Hi, all colleague,

I want to simulate the unsteady flow in turbomachinery, but I don't know which comercial softwares are the best in simulating it. Any one can give me some suggetions?

Regards

Chen Jiang

You have two major codes in this area: (1) CFX and more specifically Tascflow, which has a base in Waterloo, Canada, if I am correct and (2) Numeca. I use the first one to look at pumps, including pumps with mobile blades, and compressors.

Which one is the best in simulating unsteady flow in Turbomachinery between Tascflow and Numeca?

Thanks

There is nothing like the the best code unless you tell us a bit more about your problem. I am using StarCD for transient moving mesh calculation and it works quite well for all kind of pumps and compressors (not only axial and radial machines).

The problem with tascflow is, that it will be merged into CFX and you might have problems to get it as tascflow. I'm not sure about the transient capabilities of Numeca. They don't have anything about this on the homepage. Once I asked about this here during a discussion it suddenly stopped ... ;-)

We are studying a techniques called Edge-matching which will be used in unsteady design of turbomachinery. we need a robust, accurate and rapid unsteady simulation software.

Look at CFX Tascflow (used by most pump constructors worldwide for example) and its module in CFX-5.6, which include predefined tools amd libraries to his effect.

Hi guys,

As an employee of Numeca, I am not objective in this discussion, of course. But I just want to let you know that our software FINE/Turbo has transient capabilities, including for multi-stage axial or radial turbomachinery.

The prupose here is not to enter into a commercial discussion so I will not argue too much about our code vs. competition. The only thing I would like to say is that, in recent European Project called HPNURSA (acronym of High Performance Navier-Stokes Unsteady Rotor-Stator Analysis) which gathered different industrial companies and CFD vendors, the NUMECA software proved to be excellent on different points of view and more specifically: - the accuracy of the unsteady results - the performances of the parallel capabilities of the flow solver: low RAM per node, excellent scalability up to 16 processors (which was the maximum of proc. available for this project), limited number of sub-iterations needed for each physical time step, leading to a very reasonable global CPU time.

During this project no formal comparison was made between the different CFD codes: different solvers were tested on different testcases. However, objectively I can say that our industrial partner (Snecma) was extremely pleased with our final results and performances of our software FINE/Turbo.

If you are interested to know more about the transient capabilities of FINE/Turbo, do not hesitate to contact me by e-mail. In the meantime, I can already give you the reference of an interesting paper presented in Lausanne in September 2002 where, in particular, our results obtained in this HPNURSA project are presented:

"APPLICATION OF THE DOMAIN SCALING APPROACH TO THE CALCULATION OF THE UNSTEADY FLOW IN A ROCKET ENGINE TURBOPUMP", A. Demeulenaere, E. Lorrain, D. Lemoal, JM. Nguyen, S. Laffite, Proceedings of the Hydraulic Machinery and Systems 21st IAHR Symposium, September 9-12, 2002, Lausanne

Season's Greetings to everyone !

Bruno

I have been away from this excellent website suffering severe rheumatoid arthritis. I am pleased the cfd world of turbomachinery continues to thrive with most codes used wisely of proven worth. CFX Tascflow is one of the very best.

My world of rotodynamic pumps can be found at this address: http://fortunecity.com/meltingpot/broadway/1211/

I would welcome examples -papers from those who can add to my list of updated information references.

Bruno,

Thanks a lot for your detailed information.

we will try the code NUMECA to do our research.

Regards

Chen Jiang & Ji Lucheng