CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Main CFD Forum

CFD in a nutshell

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Display Modes
Old   June 26, 2000, 10:44
Default Re: CFD in a nutshell
  #21
John C. Chien
Guest
 
Posts: n/a
(1). I think the drag is an important issue for the commercial aircraft. (2). But it is a different story for a fighter aircraft. (3). If you are going to test it, it is a good idea to convince someone to do the cfd analysis for you. In this way, you still have a chance to find out whether cfd is useful or not. It will not give you the accurate value, but it will tell you whether there is something wrong with your design.
  Reply With Quote

Old   June 26, 2000, 11:01
Default Re: CFD in a nutshell
  #22
John C. Chien
Guest
 
Posts: n/a
(1). Without knowing your specific type of aircraft model, the size of the model, the Mach number range of the test conditions, etc. It is hard to know which wind tunnel to use. (2). Sometimes ago, there was a free multi-block structured mesh code available on this forum, specifically written for the commercial aircraft applications. You may want to dig out that old piece of information. I am sure that there is also a piece of the solver part. Whether the solver is free or not, you need to check into it. (3). I am sure that it is for commercial aircrafts. (from Germany I think.)
  Reply With Quote

Old   June 27, 2000, 12:13
Default Re: CFD in a nutshell
  #23
CFD Dummy
Guest
 
Posts: n/a
A fighter!? Well, I am not sure a fighter needs even aerodynamics nowadays, and designers of those things use CFD now probably only to reduce the radar cross section.

In any case, twice in this thread you've mentioned about convincing someone to do the CFD analysis for me for free. I'd like to see what this CFD thing can really do and compare it to the wind tunnel results, but as I am CFD Dummy I don't know too many people in the field. Do you know of anyone to approach? Any volunteers?
  Reply With Quote

Old   June 27, 2000, 18:57
Default Re: CFD in a nutshell
  #24
John C. Chien
Guest
 
Posts: n/a
(1). Why don't you propose a joint effot, in which you provide the test data to validate the vendor's code. (2). Then e-mail to the vendors to get them interested in it. There is always a 50-50 chance to get good results. This will give the vendor some confidence and publicity. At the same time, you have both the test data and the cfd results.
  Reply With Quote

Old   June 27, 2000, 21:43
Default Re: CFD in a nutshell
  #25
CFD Dummy
Guest
 
Posts: n/a
John -- A couple answeres up you stated that "It is unlikely that you will get anything from the commercial cfd codes." So who are the vendors here?
  Reply With Quote

Old   June 27, 2000, 21:52
Default Re: CFD in a nutshell
  #26
CFD Dummy
Guest
 
Posts: n/a
Tullahoma? Heh? There used to have a couple of lovely barmaids in the local pub there. I tell you, spending your time there beets watching a computer.
  Reply With Quote

Old   June 28, 2000, 03:57
Default Re: CFD in a nutshell
  #27
Jonas Larsson
Guest
 
Posts: n/a
What kind of questions is it that you want to have answers to? Are you looking for exact drag-predictions, stall-angles etc. or are you just looking for basic pressure-distributions?

You can probably get decent pressure-distributions with CFD much cheaper and quicker than in a wind-tunnel. And in CFD you can easily make small modification of "flaps", "wing-geometry" etc. and see the result in a day or two.

If you want to validate a finished design, get exact drag-predictions etc CFD is difficult and you probably need to do measurements anyway.
  Reply With Quote

Old   June 28, 2000, 07:22
Default Re: CFD in a nutshell
  #28
John C. Chien
Guest
 
Posts: n/a
(1).Take a look at the listing of the sponsors of this forum, they are cfd code vendors. (2). Some have experience with government labs in the aerospace applications. (3). Not all vendors are listed here, but you can search the Resources sections for other cfd code vendors, developers.
  Reply With Quote

Old   June 28, 2000, 18:48
Default Re: CFD in a nutshell
  #29
CFD Dummy
Guest
 
Posts: n/a
Can you recommend a good code for that? Preferably one that doesn't require a supercomputer and even a dummy can learn before Christmas?
  Reply With Quote

Old   June 29, 2000, 04:23
Default Re: CFD in a nutshell
  #30
Jonas Larsson
Guest
 
Posts: n/a
CFD is not for Dummies ;-) It is difficult to give any general advice without further information. How complex is your geometry? What kind of CAD format do you have it on? What kind of information do you want from your CFD simulation (see my previous msg)? What kind of computers do you have access to? How big is your budget? Are you a US citizen with access to NASA codes? If you want low-cost CFD software you should take a look at Bert Laney's excellent overview of Free and Low-Cost CFD software - there is a link to it in the "Resources/Software" section of CFD Online.

  Reply With Quote

Old   June 29, 2000, 16:50
Default Re: CFD in a nutshell
  #31
CFD Dummy
Guest
 
Posts: n/a
CFD is not for Dummies? Well, there lies my problem and the reason why I want to stick with a wind tunnel. You can let even dummies in there and you still get pretty decent results.

But to answer your questions: 1 My geometry is fairly complex. (Like an airplane ... You know...) 2 I can put it pretty much in any cad format. 3 My ultimate goal is to get accurate L/D and stability data, but if it's free I'll take less. (In other words, depends on the cost). 4 I have access to 486SX25 with 4 meg RAM. (No, seriously, I can probably come up with a decent PC.) 5 Budget? Well, I'll need to get the data one way or another, so what I can save on a wind tunnel I can spend on CFD. (Let's say up to 20k). 6 Last time I've checked they haven't revoked my U.S. citizenship yet. 7 I've posted my question here after looking at Bert Laney's excellent overview of Free and Low-Cost CFD software (i.e., I am not much smarter from that).
  Reply With Quote

Old   June 29, 2000, 20:35
Default Re: CFD in a nutshell
  #32
John C. Chien
Guest
 
Posts: n/a
(1). Because of your interest in cfd, and also wind tunnel testing, you maybe able to get both at the same place, that is the wind tunnel testing facility. (2). What I am saying is, most wind tunnel facility by now, should also have a 3-D cfd code to help them validate the test results and improve the facility. (3). One such example is the NPARC code and Air Force AEDC, and NASA/Glenn(?). (4). Many wind tunnel facilities have long been using cfd codes to help validate the test results. So, you can also request the engineer there to do the same validation on your test results. (in this way, you can stay happily forever as a cfd dummy. What do you think?)
  Reply With Quote

Old   June 30, 2000, 18:02
Default Re: CFD in a nutshell
  #33
CFD Dummy
Guest
 
Posts: n/a
No can do. The two wind tunnel facilities I am thinking of do not have high enough opinion of CFD to use it. I wonder why?
  Reply With Quote

Old   July 7, 2000, 10:17
Default Re: CFD in a nutshell
  #34
Ed Blosch
Guest
 
Posts: n/a
$30K + $300 hours * O(100) labor hours, is < $100K. CFD (you doing it, in-house) will cost you $code + $computer + $time, and will definitely be >> $100K. CFD outsourced as a contract, to the right group (I know of a few), might be as low as $100K (only paying for $time). Assuming you're talking about 100 simulations, $100K is ballpark for a CFD group with prior experience in wing-body simulations and a mini-supercomputer (youve got no chance with a PC on this one; you'll need O(1million) nodes and run-time per case of 3 hours or less to make $100k).

So in your position, wind tunnel looks best, unless you have a strong desire for detailed understanding of the flowfield. Or a need for fast turnaround time. If you can't book time at the tunnel until 2002, CFD may be more expensive but will get you there on schedule.

You need to take the "hype" in the proper context. $CFD < $wind_tunnel is a perspective that is valid only for an airframer or other such large company, in the event that both ventures would be pursued and supported in-house. It was never aimed at a potential individual user. For the large company, wind-tunnel costs include the cost of buildings and power and maintainance. But CFD only requires floor space and computers. That's the steady-state scenario, which comes true after 5 to 10 years of CFD development. That's the basic argument. It doesn't apply to you.
  Reply With Quote

Old   July 8, 2000, 00:29
Default Re: CFD in a nutshell
  #35
John C. Chien
Guest
 
Posts: n/a
(1). I think, there is a long way to go before everyone is convinced that CFD is a viable approach in design and analysis. (2). A project manager had asked me that "what is cfd ?" (I can't tell you the time and the place). For me to hear that, it makes me to feel that I am actually living in a virtual world. (3). And in one technical review, one manager think that to do mesh refinement, to use Y+=<1.0 for low Re model are taking too much time. And he was surprised that I said, in a similar case, someone has already established the mesh sensitivity effect, that is even at the one million cells, it was still not adequate to guarantee the mesh independent solution. (4). So, the wide spread use of cfd code is really not a good sign. You can derive the real befenit only if parties involved all understnad the value and the practice of cfd. (5). In other words, "do it right in cfd" seems to be the only way to guarantee a reliable solution. And thus promote the quality of the solution, and also the image of cfd as a viable design and analysis approach. (6). The negative effect was basically because cfd is not being practiced properly. This also includes the necessary training in using a code. This is particularly troublesome in a large company, where there are many cfd or cae codes. (7). What I am trying to say is , the current environment is still treating cfd as equivalent to word processing. And when the "easy to use", "user-friendly", "automatic meshing", etc... got into the mind of managers, the end result is most likely the "chapter-11". (8). So, I would like to promote old fashion approach, that is, let the professional consultants (or companies) to handle the cfd work, unless the company has the full staff of cfd engineers at the PhD levels.(senior engineers can work with these experts) (9). So, like the rise and fall of Internet .com companies, the situation applies equally well to the product oriented companies. They can't survive, unless they are really good. (10). So, the idea is to help the professionals in the right way, instead of trying to help everybody understand cfd. And we know that it is impossible.
  Reply With Quote

Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Future CFD Research Jas Main CFD Forum 10 March 30, 2013 13:26
CFD Wiki - We Need More Help! Jonas Larsson Main CFD Forum 0 September 26, 2005 08:11
CFD JOBS and Expected Salary.... Noel Harrison Main CFD Forum 11 November 22, 2000 08:15
Since Last June John C. Chien Main CFD Forum 3 July 12, 1999 09:38
Which is better to develop in-house CFD code or to buy a available CFD package. Tareq Al-shaalan Main CFD Forum 10 June 12, 1999 23:27


All times are GMT -4. The time now is 11:51.