CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Siemens (
-   -   spray injection (

Johannes P. January 8, 2009 06:55

spray injection

I wanna simulate an injection of fuel(Dodecane(L)) into a high pressure cell. I wanna have a look to the length of the injected fuel jet and to the diameter and velocity of the fuel droplets.

So, the problem is that the length of the jet is still constant. Doesn't matter if the pressure in the cell is initialized to 1e+5Pa or to 1000e+5Pa. So, what's wrong with my simulation?

Maybe you've an idea;-) Thanks a lot.


James January 8, 2009 08:39

Re: spray injection
You don't say which code. Is momentum coupling turned on ?

Johannes P. January 8, 2009 10:59

Re: spray injection
Hey! what do you mean with "code"? It's an Lagrangian Multi-Phase spray injection with atomization.The Atomization model is the "Huh-Model", the "MPI-2" is the nozzle model. Heat, Mass and Momentum Transfer are Standard with 0 for mass coef. and 1 for the slip factor. there is one inlet and one pressure boundary created. I wanna simulate pressures between 1e+05 and 100e+05Pa. That I can see how the pressure influences the length of my jet and the velocity and diameter of my parcels.

Thank you very much, johannes

Johannes P. January 8, 2009 11:23

Re: spray injection
ah, ok, it's the code of version 3.26

sorry, don't have the best english ;-)

Pauli January 8, 2009 13:39

Re: spray injection
Sounds like your doing a bunch right. But something is definitely wrong. You should get liquid length changing with pressure. (I did.)

Don't use the MPI nozzles unless you have the appropriate licenses. They appear to work but give strange results. Use the default nozzle.

Are you using variable properties? You should.

Are you sure you are getting the pressure you expect? Are you running compressible or incompressible? Do you have the correct reference pressure for your pressure boundary? Where is your reference cell. I suspect this is where your problem lies.

James January 9, 2009 13:54

Re: spray injection
Do you see the influence of the spray on the fluid ?

Pauli January 9, 2009 14:27

Re: spray injection
Yes I have momentum transfer from the liquid to the gas. Maximum gas velocities are a fraction of maximum liquid velocity. The exact value is grid dependent. I've seen values between 1/3 & 1/2.

Johannes P. January 14, 2009 05:43

Re: spray injection
Hey, thank you very much for your answers!

I'm a student and i'm working at the university one day a week.So, sorry for the late answer, but i'm just at the working place once a week.

Yeah,somthing is wrong...but don't know what ;)

Ok, for the next calculation i will use the standard nozzle model. but i guess that the university has the appropriate licenses for MPI model.

With variable properties you mean the molecular properties of the fluid inside the pressure cell, or? But you're right, i use constant ones.

I'm pretty sure that i get the right pressure, cause i made a pressure plot and i get the pressure i initialized.

For my calculation, the vapor and the liquid phase are incompressible.

What's the correct pressure for my pressure boundary? My reference cell is a cell inside my volume...

So, hope you can help me...

have a nice day and thaks a lot for your answers.

Best regards,


James January 14, 2009 09:38

Re: spray injection
If you have a pressure boundary then it overrides the pressure cell location but its value is relative to the pressure reference value.

All other things being equal if you only have one pressure boundary then STAR typically initializes best if you make the pressure reference value the same value as the desired value for the boundary and make the boundary value 0.

All times are GMT -4. The time now is 18:43.