|
[Sponsors] |
September 28, 2007, 06:47 |
CFD PhD
|
#1 |
Guest
Posts: n/a
|
I am writing a code/solver for my PhD to simulate gas-particle two phase supersonic impinging jets. What are you views on writing your own codes during a PhD canditures? Is it better to modify existing codes and focus on the analysing the results instead?
Shuo |
|
September 28, 2007, 10:53 |
Re: CFD PhD
|
#2 |
Guest
Posts: n/a
|
I personally consider modifying an existing code to be masters' level research.
I have seen Ph.D. candidates using commercial codes and providing user subroutines for their dissertation, and I feel that route provides neither the insight and thorough understanding of the numerical methods, nor the original research that a Ph.D. is supposed to require. Just my own opinion, take it at face value. |
|
September 28, 2007, 12:17 |
Re: CFD PhD
|
#3 |
Guest
Posts: n/a
|
It depends. If your focus is on numerical alogorithm development, then it makes sense to write your own code. However if you have some code available to you, I would suggest modifying it. Nowadays there are many good codes available, see the cfd wiki for links. If your work is more on understanding the physics, then writing your own code may not be important, as long as you have access to a good code and you understand what it does.
|
|
October 1, 2007, 14:48 |
Re: CFD PhD
|
#4 |
Guest
Posts: n/a
|
Yeah, it really depends on what your research focus is. For instance, if you are dealing with reactive flow modelling, it may make sense to develop your own code for the chemistry, and (only if you have time), the flow itself (where well-established codes are available at your disposal). The same goes for projects involving multiple components (experimental-numerical, for instance).
Ideally, you would want to do everything yourself from scratch, but you don't really want to drag your ph.d. out too long. Life is really too short. BW. |
|
October 2, 2007, 19:36 |
Re: CFD PhD
|
#5 |
Guest
Posts: n/a
|
I personally don't think there is a sweeping statement that covers every case. It is a rather case specific thing. I think you also would need to decide of the degree is focusing on a fluid mechanics problem and associated interpretation vs. becoming doing a research in applied mathematics. The analogy would be like doing an experimental investigation. If one is not careful and want to use only the latest non-intrusive techniques, then one risks the possibility of devoting most of his/her time becoming an electrical engineer developing instrumentation, instead of investigating a fluids problem using the best information within a given time frame. Similarly, if you do a degree in a theoretical framework, you do not need to invent new math.
|
|
October 3, 2007, 15:59 |
Re: CFD PhD
|
#6 |
Guest
Posts: n/a
|
<blockquote>If one is not careful and want to use only the latest non-intrusive [experimental] techniques, then one risks the possibility of devoting most of his/her time becoming an electrical engineer developing instrumentation, instead of investigating a fluids problem using the best information within a given time frame. Similarly, if you do a degree in a theoretical framework, you do not need to invent new math.</blockquote>
FYW ... this is so true. I couldn't agree more with you. BW. |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
PhD in China (part 2) | Rui | Main CFD Forum | 9 | June 9, 2007 09:40 |
Question regarding CFD PhD | Mateo | Main CFD Forum | 4 | January 5, 2006 04:52 |
CFD JOBS and Expected Salary.... | Noel Harrison | Main CFD Forum | 11 | November 22, 2000 08:15 |
ASME CFD Symposium, Atlanta, 22-26 July 2001 | Chris R. Kleijn | Main CFD Forum | 0 | August 1, 2000 11:07 |
Which is better to develop in-house CFD code or to buy a available CFD package. | Tareq Al-shaalan | Main CFD Forum | 10 | June 13, 1999 00:27 |