CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   MISES code for analysis of cascade flows (http://www.cfd-online.com/Forums/main/3631-mises-code-analysis-cascade-flows.html)

Quamber H. Nagpurwala June 22, 2001 08:25

MISES code for analysis of cascade flows
 
Friends,

I am looking for the MISES code for analysis of 2-D viscous flow through compressor and turbine cascades. Could you please transmit the same to me if you have this, or else could you please tell me the source from where this code could be obtained, preferably free of cost. Thanks.


andy June 22, 2001 09:04

Re: MISES code for analysis of cascade flows
 
MISES is an MIT code that is in use (to some extent) at Rolls-Royce who may well have paid for its development (I am not certain about that). I have used it briefly at DRA and it seems to work well but it could probably benefit from an improved trailing edge treatment for turbines (this is not a major criticism - it is a useful code).

As far as I know it is not publicly available (or listed at MITs web site but I may have looked in the wrong place). I too would be interested in obtaining a copy if anyone knows that is available for student/research use. My guess (and it is only that - I do not know) is that Rolls-Royce have some ownership rights and you would probably have to get permission from both RR and MIT.


Matt June 26, 2001 06:42

Re: MISES code for analysis of cascade flows
 
MISES is available from Analytical Methods Incorporated, or one of their distributors.

Visit the AMI web page at www.am-inc.com, or visit our web page at www.acel.co.uk which has a link from one of the CFD pages. (we distribute AMI codes in Europe)

Matt

andy June 26, 2001 07:42

Re: MISES code for analysis of cascade flows
 
Matt,

There is a server error at AMI.

Can you confirm that this is the MIT MISES code or something else?

Has the code been developed further in the last few years and, in particular, what is the current treatment of the trailing edge?

Andy.


Quamber H. Nagpurwala June 26, 2001 08:11

Re: MISES code for analysis of cascade flows
 
I too could not open the website www.am-inc.com. The other suggested site www.acel.co.uk also does not seem to give any info on MISES. I am sure the code is being used by many researchers for analysis of cascade flows. Would any of the user like to share the code with me and Andy? Thanx.

andy June 26, 2001 08:49

Re: MISES code for analysis of cascade flows
 
Quamber,

AMI and ACEL are commercial companies which sell code. MIT has presumably sold AMI some rights to the code in the same way they did with Visual3 and ICEM. I think there is no chance that anyone is going to send you the source code or even a binary that you can use without a license. I certainly do not want anyone to send me the code unless the students here have a right to use it.

I am curious about where you saw the code and what persuaded you it was a good code to use since I am not as sure as you about its widespread use.


Peter June 27, 2001 17:03

Re: MISES code for analysis of cascade flows
 
I thought also MISES was a commercial code which has adopted several releases dependending on the company. It is not only Rolls-Royce who is using MISES at the moment.

Surely, you will have to pay for a release since I doubt people of RR want others to use their tools. Whether there is or not a academic release at low cost, I can not tell. If it exists, it will be good to have it.

Researchers see pros and cons in MISES: In one hand, because of the model being used (Euler) it calculates increadibly fast, something that is crucial during arfoil design. On the other, the validity of the model must be taken into account: it doesn't calculate flow features on the BL (it just calculates the displacement thickness). Forget any attempt to calculate radial forces since it is 2D. Anyway, what it is sure is that is currently used to make blades.

Peter June 27, 2001 17:06

Re: MISES code for analysis of cascade flows
 
About inconsistences in the trailing edge?. Well, this is an Euler code and therefore there will be always an open trailing edge. If you want to see what is happening in the wake it is better to use and NS code.

andy June 28, 2001 07:16

Re: MISES code for analysis of cascade flows
 
When I initially responded to this thread I only knew about MISES as a research program written by MIT: no GUI, driven by a few simple text files and not as developed as the current commercial version. This is the version I had used briefly a few years ago. The commercial exploitation was news to me and I may have muddied the waters somewhat. I doubt Rolls-Royce use the commercial version given their internal design system but I could be wrong. At the time, a few years ago now, MISES was only partially plugged into their internal system and hence not greatly used. I do not know the current situation and certainly do not want to be seen to be plugging "MISES as used by Rolls-Royce".

I have enquired about academic supply but since we are dealing with distribution companies and a design-orientated product I doubt they will be able to help at around media-cost prices.

I am not sure I would describe a full Newton scheme as incredibly fast. MISES was certainly substantially slower than the 2D finite element inverse method potential flow solver that was used at the time. It typically required at least a minute per point and sometimes more depending on the flow condition. It could also fail for the more challenging flow conditions requiring a degree of parameter tuning, smaller steps towards target distributions, etc... One was not able to stick it in the background to calculate a curve or two and get on with something else. This may be better developed in the commercial versions.

However, given the limitations of any 2D design approach for modern blading it seemed, to me, to include about the right amount of physics, was fairly robust and sufficiently fast to be an appropriate tool for the task.

Quamber H. Nagpurwala June 28, 2001 08:15

Re: MISES code for analysis of cascade flows
 
Andy,

I do not want to eleborate who are the users of MISES, because I do not know how they have obtained the code. To name one, there is published literature from DLR, Cologne, discussing computed results of compressor cascades obtained using MISES.

As far as I know it is a fairly good tool for obtaining quick results on the performance of the cascade in terms of Mach number distribution and loss parameter. It may not be as accurate as an NS code, but is okay for quickly comparing various blade designs.

I agree that it is too much to ask for a commercial code free of cost. What I need is the information on proper source from where the code can be obtained at nominal cost.

andy June 28, 2001 09:13

Re: MISES code for analysis of cascade flows
 
This, I believe, is where it came from:

http://raphael.mit.edu/

I see there is a note from 1994 on MSES indicating that it is freely available for academic teaching and research from Mark Drela (one of the authors of MISES) and also for commercial use from a licensing officer. There is no mention of MISES but this looks like a pretty good place to try.


Peter June 28, 2001 14:14

Re: MISES code for analysis of cascade flows
 
Andy,

surely we are talking about different versions. When I said incredibly fast, I meant MISES can do arfoil calculations in much lower time than any NS solver. That is why it is used in the early stages of development process in some industries. This is possibly the key point why 3D solvers have not achieved full development in early steps of the design process: It takes too long to compute. Alternatively, some simplified tools with simple flow models can provide to some extent close-to-reality predictions with moderate computational time


All times are GMT -4. The time now is 22:58.