I am a structures engineer with a lot of FE experience but I am being asked to investigate pressure distributions on an antenna radome on an aircraft fuselage. I need these results primarily to generate net drag force values but also to assist in the structural design of the radome itself. Where should a beginner go to find out what commercial packages are out there and can someone suggest what codes they like for this kind of work?

Brian MacKay

The proper way to ask the question is: For an antenna radome of unknown shape located somewhere on an unknown aircraft fuselage, is it possible to compute the net drag force on it for structure design purposes? And the second question is: If it is possible, then is it practical for a structure engineer with FE experience ( most structure engineers have FE experience anyway.) to solve this problem using a commercial package? The last one is: If the answer to the above two questions are positive, then, where can he start? The answer to the first question is: one may have to compute the 3-D flow field over a complete aircraft including the radome to determine the worst condition. The answer to the second question is: A few months is needed to learn the code, the 3-D geometry, 3-D mesh, and the time to obtain the solution. The answer to the last question is: Any general purpose code should be able to give you the 3-D solution. ( I mean, any code which can handle flow over 3-D body should be able to do the job.) But, the ability to obtain a 3-D solution does not necessarily mean that the solution is accurate. In this case, a structure engineer with FE experience is not really an adavantage. An engineer with strong aerodynamics experience is really required in this case.

It's probably easier to ask a consulting company to do the analysis for you

Mr. Chien,

You have spent a considerable amount of effort NOT answering my question. You went on to state the obvious at least 5 different ways. Finally you conjured up your own questions and answered those instead. Thanks for your complete lack of help.

I take it from your yahoo.com email address that you are not actually gainfully employed in this field or possibly this is the official opinion of the CFD department of the YAHOO company?

For anyone with any constructive suggestions I should clarify one thing. Over the long term my company is looking to build analytical experience in this area on top of our already extensive experience in building antenna products and installing them on private and commerical aircraft. Until this point in time however we have focused on the structural aspects of the work while farming out the aerodynamic details to outside consultants.

I realize that the problems I need to solve do not push the state-of-the-art in any way. I am looking for basic suggestions on what kinds of codes are best suited to external fuselage type flow as opposed to a code designed for mold filling or electronics cooling.

Thank you

Brian MacKay

I think this is a good answer, and I agree with you.

I am sorry that I can not answer your question directly, but I think I have provided enough hints for your question. I am not with any software company, therefore, I can not evaluate your problem in terms of commercial codes. The direct answer should come from the commercial code companies. I am sure that you are free to contact them directly. But I can say that the problem you are trying to solve is not a simple problem because I have many years of direct experience in aircraft aerodynamics. Hope someone else will provide the direct answer.

Two potential sources of good codes for you are

(1) Aerosoft, Inc. www.aerosft.com - Makers of the GASP and GUST codes

(2) the NPARC Alliance WIND code - www.arnold.af.mil/nparc a code that is freely available to U. S. companies and was derived from an IRAD code developed by McDonnell Douglas aircraft

I have had direct experience with both codes and can tell you that they both do a good job. Both are well-documented and have been extensively used within the U. S. defense industry.

For grid generation, you might want to check with Pointwise, Inc. ( www.pointwise.com). Their GRIDGEN codes have been extensively used and are good products.

There are certainly other codes out there, but these are ones with which I am familiar and belive will fit the bill for you.

Dear Brian:

The only commercial code available that offers a tight Fluid-Structure coupling is CFD-ACE+ from CFDRC to my knowledge. It allows you to simulate structural stress as generated by the surrounding fluid, and also, if required , the structural deformations which in turn modify the fluid flow. Send me an email if you need more infos.

Best Regards,

Cornel

Brian,

First, let me state that I have performed the types of CFD analysis you are interested in. I am currently investigating the addition of aerodynamic fairings to the Boeing 737 and may soon do a radome analysis on the B767. However, I have not performed any structural anlaysis. Based on this experience, I offer the following advice.

For a basic estimation of net drag increments due to the radome, you do not need CFD. There is an encyclopedic reference on drag titled "Fluid Dynamic Drag" by Hoerner. It was published by the author, but can be found in most government scientific libraries. Hoerner was a german aerodynamicist who came to the US after WWII. This text includes an exhaustive database of drag coefficients for all classes of problems. An entire chapter is devoted to the types of protuberances you are interested in, e.g., addition of radomes, gun turrets, antennas, etc. Hoerner is a good place to start.

For detailed pressure distributions, you will need CFD. However, you may not need to model the complete aircraft. If the radome is located on top of the fuselage, away from the wings, you can probably get acceptable results by modeling the fuselage only. This will greatly simplify the problem. This is exactly the approach I am using on the 737. I am using the commercial code, CFD-FASTRAN.

If your problem is similar to the one I've discussed above, then it is very feasible for a new user. I would estimate you could begin producing results within a few weeks. An experienced user could have the first solution in a week. If you have to model the complete aircraft, then it is probably beyond the scope of a new user.

If you would like additional information, I would be happy to follow up with you.