, 01:25		#6
Onig New Member GPJ Join Date: Feb 2015 Posts: 27 Rep Power: 11	I am doing heat transfer in C-D nozzle..Conditions: Inlet mass flow and outlet supersonic...And am giving rotational periodicity(in domain interfaces) for each sides as i have modelled only a sector of 60deg..

February 26, 2015, 01:37		#7
Onig New Member GPJ Join Date: Feb 2015 Posts: 27 Rep Power: 11	Also if i am using a rotational periodicity for the sides in Domain interfaces...At inlet whether i have to give total flow rate or flow rate corresponding to that sector only...

February 26, 2015, 04:06		#8
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	You specify the flow rate for the segment you have modelled only. Are you sure your boundary conditions are valid? If your outlet is pulling harder than your inlet can supply gas then you will get a negative absolute pressure. Your simulation has progressed quite a way before crashing. So rerun it and output a backup file just before it crashes. Then you should be able to see where the region of low pressure is and that should help debug it.

February 26, 2015, 06:12		#9
Onig New Member GPJ Join Date: Feb 2015 Posts: 27 Rep Power: 11	Thanks for you peoples suggestions..I have set pr as inlet condition and supersonic at outlet..now works fine... Now two flows are seperated by solid...In one reaction takes place(fuel n oxi) and in other only coolant How to define coolant as it is not taking part in reactions..whether coolant has to be mentioned as seperate domain...Say it is (gaseous H2- O2) reaction and coolant is Liquid H2... Pls look in to my prob

February 26, 2015, 16:47		#10
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	If you disable the "constant domain physics" option then you should be able to have one domain a single phase single component flow and the other domain multiphase/multicomponent.

March 2, 2015, 01:38		#11
Onig New Member GPJ Join Date: Feb 2015 Posts: 27 Rep Power: 11	Thank you..i could do different domains with that.. I have modelled only sector of 60 deg of the C-D nozzle...If i am giving both the sides 'rotational periodicity' in 'domain interfaces' OR if i am giving 'symmetry' in 'boundary conditions' for the sides... 1. What is the physical difference and which one is correct for my problem...Rotational periodicity or symmetry 2. Since am using 60 deg..whether symmetry conditions on sides is valid 3. Pls explain the "symmetry" and "rotational periodicity" difference? Pls comment on my problem

March 2, 2015, 04:34		#12
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	You should read the documentation and a CFD textbook on this. You really need to understand the basics to set up a good CFD simulation. Symmetry implies no flow normal to the boundary and zero normal gradient for all variables. Periodicity implies that the values at one side of the interface get mapped to the equivalent location on the other side of the interface.

March 11, 2015, 00:49		#13
Onig New Member GPJ Join Date: Feb 2015 Posts: 27 Rep Power: 11	Hi Got a warning when i opened my result file...Did nt understood..Pls help WARNING Object(s) 'Domain 1 Default' referenced in object 'EXPORT/Location List' either do not exist or are currently unavailable for use. During state load this could be because the referenced object(s) have yet to be fully initialized.

March 11, 2015, 00:51		#14
Onig New Member GPJ Join Date: Feb 2015 Posts: 27 Rep Power: 11	Where can i find the documentation

March 11, 2015, 05:24		#15
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Look under help.

March 12, 2015, 02:06		#16
Onig New Member GPJ Join Date: Feb 2015 Posts: 27 Rep Power: 11	Got it...When i do conjugate heat transfer (fluid domain where combustion occurs and solid domain of 1.2 mm thick simulating chamber) the results doesnt shows any temp variation in the solid outer wall..?? Conditions given were 1. fluid -solid domain interface - general connection 2. Heat transfer- Thermal energy for solid domain 3. No boundary conditions on solid domain..only initial temp ~ ambient 300K Is there anything i am missing....Conduction has to takeplace right? Gone thru documentation...but as you can see very little they mentioned abt this Suggestions pls...(My actual objective is to know the coolant temp variation along the CD nozzle ...that is two fluid domains and in between the solid domain)

March 12, 2015, 04:00		#17
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Please post your CCL and an image of what you are modelling.

March 17, 2015, 03:50		#19
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	You have not provided much information so I cannot say anything for certain. Your minimum temperature is 88K. That's pretty cold. Are you sure that is correct? If the starting temperature is way too cold that might explain why the flame never gets too hot.

March 17, 2015, 08:39		#20
Onig New Member GPJ Join Date: Feb 2015 Posts: 27 Rep Power: 11	What else inf is needed..can u specify.. Lower temp is 88K ...But the comb of H2-O2 should provide higher temp rt?