|
[Sponsors] |
September 28, 2012, 07:54 |
Problem with Secreening
|
#1 |
Member
Join Date: Jun 2012
Posts: 32
Rep Power: 14 |
Hello,
I'm new in Numeca world and I hope somenone could help me I have a problem launching Autoblade CFD-Screening. When I start the process (geometry made in autoblade, mesh ni AG5), the Fine Task Manager Shows: Launching IGG/Autogrid Attempt number 1 over 100 IGG/Autogrid STARTED IGG/Autogrid FINISHED Launching EuranusTurbo Attempt number 1 over 100 EuranusTurbo STARTED But it does nothing... I've looking the windows task manager and in autogrid process it works (it uses a core and about 400Mb memory), but the process in euranus (suposed nitask... in windows task manager) works at less than a half than autogrid, and only 8 Mb... Furthermore I see no action in any Module... The same occurs when I launch Fine Turbo model... I use Windows 7 64bits... Any help would be very welcome.... Thank you!! |
|
September 29, 2012, 06:21 |
|
#2 |
Senior Member
Hamid Zoka
Join Date: Nov 2009
Posts: 293
Rep Power: 19 |
Is this an optimization process?
Are you about to run an optimization? |
|
October 1, 2012, 03:12 |
|
#3 |
Member
Join Date: Jun 2012
Posts: 32
Rep Power: 14 |
Hi Hamid,
Thank you for your reply. Yes, I want to run an optimization process. So I wanted to follow the steps cited in this post: http://www.cfd-online.com/Forums/num...imization.html Is there anything I'm doing wrong? I don't know what to do to solve this problem, so I would be very gratefully to have an advise... Thank you! |
|
October 1, 2012, 04:51 |
|
#4 |
Senior Member
Hamid Zoka
Join Date: Nov 2009
Posts: 293
Rep Power: 19 |
Dear DonQuijote;
when "EuranusTurbo STARTED" is shown on Eranus window, it means that your flow domin is being solved. no other command will appear till the convergence criteria is met. in case you wish to check your solver, use the 'Monitor' which is included in your numeca package. simply open it and introduce your *.run file. It will let you track the convergence, residuals, mass flow rate, etc. does it help? regards |
|
October 1, 2012, 05:29 |
|
#5 |
Member
Join Date: Jun 2012
Posts: 32
Rep Power: 14 |
Dear Hamid,
Thanks again for your reply. I've tried what you see, but when I open the monitor (with 'euranus started'), the monitor requests a .res (residual) file, which I don't have in any folder. If I try to introduce the .run file, in the ms-dos monitor window apperears unknown module project_type, and no action is taken in the monitor... Furthermore, when I try to launch the Fine Turbo solver the same problem happens. There is no iterations in Convergence History, and in Task Manager the last two sentences are start_status_ok RECONNECTED_TO_OTHER_FINE_SESSION I really don't know where can be the problem... Thank you for your help!! Last edited by DonQuijote; October 1, 2012 at 07:05. Reason: Furthermore... |
|
October 1, 2012, 07:43 |
|
#6 |
Senior Member
Hamid Zoka
Join Date: Nov 2009
Posts: 293
Rep Power: 19 |
Dear DonQuijote;
Forget about Fine/Turbo. No data is accessible through the Fine/Turbo when you are running Design3d. Follow the steps in the below to track the residuals: -click on "add" in Monitor -Suppose your Design3D project folder's name is "A". -open it -two folders exist namely "A_computation-1" and "A_Design-1" -click on "A_Design-1" folder. -click on "_trk" folder. -click on "_flow" folder. -click on "computation_1" folder -select the *.res file this will let you track what is going on with your solver. i wish it helps regards |
|
October 1, 2012, 08:21 |
|
#7 |
Member
Join Date: Jun 2012
Posts: 32
Rep Power: 14 |
Dear Hamid,
I searched this file before and it wasn't in any folder. In the folder you mention only appears a computation1.run and job_euranus. There is a similar problem in this post: http://www.cfd-online.com/Forums/num...h-fitting.html What do you think? I will try to reinstall the software if I'm not illuminated... Than you again Hamid |
|
October 1, 2012, 09:44 |
|
#8 |
Senior Member
Hamid Zoka
Join Date: Nov 2009
Posts: 293
Rep Power: 19 |
Dear DonQuijote;
another important thing regarding Design3D is that do not use complicated addresses for your work folder. use simple adresses such as f:/ Design3D may have difficulties with complicated adresses. try placing your work folders in root of your hard drive. regards |
|
October 1, 2012, 10:35 |
|
#9 |
Member
Join Date: Jun 2012
Posts: 32
Rep Power: 14 |
I've tried it but the result is the same...
I'm getting mad... ok, I will explain my process for if I'm missing any step: 1-create my geometry and export in .geomturbo file extension 2- Open autogrid and make a new project with the geomturbo file created in 1. Save the case 3- Make a new project in Screening mode. Here I use the file .trb created in step 2 (Computation management switch to 3D associated computation) 4- Set up flow conditions 5- Solve (start euranus) Are this steps right, or am I missing anything? Thank you for your help! Last edited by DonQuijote; October 1, 2012 at 12:05. Reason: process explain |
|
October 2, 2012, 04:31 |
|
#10 |
Senior Member
Hamid Zoka
Join Date: Nov 2009
Posts: 293
Rep Power: 19 |
A couple of things come to my mind regarding your case.
1- you need to make a parametric model for your blade (with *.par extension) through a fitting process in Design3D prior to CFD-Screening. Then you need to introduce this *.par file to your case in CFD-Screening mode. 2- when you select the "3D associated computation" you need to go back to Fine/Turbo module and assign you mesh file as well as boundary conditions. In general I must say that: - If you are new to NUMECA it is rather difficult to start with Design3D module. You had better to start with Autogrid and Fine/Turbo and run some CFD cases and then switch to more complicated processes like optimization. - You also need to find experiences in "autoblade fitting process" which is very crucial in optimization processes. through this step you will find parametric models of your blade geometries. - Another tip is to study NEMCA documentations carefully. you should develop a basic understanding of the method that NUMECA applies to optimize the geometries, namely: algorthms, database generations, etc. Regards |
|
October 2, 2012, 05:18 |
|
#11 |
Member
Join Date: Jun 2012
Posts: 32
Rep Power: 14 |
Thank you for your advice.
The two steps you mention are ok, so here, unfortunately, is not the problem. In other way, I have tried to run a simulation in fine/turbo (made a new project without optimization), and the problem persists. Searching possible reasons for this problem I found that, when I open the host definition in task manager module, the host architecture states that it is a WIN32, but my computer is a WIN64 based. I don't know if it is normal, or how to change it... Thank you for your help!! |
|
October 2, 2012, 08:42 |
|
#12 |
Senior Member
Hamid Zoka
Join Date: Nov 2009
Posts: 293
Rep Power: 19 |
the Win32 is not the issue.
32-bit programs are executed in 64-bit systems quite normally. all you need is to generate a mesh, open a new project in Fine/Turo, link the generated mesh to the new project, defining boundary and initial conditions and simply running it. if you control the running from the task manager, try running it through the Fine/Turbo page itself. nothing else come to my mind regarding your case. regards |
|
October 15, 2012, 04:23 |
|
#13 |
Member
Join Date: Jun 2012
Posts: 32
Rep Power: 14 |
Problem solved.
The problem, as other posts says, is an installation problem, in my case regarding my windows 7 version. I solved the problem reinstalling the sotware in a windowsXP computer. Special thanks to Hamid, who helped me to find the problem. Best regards |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
UDF compiling problem | Wouter | Fluent UDF and Scheme Programming | 6 | June 6, 2012 05:43 |
Gambit - meshing over airfoil wrapping (?) problem | JFDC | FLUENT | 1 | July 11, 2011 06:59 |
natural convection problem for a CHT problem | Se-Hee | CFX | 2 | June 10, 2007 07:29 |
Adiabatic and Rotating wall (Convection problem) | ParodDav | CFX | 5 | April 29, 2007 20:13 |
Is this problem well posed? | Thomas P. Abraham | Main CFD Forum | 5 | September 8, 1999 15:52 |