error number 76

Tattaraaaaaaa · March 12, 2013, 03:25

Hi All,

I just started simulating engine ports-flow with starCD. I made the model in es-ice, I have set-up all the boundary and solvers. The model hasn´t shown any mesh problem during the mesh check, but while It was running for the first time an error occurred:

error 76, negative densities fond at more then 100 cells.

Since It is my first real run, after the tutorials does anyone know how to overpass this error?

cheers

Ed_89 · March 12, 2013, 09:58

Have you tried using a debugger ?? I'd check the User guide for more information

Tattaraaaaaaa · March 12, 2013, 10:38

Hi,

right now I have seen many debugfiles in the output of the run which crashed so I think is due to mesh problems. As a metter of fact I also have checked the worning and errors present in the star.info file of output and I have found bad splines definition in my model!

I guess the geometry has to be done more precisely with less splines.

Cheers

JOEROCKER · March 22, 2013, 05:29

Hi,
Even i get the same problem of negative densities. How to solve it? I checked the manual, it says "Check the result shown in the .ccmp file at the time the calculation crash occurred. Look for abnormal velocity magnitudes, or very high value of absolute pressure or temperature. The quality of cells associated with such abnormalities is generally not very high. You can also look at the .info file and search for cells associated with a high Courant number. Locate these cells in the es-ice model and fix the mesh." How to identify these cells, thje star.info file seems to be too big. Can u please help?

Ed_89 · March 22, 2013, 05:37

How far in does your simulation run ? what are you trying to simulate ?? if you can try and produce a vector plot in pro-STAR to conduct a visual check for velocity problems, if not try and check the mesh in the create results panel of es-ice, failing that you can try reducing the pressure under relaxation and your time step, if you had a simulation that run conduct post-processing on what little data you have to identify the problem (temp, velocity, pressure).

Because without having your files to look at I can't do any of this for you.

Ed_89 · March 22, 2013, 05:39

It could also be a mesh problem.. take your time and check your boundary conditions and initial conditions to ensure you have not made any silly mistakes there and let me know how you get on.

Ed

JOEROCKER · March 22, 2013, 05:51

Hi Ed, i'm trying to simulate an IC engine with methane as a fuel. The simulation starts at 348 degrees of crank angle. The simulation stops at 872 degrees. I had given an initial time step of 0.1 and later i tried to increase it to 0.05.

Ed_89 · March 22, 2013, 06:05

Have you read the best practices file ?? if you follow that as to your initial conditions etc, you should be fine, try post processing the data and check the mesh as I said at say 871 deg CA to 872 deg Ca for an interval 0.1, if the mesh has problems there you will spot them. If not drop your pressure under relaxation.

Ed

skylent · March 22, 2013, 07:30

if your engine speed is high enough, timestep=0.1deg, is always ok. maybe the mesh around 872 not very good. you can decrease the timestep to a half to decrease the courant number.

Quote:

Originally Posted by JOEROCKER

Hi Ed, i'm trying to simulate an IC engine with methane as a fuel. The simulation starts at 348 degrees of crank angle. The simulation stops at 872 degrees. I had given an initial time step of 0.1 and later i tried to increase it to 0.05.

Tattaraaaaaaa · April 12, 2013, 08:37

Hi if can help for this discussion, even if I am very late in replying, I will describe you what I found.
In my case (I simulate flow into ICE ports during compression stroke) I have looked into the debug files and there where too many splines defined at the ports and valves inlet. For this reason I had very small cells creating problems.

Hope this information can help!

March 12, 2013, 03:25	error number 76	#1
Tattaraaaaaaa New Member Join Date: Mar 2013 Posts: 3 Rep Power: 2	Hi All, I just started simulating engine ports-flow with starCD. I made the model in es-ice, I have set-up all the boundary and solvers. The model hasn´t shown any mesh problem during the mesh check, but while It was running for the first time an error occurred: error 76, negative densities fond at more then 100 cells. Since It is my first real run, after the tutorials does anyone know how to overpass this error? cheers

March 12, 2013, 10:38	answer et Ed-89	#3
Tattaraaaaaaa New Member Join Date: Mar 2013 Posts: 3 Rep Power: 2	Hi, right now I have seen many debugfiles in the output of the run which crashed so I think is due to mesh problems. As a metter of fact I also have checked the worning and errors present in the star.info file of output and I have found bad splines definition in my model! I guess the geometry has to be done more precisely with less splines. Cheers

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

March 12, 2013, 09:58		#2
Ed_89 New Member Edward Hopkins Join Date: Dec 2012 Posts: 10 Rep Power: 2	Have you tried using a debugger ?? I'd check the User guide for more information

March 22, 2013, 05:29		#4
JOEROCKER New Member JOSEPH SAJAN Join Date: Mar 2013 Posts: 2 Rep Power: 0	Hi, Even i get the same problem of negative densities. How to solve it? I checked the manual, it says "Check the result shown in the .ccmp file at the time the calculation crash occurred. Look for abnormal velocity magnitudes, or very high value of absolute pressure or temperature. The quality of cells associated with such abnormalities is generally not very high. You can also look at the .info file and search for cells associated with a high Courant number. Locate these cells in the es-ice model and fix the mesh." How to identify these cells, thje star.info file seems to be too big. Can u please help?

March 22, 2013, 05:37		#5
Ed_89 New Member Edward Hopkins Join Date: Dec 2012 Posts: 10 Rep Power: 2	How far in does your simulation run ? what are you trying to simulate ?? if you can try and produce a vector plot in pro-STAR to conduct a visual check for velocity problems, if not try and check the mesh in the create results panel of es-ice, failing that you can try reducing the pressure under relaxation and your time step, if you had a simulation that run conduct post-processing on what little data you have to identify the problem (temp, velocity, pressure). Because without having your files to look at I can't do any of this for you.

March 22, 2013, 05:39		#6
Ed_89 New Member Edward Hopkins Join Date: Dec 2012 Posts: 10 Rep Power: 2	It could also be a mesh problem.. take your time and check your boundary conditions and initial conditions to ensure you have not made any silly mistakes there and let me know how you get on. Ed

March 22, 2013, 05:51		#7
JOEROCKER New Member JOSEPH SAJAN Join Date: Mar 2013 Posts: 2 Rep Power: 0	Hi Ed, i'm trying to simulate an IC engine with methane as a fuel. The simulation starts at 348 degrees of crank angle. The simulation stops at 872 degrees. I had given an initial time step of 0.1 and later i tried to increase it to 0.05.

March 22, 2013, 06:05		#8
Ed_89 New Member Edward Hopkins Join Date: Dec 2012 Posts: 10 Rep Power: 2	Have you read the best practices file ?? if you follow that as to your initial conditions etc, you should be fine, try post processing the data and check the mesh as I said at say 871 deg CA to 872 deg Ca for an interval 0.1, if the mesh has problems there you will spot them. If not drop your pressure under relaxation. Ed

April 12, 2013, 08:37		#10
Tattaraaaaaaa New Member Join Date: Mar 2013 Posts: 3 Rep Power: 2	Hi if can help for this discussion, even if I am very late in replying, I will describe you what I found. In my case (I simulate flow into ICE ports during compression stroke) I have looked into the debug files and there where too many splines defined at the ports and valves inlet. For this reason I had very small cells creating problems. Hope this information can help!