Jump to content

Adverse yaw moment during taxi with high NWS sideslip angles


Recommended Posts

Hey there :-)

 

I came across some strange behaviour in the Hornet when the nose wheel has a large sideslip angle. Take a look at this:

 

 

<iframe width="560" height="315" src="https://www.youtube.com/embed/1PjliCpQgXo?rel=0" frameborder="0" allow="autoplay; encrypted-media" allowfullscreen></iframe>

 

To make it very clear: This is different from the threads that are concerned with the nose wheel steering logic (NWS vs. NWS HI). AFAICT the HI/LO logic behaves as stated in the NATOPS manual.

 

This is purely about the aircraft ground model. The only reason I pointed out the HI/LO modes in the video is because you can only reproduce this behaviour in the HI mode of the NWS.

 

I was unable to reproduce this in any other model I own, still I think it is NOT Hornet-specific! I think it only SHOWS in the Hornet because it is the only model that has such huge NWS angles.

 

I already have somewhat on an idea as to why the ground model produces this and how to fix it, but I'd like to hear you guys' unbiased opinion on this. Kinda like you shouldn't talk too much on "who wants to be a millionaire" before asking the audience :-D

 

Dirty :-)

Adverse Yaw during NWS Skid.trk

Link to post
Share on other sites

The very hi gain nws, you would and should only use when at a very slow walking pace when it can turn to 75 deg. This would not be good to do IRL.

 

Your doing around 15 km/h or 10 miles and hour

 

Even the low gain nws you need to be careful with a fully fuel aircraft, there not built or designed that well to be raced on the ground. You asking a lot of 2 very small wheels.

 

Light aircraft 32000 pounds or 16 tons

 

Can be 49,000+


Edited by David OC

i7-7700K OC @ 5Ghz | ASUS IX Hero MB | ASUS GTX 1080 Ti STRIX | 32GB Corsair 3000Mhz | Corsair H100i V2 Radiator | Samsung 960 EVO M.2 NVMe 500G SSD | Samsung 850 EVO 500G SSD | Corsair HX850i Platinum 850W | Oculus Rift | ASUS PG278Q 27-inch, 2560 x 1440, G-SYNC, 144Hz, 1ms | VKB Gunfighter Pro

Chuck's DCS Tutorial Library

Download PDF Tutorial guides to help get up to speed with aircraft quickly and also great for taking a good look at the aircraft available for DCS before purchasing. Link

Link to post
Share on other sites
The very hi gain nsw you would and should only use when only at a very slow walking pace when it can turn to 75 deg. This would not be good to do IRL.

 

Yes, that's correct! :-)

 

If this was a professional simulation product these high sideslip angles would probably be discarded as "out of scope" and that would be the end of it. And to a certain degree I could accept ED to simply consider it out-of-scope as well. No big deal. After all it is a flight-sim, not a drift-sim.

 

However, we expect a correct system response from an incorrectly handled system in many other cases. Like the alignment sequence of the INS or the correct warnings being triggered from shutting down an engine in flight, don't we? Has anyone ever tried to switch off your pilots oxygen flow on the left aft side console?

 

The reason why I pointed this out to the devs is because it might be an indication of an underlying more fundamental error in the modelling of dynamic surface friction in DCS as a whole, not just the Hornet.

 

My hypothesis: This adverse yaw comes from the drag of the wheels being modelled as acting opposite to the direction of TRAVEL of the vehicle. Instead it should act in a direction opposite to the direction of the RELATIVE MOTION of the surfaces against each other. That is not a big factor at low angles, but it is at 75° NWS angle.

 

Long story short: A freely spinning wheel (very similar to a polarisation filter) transmits only force-components along its axis and "ignores" (almost) all force-components within the plane of rotation. That applies even to a wheel at a sideslip angle of 75°.

 

The good news: An incorrectly modelled surface friction COULD explain this behaviour and IF(!) that is the cause, it could be corrected with 3 lines of code.

 

Again, this is not something I'd call "urgently pressing" for us gamers/simers, but if I were a company planning on using a vehicle dynamics simulation engine commercially, I'd love to know where this behaviour comes from to be sure as to wether it is a bug, an artefact or a feature :-)

 

Anyways,... can someone make sense of this adverse yaw? Is there a logical explanation for it?

 

Dirty :-)

Link to post
Share on other sites

Turning NWS HI at high Speeds, you loose traction and nose gear slides.

 

The Side Slip/Grip of the Nose Gear Tire isnt certified for forces that high., you turn the wheel and the tire's grip loses traction and slides.

 

It's the same as driving 50 MPh, with STM and TCS off in a Car, and turning the wheel to the lock, eventually, the tire's grip is exceeded by force, and the car no longer turns, either just goes straight, or slides uncontrollably, until the car slows down enough and/or the tires are centered enough for the tires to re-gain traction.


Edited by SkateZilla

Windows 10 Pro, Ryzen 2700X @ 4.6Ghz, 32GB DDR4-3200 GSkill,

X470 Taichi Ultimate, 2x R7970 Lightnings Crossfired @ 1.1/6.0GHz,

3x ASUS VS248HP + Hanns·G HZ201HPB + Oculus HMD,

Thrustmaster Warthog HOTAS + MFDs, CH Fighterstick+Pro Throttle+Pro Pedals

Link to post
Share on other sites
Not a Bug,

 

Turning NWS HI at high Speeds, you loose traction and nose gear slides.

 

The Side Slip/Grip of the Nose Gear Tire isnt certified for forces that high., you turn the wheel and the tire's grip loses traction and slides.

 

It's the same as driving 50 MPh, with STM and TCS off in a Car, and turning the wheel to the lock, eventually, the tire's grip is exceeded by force, and the car no longer turns, either just goes straight, or slides uncontrollably, until the car slows down enough and/or the tires are centered enough for the tires to re-gain traction.

All true, although I disagree with this:

either just goes straight, or slides uncontrollably

The vehicle will go straight. It will slide uncontrollably only if the rear wheels are also locked and start skidding, which is not the case presented by OP. As I understand it, the operative word is ADVERSE yaw. Even considering what OP is observing, that the contact point of the outside nose wheel is outside the turn relative to the CG (even then, the inside nose wheel is still inside the relative to the CG), I can't imagine or draw of force-vector diagram which would systematically cause this heavy airplane to significantly change its course the opposite way.

 

I've driven a few Hornets off the Stennis. I respectfully submit it's worth investigated a bit more.

 

Could you indicate the static (non-sliding) and kinetic (sliding) friction coefficient values used for the Hornet in the case of rubber-on-dry-asphalt?


Edited by Jack McCoy

i7-7700K@4.8GHz, 16Gb-3200, GTX-1080Ti-Strix-11Gb, Maximus IX Hero, Oculus Rift, Thrustmaster Warthog+F/A-18C, Logitech G940 Pedals.

Link to post
Share on other sites
All true, although I disagree with this:

 

The vehicle will go straight. It will slide uncontrollably only if the rear wheels are also locked and start skidding, which is not the case presented by OP. As I understand it, the operative word is ADVERSE yaw. Even considering what OP is observing, that the contact point of the outside nose wheel is outside the turn relative to the CG (even then, the inside nose wheel is still inside the relative to the CG), I can't imagine or draw of force-vector diagram which would systematically cause this heavy airplane to significantly change its course the opposite way.

 

I've driven a few Hornets off the Stennis. I respectfully submit it's worth investigated a bit more.

 

Vehicle will go straight if it's Rear Wheel Drive and the Differential is Locked.

 

otherwise, the vehicle will start sliding uncontrollably in either direction depending on which way tires are turned, which rear wheels have better bearings, asymmetrical weight, angle of the surface, grade of the surface etc.

 

I didnt mean sliding as in all the tires lose traction, i meant sliding as in the tires that locked will begin to shift direction.

 

 

 

 

The Opposite movement in the video is off though, if anything nose should be skidding to the side the tires are pointing to.


Edited by SkateZilla

Windows 10 Pro, Ryzen 2700X @ 4.6Ghz, 32GB DDR4-3200 GSkill,

X470 Taichi Ultimate, 2x R7970 Lightnings Crossfired @ 1.1/6.0GHz,

3x ASUS VS248HP + Hanns·G HZ201HPB + Oculus HMD,

Thrustmaster Warthog HOTAS + MFDs, CH Fighterstick+Pro Throttle+Pro Pedals

Link to post
Share on other sites
Vehicle will go straight if it's Rear Wheel Drive and the Differential is Locked.

 

otherwise, the vehicle will start sliding uncontrollably in either direction depending on which way tires are turned, which rear wheels have better bearings, asymmetrical weight, angle of the surface, grade of the surface etc.

 

I didnt mean sliding as in all the tires lose traction, i meant sliding as in the tires that locked will begin to shift direction.

I'm not going to over-analyse this statement, which sounds rational anyway. I just don't see how RWD and locking differential has anything to do with the Hornet. And I bet the bearing quality on the Hornet is mathematically identical on both main, undamaged wheels.

 

Of course, if differential braking action is involved in the experiment, then it's anybody's guess, without telemetry.

 

The Opposite movement in the video is off though, if anything nose should be skidding to the side the tires are pointing to.

Precisely to the point of OP. I think you were a little quick to dismiss as "Not a bug" in this case.

i7-7700K@4.8GHz, 16Gb-3200, GTX-1080Ti-Strix-11Gb, Maximus IX Hero, Oculus Rift, Thrustmaster Warthog+F/A-18C, Logitech G940 Pedals.

Link to post
Share on other sites

Hey :-)

 

I should probably have added:

1. I did not use differential braking. In fact, I did not use any braking at all. Should be visible in the track (CRTL+ENTER).

2. No wind present

3. I tried this on different runways/taxiways/aprons to make sure that a sloping ground does not play a role here.

 

If you want to use the analogy of a car, I'd find that very resonable! If we leave out the fact that this vehicle has wings attached to its fuselage for a monent, we could consider it practically an unpowered car with a high center of gravity that seems to suffer from severe(!!!) understeering.

 

And to all those who want to swipe this under the out-of-scope-carpet :-))))) :

Imagine you taxi your Hornet onto the active runway just a tad bit faster than what you learned in flightschool. When you realise that you have quite a big turn radius even with full NSW deflection, you might be tempted to tighten the turn by switching NWS to HI momentarily, giving you access to those extra ~50° of NWS angle. You CAN get away with this at a groundspeed of 11Kts or less, but as soon as you try this at a speed of 12Kts, you will see adverse yaw. It is not THAT far fetched! :-)

 

BTW, I am NOT advocating to reduce NWS angle or even deflection rates. The angle and rates are realistic/plausible. I guess even the fact that there is no logic-circuit to prevent the pilot from doing this is realistic.

 

What I would hope for the devs to do:

- Take a look into the friction model, especially into what determines the direction of the force generated by it.

- If I'm wrong just let me know that someone worth his salary took a look at it, reproduced the behaviour and decided that the model behaves as desired.

- The icing on the cake for me would be, if you could drop me a line here where the adverse moment comes from, cause I have absolutely no clue.

 

Oh,... and because I have not mentioned it: I think the Hornet is a great model!!! And the developers are doing a great job with it! After more than a decade of "abstinence", I'm back in flightsimming because of IT :-)

 

Dirty

Link to post
Share on other sites

Please fix it

 

The adverse yaw is must be addressed.

At first I encouter with this on Stennis.

The unnatural yaw is sustainable on grass.


Edited by discwalker
track removal due incorrect replay

GTX 1070 8GB, 16GB DDR3, W8.1 on SSD, DCS on another SSD

Link to post
Share on other sites
  • 2 months later...

I think we can all agree that whatever it does it shouldn't be turning in the opposite direction of the commanded turn direction. I've never experianced a car understeer left while I'm turning the wheel right.

Proud owner of:

PointCTRL VR : Finger Trackers for VR -- Real Simulator : FSSB R3L Force Sensing Stick. -- Deltasim : Force Sensor WH Slew Upgrade -- Mach3Ti Ring : Real Flown Mach 3 SR-71 Titanium, made into an amazing ring.

 

My Fathers Aviation Memoirs: 50 Years of Flying Fun - From Hunter to Spitfire and back again.

Link to post
Share on other sites

That's because of the steering geometry in your car (in that it's symmetrical around the car's axis)

 

Watch the video, and the geometry of the nose wheel steering.

 

At the point you turn the NWS to high, and turn the wheel left, it stops providing any lateral turning force (it's locked, not turning and just skidding), but that skid is still providing retarding drag.

 

The wheel trails the suspension strut, so at the point you apply full left rudder, the wheels are to the right of the strut.

 

No turning force from the tyre + a retarding force to the right of the suspension strut (& so to the right of the aircraft's axis) = a slow right turn.

 

As the aircraft yaws relative to its direction of travel, the force becomes more asymmetrical, and the aircraft will 'spin out' - just as a car will.

 

No pilot would actually allow this circumstance to occur.

NWS.JPG.56c0c50e58d42deccf85aa3e9d8b8704.JPG


Edited by Weta43

Cheers.

Link to post
Share on other sites

Weta43 seems right to me. The nosewheel has trail--the wheels themselves are behind the steering axis, so when you crank them over and they skid, they produce more drag on the side opposite your steering input, and the plane turns the other way.

Of course, doing this in real life would destroy the tires, which is why they don't do this in real life. But, the real F/A-18C may very well behave like this.

 

 

Oh man, lots of discussion over this little detail... Let the ED dudes work on stuff that matters :)

 

 

AD

Kit:

Asus Z170-P, i5-6600k at 4.5 GHz, 1.35V w/ CM Hyper 212 EVO, 2 x 8GB GSkill Ripjaws 4 @ 3000MHz 15-15-15-35, Zotac GTX 1080 Ti AMP Extreme Core Edition, EVGA SuperNova 750 G2 PSU, HP Omen 32" 2560x1440, Thrustmaster Cougar HOTAS fitted with Leo Bodnar's BU0836A controller.

--Flying is the art of throwing yourself at the ground, and having all the rules and regulations get in the way!

If man was meant to fly, he would have been born with a lot more money!

Link to post
Share on other sites

This IS stuff that matters! :-)

 

..yea yea,... I know...

 

I agree, that there are more pressing issues on the table for the developers right now. It's not like I can no longer enjoy the beauty of this module. But isn't this forum also a little bit about education, technical understanding and a fair appreciation of the insane amount of detail a modern PC-based flightsim can provide? To me, that is what flightsimming is all about. I want to understand this behaviour at least as much as I want the developers to fix it (IF it needs fixing).

 

Weta43 picked up the point that I made in my second post and what I already said in the video: The nosewheel has a caster. That shifts the contact surface of the wheels over to the side opposite of where you are steering. The point where wheel drag applies shifts with it. So far,... everything appears make sense :-) But that is only the first step of the argument!

 

The second is this: The point where the drag ACTS on the airframe shifts over adversely (due to caster), but the direction where it POINTS is still way over to the provers side. Or to be more precise: The vector of that force still passes the centre of gravity on the provers side --> huge provers yaw moment.

 

All these arguments are founded on the premise, that friction is a force opposite to the direction of the relative motion of the SURFACES involved. Keep in mind: The relative motion of the surfaces is a superposition of the vehicle motion and the motion due to the freely spinnig wheel. That wheel still spins at cos(NWS_angle°)*Groundspeed.

The behaviour that can be observed however could be nicely explained if friction was (wrongfully) modelled as a force acting opposite to the direction of motion of the VEHICLE alone. I don't think this is a coincidence. I think it points directly to where the devs should look for a fix :-)

 

I tried my best to explain this in simple words and as math-free as possible. I know, that this topic will not raise huge attention with 99% of the forum members, but maybe there is someone among the other 1% who can confirm or correct my hypothesis. I'd be thrilled either way!

 

As long as I go to bed tonight a little smarter then I woke up,... it was a day worth living :-)

 

Dirty :-D

Link to post
Share on other sites

 

Oh man, lots of discussion over this little detail... Let the ED dudes work on stuff that matters :)

 

OMG . . . yes AD, if people start complaining about issues like “NWS HI isn’t simulated correctly” leads to the conclusion,

that all major systems of the hornet are already simulated perfectly and only minor issues have to be solved.

 

AD, you’re 100% right. Let the ED team solve real issues and don't let them get headaches from nullities.

Link to post
Share on other sites
OMG . . . yes AD, if people start complaining about issues like “NWS HI isn’t simulated correctly” leads to the conclusion,

that all major systems of the hornet are already simulated perfectly and only minor issues have to be solved.

 

AD, you’re 100% right. Let the ED team solve real issues and don't let them get headaches from nullities.

 

While this NWS business may not even be a bug, it's unwise to suggest that small bugs aren't important to fix. The irritations of multiple small bugs can add up to ruin the feel of any game.

Link to post
Share on other sites

Hey Weta43,

 

I'd like to comment on your post, if I may :-)

Please take this in a friendly way. Open-minded and respectful. After all, we wouldn't be here if we weren't infected with the same borderline insane obsession with simulated flying. I really don't care WHO's right, but I do care very much WHAT's right :-)

 

 

That's because of the steering geometry in your car (in that it's symmetrical around the car's axis)

 

Cars also have caster wheels. It is what lets you feel the lateral forces in your steering wheel when cornering. Even modern power-drive steering provides such feedback. It is essential for controlability in any car.

 

 

At the point you turn the NWS to high, and turn the wheel left, it stops providing any lateral turning force (it's locked, not turning and just skidding), but that skid is still providing retarding drag.

 

It is not static. It does still spin. It spins at cos(NWS_angle)*groundspeed. It even does so in DCS. Exactly how it should be. If it weren't spinnig you'd be right, but it is :-)

 

 

As the aircraft yaws relative to its direction of travel, the force becomes more asymmetrical, and the aircraft will 'spin out' - just as a car will.

 

Correct! ....for a vehicle in an OVERsteering situation :-) A situation where the angular rate is higher than the wheel deflection demands. A situation where PROverse yaw is being induced from differing friction coefficients between the front and rear tyres. This situation is indeed unstable (at least in fwd motion) but that is not the case here. It is not OVERsteering, it is not even UNDERsteering,... it is ADVERSELY steering :-)

 

 

No pilot would actually allow this circumstance to occur.

 

...hmmmm... *blushing_slightly* ...let me PM you :-)

 

 

 

I guess many of you understand, that the caster wheel will offset to the adverse side when deflected. What I'm trying to convince you guys to do, is take the "next step" and ask yourselves: where does the resulting force point?

 

Dirty :-D

Link to post
Share on other sites
While this NWS business may not even be a bug, it's unwise to suggest that small bugs aren't important to fix. The irritations of multiple small bugs can add up to ruin the feel of any game.

 

Unfortunately the Hornet sim has quite a few major bugs or open issues to be solved, these may ruin the feel of the game.

The “NWS business” is the first minor bug and it surely won’t ruin my feel of the game at all.

Link to post
Share on other sites

The “NWS business” is the first minor bug and it surely won’t ruin my feel of the game at all.

That depends on how many times killed you this bug on the Stennis.

Sim is about for me plane flying in the air+rolling on the ground, about controllability.

GTX 1070 8GB, 16GB DDR3, W8.1 on SSD, DCS on another SSD

Link to post
Share on other sites

AD, you’re 100% right. Let the ED team solve real issues and don't let them get headaches from nullities.

 

Yes, yes,... I said it before, I say it again: This is NOT pressing, major-issue, important, urgent, deal-breaker, necessary,... etc.

 

This is simply an attempt to find like-minded people to discuss with and exchange ideas about something we are all somewhat interested in and passionate about. In mutual respect and with full knowledge and acceptance of the others thoughts and ideas being their own and not a copy of mine.

 

If someone finds other aspects of the sim worthy of discussion, they may start a thread and I'd be willing to add my point of view to it, if I have anything substantial to say about the topic of discussion.

 

But at no point will you ever see me questioning the very existence of the discussion itself. If I ever did that, I'd have to question my own participation in it first!

 

Dirty

Link to post
Share on other sites
That depends on how many times killed you this bug on the Stennis.

Sim is about for me plane flying in the air+rolling on the ground, about controllability.

 

Yes, discwalker, you have a valid argument, carrier op could be an issue if NWS is not reliable.

But it simply isn’t unreliable, it's working well.

 

I have bound “HI” function of NWS to the pinky lever of my TM Warthog joystick.

In order to maneuver the Hornet slowly on the carrier deck I keep the “HI” function continuously to ON by holding the pinky lever down.

With little or more inputs to the rudder I’m able to fine control a gradual deflection of the nose wheel hence the fine control of the movement on deck.

Of course, with full rudder input the nose wheel will deflect 90 deg and block any controlled movement.

Even if this happened I may simply release the rudder input and I’m immediately fine.

 

I really don’t understand the issue here.

Yes, there could be a limitation for the nose wheel max deflection, e.g. to a max of 75 degrees.

But I can live with situation as it is right now without any headaches for me - and for the ED team.


Edited by wernst
Link to post
Share on other sites

Except in real life the NWS does not slide like that EXCEPT on BADLY worn or very greasy carrier decks.. on the beach or on a well maintained deck they just do not push like they do here in this sim.. 3 Med cruises as a Flight Deck TS/FC with Hornets A-7Es and S-3Bs and the only time they slip like this is when the deck is in BAD shape.

US Navy AT3 VA-105, VFA-105, VFA-106, VS-27

1987-1997 A-7E, F/A-18A-D , S-3B.

Link to post
Share on other sites
Except in real life the NWS does not slide like that EXCEPT on BADLY worn or very greasy carrier decks.. on the beach or on a well maintained deck they just do not push like they do here in this sim.. 3 Med cruises as a Flight Deck TS/FC with Hornets A-7Es and S-3Bs and the only time they slip like this is when the deck is in BAD shape.

 

So you've seen a pilot suddenly crank the NWS hard over? I doubt they'd do that. Just because you never saw it doesn't mean it couldn't happen.

Link to post
Share on other sites
  • 2 months later...

please watch from 22:48 https://youtu.be/hlDdknkSNPU?t=1368 Interview with real pilot: all turns make at idle power, gain forward momentum and then turn -> this is not true to the current DCS implementation, NWS HI is punch you in the face and works only from stop condition and powered

Whole video:

If just slip forward the plane in maximum deflection (like Su-33) noone would complain, but in maximum deflection plane rolls to the opposite direction.


Edited by discwalker
NWS

GTX 1070 8GB, 16GB DDR3, W8.1 on SSD, DCS on another SSD

Link to post
Share on other sites
  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...