• Which the release of FS2020 we see an explosition of activity on the forun and of course we are very happy to see this. But having all questions about FS2020 in one forum becomes a bit messy. So therefore we would like to ask you all to use the following guidelines when posting your questions:

    • Tag FS2020 specific questions with the MSFS2020 tag.
    • Questions about making 3D assets can be posted in the 3D asset design forum. Either post them in the subforum of the modelling tool you use or in the general forum if they are general.
    • Questions about aircraft design can be posted in the Aircraft design forum
    • Questions about airport design can be posted in the FS2020 airport design forum. Once airport development tools have been updated for FS2020 you can post tool speciifc questions in the subforums of those tools as well of course.
    • Questions about terrain design can be posted in the FS2020 terrain design forum.
    • Questions about SimConnect can be posted in the SimConnect forum.

    Any other question that is not specific to an aspect of development or tool can be posted in the General chat forum.

    By following these guidelines we make sure that the forums remain easy to read for everybody and also that the right people can find your post to answer it.

FS2004 Auto Throttle vs Wind

G

geoffco

Messages
206
Country
unitedkingdom
Hi,

I have encountered some strange behaviour regarding the FS9 auto throttle and turning at altitude...

At FL360 with speed M0.76 (pretty typical B712 cruise setup):

With a 20 knot headwind, if I turn through 180 degrees, round so that I then have 20 kts tailwind, I find that the speed falls well below (down to ~ M0.71) the set value until after completing the turn. Then it recovers.

When making the opposite 180 turn back from tailwind to headwind, I get a corresponding increase in the speed and this can easily be enough to go above M0.81 (which is too close to the pole for my liking).

With no wind component, the autothrottle seems to handle the changes required for the turn quite well, so I wonder why it is not also compensating for the wind properly... I undertand that this is all to do with the inertia of the aircraft vs the IAS having a wind component, and if I was using manual throttles where I didn't change the setting during the turn it would be forgivable. But this is with the AT engaged!!

Obviously the above 180 turn regime is an extreme example for a demonstration, but I have seen overspeeds due to this in normal cruise flight before and that is clearly abnormal. Surely the AT should be able to cope with this - it's a normal 3 deg/sec turn rather than some high G manouvre!

What tweaks might be recommended to try and fix it? I am happy to experiment and try changing the AP section of the aircraft.cfg or the autopilot SPD_ variables in the .air file, but I'm not sure where to start...

Cheers,
Geoff
 
It is a matter of inertia,

At a constant true airspeed (inertial speed) the ground speed of an aircraft flying with a 20 kt headwind is 20 kt less than the airspeed – flying with a 20 kt tailwind the ground speed is 20 kt more than the air speed. The aircraft has to accelerate by 40 kt when making a 180 deg turn. This needs a thrust increase while the aircraft is accelerating. A further thrust increase in needed to compensate for the increased drag in the turn. Thrust doesn’t respond immediately to a commanded change: nor does the aircraft speed to a change in thrust. As a result, some lag is inevitable even on real aircraft that are likely to have much more sophisticated controllers than FS. I don’t know FS precision compare with real aircraft.

On the question of changes, the PID constants in the .air are only for AI aircraft. The [autopilot] section .cfg file contains the constants and limits for navigation (heading) and glide slope controllers. The only variables for the autothrottle are autothrottle max_rpm and max_throttle_rate
 
I guessed that would be the case. Anyway to get around it I wrote some code to shif the AP mach variable temporarily if an overspeed is likely...

Ahaa, so the section in the screenshot below (part of Autopilot 1199) is of no use in FS9 for user aircraft..? Good to know so I don't waste any more time trying tweaking those!

Thanks,
Geoff
 

Attachments

  • Image2.jpg
    Image2.jpg
    20.9 KB · Views: 584
Microsoft never document the .air file for FS, but the ESP/P3D SDK .air file documentation states:
PID Controllers

PIDs are only used to aid in the handling of AI (computer) controlled aircraft.
Click to expand...

It makes sense, otherwise there'd be a conflict between them and the [autopilot] section in the .cfg file.
 
Aircraft.cfg

[autopilot]
autopilot_available=1
flight_director_available=1
default_vertical_speed=0.000000
autothrottle_available=1
autothrottle_arming_required=1
autothrottle_takeoff_ga=1
autothrottle_max_rpm=95.000000
pitch_takeoff_ga=8.000000
autoland_available=1
use_no_default_bank=1
use_no_default_pitch=0
default_pitch_mode=1
default_bank_mode=0
max_pitch=7.000
max_pitch_acceleration=0.15
max_pitch_velocity_lo_alt=0.15
max_pitch_velocity_hi_alt=0.15
max_pitch_velocity_lo_alt_breakpoint=8000.00
max_pitch_velocity_hi_alt_breakpoint=28000.00
max_bank=25.000000
max_bank_acceleration=1.0
max_bank_velocity=1.75
max_throttle_rate=0.02 // Reaction rate of the thrust lever
nav_proportional_control=10.000000
nav_integrator_control=0.10000
nav_derivative_control=10.000000
nav_integrator_boundary=0.500000
nav_derivative_boundary=0.000000
gs_proportional_control=20.00
gs_integrator_control=0.10000
gs_derivative_control=10.000000
gs_integrator_boundary=10.100000
gs_derivative_boundary=1.000000
yaw_damper_gain=1.000000


[TurbineEngineData]
fuel_flow_gain=0.0015 // spool rate engine
inlet_area=47.080
rated_N2_rpm=16383
static_thrust=59000.000
afterburner_available=0
reverser_available=1


It must be between two values​​, an optimal compromise be found.
You can test the best, if the plane goes after the descent to horizontal flight, or on final approach (ILS; APR).
 
The SDK states:

max_throttle_rate - max_throttle_rate=0.10 - This value sets the maximum rate at which the autothrottle will move the throttle position. In the example, the maximum rate is set to 10% of the total throttle range per second.
Click to expand...
 
Yeah, I don't actually think the throttle rate is a problem... When this problem occurs, the rate is well below 10% per second.

Geoff
 
Geoff,
Have you tried the turn with autothrottle not selected? That should give you an idea of what the autothrottle is trying to correct.

Roy
 
Geoff,
Another thought or two.
The FS autopilot is a one size fits all implementation.

I suspect it was originally designed for piston prop aircraft. The autothrottle factors you can alter in the config file only concern throttle movement. (ignoring spool-up rates)

With a piston prop, thrust is proportional to throttle position, so a 10% throttle movement will give you a 10% thrust increase or decrease, regardless of throttle setting.

With a jet, thrust versus throttle position is not linear. At low throttle settings thrust increases slowly as it is moved forwards and at high throttle positions it increases rapidly, when moved forward. For example, at 50% throttle in a piston prop, moving it to 60% would give a 10% thrust increase. In a jet it would give you about 7% thrust increase. At 80% throttle in a piston prop, moving it to 90% would give you a 10% thrust increase, but in a jet it would give about 20% thrust increase. Conversely, at 80% throttle setting, moving it to 70% will give you a 10% decrease in thrust, and about the same in a jet.

I suspect the autothrottle speed set error triggers a linear throttle response. Assuming it is a PID system the non-linear thrust response of a jet could lead to large integral error build up.

If so, this could explain the autothrottle controlled speed drop-off turning downwind and the over-compensation in thrust when turning up-wind again.

A second point concerns inertia or momentum. Generally, this is described as mass times true airspeed squared. If that was absolutely true, wind would have no effect on momentum. The airplane's momentum would be the same with no wind or with some wind.

A sailplane flying at 50 kts into a 0 kt headwind will have 50kts ground speed. If it turns rapidly downwind it will still have 50 kts ground speed (assuming no speed loss because of the turn and increased induced drag). It will also still have 50 kts airspeed.

If the sailplane at 50 kts is flying into a 50 kt head wind it will be stationary with respect to the Earth, zero kts ground speed. If it turns rapidly downwind, sort of spins around a point on the ground, it will still be at about zero groundspeed with respect to the Earth, but now the wind is coming from behind. To stay at 50 kts airspeed it will have to accelerate to 100 kts ground speed. Sure the wind will help, but airspeed will drop for a while unless it dives. The 100 kts momentum will still be there if it turns up wind, so it should have more than 50 kts airspeed and more than zero kts ground speed. It will end up lower because of the dive, but some of the altitude loss can be regained by trading speed for altitude. Most birds know about this effect.

Think about that in the context of your original question. Airspeed drops when you turn down wind, because the wind is behind you. Airspeed increases when you turn upwind, because the wind is against you. Nowhere as extreme as the sailplane example, but similar. That is why I suggested doing the turn with autothrottle unselected to see if the speed changes still happen.

Just two thoughts
Roy
 
Momentum is mass times groundspeed. In your example, a sailplane flying at 50kt airspeed into a 50 kt headwind would have zero ground speed and zero momentum. If it then turned instantly through 180 deg, it would still be at zero groundspeed and zero momentum (Law of Conservation of Momentum) Hoever its airspeed would be minus 50 kt and the aircraft would be unflyable. The pilot would have to make a combined turn and dive manoeuvre maintain airspeed by exchanging height for speed..

I saw the Breguet Atlantique accident at the 1968 Farnborough air show. The pilot demonstrated a very low-speed, low-level pass into wind with one engine feathered. He then opened up both engines and attempted a 180 turn but without waiting for sufficient ground speed build up. The aircraft lost air speed and couldn’t maintain height in the turn. The inner wing clipped the roof of a building resulting in a crash that killed everyone onboard.
 
Momentum is mass times groundspeed
Click to expand...

Thanks. I was struggling a bit to understand how momentum played out. I had the wrong frame of reference (air) instead of ground. I also confused kinetic energy with momentum.

Making momentum play out with respect to ground does explain the original question of why the plane slowed when turning downwind and sped up when turning upwind.

The autothrottle was just behind the game and I gave some suggestions on why that could be.

I chose the sailplane example because I had experienced the dangers of stalling when turning sharply down wind in a sailplane.

Roy
 
There are three fundamental conservation laws - mass, energy, and momentum - but momentum often gets overlooked for some reason.

On the original point, my feeling is that Flight Simulator auto-throttle copes well enough in steady flight but has its limitations elsewhere. Real world aircraft have much more sophisticated systems than FS so its only to be expected.
 
You must log in or register to reply here.
Back
Top