Board index FlightGear Development Aircraft

JSBSim Piper J-3 Cub  Topic is solved

Questions and discussion about creating aircraft. Flight dynamics, 3d models, cockpits, systems, animation, textures.

Re: JSBSim Piper J-3 Cub

Postby Thorsten » Sun Jul 30, 2017 5:34 pm

https://en.m.wikipedia.org/wiki/Angle_of_attack


Yeah - that says 'reference line' and not 'definition changes with airfoil deflection' - did you actually read it? A reference line that changes all the time wouldn't be much of a reference, would it now?

In my simple mind and after Richard's phrase I firstly thought of mere damping


That's what other coefficients stand for, yes - but, see above, it's not the complete story.
Thorsten
 
Posts: 11580
Joined: Mon Nov 02, 2009 8:33 am

Re: JSBSim Piper J-3 Cub

Postby dany93 » Sun Jul 30, 2017 5:44 pm

[EDIT] (written as the same time as Thorsten's response) [/EDIT]

Thorsten wrote in Sun Jul 30, 2017 12:13 pm:Angle of Attack: Direction of the undisturbed airstream (aka, what hits the leading edge) relative to a reference axis.

Bomber wrote in Sun Jul 30, 2017 2:59 pm:The AoA is an angular measurement between an airfoils trailing edge and it's leading edge....
(= the reference axis)
...... against the relative wind..

You seem to agree ?? Maybe only to take the relative wind direction where it is undisturbed ?
dany93
 
Posts: 785
Joined: Mon Sep 07, 2009 3:43 pm
Location: France (Paris region)
Version: 2020.2.0
OS: Linux Mint 18 (64 b)

Re: JSBSim Piper J-3 Cub

Postby Bomber » Sun Jul 30, 2017 5:51 pm

yes, but Thorsten would have everyone believe that everything I write is wrong and everything he writes is correct... that way everyone reading starts with a prejudiced view of what I've written and challenge it as being wrong and a rose tinted view of what he's written and accept it as being correct.

Case in point , not one person challenged his statement that throwing physics at the problem wouldn't solve it.
Last edited by Bomber on Sun Jul 30, 2017 6:00 pm, edited 1 time in total.
"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 Mitchel
Bomber
 
Posts: 1934
Joined: Fri Dec 14, 2007 7:06 pm
OS: Windows XP and 10

Re: JSBSim Piper J-3 Cub

Postby Bomber » Sun Jul 30, 2017 5:59 pm

Richard wrote in Sun Jul 30, 2017 2:23 pm:
There will be effects from downwash; but this also depends on the htail arm, the incidence of the wing and htail, deflection of ailerons, flaps, spoilers, beta, and maybe other things I can't think of. However often downwash isn't directly considered because the geometry is fixed and therefore it can be in the base coefficient.
.


Well I agree up until separation at which point downwash wont effect the h-stab and the AoA seen by the h-stab airfoil will become the same as seen by the front wing
"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 Mitchel
Bomber
 
Posts: 1934
Joined: Fri Dec 14, 2007 7:06 pm
OS: Windows XP and 10

Re: JSBSim Piper J-3 Cub

Postby dany93 » Sun Jul 30, 2017 6:02 pm

Bomber wrote in Sun Jul 30, 2017 11:11 am:All of the above occurs with zero deflection of the elevators.. deflection of the elevators in effect increases/decreases the AoA that the h-stab see's.

I think to understand that your disagreement is here: definition of the reference line, with or without elevator deflection.
The h-stab alone, or h-stab+deflected elevator...
Last edited by dany93 on Sun Jul 30, 2017 6:04 pm, edited 2 times in total.
dany93
 
Posts: 785
Joined: Mon Sep 07, 2009 3:43 pm
Location: France (Paris region)
Version: 2020.2.0
OS: Linux Mint 18 (64 b)

Re: JSBSim Piper J-3 Cub

Postby Bomber » Sun Jul 30, 2017 6:02 pm

Richard wrote in Sun Jul 30, 2017 2:23 pm:Also needs to be considered are the derivatives, pitch damping due to q dot (CMq), change in pitch moment due to alpha dot. Last time I looked at one of your models you didn't have any derivatives - .


They are there, just that I have them effecting the Angle of Attack.
"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 Mitchel
Bomber
 
Posts: 1934
Joined: Fri Dec 14, 2007 7:06 pm
OS: Windows XP and 10

Re: JSBSim Piper J-3 Cub

Postby Bomber » Sun Jul 30, 2017 6:12 pm

dany93 wrote in Sun Jul 30, 2017 6:02 pm:
Bomber wrote in Sun Jul 30, 2017 11:11 am:All of the above occurs with zero deflection of the elevators.. deflection of the elevators in effect increases/decreases the AoA that the h-stab see's.

I think to understand that your disagreement is here: definition of the reference line, with or without elevator deflection.
The h-tab alone, or h-tab+elevator...


The point I would make Dany is that the h-stab can be stalled out..and it will do so at a defined value of fuselage AoA depending upon it's incidence......agreed ?

And that if the elevator is deflected such that it's trailing edge is lowered.... the h-stab airfoil will stall at a lower fuselage AoA.

I simply ask.... and lets be straight about this, I'm insulting no one's work by asking this question, contrary to Thorsten's accusation that I am.... Where is this effect modelled in the current code... ? I can see the h-stab stalling with AoA but not see the effect of elevator deflections increasing the likelihood of the airfoil stalling out.

You might be able to trawl through ALL existing planes and find it....but I'm asking about THIS 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 Mitchel
Bomber
 
Posts: 1934
Joined: Fri Dec 14, 2007 7:06 pm
OS: Windows XP and 10

Re: JSBSim Piper J-3 Cub

Postby Bomber » Sun Jul 30, 2017 6:43 pm

Thorsten wrote in Sun Jul 30, 2017 5:34 pm:
Yeah - that says 'reference line' and not 'definition changes with airfoil deflection' - did you actually read it? A reference line that changes all the time wouldn't be much of a reference, would it now?


But after the elevator has been deflected what you actually have in reality is a different airfoil shape... you could actually swap it out with a new single piece, ie no elevator and you'd draw it's reference line from the trailing edge to leading edge and it's AoA would be different.

Consider that most leading aircraft designers are experimenting with 'morphing' wings using modern materials, to change a wings thickness to cord, and dropping or raising the trailing edge to give control.
"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 Mitchel
Bomber
 
Posts: 1934
Joined: Fri Dec 14, 2007 7:06 pm
OS: Windows XP and 10

Re: JSBSim Piper J-3 Cub

Postby Thorsten » Sun Jul 30, 2017 8:23 pm

Simon, it's late and if needed I'll explain this in more detail tomorrow, but just briefly:

You're mixing a macro and a micro description of the physics, that's always bad, and bad physics doesn't solve problems.

In the micro description of CFD, you don't have anything like an AoA. You have air volume elements moving and hitting surface elements (think turbulent airstream, the angles under which they all hit surfaces can vary a lot). And yes, if you use the elevator, the airstream will alter a lot, you'll in essence alter the profile of the surface.

But - in the macro description where you parametrize what is happening microscopically in terms of forces and moments (and yes, there's an assumption involved that you don't deform dynamically too much), you just define one reference line, measure undisturbed airstream relative to that line, call the angles alpha (AoA) and beta (sideslip) and express what happens microscopically as functions of these angles (and the other parameters we know and love - qbar, elevator deflection angle,...)

It doesn't matter what your reference surface is, just as it doesn't really matter what your reference line is - as long as you pick one and stick with it. It's just a convention you select to parametrize.

You can't get more realistic defining an 'AoA' per surface, because the airstream actually reaching the tail may be very different even across the surface already - there's a whole plane in front of it, possibly a propeller, the airfoil looks different across its length - once you enter this game it doesn't stop before doing CFD.

What you're trying to do is cooking up a hand-waving micro description using parametrized macro concepts - which gets you up to a point, but then it misleads you - badly. Once you are in a description where you want a per-surface AoA, you have no clue what that should be for the tail of the aircraft, because you have no idea how everything that comes before has disturbed the airstream already.

Once again, that's the HUGE achilles heel of your approach - you don't need to guess effective descriptions as the approaches you criticize (which you, as Dany pointed out, one can do to 10-20% accuracy from similar aircraft), you have to guess a few free functions with dozens of parameters. Which is where it all the accuracy goes down rapidly (look at your ASK-13 JSBSim model - it has trouble getting equilibrium glide right...)
Thorsten
 
Posts: 11580
Joined: Mon Nov 02, 2009 8:33 am

Re: JSBSim Piper J-3 Cub

Postby Bomber » Sun Jul 30, 2017 9:05 pm

If you wish to critic my Ask13 then by all means I'm all ears. I believe this is your first mention of it ?
"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 Mitchel
Bomber
 
Posts: 1934
Joined: Fri Dec 14, 2007 7:06 pm
OS: Windows XP and 10

Re: JSBSim Piper J-3 Cub

Postby UsAir26 » Mon Jul 31, 2017 3:42 am

Why is this the only aircraft that crashes my fg on spawn? Any ideas on how to fix it? I would love to fly this.
Favorite planes:

777-300ER Seattle Project
777-300 Seattle Project
Boeing 707 from Lake of Constance
Cessna 172P Detailed Cockpit
UsAir26
 
Posts: 45
Joined: Mon Jun 15, 2015 7:51 pm
Location: Ukiah, CA
Callsign: UsAir26
Version: 2016.2.1
OS: Mac OS X Yosemite

Re: JSBSim Piper J-3 Cub

Postby wlbragg » Mon Jul 31, 2017 3:55 am

Do you have the latest version? The one that was updated a few days ago.

How much RAM do you have?

Do you feel comfortable commenting out a line or two of code if I give you some instruction?

Could you post the debug log report, I don't know where it is or what it is called on a Mac.
Kansas(2-27-15)/Ohio/Midwest scenery development.
KEQA (2-27-15), 3AU, KRCP Airport Layout
Intel i5 3570K AMDRX480
User avatar
wlbragg
 
Posts: 5452
Joined: Sat Aug 25, 2012 11:31 pm
Location: Kansas (Tornado Alley), USA
Callsign: WC2020
Version: next
OS: Win10/Linux/AMDRX480

Re: JSBSim Piper J-3 Cub

Postby Thorsten » Mon Jul 31, 2017 3:58 am

If you wish to critic my Ask13 then by all means I'm all ears. I believe this is your first mention of it ?


No, I asked you right before committing it whether you are sure the version you pointed me to is the correct one because the FDM has way too much sink in equilibrium glide and you confirmed the version.
Thorsten
 
Posts: 11580
Joined: Mon Nov 02, 2009 8:33 am

Re: JSBSim Piper J-3 Cub

Postby Bomber » Mon Jul 31, 2017 11:39 am

I don't remember you saying it had way to much sink in equilibrium glide....
"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 Mitchel
Bomber
 
Posts: 1934
Joined: Fri Dec 14, 2007 7:06 pm
OS: Windows XP and 10

Re: JSBSim Piper J-3 Cub

Postby Richard » Mon Jul 31, 2017 2:20 pm

Thorsten wrote in Sun Jul 30, 2017 4:57 pm:
Maybe Thorsten can explain this as I can't.


Generically they're transient effects, the airflow adapting to a change and generating forces in the process. For instance one driving effect for pitching moment due to alpha dot is delayed downwash.


Thanks for the explanation of the aero effects - which is almost how I understood it; but it's the mathematics that I was struggling with. However after one of those magic sleeps[1] I think I now almost understand what the derivative is.

For a long time I've understood that the derivatives are second order functions[2]. My realisation this morning is that all of the coefficients must be differential equations - because of the nomenclature in the Boeing document., and what's happening with these equations is to include values into the coefficients based on an acceleration or rate (e.g. q or alpha dot). But I'm not exactly sure of how all of this works mathematically.

Bomber wrote in Sun Jul 30, 2017 5:59 pm:
Richard wrote in Sun Jul 30, 2017 2:23 pm:There will be effects from downwash...


Well I agree up until separation at which point downwash wont effect the h-stab and the AoA seen by the h-stab airfoil will become the same as seen by the front wing


Depends on what you mean by separation; if you're talking about flow separation; the resulting turbulent air is going to largely destroy the effects of the htail, but the downwash is still there, just turbulent.

It's also wrong to add downwash onto the alpha(frl[3]) for the tail and expect the result to be correct; I'm sure I've drawn this out once already, it included vector maths, alphadot, q, and quite a few other terms.


Bomber wrote in Sun Jul 30, 2017 6:02 pm:
Richard wrote in Sun Jul 30, 2017 2:23 pm:Also needs to be considered are the derivatives, pitch damping due to q dot (CMq), change in pitch moment due to alpha dot. Last time I looked at one of your models you didn't have any derivatives


They are there, just that I have them effecting the Angle of Attack.


Unfortunately that won't work and can lead to a twitchy model; pitch damping (due to pitch rate) needs to use the pitch rate (Q); without this the effects will be instantaneous. You need the aerodynamic derivatives.


Bomber wrote in Sun Jul 30, 2017 6:12 pm:
dany93 wrote in Sun Jul 30, 2017 6:02 pm:The point I would make Dany is that the h-stab can be stalled out..and it will do so at a defined value of fuselage AoA depending upon it's incidence......agreed ?

And that if the elevator is deflected such that it's trailing edge is lowered.... the h-stab airfoil will stall at a lower fuselage AoA.


That seems wrong to me. If we're talking about an all moving surface then lowering the trailing edge will increase the incidence of the surface. For a surface with a moving tail edge section (i.e.elevator) the shape of the airfoil changes, but we don't normally model that, instead the entire horizontal tail would be at a certain incidence, which defines the amount of airflow, and the available pitching moment becomes a function of this - and with the conventional way of treating an airplane as an entire set of geometry [4] this will be correct[5] for each given alpha (frl) - because the measurements are taken (or calculated) in the freestream.


-----------------
[1] woke up this morning and it all became much clearer
[2] although I probably still don't properly understand what a second order function actually is
[3] fuselage reference line, or could be wing design plane; i.e. the alpha that is used for the wings.
[4] from which the forces and therefore the coefficients are derived.
[5] subjective to some sort of error, either interpolation or measurement.
Richard
 
Posts: 765
Joined: Sun Nov 02, 2014 10:17 pm
Version: Git
OS: Win10

PreviousNext

Return to Aircraft

Who is online

Users browsing this forum: No registered users and 3 guests