CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)

- Main CFD Forum (https://www.cfd-online.com/Forums/main/)

- - NPARC User (https://www.cfd-online.com/Forums/main/1817-nparc-user.html)

I want to share my experiences with NPARC users. Please reply

I am using NPARC on a PC and am willing to share my experiences. It seems that most NPARC users have converted to WIND but WIND will not run on a PC if you do not have access to a unix machine to run the pre and post processor. Also most NPARC users are either government users or work on contracts to the government (except for Boeing which does commercial work with NPARC and WIND).

Because NPARC is no longer officially supported by the NPARC Alliance it would be nice to get a internet-based forum/email list/web board established.

I am glad that you want to share your experience with CFD community. Few of my experiences with NPARC are;

1) Static aerodynamics coefficient except drag are within 5% to Wind Tunnel results. I attribute incorrect drag value to Turbulence Model. I use Baldwin Lomax Turbulence Model. It doesnot show rapid convergence. I tried other Models without any success. What is your experience.

2) I use Gridgen as preprocessor and FAST as postprocessor. I have made some special purpose Grid Generators and Post Processors

3) I want to apply NPARC for dynamic simulations. Can you help me.

4) Subsonic results show slow convergece than the Supersonic simulations.

5) Have you done any Transonic validation?

6) Please tell me something about your work.

7) I will be more than willing to send you some of my results alongwith experimental results.

Regards

William, 1)your results of 5% for static aero coefficients are similar to what I see. It's been some time since I did comparisons with test data; I usually compare my CFD results to a baseline CFD case which has some test data. It seems that CFD gives more reliable results when comparing one CFD-analyzed design versus another CFD-analyzed design. Poorer CFD results with regards to drag may be attributed to insufficient grid density, inadequate turbulence model,slow convergence and probably others factors that I now have forgotten, and other unknowns. I think that using deltas for drag predictions is preferred to predicting absolute levels of drag. I use the Spalart-Almaras (sp?) turbulence model; this is supposedly a better model for wall-bounded flows. This has given me good results when predicting engine-nozzle performance.

2)I used Gridgen about 5years ago. I now use ICEM-CFD; it's buggy but seems to have more capability than Gridgen; but keep in mind that the version of Gridgen, that I used, was the version before they went public. I'vd also used FAST which is a good post processor but I now use Tecplot since it comes in Windows and I am confined to a PC/Windows98 and NT. Tecplot is very good; it has some advantages (can integrate flow quantities over streamlines) over Fast and some disadvantages (cannot easily choose portions of flow field to load into RAM).

3)There is version of PARC, version 3.1, developed by NASA/Glen, that seems to be tailored for unsteady flow. If interested I can give you a contact.

4)Yes, that seems to be a case and that may be due to the nature of the supersonic flow being hyperbolic with zones of influence.

5)I have not done any transonic validation. Have you checked the NPARC Alliance web page. Although they no longer officially support Nparc, they may have some documenation: http://www.arnold.af.mil/nparc/

6)Now, I am using Nparc to design and analyze a supersonic inlet for supersonic cruise aircraft. In the past I have used Nparc for commercial turbofan engine nozzle design, and high-angle fighter aircraft inlet analysis.

7)What kind of work are you involved in?

steve podleski, Tue, 15 Feb 2000, 11:49 a.m. WROTE:

6)Now, I am using Nparc to design and analyze a supersonic inlet for supersonic cruise aircraft. In the past I have used Nparc for commercial turbofan engine nozzle design, and high-angle fighter aircraft inlet analysis.

Hi Steve:

I am also simulating ramjet-inlet. Would you like, or any one else there we share our experiences. Although I am pretty sure, i can learn a lot of things from you or the people there about the inlet. But i don't how to start!!

I am developing a code based on roe scheme for my phd, then I will use the code to simulate the inlet. Any one there can tell me whether this is a good way to do this project.

What is the numerical scheme used in Nparc?

THanks. MJ

Mohammad, NPARC is based on central-differencing pentadiagonal system. I will give you more info when I'm in my office to access the user guide.

Is your goal to design a ramjet inlet or to develop a code and use a ramjet inlet as a test case?

Mohammed, according to the user manual, NPARC uses the pentadiagonal form of the Beam and Warming approximate factorization algorithm modified by Pulliam for ADI.

Steve, I have been involved in CFD application in design process of Flight Vehicle. I have successfully applied CFD to calculate aerodynamic coeficients on the flying vehicle to reduce the number of Wind Tunell tests- which reduce the project cost. Now-a-days I am facing a challenging problem to analyse STAGE SEPARATION, using NPARC. NPARC or any other CFD codes are not accurate in the wake. So my problem is complicated. Today I got some results. I can share them with a person having interest in this area. In the near future, I will be working on a subsonic cruise. I will successfully apply NPARC in calculting Aerodynamic forces but the engine intake and exhaust in a new thing for me. I will certainly need your help. I want NPARC results in the following cases; 1- Forebody drag of 10 degree cone at M=2 AOA=2

2- Cn,Cm,Xcp of 10 degree cone at M=0.6 AOA=2

3-Base Drag of 10 Degree cone at various Mach #

4-Temperature value at the nose at Hypersonic Mach #

Steve please suggest some sample problem for Engine intake and exhaust analysis. Tell me the contact of PARC Can somebody tell me the advantages of Wind over NPARC

Regards

Wiiliam,

From the little that I know of WIND:does not require grid overlap; variable gamma; variety of algorithms; grid sequencing; may have multi species; uses cgns(sp?) format; NPARC Alliance support and development.

One contact for Version 3.1 of NPARC (time accurate version) is Scott Townsend (fsset@lerc.nasa.gov).

A contact for some test or validation cases for NPARC: Charles Towne MS 5-11 NASA GRC 21000 Brookpark Rd Cleveland, OH 44135 E-mail: towne@grc.nasa.gov

Although Charlie may now only actively support WIND, he may offer some help with NPARC.

The home page of the NPARC Alliance is: http://www.arnold.af.mil/nparc/index.html

We did some test cases on cones but at supersonic Mach numbers. The test data is very comprehensive; the paper describing the test data may have references on tests with lower Mach numbers. Our CFD paper's abstract is shown below (more papers on NPARC are listed in http://www.arnold.af.mil/nparc/NPARC_TRS/NPARC_TRS.html):

Smith, C. F., and Podleski, S., Thin-Layer and Full Navier-Stokes Calculations for Turbulent Supersonic Flow Over a Cone at an Angle of Attack, NASA CR-189103, November 1993. Keywords: Computational Fluid Dynamics; Computational Grids; Conical Flow; Multigrid Methods; Navier-Stokes Equation; Supersonic Flow; Turbulent Flow

Abstract: The proper use of a computational fluid dynamics code requires a good understanding of the particular code being applied. In this report the application of CFL3D, a thin-layer Navier-Stokes code, is compared with the results obtained from PARC3D, a full Navier-Stokes code. In order to gain an understanding of the use of this code, a simple problem was chosen in which several key features of the code could be exercised. The problem chosen is a cone in supersonic flow at an angle of attack. The issues of grid resolution, grid blocking, and multigridding with CFL3D are explored. The use of multigridding resulted in a significant reduction in the computational time required to solve the problem. Solutions obtained are compared with the results using the full Navier-Stokes equations solver PARC3D. The results obtained with the CFL3D code compared well with the PARC3D solutions.

I am not sure if there are sample problems for engine intakes and nozzles. Maybe Charlie can help. One word of advice on supersonic inlet analysis: make sure that the boundary layer is accurately captured as it will have a siginificant effect on the compression field. As you may know, one should do a short study of the effect of grid density on the variable of interest.

Hi Steve:

> Is your goal to design a ramjet inlet or to develop a code and use a ramjet inlet as a test case?

It is not intended to design an inlet. However, it is mostly planned to apply the code to inlet in order to analyze and better understand the physics of the flow in the inlet.

Right now, I am benchmarking the code for some laminar cases, containing shock boundary layer interaction.

I don't know what are really good to look at the inlet.

Regards. MJ

Mohammed,

I'm not sure if the Mach 5 inlet test data developed by Nasa/Glenn is available to the public but you may try your luck at the NASA/Glenn Research Web site for email addresses and phone numbers of people in the propulsion division. Another good source of contacts are the AIAA Joint Propulsion Conferences papers or the AIAA Journal of Propulsion. (AIAA=Association of Aerospace and Aeronautics, I think). I wonder whether the NRC or Carleton, in the aerospace heydey, did any high speed tunnel tests; have you check that out?