Hi there,

I am a PhD student studying the optimisation of fluid flow in Grinding. I have been investigating internal fluid flow in grinding coolant nozzles for the past year or so. I've tried to advance this to look at nozzle exit flows and examine the point at which they "break up" but have met with several problems.

I was wondering if anyone on the forum had knowledge of this kind of simulation?

The first problem Is a relatively simple problem I believe but I just can not solve it. When I mesh in CFX mesh then take it to CFX-Pre through Workbench the model transfers no problem. However, when transfering from ICEM to CFX-Pre the model units change from mm to m for some unknown reason. Does anyone have the answer to this?

My next problem refers to my simulation of exit flows. I have been modelling this as both a free surface, and a multiphase model. The basic settings are Water at the inlet at 3 m/s and air in the box surrounding. This leads me to my next question. Do I need to use this box as it apears to restrict the flow and if its bigger the computational power increases? Is there someway of meshing "free space" so to speak?

The next questions is would anyone know why the results are like the picture shown here? ....

http://img526.imageshack.us/img526/6175/str8velxm8.jpg

...

They exit jet from experimental work is 10 times this size.

Any help would be really apreciated. Does anyone have any examples or papers on this type of CFD work. Jets exiting into air etc?

Yours

Vadim

PS. Im using ANSYS 10 with CFX and ICEM.

Hi,

ICEM files do not contain units. CFX-Pre just assumes an ICEM mesh file is in metres by default (or whatever you have set as default) so if it is another unit system you have to tell it.

Your spray break-up simulation is not a trivial problem. You should be able to find plenty of general literature on it. Also check on the CFX community page for literature using ANSYS software.

Also you may find the multiphase enhancements in CFX11 will assist in this type of simulation. In particular I have found the new coupled momentum/volume fraction solver to help for similar simulations.

Glenn Horrocks

Thanks for the comments so far Glenn.

RE> The units. How do I set the default in CFX-Pre then. I start through workbench and see nowhere how to select (m) or (mm) like the options I get in design modeller for instance. This isnt a problem however as If this is the case I will jsut model the system in (m).

I am only jsut starting my Literature review in this area. Ive actually wrote my PhD up but need this to confirm my experimental program, left it a bit late but oh well. I've searched the literature on many sites but have jet found this type of simulation. Could you advise as to possible papers?

Thanks for your advice on CFX 11, we are looking into purchasing the new software with support foir this my final year.

Thanks once again

Vadim

Hi Vadim,

you can scale the geometry in Pre by a uniform factor of 0.001. Then everything appears to be in mm.

Gert-Jan

Hi Gert-Jan,

Thanks for your response. I have solved that problem now due to yours and Glenns help. The problem I face now is how to overcome this difference between experimentation and simulation. I know it must be a fundamental problem, but guess its going to take lots of research before im anywhere close.

Thanks for your help so far. Any suggestions on how to model the legnth of break up for jets of water would be gratefully apreciated.

Regards

Vadim