Learning Forum

[saima]

 hi  i am simulating cavitation in fuel injector using fluent and used Schnerr and Sauer model to reproduce Mohan results published in 2014, in which 1e+07 Pa pressure at inlet and at outlet 3e+06 Pa is specified. I used diesel liquid and vapor as primary and secondary phases but the cavitation appeared in the nozzel is far less than shown in the paper . can anybody help me in this regard. 

[Rob]

Check over ALL of the settings, and also whether the mesh is suitably resolved. 

Edit: I'll also move to Fluids. 

[saima]

thanks for responding, i tried the same problem with different meshes but unable to detect the problem  

[saima]

at the beginning the model and cavitation region seems in good arrangement with theory but when the simulation proceeds the vapor fractions starts to diminish and tends towards zero i  used diesel liquid with 832 kg/m3 density, viscosity 0.0021kg/ms and diesel vapor with density 0.1361 kg/m3, viscosity 0.000005953 kg/ms, vaporization pressure 2000 pa

kindly inform me will my flow cavitate?

[Amine Ben Hadj Ali]

Check model settings and boundary conditions. If the authors used Fluent check with them. Which bcs are used in the paper?

[saima]

yes author used fluent in the paper with pressure inlet and pressure outlet boundary conditions he fixed the inlet pressure and at 1e+07 and varies the outlet pressure between 4e+06 t0 1e+06 . i'm trying to simulate with oulet pressure at 3e+06 with k-epsilon model and non equilibrium wall function (same as in the paper) but author did not mention the turbulence specification method so i take turbulence intensity as 5% and hydraulic diameter as specification method

 

[saima]

https://www.tandfonline.com/doi/abs/10.1080/19942060.2014.11015498

are these specification methods are enough to make fluid cavitate? attached is  the considered paper if anyone can help me in this regard

[Amine Ben Hadj Ali]

To know if your fluid might cavitate you can calculate the cavitation number or cavitation coefficient (denominator is the pressure difference between inlet and outlet). Are you sure that the author fixed the inlet pressure? Generally one fix the outlet pressure and varies the inlet pressure

[saima]

yes for validation purpose author compared the results with Winklhofer(2001) in which inlet pressure is kept constant and outlet pressure is varied

[Amine Ben Hadj Ali]

Then stick to the paper and try to reproduce it: transient if it is probably transient.

[saima]

thanks for your reply and i tried a lot but did not find the solution to my problem can you tell me how to check my mesh so that it is suitable for this type of flows

[Amine Ben Hadj Ali]

Ensure a mesh with good quality: ortho. quality (>0.2), skewness (<0.7) and aspect ratio (<100). Just stick to the paper as it contains almost all information. Use same Boundary Conditions. 

 

[saima]

thank you so much for your quick response i will check the mesh and come back to you if any help would be required

[saima]

hi, i am back with the same problem as i did not find time to do simulation. i started simulation yesterday and came to know that i have to put outlet far from original one to ensure that the boundaries have no impact on the flow so i put it 20mm  away from nozzel outlet (which is of 1mm only). i also ensure the mesh quality is as you described in your reply. i noted that inlet pressure is decreased by dynamic pressure during simulation (the only thing that can makes sense for me) but after some 3000 iteration reverse flow is started,  outlet static pressure (which i fix 3e+06 became 2942978 and was decreasing continuously) and vapor fractions again started to decrease because of no reason. can you tell me whats going on.

any suggestion will be highly appreciated

 

 

 

[saima]

i have added some images of simulation when reverse flow started kindly consider them and suggest something because i have stuck in there.