Design process outline: airfoils/engines

[Nexim]

Hi, I want to design a remote controlled airplane that uses a small gas turbine for propulsion. I've found a place in my city that will cast specific metal parts for a price, given a CAD file of the part.

I am an electrical engineer primarily, but I have always been fascinated with airplanes too, and consider myself to have an ample general knowledge of how they (And gas turbine engines) work.

What I wanted was a bulletpoint list of the general process in designing:

1.) A airfoil wing,

and (more complicated):

2.) A gas turbine engine, including inlet, axial compressor, combustion chamber, turbine assembly, and nozzle.

...Actually maybe thats alot to ask?

I'm aware of CFD for optimizing geometric designs, to better suit a certain range of operability with different flow conditions, but I've no idea where to start with this. For example do you just enter a CAD design, and set of parameters describing the flow, into the program, and after a while of simulation it pops out a more efficient shape for your airfoil (or whatever the part is).

Also what CFD is available for amatuers like me (maybe even free?), as opposed to industry specific private software, and what can I expect from it?

I have access to solidworks 2007, and I'm willing to put in time to work on this (It would be a one man project), as it's kind of a hobby/goal for me. I know miniature gas turbines can be built and designed even by the amatuer, as I've seen a few projects dotted around in the past.

So basically can someone list the steps in designing/making these parts, so that I can then go away, and look into certain steps in further depth on my own? Also to get a general grasp of what is required.

Thanks alot, Mat

 

[rb1957]

"...Actually maybe thats alot to ask? "

no maybe about it !

if you're interested in the goal, i'd recommend torenbeek "synthesis of subsonic airplane design" or raymer.

if you're interested in the process, draw the most complicated flow-chart you want, include lots of iterative loops, particularly ones that go nowhere and ones that have "management" input, oh, and be sure it include the "miracle happens here" step !

 

[Nexim]

Thanks,

"if you're interested in the process, draw the most complicated flow-chart you want, include lots of iterative loops, particularly ones that go nowhere and ones that have "management" input, oh, and be sure it include the "miracle happens here" step !"

I realise that in practice there may not be a very well defined or structured method to it, and from what I have read it seems it can be very empirical and somewhat of a 'dark art'. Still, there must be some kind of rough step-by-step process that aerospace engineers aspire to follow at least? (With hiccups along the way of course)

Theres so much to consider, I don't have a clue where to start. What is a good starting point at least?

Another key point is about CFD. I mean, will a CFD program encapsulate the entire design process, given a handful of parameters, or is it just a single stage of the design? (I would imagine CFD is used almost universally nowadays for these things at some point in the design stage).

"if you're interested in the goal, i'd recommend torenbeek "synthesis of subsonic airplane design" or raymer.
"

Thanks, I'll check to see if my local library, or bookshops have it.

 

[rb1957]

the process is incredibly complex, with a bizillion interactions, and data changing by the minute ... its amazing that we get anything done !!

i know bombardier has been working for years (seriously) to try to codify the design process. seriously, try some airplane oriented project engineering courses (U.Kansas comes to mind).

"will a CFD program encapsulate the entire design process" ... no, aerodynamics is only part of the problem (and the solution) ... aero., stress, operating economics, manufacturability, materials, company history, the weather ... all go into the mix

 

[Nexim]

Well it seems intuitve to me that outlinig the operating conditions (i.e. weather) would be a suitable place to start. i.e. I can say that the main operating conditions will be at 293K with an atmospheric pressure of 101.1kPA, with a chemical composition of 70% Nitrogen 21% oxygen etc, then choose a suitable range, of pressures/temperatures/altitudes etc.. that I want the plane to be able to operate within.

I think the main difference with what I'm doing, and what someone like you would do for work, is that I have no real specific criteria for how I want the plane...no customer demands...management problems...consultancy...company history considerations...rather, for me it's just a case of using the physics and maths I know, and applying it into a working engine. Even so, I accept there isn't a real straightforward method to it, but that alot of things will have to be tweaked, and trial and error/empiral methods will come into it. Though I think for me it will be alot MORE straightforward.

So once I have outlined my desired operating conditions, I can make a simple design, and input my conditions as parameters to a CFD simulation, and have it process things like shape optimization on the airfoils?

Then I can take the more efficient aerodynamic design outputted from the program, choose some starting materials, and calculate stresses etc within the parts....then if it isn't suitable, select what specifications the material for each part would need, and try to match them with existing materials... then run stress simulations of that?

Then overview the entire design, and whether it's at all easy to manufacture...if not, then re-evaluate it into something more practical, and run simulations again?

To me that sounds like a pretty intuitive way to go about it...start with a basic concept...simulate it...tweak it...simulate it...tweak it...simulate some more...tweak some more...etc..make it.

What I'm unsure of is how to use any of the programs involved, to run the simulations, and see the resulting data....In fact, I'm not even sure what programs to use...or what programs are even available to me.

It's not convenient for me to go off and take any formal courses in aircraft design, but I'm quite happy to read, and teach myself, as I have done already with alot of stuff I've learned.

Thing is, many texts (In all areas of engineering in fact) can tend to rabbit on about useless stuff alongside the context (History of the subject is an annoying one for example), and it's hard to pick a suitable text with good, clear, to-the-point information).

I checked my local library for the book you recommended, but they didn't have it. I read a little on amazon.com looked OK, and may consider buying it, as well as looking around for more.

Thanks, Mat

 

[wes616]

Mat,

Decide which of these two things you want to do. 1. DESIGN a rc airplane, or 2. Design and FLY an rc airplane. If you are like me you probablly choose the latter.

Get yourself a copy of a book of airfoil sections. Since you already have your performance criteria, pick one out that meets your needs. Don't worry about optimizing at this point.

Build a model from foam. Glue some lengths of thread to the wing and attach it to your car, with your handycam aimed at it, then go drive your car as fast as you need to in order to validate your design. Watch the video and change the shape according to your needs until you get the correct airfoil shape, desired L/d etc etc.

Next, I say that you get an OTS engine, because well, save the engine design for your next project.... At least I like to focus on one hobby at a time. So buy an engine with your performance charactaristics in mind. (feel free to do this first to ensure your desired performance can be achieved with an off the shelf RC engine.

Next, size your fuse to fit the engine. Attach the wing to the engine and size the tail to balance the rolling moment of your airplane.

Next design your control surfaces.

Hook up the engine and fly... oh, and don't forget the superglue, so you can make a second flight.





Wes C.
------------------------------
No trees were killed in the sending of this message, but a large number of electrons were terribly inconvenienced.

 

[wes616]

Oh, and exaust the web before buying a textbook, especially for a hobby. Here are just a few.

http://adg.stanford.edu/aa241/AircraftDesign.html

http://www.aoe.vt.edu/~mason/Mason/ACinfoTOC.html

http://www.ae.uiuc.edu/m-selig/ads/aircraft.html

http://ntrs.nasa.gov/search.jsp?N=17&Ne=2

Wes C.
------------------------------
No trees were killed in the sending of this message, but a large number of electrons were terribly inconvenienced.

 

[Nexim]

Thanks for the links. The first one looks really good. I'm just gonna get reading, and see how far I can get with the project...

 

[Nexim]

I've learnt the basics about flow classifications now viscid/inviscid (boundary layers and no-slip condition), compressible/incompressible, laminar/turbulent etc...

The plane will be travelling at subsonic speeds (I don't expext its possible to get enough power for sonic/supersonic speeds at this scale), so the mach number < 1. I read that if Ma < 0.3, in air at atmospheric conditions, then the flow can be considered to be incompressible, and the Navier-Stokes laws can be simplified, and still be fairly accurate.

For external flow on the wings, can I give the speed of the flow, V, as simply the speed of the plane (thorugh the fluid...assuming no wind)?...So that if v < 100m/s approx, the flow will be incompressible? Thanks

 

[rb1957]

yeah, mind you an RC model isn't likely to go much more 50 m/s = 110 mph ... i'm giessing that that would be hard to track

 

[Nexim]

Thanks, you Reckon 50m/s is really the fastest you can get an RC model even with a jet engine? I've seen videos of some of the planes made with commercially available RC jet engines, and they absolutely rocket.... I'm not very good at judging things like that just by sight, but they seem to be going immensly fast at take off speeds.

I think if the max speed was 50m/s, then the mach numbers would still be less than 0.3, because the speed of sound (at sea level) is 340.29 m/s, so if Ma = v/c, then 50/340 = 0.15, so this is less than 0.3, and the flow can be considered incompressible can't it?

Are you saying that if the mach number is 0.3, or is very close to 0.3, then it is hard to say with any accuracy whether the flow is compressible/incompressible? In which case, you might as well include the compressibility terms in the calculations just to be sure hadn't you?

Thanks

 

[rb1957]

100m/s = 225 mph ... WW2 fighters were quite happily going fastener than that ... tho' they did get into problems about 200 m/s.

yes, i think M<0.3 is reasonably incompressible

my point was that RC are controlled by the operator on the ground, in line-of-sight. maybe you plan to mount a video camera to get the pilot's perspective ?

 

[Nexim]

Oh yeh I see what you mean. I'm planning to put a PAL spycam in there, and have a wireless video link to a controller, so I would have a pilots perspective.

What about the flow within the engine? I'm assuming this flow would be treated seperately as an internal forced flow, and could therefore have a much higher mach number?...even though it would be depending on the speed of the plane and the design of the nozzle anyway.

I would certainly imagine that compressibility terms have to be used there since thats a major function of the engine?

 

[Nexim]

BTW I found a good book on fluid mechanics (mostly dynamics)..."Fluid Mechanics, Fundamentals And Applications" by Yunus A.Cengal, and John M. Cimbala

 

[rb1957]

i'd've thought you'd've picked an off-the-shelf engine (rather than trying to design one yourself).

you could start with your plane, and determine the engine you need (but it might not be available). or you could start with an engine and figure out what your plane can do. either way, your 1st pick probably won't work (defeatist, or experienced?), and either way you'll have some information to get closer to a working solution ... if you pick an engine 1st, your plane might not have the performance you need so you can improve your plane or change your engine; if you design the plane first, you'll see what engines come closest to fitting your needs, and in the same way you'll either change the performance required or the engine installed.

i don't think cast parts will work for a gas turbine engine. i think designing a plane should be challenge enough ! how much experience do you have with RC planes and such ? (are you jumping in at the deep end ??)

good luck

 

[GregLocock]

AFAIK it is illegal to fly a hobbiest RC plane out of line of sight in the USA

for obvious reasons

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[rb1957]

whilst a gas turbine is probably beyond the scope of a hobbist to build (maybe i'm seeling you guys short, based on my own ineptness !) ... i'd think that a ram jet (like the V1 bomb) would be simple enough to make in a simple workshop.

 

[LateApexEE]

Wasn't the V1 a pulsejet power plant? If so, that's a heck of a lot simpler than a turbine for a hobbyist to build.

But why turbine? There are electric (brushless) ducted fan jets that do over 200 MPH, and they are QUIET and don't require a multi thousand dollar investment.

http://www.bvmjets.com

for info on the electric fans.

 

[KENAT]

1. Go to your local library. Back in the UK my local library had several books about building RC planes that were simple enough for someone with your background to follow. I'd start with a simple glider and go from there.

2. If you want to get an idea of the real design process look at Raymer but it's still a simplification as he's only doing the top level stuff.

3. Forget about building your own gas turbine. This was a project at my Uni. They strapped couple of diesel super or turbo chargers together and ended up with a very heavy inefficient engine. If you insist age least go centrifugal.

KENAT, probably the least qualified checker you'll ever meet...

 

[Nexim]

Hi guys, despite your criticism, I've actually found a community of hobbyist jet turbine builders now, and have already gained a huge amount of support from them. It's not beyond the scope of a hobbysit, and these guys are living proof. I have seen pictures of the engines they've built, and they ARE jet turbines and not ramjets or pulse jets.

The hobbyist engines are of course dumbed down versions of a real jet turbine, but they are definately still jet turbines, having a compressor, flame can, and turbine assembly. They run on kerosene.

LateApexEE: Thanks, I hadn't seen those fans before, I'll definately check em out. However, I still want to build a jet turbine.