Force instead of lift

[Oleksiy Frolov]

Hello

I am trying to simulate the effects of the differential wing lift. Such
as single wing stall due to the propwash loss at a low speed, and the
effect of the icing on a single wing (which can happen during a deicing
failure on that wing). All of these are important aerodynamic effects.

Now, we know we only have one wing in jsbsim currently, not 2. My idea is
therefor to try using 2 <force> records to simulate the wings separately,
instead of the usual CL record.
This however would require some kind of algorithm to define how the force
origin moves to overlay with center of lift for each wing, which is
probably not an easy task

I was wondering if anyone have tried a similar approach before, and if
this problem pechaps has an easier solution ?

Thank you

regards,
Oleksiy Frolov

[Ron Jensen]

Post by Oleksiy Frolov
Hello
I am trying to simulate the effects of the differential wing lift. Such
as single wing stall due to the propwash loss at a low speed, and the
effect of the icing on a single wing (which can happen during a deicing
failure on that wing). All of these are important aerodynamic effects.
Now, we know we only have one wing in jsbsim currently, not 2. My idea is
therefor to try using 2 <force> records to simulate the wings separately,
instead of the usual CL record.
This however would require some kind of algorithm to define how the force
origin moves to overlay with center of lift for each wing, which is
probably not an easy task
I was wondering if anyone have tried a similar approach before, and if
this problem pechaps has an easier solution ?
Thank you
regards,
Oleksiy Frolov

A better place to ask might be on the JSBSim user list
JSBSim user questions <jsbsim-***@lists.sourceforge.net>

In the aerodynamics section of a JSBSim configuration you are only allowed one
force vector, applied at aerorp, and consisting of lift, drag and side or
normal, axial and side. You can also specify the moments for yaw, pitch and
roll.

An alternative is to use the 'external force' section of JSBSim. This allows
you to specify an arbitrary number of forces, each of which can be applied at
separate points on the airframe and can be computed in different axis
systems.

I've played a little with doing the vertical stabilizer as external forces in
the JSBSim config I did for the Gloster Meteor:
https://gitorious.org/ron-s-hanger/gloster-meteor

The JSBSim config is here:
https://gitorious.org/ron-s-hanger/gloster-meteor/blobs/master/gloster-meteor-jsb.xml
And the systems file that computes the external force is here:
https://gitorious.org/ron-s-hanger/gloster-meteor/blobs/master/Systems/external_reactions.xml

This was a quick exercise, and I didn't fully validate the trig used to
compute the axial/normal/side components.

Ron

[Oleksiy Frolov]

Thank you.

You are right, I meant the <force> clause in the external force section.
Thank you for the links, I will study your work on this issue.

With best regards,
Oleksiy Frolov

Post by Ron Jensen

Post by Oleksiy Frolov
Hello
I am trying to simulate the effects of the differential wing lift. Such
as single wing stall due to the propwash loss at a low speed, and the
effect of the icing on a single wing (which can happen during a deicing
failure on that wing). All of these are important aerodynamic effects.
Now, we know we only have one wing in jsbsim currently, not 2. My idea is
therefor to try using 2 <force> records to simulate the wings
separately,
instead of the usual CL record.
This however would require some kind of algorithm to define how the force
origin moves to overlay with center of lift for each wing, which is
probably not an easy task
I was wondering if anyone have tried a similar approach before, and if
this problem pechaps has an easier solution ?
Thank you
regards,
Oleksiy Frolov

A better place to ask might be on the JSBSim user list
In the aerodynamics section of a JSBSim configuration you are only allowed one
force vector, applied at aerorp, and consisting of lift, drag and side or
normal, axial and side. You can also specify the moments for yaw, pitch and
roll.
An alternative is to use the 'external force' section of JSBSim. This allows
you to specify an arbitrary number of forces, each of which can be applied at
separate points on the airframe and can be computed in different axis
systems.
I've played a little with doing the vertical stabilizer as external forces in
https://gitorious.org/ron-s-hanger/gloster-meteor
https://gitorious.org/ron-s-hanger/gloster-meteor/blobs/master/gloster-meteor-jsb.xml
https://gitorious.org/ron-s-hanger/gloster-meteor/blobs/master/Systems/external_reactions.xml
This was a quick exercise, and I didn't fully validate the trig used to
compute the axial/normal/side components.
Ron
------------------------------------------------------------------------------
All of the data generated in your IT infrastructure is seriously valuable.
Why? It contains a definitive record of application performance, security
threats, fraudulent activity, and more. Splunk takes this data and makes
sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-d2dcopy2
_______________________________________________
Flightgear-flightmodel mailing list
https://lists.sourceforge.net/lists/listinfo/flightgear-flightmodel

[Bill Galbraith]

-----Original Message-----
Sent: Friday, September 30, 2011 3:43 PM
Subject: [Flightgear-flightmodel] Force instead of lift
Hello
I am trying to simulate the effects of the differential wing
lift. Such as single wing stall due to the propwash loss at a
low speed, and the effect of the icing on a single wing
(which can happen during a deicing failure on that wing). All
of these are important aerodynamic effects.
Now, we know we only have one wing in jsbsim currently, not
2. My idea is therefor to try using 2 <force> records to
simulate the wings separately, instead of the usual CL record.
This however would require some kind of algorithm to define
how the force origin moves to overlay with center of lift for
each wing, which is probably not an easy task
I was wondering if anyone have tried a similar approach
before, and if this problem pechaps has an easier solution ?
Thank you
regards,
Oleksiy Frolov

No reason that you couldn't build this inside of the existing JSBSim format
without using the external forces. Divide your wing into panels, maybe (on
each side) one that is influenced by the propeller slipstream, one not. Sum
all of your lifts to apply them at the aero reference point. So with a two
engine aircraft, you might have a total of four panels, two per wing half,
one inboard in the slipstream, one outboard in clean air. A mismatch of
forces left/right would normally shift the aero reference point left or
right, but that can be resolved with a moment.

You have a couple advantages doing this. A force at a distance is a moment,
so if the outer panels are experiencing different lift forces, that lift
times its arm creates a moment. If the aircraft is rolling, each wing sees a
different angle of attack, and therefore different lift. This is usually
handled other ways, but your method has the advantage of spinning properly.
There are disadvantages, though.

If you are going to that trouble, you might consider keeping the flaps and
ailerons seperate as well, unless the actuator mechanism is guaranteed to
never give you different flap or aileron angles.

Bill

[Bill Galbraith]

-----Original Message-----
Sent: Friday, September 30, 2011 3:43 PM
Subject: [Flightgear-flightmodel] Force instead of lift
Hello
I am trying to simulate the effects of the differential wing
lift. Such as single wing stall due to the propwash loss at a
low speed, and the effect of the icing on a single wing
(which can happen during a deicing failure on that wing). All
of these are important aerodynamic effects.
Now, we know we only have one wing in jsbsim currently, not
2. My idea is therefor to try using 2 <force> records to
simulate the wings separately, instead of the usual CL record.
This however would require some kind of algorithm to define
how the force origin moves to overlay with center of lift for
each wing, which is probably not an easy task
I was wondering if anyone have tried a similar approach
before, and if this problem pechaps has an easier solution ?
Thank you
regards,
Oleksiy Frolov

No reason that you couldn't build this inside of the existing JSBSim format
without using the external forces. Divide your wing into panels, maybe (on
each side) one that is influenced by the propeller slipstream, one not. Sum
all of your lifts to apply them at the aero reference point. So with a two
engine aircraft, you might have a total of four panels, two per wing half,
one inboard in the slipstream, one outboard in clean air. A mismatch of
forces left/right would normally shift the aero reference point left or
right, but that can be resolved with a moment.

You have a couple advantages doing this. A force at a distance is a moment,
so if the outer panels are experiencing different lift forces, that lift
times its arm creates a moment. If the aircraft is rolling, each wing sees a
different angle of attack, and therefore different lift. This is usually
handled other ways, but your method has the advantage of spinning properly.
There are disadvantages, though.

If you are going to that trouble, you might consider keeping the flaps and
ailerons seperate as well, unless the actuator mechanism is guaranteed to
never give you different flap or aileron angles.

Bill

[Oleksiy Frolov]

Hello Bill

This is a good suggestion, thank you.

I understand about calculating 2(4) different CL-s.

Do I also understand correctly we will than have use something you guys
did to the flaps once:

<axis name="ROLL">
.................
<function name="aero/coefficient/CldF3">
<description>
Roll Moment Coefficient due to Asymetrical Fowler Flaps
Deflection
calculated as difference between left and right flap lift
coef,
times distance from centerline to MAC of surface.
</description>
<product>
<value> 21.28</value>
<difference>
<property>aero/coefficient/CLdF3R</property>
<property>aero/coefficient/CLdF3L</property>
</difference>
</product>
</function>


in order to calculate the moments due to difference in individual CLs ?

With best regards,
Oleksiy Frolov

Post by Bill Galbraith

-----Original Message-----
Sent: Friday, September 30, 2011 3:43 PM
Subject: [Flightgear-flightmodel] Force instead of lift
Hello
I am trying to simulate the effects of the differential wing
lift. Such as single wing stall due to the propwash loss at a
low speed, and the effect of the icing on a single wing
(which can happen during a deicing failure on that wing). All
of these are important aerodynamic effects.
Now, we know we only have one wing in jsbsim currently, not
2. My idea is therefor to try using 2 <force> records to
simulate the wings separately, instead of the usual CL record.
This however would require some kind of algorithm to define
how the force origin moves to overlay with center of lift for
each wing, which is probably not an easy task
I was wondering if anyone have tried a similar approach
before, and if this problem pechaps has an easier solution ?
Thank you
regards,
Oleksiy Frolov

No reason that you couldn't build this inside of the existing JSBSim format
without using the external forces. Divide your wing into panels, maybe (on
each side) one that is influenced by the propeller slipstream, one not. Sum
all of your lifts to apply them at the aero reference point. So with a two
engine aircraft, you might have a total of four panels, two per wing half,
one inboard in the slipstream, one outboard in clean air. A mismatch of
forces left/right would normally shift the aero reference point left or
right, but that can be resolved with a moment.
You have a couple advantages doing this. A force at a distance is a moment,
so if the outer panels are experiencing different lift forces, that lift
times its arm creates a moment. If the aircraft is rolling, each wing sees a
different angle of attack, and therefore different lift. This is usually
handled other ways, but your method has the advantage of spinning properly.
There are disadvantages, though.
If you are going to that trouble, you might consider keeping the flaps and
ailerons seperate as well, unless the actuator mechanism is guaranteed to
never give you different flap or aileron angles.
Bill
------------------------------------------------------------------------------
All of the data generated in your IT infrastructure is seriously valuable.
Why? It contains a definitive record of application performance, security
threats, fraudulent activity, and more. Splunk takes this data and makes
sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-d2dcopy2
_______________________________________________
Flightgear-flightmodel mailing list
https://lists.sourceforge.net/lists/listinfo/flightgear-flightmodel

[Bill Galbraith]

In this case, the Cl is roll moment coefficient, not CL, the lift
coefficient

The attached file shows a new format for JSBSim files that I came up with.
It works just as well as the older format, but the formulation of the
coefficients more closely matches classical aerodynamic textbook equation
buildup. Feel free to use this, or the older format. They are both equally
valid.

Bill

-----Original Message-----
Sent: Saturday, October 01, 2011 12:03 AM
To: Flight dynamics model discussions
Subject: Re: [Flightgear-flightmodel] Force instead of lift
Hello Bill
This is a good suggestion, thank you.
I understand about calculating 2(4) different CL-s.
Do I also understand correctly we will than have use
<axis name="ROLL">
.................
<function name="aero/coefficient/CldF3">
<description>
Roll Moment Coefficient due to Asymetrical
Fowler Flaps Deflection
calculated as difference between left and
right flap lift coef,
times distance from centerline to MAC of surface.
</description>
<product>
<value> 21.28</value>
<difference>
<property>aero/coefficient/CLdF3R</property>
<property>aero/coefficient/CLdF3L</property>
</difference>
</product>
</function>
in order to calculate the moments due to difference in
individual CLs ?
With best regards,
Oleksiy Frolov

Post by Bill Galbraith

-----Original Message-----
Sent: Friday, September 30, 2011 3:43 PM
Subject: [Flightgear-flightmodel] Force instead of lift
Hello
I am trying to simulate the effects of the differential

wing lift.

Post by Bill Galbraith

Such as single wing stall due to the propwash loss at a low speed,
and the effect of the icing on a single wing (which can

happen during

Post by Bill Galbraith

a deicing failure on that wing). All of these are important
aerodynamic effects.
Now, we know we only have one wing in jsbsim currently, not 2. My
idea is therefor to try using 2 <force> records to

simulate the wings

Post by Bill Galbraith

separately, instead of the usual CL record.
This however would require some kind of algorithm to

define how the

Post by Bill Galbraith

force origin moves to overlay with center of lift for each wing,
which is probably not an easy task
I was wondering if anyone have tried a similar approach

before, and

Post by Bill Galbraith

if this problem pechaps has an easier solution ?
Thank you
regards,
Oleksiy Frolov

No reason that you couldn't build this inside of the

existing JSBSim

Post by Bill Galbraith
format without using the external forces. Divide your wing into
panels, maybe (on each side) one that is influenced by the

propeller

Post by Bill Galbraith
slipstream, one not.
Sum
all of your lifts to apply them at the aero reference

point. So with a

Post by Bill Galbraith
two engine aircraft, you might have a total of four panels, two per
wing half, one inboard in the slipstream, one outboard in

clean air. A

Post by Bill Galbraith
mismatch of forces left/right would normally shift the aero

reference

Post by Bill Galbraith
point left or right, but that can be resolved with a moment.
You have a couple advantages doing this. A force at a distance is a
moment, so if the outer panels are experiencing different

lift forces,

Post by Bill Galbraith
that lift times its arm creates a moment. If the aircraft

is rolling,

Post by Bill Galbraith
each wing sees a different angle of attack, and therefore different
lift. This is usually handled other ways, but your method has the
advantage of spinning properly.
There are disadvantages, though.
If you are going to that trouble, you might consider

keeping the flaps

Post by Bill Galbraith
and ailerons seperate as well, unless the actuator mechanism is
guaranteed to never give you different flap or aileron angles.
Bill

----------------------------------------------------------------------

Post by Bill Galbraith
-------- All of the data generated in your IT infrastructure is
seriously valuable.
Why? It contains a definitive record of application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-d2dcopy2
_______________________________________________
Flightgear-flightmodel mailing list
https://lists.sourceforge.net/lists/listinfo/flightgear-flightmodel

--------------------------------------------------------------
----------------
All of the data generated in your IT infrastructure is
seriously valuable.
Why? It contains a definitive record of application
performance, security threats, fraudulent activity, and more.
Splunk takes this data and makes sense of it. IT sense. And
common sense.
http://p.sf.net/sfu/splunk-d2dcopy2
_______________________________________________
Flightgear-flightmodel mailing list
https://lists.sourceforge.net/lists/listinfo/flightgear-flightmodel