Critical angle of attack

KL-666
Posts: 1602
Joined: Mon Sep 28, 2015 8:42 am

Re: Critical angle of attack

Postby KL-666 » Thu Oct 19, 2017 10:20 pm

Hello Bomber,

There are many actual numbers generated in the wonderful nasal properties tree. I always said: *If* the actual AoA is available, it is easy to compare it to the critical AoA. But no one ever said that the actual AoA is not available. You are the first one to say so. In the case that you have to calculate the actual AoA yourself, things get a lot more complex indeed.

Edit (FYI, after Jwocky's next response):
Just wondering about something: AoA is kind of the essence of flight. Boeings and Airbuses are designed based on AoA's. Now you tell me that the Flightgear fdm engine does not care about AoA. I believe you, when you say so. But it is kind of worrying for the quality of the fdm's we use in Flightgear.


Kind regards, Vincent

User avatar
jwocky
Site Admin
Posts: 1807
Joined: Sat Sep 12, 2015 12:04 pm
Contact:

Re: Critical angle of attack

Postby jwocky » Thu Oct 19, 2017 10:42 pm

On the other hand, JSB gives you the stall-status for the whole plane based on similar internal calculations ... just saying, Vincent.
Free speech can never be achieved by dictatorial measures!

KL-666
Posts: 1602
Joined: Mon Sep 28, 2015 8:42 am

Re: Critical angle of attack

Postby KL-666 » Thu Oct 19, 2017 11:33 pm

Hello Jwocky,

That could be a good usable number too. Just trying to brainstorm a bit about what would be the easiest way to measure for a stall protection within the limitations of Flightgear here.

Kind regards, Vincent

bomber
Posts: 834
Joined: Mon Nov 30, 2015 3:40 pm

Re: Critical angle of attack

Postby bomber » Fri Oct 20, 2017 8:51 am

It's normally referred to as 'Angle of Attack is king.

The conventional way of modelling flight is by using a table that determines the planes coefficient of lift and drag according to its present AoA which is based on its W and U velocities. (You know all this I'm sure)

This will give us the lift and drag of the plane....

Control surface responses are then implemented using the current angle of defection and damping responses are implemented using the negative of the planes current rotation...

Stall responses are implemented by add additional roll and yaw when the plane passes this known critical angle.

It's all works in a kind of a way...and if that's a person's bag well good luck to ya.....

Yet what I've talked about so far is the 'whole plane'... But what you're interested in is the AoA of each wing which no one (apart from my flight models) makes any attempt to calculate and from this value determine flight and roll response.

Stick with jwocky solution as it applies to the whole plane.

Simon.
"If anyone ever tells you anything about an aeroplane which is so bloody complicated you can't understand it, take it from me - it's all balls" - R J Mitchell

Richard
Posts: 76
Joined: Sun Sep 11, 2016 5:57 pm

Re: Critical angle of attack

Postby Richard » Fri Oct 20, 2017 3:44 pm

bomber wrote:The conventional way of modelling flight is by using a table that determines the planes coefficient of lift and drag according to its present AoA which is based on its W and U velocities. (You know all this I'm sure)


The conventional is indeed to use a set of tables, and is often called the coefficient build up method. This is basically a set of lookup tables for the forces and moments (x,y,z force, l,m,n moments), this gives us 6 components for the aero model.

The way that you described it is misleading, possibly I'm being picky, but there should be a base table for each of the drag, lift, side, roll, pitch, raw - and these base tables will be added with other tables for each of the 6. The base tables for pitch, roll and yaw determine the stability (stable, unstable). Control surfaces are simply another lookup table, usually alpha, beta, deflection, but sometimes this can be simplified.

The damping is, or rather should be, implemented using a set of first order derivatives[1] in pitch, roll, yaw. These need to be derivatives because they use the angular velocities p, q, r. These are not easy to figure out, and make a huge difference to the handling and stability. I don't think Simon's approach has these derivatives. There are derivatives for moments due to rates, as well as the damping, for pitch this will be two tables, one that is the damping that should be CMLP(alpha) * p in body axes, and the roll moment due to yaw rate CMLR(alpha) * r in body_axes

Stall response like anything else is inside of the tables. You will often see this as a subtle change in direction or magnitude within the yaw or roll tables. Some aircraft also need an extra table that will introduce an effect that isn't seen until a certain alpha, and thus will have a lot of zeros until this point.

jsbsim has fdm/jsbsim/aero/alpha-deg that can be used - I can't find a definitive reference however I assume this is alpha based on the fuselage reference line. Also the values reported in the cockpit (e.g. from an AoA sensor) will be offset a few degrees from this value.

Calculating the alpha for a single wing, or section of a wing is just a geometry equation to solve, but there isn't really a need to do this.

[1] http://fr.mathworks.com/matlabcentral/f ... ofacets4-1
[2] http://www.dept.aoe.vt.edu/~mason/Mason ... Issues.pdf

bomber
Posts: 834
Joined: Mon Nov 30, 2015 3:40 pm

Re: Critical angle of attack

Postby bomber » Sat Oct 21, 2017 2:06 pm

I don't think you're being fair to the flight models I create...
"If anyone ever tells you anything about an aeroplane which is so bloody complicated you can't understand it, take it from me - it's all balls" - R J Mitchell


Return to “FDM Modelling”

Who is online

Users browsing this forum: No registered users and 1 guest