bum steer

[Richard]

forgot about Equivalent Force Systems.

My mathematics isn't that great; I can do vectors and most mechanics but struggle with calculus, so differential equations aren't something that I properly understand. I've got a working knowledge of mostly what I need to know - but sometimes my lack of understanding can be a real obstacle to modelling something new (such as the turbine model I've been working on for years).

I've never come across the term "Equivalent Force Systems" before; and in the context of this discussion I don't see how it is relevant - because what I thought we were discussing was what JSBSim expects of AERORP and how JSBSim applies CG shift calculations.

When I did the F-14 I failed to understand what JSBSim was doing with AERORP and CG which is what my posting to the JSBSim list you refer to was about.

Theoretically you could put the datum anywhere if you weren't using JSBSim - but we are using JSBSim so it needs to be defined in the way that JSBSim expects.

Richards publicly critical that my flight models don't have rotational damping...

Yes I have criticised your flight models, but always with the intention of pointing out how your technique could be improved. The way you explain it above shows that you haven't understood my criticism.

Your aero modelling technique has limitations in the accuracy that can be achieved because the airflow isn't modeled; and the surfaces don't interact with each other or the fuselage.

There are also important aerodynamic effects that you don't fully model (last time I saw one of your spreadsheets) such as downwash. The equations for downwash need to take into account the freestream velocity as well as the angle of the airflow coming off the wing. It is the freestream velocity that determines how long the airflow takes to reach the horizontal tail. Also when the aircraft's pitch is changing the rotation of the horizontal tail around the centre point will itself produce resistance because of the horizontal surface rotating into (or out of) the freestream. Also this rotation will change the angle of incidence of the horizontal tail.

[bomber]

Richards publicly critical that my flight models don't have rotational damping...

Yes I have criticised your flight models, but always with the intention of pointing out how your technique could be improved. The way you explain it above shows that you haven't understood my criticism.

Really ?

When you criticized my flight model saying it had no rotational damping... I tried to explain where the rotational damping was applied to the flight model, even others tried.... you didn't look at it and retract your criticism, it still stands as your opinion of my flight model.

Your aero modelling technique has limitations in the accuracy that can be achieved because the airflow isn't modeled; and the surfaces don't interact with each other or the fuselage.

You just make this up ?
where in 99% of the vanilla JSBsim flight models does this happen ?
'glass houses' spring to mind.

There are also important aerodynamic effects that you don't fully model (last time I saw one of your spreadsheets) such as downwash. The equations for downwash need to take into account the freestream velocity as well as the angle of the airflow coming off the wing. It is the freestream velocity that determines how long the airflow takes to reach the horizontal tail. Also when the aircraft's pitch is changing the rotation of the horizontal tail around the centre point will itself produce resistance because of the horizontal surface rotating into (or out of) the freestream. Also this rotation will change the angle of incidence of the horizontal tail.

There are always aerodynamic effects that aren't fully modeled in both the vanilla JSBsim method and mine.. I cannot deny that I've had to say "I'm going to ignore that, as it's effects are so small at the speeds we're going at compared to the effects of other things, like the major vectors of lift, side and drag.... like 1/1000th of the effect....

In my opinion there are no two streams like you've expressed above, have a look yourself at flow diagrams for various angles of attack...
viewtopic.php?f=54&t=1402
You can look at this topic to see how I do downwash

The h-stab does not rotate through and out the over side of the airstream coming off the back of the wing... maybe at the very extreme of stall (which isn't modeled in vanilla JSBsim techneques) but here again there are far greater effects elsewhere to worry about something that is extremely transient. Yes a rotating h-stab will have a change in AoA and this is calculated.

Here are the changes in flow stream between what the wing see's as it's AoA and the H-stab... with not rotational complications as you've described above

Code: Select all

   <fcs_function name="T4T/aero/h-stab_left/downwash-deg">                                    
      <function>                                 
         <todegrees>                              
            <table name="T4T/aero/h-stab_left/downwash-rad">                           
               <independentVar lookup="row">T4T/aero/wing_left/_1_AoA-rad</independentVar>                        
               <independentVar lookup="column">T4T/aero/h-stab_left/beta-deg</independentVar>                        
               <tableData>                        
                     -45   0   45            
      <!--   90   -->      -1.570   0   -1.570   0            
      <!--   32   -->      -0.559   0   -0.483   0            
      <!--   16   -->      -0.279   0   -0.247   0            
      <!--   8   -->      -0.140   0   -0.137   0            
      <!--   4   -->      -0.070   0   -0.083   0            
      <!--   2   -->      -0.035   0   -0.056   0            
      <!--   1   -->      -0.017   0   -0.043   0            
      <!--   0   -->      0.000   0   -0.025   0            
      <!--   1   -->      0.017   0   -0.012   0            
      <!--   2   -->      0.035   0   0.000   0            
      <!--   4   -->      0.070   0   0.020   0            
      <!--   8   -->      0.140   0   0.079   0            
      <!--   16   -->      0.279   0   0.200   0            
      <!--   32   -->      0.559   0   0.437   0            
      <!--   90   -->      1.570   0   1.570   0            
               </tableData>                        
            </table>                           
         </todegrees>                              
      </function>                                 
   </fcs_function>                                    
   

[mue]

I've never come across the term "Equivalent Force Systems" before; and in the context of this discussion I don't see how it is relevant - because what I thought we were discussing was what JSBSim expects of AERORP and how JSBSim applies CG shift calculations.

OK, let me try to explain.

"Equivalent Force Systems" means that force and moment at one point can be "transformed" into an equivalent system of force and moment at another point.

That's also what JSBSim does internally: It takes the force and moment (from the <aerodynamics> section) at the AERORP and transformes them into force and moment at the (potentially moving) current CG. JSBSim does NOT make any stipulation where the AERORP is located or if it has a meaning (e.g. aerodynamic center, neutral point, ...). For JSBSim it's just a point for that it gets the force and moment and then transformes them into force and moment at the CG and then computes accelerations, etc. ... . For JSBSim the AERORP can be anywhere.

Where the AERORP point is, only matters for the flight modeller, because he has to provide the correct force and moment at that point (in the <aerodynamics> section). And that's why the AERORP is often chosen so that force and moment at that point can be more easily computed or measured (e.g. AERORP at aerodynamic center because pitching moment is here mostly invariant to alpha..., or AERORP at moment reference point of wind tunnel measurements).

Again: JSBSim does NOT stipulate any special location of AERORP. If you have a mathematical model that gives you force and moment at the aircraft nose or say you have measured force and moment at the aircraft nose in the wind tunnel, you can set AERORP at the aircraft nose. JSBSim does NOT care.

[Richard]

Yes I've spent a considerable amount of time looking at your spreadsheets, XML and flying a few examples of your models.

Your model for downwash is too simple; it doesn't take into account pitch rate, the damping due to pitch rate or the time it takes for the air to flow from the wing to the tailplane; (at 50m/s on a C172 it will take around 0.1 second). The tailplane rotates through the freestream due to pitch rate.

[Richard]

Again: JSBSim does NOT stipulate any special location of AERORP

Except for the mass calculations that are applied based on the difference between the reference point and the computed CG - so there must be some correlation between the two - or the two must be set equal.

[mue]

Again: JSBSim does NOT stipulate any special location of AERORP

Except for the mass calculations that are applied based on the difference between the reference point and the computed CG - so there must be some correlation between the two - or the two must be set equal.

Mass calculation should only influence the position of CG and the inertia tensor.
And I don't mean that the AERORP is ignored by JSBSim. Of course the AERORP is used for computing the moment at the CG: M_cg = M_aerorp + (CG - AERORP) x F_aerorp. But you are free to choose the position of the AERORP. You only have to make sure to provide (in the <aerodynamics> section) force F_aerorp and moment M_aerorp that act on that choosen position.

[bomber]

And if you didn't provide a m_aerorp...... what would happen then ?

[bomber]

Your model for downwash is too simple;

show me where 99% of JSBsim flight models are more complex in downwash

it doesn't take into account pitch rate,

yes it does

the damping due to pitch rate

yes it does

Code: Select all

<fcs_function name="T4T/velocities/h-stab_right/w-aero-fps">                        
   <function>                     
      <product>                  
         <sum>               
            <property>velocities/w-aero-fps</property>            
            <product>            
               <sin><property>T4T/aero/h-stab_right/downwash-rad</property></sin>         
               <property>velocities/u-aero-fps</property>         
            </product>            
            <product>            
               <v>-1</v>         
               <property>T4T/velocities/h-stab/p_rotational-fps</property>         
            </product>            
            <property>T4T/velocities/h-stab/q_rotational-fps</property>            
         </sum>               
         <property>T4T/velocities/ground_effect-switched</property>               
      </product>                  
   </function>                     
</fcs_function>                        

or the time it takes for the air to flow from the wing to the tailplane; (at 50m/s on a C172 it will take around 0.1 second).

Angle is an angle... how fast the air is moving is a product of the qbar calculation.

The tailplane rotates through the freestream due to pitch rate.

The tailplane doesn't rotate through any freestream as the wing downwash effects the freestream for some considerable distance above and below the airfoil.

[mue]

And if you didn't provide a m_aerorp...... what would happen then ?

If you ignore an existing moment at the aerorp then of course JSBSim will calculate wrong kinematics.

[bomber]

not what I asked... it was a simple question... so here another...

This is the pitch moment section of the C172.... point out this m_aerorp as it's got to be there or as you say JSBsim will be producing wrong kinematics

Code: Select all

        <axis name="PITCH">
            <function name="aero/coefficient/Cmo">
                <description>Pitching_moment_at_zero_alpha</description>
                <product>
                    <property>aero/qbar-psf</property>
                    <property>metrics/Sw-sqft</property>
                    <property>metrics/cbarw-ft</property>
                    <table>
                        <independentVar lookup="row">aero/qbar-psf</independentVar>
                        <tableData>
                            13.6  0.09
                            21.2  0.04
                        </tableData>
                    </table>
                </product>
            </function>

            <function name="aero/coefficient/Cmalpha">
                <description>Pitch_moment_due_to_alpha</description>
                <product>
                    <property>aero/qbar-psf</property>
                    <property>metrics/Sw-sqft</property>
                    <property>metrics/cbarw-ft</property>
                    <sin>
                        <property>aero/alpha-rad</property>
                    </sin>
                    <value>-1.8</value>
                    <!--
                        stall and spin (5):
                        horizontal tail stall (for flat spin)
                    -->
                    <table>
                        <independentVar lookup="row">aero/alpha-deg</independentVar>
                        <tableData>
                            20    1
                            25    0.6
                            35    0.4
                            45    0.5
                            55    0.4
                            65    0.2
                            90    0.1
                        </tableData>
                    </table>
                </product>
            </function>

            <function name="aero/coefficient/Cmq">
                <description>Pitch_moment_due_to_pitch_rate</description>
                <product>
                    <property>aero/qbar-psf</property>
                    <property>metrics/Sw-sqft</property>
                    <property>metrics/cbarw-ft</property>
                    <property>aero/ci2vel</property>
                    <property>velocities/q-aero-rad_sec</property>
                    <value>-12.4000</value>
                </product>
            </function>

            <function name="aero/coefficient/Cmadot">
                <description>Pitch_moment_due_to_alpha_rate</description>
                <product>
                    <property>aero/qbar-psf</property>
                    <property>metrics/Sw-sqft</property>
                    <property>metrics/cbarw-ft</property>
                    <property>aero/ci2vel</property>
                    <property>aero/alphadot-rad_sec</property>
                    <value>-7.2700</value>
                </product>
            </function>

            <function name="aero/coefficient/Cmde">
                <description>Pitch_moment_due_to_elevator_deflection</description>
                <product>
                    <property>aero/function/qbar-induced-psf</property>
                    <property>metrics/Sw-sqft</property>
                    <property>metrics/cbarw-ft</property>
                    <property>fcs/elevator-pos-rad</property>
                    <value>-1.28</value>
                    <!--
                        stall and spin (6):
                        elevator loss of efficiency when stalled
                    -->
                    <table>
                        <independentVar lookup="row">fcs/elevator-pos-rad</independentVar>
                        <independentVar lookup="column">aero/alpha-deg</independentVar>
                        <tableData>
                                    18  25   35     45     55      65      90
                            -0.49   1   0.5  0.2    0.1    0.1     0.1     0.1
                            0       1   0.6  0.3    0.15   0.1     0.1     0.1
                            0.40    1   0.9  0.8    0.7    0.6     0.5     0.4
                        </tableData>
                    </table>
                </product>
            </function>

            <function name="aero/coefficient/Cmdf">
                <description>Delta_pitching_moment_due_to_flap_deflection</description>
                <product>
                    <property>aero/qbar-psf</property>
                    <property>metrics/Sw-sqft</property>
                    <property>metrics/cbarw-ft</property>
                    <table>
                        <independentVar>fcs/flap-pos-deg</independentVar>
                        <tableData>
                            0.0000      0.0000
                            10.0000    -0.0654
                            20.0000    -0.0981
                            30.0000    -0.1140
                        </tableData>
                    </table>
                    <value>0.7</value> <!-- tests, adjustments -->
                </product>
            </function>
        </axis>

And Richard can show us where the h-stab downwash is so well modeled.

Simon