CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   AVL FIRE (http://www.cfd-online.com/Forums/avl-fire/)
-   -   solution stopped (http://www.cfd-online.com/Forums/avl-fire/127155-solution-stopped.html)

giantstp2 December 4, 2013 15:19

solution stopped
 
Hi everyone!
I started a simulation with ESE Diesel, but than I countinue with the WM. I used Species Transport, Combustion and Spray module. My simulation always stoppes at a specific point, where the injection has to start. I got the following error message, that I read out from the .FLA file:

Testing proper position of injector

more than two procs found bore center
bore center shifted by about 1.e-6m.

Bores could not be located inside domain.

Check position of injector !!!

Test of injector position failed !!!

Bores could not be located inside domain.

Check position of injector !!!

Test of injector position failed !!!

Bore center at gridline or outside grid !!!

Bore center at gridline or outside grid !!!

Calculation continues.

====================================

Execution terminated! Bore seems partly outside domain.

Execution terminated! Bore seems partly outside domain.

Tried more than 1000 times to locate parcel in bore.

Tried more than 1000 times to locate parcel in bore.
more than two procs found bore center.


Can anyone help me?
Best regards, Edem

giantstp2 December 5, 2013 02:15

Actually, I find out that the program does not like that I adjust something different value from default for NOZZLE DIAMETER AT HOLE CENTER POSITIONS.
I would like to adjust 0.000184 mm nozzle hole diameter. Can anyone tell me, where can I adjust it?
thanks

chang.gaikwad December 6, 2013 03:32

Go to Modules -> Spray -> Particle introduction methods -> Nozzles -> nozzle [1] -> 'Geometry data' tab

amin_u50 December 8, 2013 14:43

Hi...
I think you should change your injector position and set it in the domain mesh. You check the nozzle position in the mesh=>Geo info=> distance and choose correct injector position. then copy this X Y Z in the nozzle position.
Best regards
Amin_u50

giantstp2 December 10, 2013 05:28

Thanks for the answers, they were useful!

But unfortunately I have one more problem. I used to do the simulation with a fifth piston model, because the injector has 5 hole, and it runs correctly. When I do a simulation with the full model, beacause of the bowl center offset, the calculation stoppes at the first timestep. I did the meshing in ESE Diesel.
From the .Fla file I got the following error message:
"Problem diverges: Momentum transport equation, variable: X-velocity, material: 1".
I read the topic about this, but there is not too much information.

I hope You can help me, best regards!

amin_u50 December 10, 2013 10:54

Hi...
ooook... I don not understand. You simulate one-fifth section of model or all??
I think if you use one-fifth of model you should reduce your fuel injection rate and divide fuel rate into five section because you are simulating one-fifth. Your fuel rate is more than real amount for one injector and you will face this error

Best regards
Amin_u50

giantstp2 December 10, 2013 12:08

I have started with an one-fifth model, which runs correctly. Yes, I applied the one-fifth amount of the injected fuel for one nozzle hole, so I adjust the one-fifth of the total mass:
NOZZLE[1]->GEOMETRY DATA->NUMBER OF NOZZLE HOLES=1
NOZZLE[1]->GENERAL NOZZLE DATA->MASS IN COMPUTIONAL DOMAIN: (total mass)/5.

The problem is that I have to do a FULL model, because on the piston there is a bowl offset. In this case MASS IN COMPUTIONAL DOMAIN=total mass, it is ok.
The simulation stoppes immediately at the first time step, at the compression part of the calculation.
All of the parameters I have adjusted is the same as in the one-fifth model, so I think the reason of the divergence is the meshing. But I do not know what is it exactly...

amin_u50 December 10, 2013 13:53

I do not know why you face this error? You execute meshing process by ESE or by WM??!!! I think you should decrease the size of meshes because fuel accumulate in meshes and your calculation go to divergence.

chang.gaikwad December 11, 2013 05:53

you might want to check the initialisation of computational domain too.


All times are GMT -4. The time now is 07:17.