Trainer Jet WIP

User avatar
Steve
Master Chief Petty Officer
Posts: 394
Joined: 29 Jun 2009, 10:20
Type the number ten into the box: 0
Location: Hiding (Like a ninja)

Trainer Jet WIP

Post by Steve »

Ack. I'd like to say "I'm back! =D" but it seems every post I make involves saying that in some form or another I might as well admit that I'm just slow to check online.

The reason? Uni's been kicking me in the teeth. But fear not, I actually have things to show for it that are (mostly) relevant to this forum. So, while I can't show everything yet until IP rights are cleared up I can at least make a start with what I've been up to.

Some of you know my chosen subject at uni is aeronautics, and as a final year student I have to carry out an individual project indicative of my learning and skills gained from the last 3 years of study. Normally it's research based or the optimization/re-design of a specific component, such as canopy seals to reduce air drag and so forth. I, like a fool, decided to design a wing for a theoretical training jet plane. That was foolish because I also decided that it had to make use of a dynamic property known as Vortex Lift. What was even more foolish was that I soon realized that in order to prove the wing is fit for purpose, I needed to factor in the plane's tails, weight, intakes and stability...so I ended up designing the whole damn plane.

Truespace 6 was actually used quite extensively so it's nice to know the program can still do its stuff. I'll also be showing some of the work done in two other modelling packages, Alias Automotive and PTC Creo, dedicated surface and solid modelling programs, as well as an aerodynamic analysis package called Ansys Fluent.

Cheer up U3DA. At least I'm modelling a plane I designed myself for once!
Last edited by Steve on 29 Apr 2013, 21:29, edited 2 times in total.

User avatar
Steve
Master Chief Petty Officer
Posts: 394
Joined: 29 Jun 2009, 10:20
Type the number ten into the box: 0
Location: Hiding (Like a ninja)

Re: Trainer Jet WIP

Post by Steve »

Okay. So a bit about the plane itself.

Many ex-military jet trainer aircraft, like the Jet Provost or L-39 have been converted to civilian General Aviation use, but as these jets continue to age the running costs will mount exponentially and repairs will become much more difficult. The problem is that aircraft currently in use by the military may not be operable under General Aviation rules, leaving a massive hole in the market. So there's potential to introduce a new aircraft which matches the aerodynamic performance of existing jets in military operation. but because it's new and specifically designed, will be of a fraction of the cost it takes to run the current war-bird fleets. It starts to get tricky because civvies don't always use big long runways, so the aircraft must be short and unprepared field capable, but to match the performance of and must be post-stall maneuverable. But also modern fighters are maneuverable because they are unstable, which requires flight computers to adjust the aircraft every micro-second as humans can't manage it. Civvies don't have that luxury either.

So, the target specs for the aircraft are worked out as:
----------------------------------------------------------------------
Type: Civilian Fast Jet Trainer
Certification: Aerobatic, IFR (night, crap weather) capable
Crew: 2 – (1 pilot, 1 Passenger/Student)
MTOW: 5,670kg or less
Payload: 600kg capacity external stores

VSO: 80kts @ Max weight (92mph)
VMax: Mach 0.6-0.8 (456mph-609mph)
Take-off distance: 300m @ Max weight
Landing distance: 250m @ Max weight
G load limit: +8g/-3g

Rate of climb: 3150ft/min
Maximum Height: 30,000ft (9144m)

Range: 8,600km
Loiter time: 30 minutes @ max range
----------------------------------------------------------------------


TL;DR version. Making a jet Trainer that civvies can use. has to be ^^^^ that good

So to work out where I'm going to put the wing, I needed to come up with a basic fuselage design. Here it is:

Image

To highlight all the funny errors, this was the quickly made in Truespace
Image

More to come a bit later
Last edited by Steve on 04 May 2013, 15:00, edited 1 time in total.

Unavailableartist
Petty Officer First Class
Posts: 58
Joined: 31 Mar 2013, 19:50
Type the number ten into the box: 10

Re: Trainer Jet WIP

Post by Unavailableartist »

Looks nice man, I like military-looking (I know this is a civi) jets, can't wait to c it finished.

jamesmc
Captain
Posts: 1776
Joined: 08 Jul 2009, 15:35

Re: Trainer Jet WIP

Post by jamesmc »

Very smooth and looks mean all at once!

Looking forward to seeing more...

:worship:
Walk by faith, not by sight.

User avatar
MikomDude
Captain
Posts: 1810
Joined: 04 Oct 2010, 11:12
Type the number ten into the box: 0
Location: Verona, Italy
Contact:

Re: Trainer Jet WIP

Post by MikomDude »

Looking cool!

User avatar
Steve
Master Chief Petty Officer
Posts: 394
Joined: 29 Jun 2009, 10:20
Type the number ten into the box: 0
Location: Hiding (Like a ninja)

Re: Trainer Jet WIP

Post by Steve »

well thank you very much. Slightly ashamed of that model as I've got the final design here and I assure you it looks a lot better than that.

TL;DR version much further below

So, the first thing I need to do, is calculate just how big the wing needs to be, this will be achieved as a function of 'Wing loading' (W/S), or how many kg of aircraft weight per meter of wing area and 'Thrust to weight' (T/W), the ratio of thrust available versus the weight of the aircraft, if T/W=1 then all of the weight can be supported by the engines, enabling sustained vertical flight. The required T/W and W/S values are calculated for takeoff performance, climb performance, stall speed performance and maximum height

We start as we mean to go on...here's how to work each out
Image
...Which was pleasant. It's then all plotted onto a line graph to show the optimum W/S & T/W which satisfies all performance criteria with minimum excess power or wing area. Which was in this case at maximum takeoff weight:

T/W = 0.64
W/S = 272kg/m^2

The wing size can be calculated by dividing weight by the wing loading, so the wing = 19.5m^2

This allows me to pick good aerofoils to generate the most lift for minimum drag. here's the one for the wing root and here's the one for the wingtip


Taking off my anorak now, TL/DR version is. Calculated the wing size, here's a mockup of how big and the shape of how I expect it to look
Image
Image

Leaving the wing 'planform' shape for now though, as the next step is to decide what landing flaps will be used as they need to be specially designed and tested for each aerofoil type.

jamesmc
Captain
Posts: 1776
Joined: 08 Jul 2009, 15:35

Re: Trainer Jet WIP

Post by jamesmc »

Looking good!

But, all that math made my head hurt. :shock:
Walk by faith, not by sight.

User avatar
clintonman
Captain
Posts: 3804
Joined: 21 May 2009, 21:08
Type the number ten into the box: 0
Location: California
Contact:

Re: Trainer Jet WIP

Post by clintonman »

Do you plan to produce it in some physical form when you're done? Like a wind tunnel model or a 3D print?
Clinton Reese

http://clintons3d.com

User avatar
Steve
Master Chief Petty Officer
Posts: 394
Joined: 29 Jun 2009, 10:20
Type the number ten into the box: 0
Location: Hiding (Like a ninja)

Re: Trainer Jet WIP

Post by Steve »

Hi Clintonman.

It's going to be 3D printed, but for display purposes only. There are two things stopping wind tunnel testing at the moment, the first one is skin roughness, which affects the shape more and more as it gets smaller, but the second is 'Reynolds number matching', the wind tunnel at the uni isn't fast enough even for the low speed tests I would like to do.

But, stay tuned, I have done other tests which I'll be showing a little later on which give me a good indication of what the plane is doing.

For those interested. Scale wind tunnel models are used for practical reasons like cost and equipment size. But a scale model isn't tested at the same airspeed as the full size object, as the flow properties of a shape is determined by how turbulent the air is, not the speed at which it is traveling. A 'Reynolds number' indicates how turbulent the flow is, below 2500 it is 'laminar' (smooth), between 2500-5000 it is transitioning to turbulent, and above 5000 is turbulent. To put it into perspective, it is fairly common to see Reynolds numbers on aircraft up to about 1,000,000 at their lowest speeds.

The factors that affect turbulence are the 'Dynamic Viscosity' (how thick and gloopy it acts when it's in motion, which varies from sea level to altitude), the airspeed and the object length. If the object length decreases, so does the turbulence. So opposite to common sense, to balance the correct reynolds number for a scale model, it needs to be tested at a higher airspeed than the full size object.


Also, James, I share your pain. My head hurt too!

Unavailableartist
Petty Officer First Class
Posts: 58
Joined: 31 Mar 2013, 19:50
Type the number ten into the box: 10

Re: Trainer Jet WIP

Post by Unavailableartist »

As someone who speaks with some experience. Math like that isn't easy... When your in a field like that... It's never easy. It's like the equivalent of back-breaking labor, but only for brains.
Brain-busting. ;-)
This may make me switch my engineering major. :D

Great-Job with the plane... It looks gorgeous.

Post Reply