Terrible ground steering

Greetings all,

I am using mfs accelerator and flying the 172. Hdwr is saitek rudder/yolk and throttle. My problem is that on the ground my acft goes everywhere but straight. I have used every sensativity setting combo I can think of and it still goes everywhere. I am not a computer geek so please dumb down your answers:) Thanks for any help.

 

Perry266

Can I assume you have:

Calibrated your Saitek Hdwr,

Fair weather no cross winds,

Aircraft control surfaces are trimmed to 0 degrees.

Example images to help verify settings and conditions.

Ron

 

 

 

 

 

And don't forget that in FS9/FSX runways and landing fields are "slippery" as if covered in ice, it's a thing that was never fixed as far as I know, so bear that in mind.

Hi Perry,

I try to emulate real world conditions as accurately as possible. I fly from the 3D cockpit, have the zoom set to 1.0 (I've left the WideViewaspect=False), and even after getting rudder pedals had the steering control problem. I have found someting that helps though; steer like you do in a car. Don't look at the ground immediately in front of the aircraft; concentrate on the what you see in the distance. This has really helped me to not oversteer.

Jim F.

FSX has an issue with ground friction. If you get FSUIPC it is good fro calibtration and also has aground friction fix + some other fixes. Cheap for the price.Its not perfect but does help a lot.

The apparent problems with "ground friction" I often see attributed to FS is not the fault of the simulator but of modelers not setting the inertia factors in the CFG and AIR files correct for the size and weight of the aircraft. Incorrect wheel locations and inertia/suspension settings (which are very difficult to get right) also play a part.

 

These things can take hundreds or thousands of hours of testing to get exactly right and most modelers will not sped that time if it "works OK."

 

Most modelers just copy these settings from other aircraft and fail to modify them correctly. A correctly configured aircraft will behave on the ground as expected which proves FS is capable (and always has been) of performing this function. If the aircraft's wings don't dip on a hard turn, this is ONE CLUE the model dynamics are not well designed. Another clue on medium to heavy aircraft is if the pushback radius is unrealistically too wide. These problems and others CAN be properly controlled by a talented and dedicated modeler.

 

Unfortunately, the plethora of poorly modeled aircraft out there which simmers will not abandon because they look so nice give FS a bad name. It's the modelers which should be wearing mud.

 

As mentioned above, over steering is the most common cause of control problems on the ground. Keep your taxi speed between 5 and 20 knots depending on taxiway turns and weight of the aircraft. The heavier it is, the harder to control. Only fuel the aircraft to the level needed to get you your planned destination plus one hour of contingency.

FS models surface friction on CORRECTLY DESIGNED SCENERY. Control is easiest on dry pavement and degrades on wet, ice, gravel, dirt, grass.

 

-Pv-

>

 

Good point, Paul. I fly the Level-D 767 and have always noticed how much better it steers on the ground than any of the stock FS aircraft do. It is much more predictable. I use the Logitech 3D Extreme flight stick. It is my second one and I think it has problems with on ground steering, anyway.

FS models surface friction on CORRECTLY DESIGNED SCENERY.

 

I guess I've never used any of that type scenery...

"Logitech 3D Extreme flight stick. It is my second one and I think it has problems with on ground steering, anyway. "

 

Joysticks and their software (drivers) don't know anything about the state of the aircraft and the environment the aircraft is in. They just send control input to the OS and program based entirely on user action and the sensitivity adjustment made by the user.

The stick and its software does not check or respond to being on the ground or in the air or on a taxiway or in slew or in wind, rain or anything else. A minor exception is FFB sticks which have SOME state detection and modification of their behavior in the form of responding to moments (forces) but that does not keep even them from sending their full range of configured control responses the user has permitted with sensitivity adjustments.

 

If your stick works good in one plane, and not another the obvious need not be stated.

 

-Pv-

"I guess I've never used any of that type scenery... "

 

Rather, I would propose you have not flown any correctly modeled aircraft.

 

Added:

Also, correctly modeled aircraft which are operated on the ground to real world specification. This brings up the point of "skidding."

 

How do we in the real world know a car we are driving is skidding? Most of us know how to introduce skidding in a car. We turn sharper and at a higher speed than the tires can hold the surface. We get a fair amount of feedback just prior to and during the skid which informs us of the severity of the skid and what we did to cause it, and when we have applied the corrective action. We feel the introduction or loss of expected G forces.

 

Now imagine some of the car skids you have been in and experiencing them from a camera in a room somewhere else sitting in a chair. There is no sound other than the car engine. You cannot see the driver's hands on the wheel. You cannot feel the car move or turn. One of the few clues you had there was any skidding going on is if the driver didn't stay exactly in the lane. You might ask how good of a driver is this? Did he turn too fast or too slow, was he going too fast for the turn? What is the condition of the road? Is there any wind and what direction and strength? You cannot FEEL the G forces affecting your body in the car as the driver controls the car. You cannot see the speedometer and there is no skid indicator on the dash. You cannot hear the tires screech or scrape.

 

In FS as pilots at the "wheel" we are missing a LOT of the information we expect to have when controlling a machine. In addition, aircraft have tripod not quad wheels.

Kind of like riding a tricycle or three wheel bicycle. They tend to "lean" more than a car.

 

Because of these many factors, we have to rely on some sublime information to judge the performance of ourselves and the model. One of those is speed. If we are moving faster than the specifications allow for taxi and turning, a correct model will begin to skid in a turn. How do we know it's skidding when we cannot feel it? If our controller is at the extreme and the model is not correcting, that is one clue. Another is the skid indicator on the HSI or AI. If you are on a dry paved bit of default scenery in a 90 degree turn to the hold short in a medium airliner at less than max take off weight at less than maximum taxi speed for the turn (usually 4 - 10 knots) using differential breaking and the skid indicator moves, then it's likely you should be looking for a better model. The tricky part our brains object to is we are 100% dependent on visual cues and those cues can lie to us if we do not mentally factor in certain data like speed, weight, wind and roll angle.

 

Alternatively, if under the same conditions you hit that turn at 14 knots or more and see the skid indicator move, then FS and the model are doing what they are supposed to. Single engine planes in the same conditions will have nearly the same speed reactions. In addition, factor in wind. We hit that same left turn at 10 knots during a 20 knot wind gust from 90 degrees to our right. That wind hits the tail pushing the nose to the right counter-acting on our control inputs making us think we are skidding, but we are not. Instead, our control inputs have been halved or less. This is normal behavior and not a defect in the core of the simulation. Some aircraft will not roll at all in the turn no matter how harsh but behave more like a car with a 30 foot wheel base. This removes one of the bits of visual feedback we need to judge the effectiveness of our control input and is a basic flaw in many models.

 

-Pv-

I have also found that when the AutoPolot is engaged, it affect the rudder control of the nose wheel.

 

Try engaging and then disengaging the AP and see if this helps.

 

RodeoGeorge

"AutoPolot is engaged, it affect the rudder control"

 

This would happen if the yaw damper were left on or the HDG/Nav modes on. Otherwise, the AP will not affect steering.

 

-Pv-

I won't attempt to go through your entire post, but:

 

Another is the skid indicator on the HSI or AI. If you are on a dry paved bit of default scenery in a 90 degree turn to the hold short in a medium airliner at less than max take off weight at less than maximum taxi speed for the turn (usually 4 - 10 knots) using differential breaking and the skid indicator moves, then it's likely you should be looking for a better model.

 

I presume you're talking about the inclinometer, or ball. It reacts to sideways force, and is not, strictly speaking, a skid indicator, though it will show in flight when you are skidding, but also when you are slipping -- anytime the sideways forces are not balanced. When on the ground, it does not show skid, only a sideways force acting on it, even though the tires have enough friction to prevent skidding at the specific rate of turn in use.

 

Rather, I would propose you have not flown any correctly modeled aircraft.

 

I guess I need some lessons in this stuff, so what is an example of a "correctly modeled aircraft?" Perhaps two or three of them so that I have choices? How about a "CORRECTLY DESIGNED SCENERY?" Should I also presume that the default aircraft and scenery were not "correctly modeled?"

 

I really want to see this in action.

What is "it" exactly you want to see in action? What specific behavior under what conditions are you looking for? Runway, taxi, clear wx or otherwise? Taxi speed, rate of turn? Controller deflection?

 

I fly a LOT of models, most of them ones I designed and I'm satisfied with the ground steering on all of them, all in default scenery.

 

It's likely you'll not be interested in the models I've produced. Comparatively few people have downloaded them when considering the ever popular 737s for instance.

-Pv-

Money where my mouth is... Actually, I wish I had some money where my mouth is ;)

 

https://www.flightsim.com/vbfs/fslib.php?searchid=32061391

https://www.flightsim.com/vbfs/fslib.php?do=copyright&fid=168549

https://www.flightsim.com/vbfs/fslib.php?do=search&fsec=0&fname=CONCALL2012_PATCH2.ZIP

 

See how the steering reacts in the Clear theme when the fuel is configured to specification for take off before you test with other weather applied.

Needless to say SENSITIVITY is subject to your controller settings. It's impossible for a modeler to control that aspect. In MY FS settings, all my JS control sliders are CENTERED as a standardized test config to accommodate the most number of people.

 

-Pv-

It might be worth juggling the 3 'Aircraft realism' slider settings to see if they have any effect on steering etc

What is "it" exactly you want to see in action? What specific behavior under what conditions are you looking for? Runway, taxi, clear wx or otherwise? Taxi speed, rate of turn? Controller deflection?

 

The "it" I want to see is the "realistic" ground friction you touted. I downloaded your Concorde, and its acceleration was anemic, to say the least. Also, it kept wandering all over the runway, instead of tracking straight as all my aircraft do (rudder was having a very delayed effect), and an attempt to turn made it slide sideways (no, I was nowhere near takeoff speed -- I couldn't get there on the 12,000 foot runway) with almost no turn occurring.

 

So I've deleted your "correct" aircraft.

 

Let's just forget it. We'll just need to agree to disagree...

You do not understand the model and you did not read the documentation. The yaw you experienced on take off roll is how the real plane behaves when engine 4 is throttled back to prevent the right wing vortex from being ingested into the engine and causing a flame out. It has NOTHING to do with the nose wheel steering and runway friction. When the engine is throttled back the pilot compensates with rudder. When the speed is such that the vortex is forced behind the engine, the throttle returns to normal and the aircraft behaves as expected.

 

This is an ADVANCED and realistic model requiring advanced knowledge, skill and study.

It cannot be flown to specification without detailed understanding of the aircraft operation and specification as provided in the documentation including the kneeboard checklist and specifications. Fuel loading, speeds and throttle settings are all critical at various altitudes and air speeds. Unlike a lot of primitive FS models, the engine power is realistic. The plane is not a fighter jet and did not have a wide margin of power at full weight take off. The plane should only be fueled to the amount needed for the distance. If you insist on always taking off full weight, then you have to accept the cost of operating that way.

 

-Pv-

I've long been concerned about ground steering. It often seems to me like the pedals are connected to the nose wheel by bungee cord - there's a delayed reaction, which can make steering difficult, and I sometimes end up swaying from side to side as a result, along the runway. Surely real planes don't handle like this? If only the steering acted like in a real car, without the sloppiness. Can anything in the aircraft cfg file be tweaked to reduce this effect?

Cars are SO much different from aircraft, different skills are required.

 

1) Cars to not have controls that gather and redirect the wind to change direction and rise off the ground.

2) Most aircraft have wheels in three positions not four.

3) Most aircraft are much heavier than cars so they have more inertia and momentum.

4) Many pilots while learning try to use joysticks/yokes like car steering wheels and over compensate. Over sensitive controls also contribute to this.

5) In cars, steering has a much more immediate reaction than aircraft and you can FEEL the response. In FS, you cannot FEEL the response of the controls. Think to your first attempts at learning to control computer race cars and how you had to re-train yourself not to over compensate.

6) In aircraft the general rule is to move the control a VERY SMALL AMOUNT and WAIT...

Observe over time how the aircraft responded to the small control action then CORRECT the action by a VERY SMALL AMOUNT and WAIT...

6) As the aircraft accelerates down the runway, it slowly transitions from a BUS to an AIRPLANE as enough air speed moves across the directional surfaces to grip and redirect the wind. While the aircraft is still a BUS, there are factors to take into consideration:

a) Weight and speed. If the speed during taxi exceeds the ability of the tires to grip the surface, the aircraft will skid. The heavier the plane, the slower you have to turn a corner on the taxiway (like a car.)

b) Wind. The aircraft is DESIGNED to grip and redirect air. This makes the aircraft more sensitive to wind than a car. One of the most misunderstood features of an aircraft on taxi and take off roll is the tail. While the plane is moving slow on the taxiway, the tail often works against the controls. The large surface at the opposite end of the plane from the nosewheel can work against the pilot's attempts to steer the plane. THink of it like a steering wheel at both ends of the car. It's important to know what the wind direction and speed is on the ground so you understand how your controls will change. I discussed this in posts above.

 

As your take off roll increases, the nosewheel becomes less effective as the tail gets enough air speed moving across it to take directional control. This transition period is the most difficult and sometimes dangerous part of the take off. With no wind, it's relatively easy. As the plane accelerates, the tail takes increasing control forcing the nose to straighten out and the roll becomes increasingly easier as speed increases until take off. At that point, the nosewheel loses all effect and there may be a few moments of transition as the pilot adjusts to wind conditions having FULL effect on the aircraft and the nosewheel none.

 

In short, take off roll for new pilots is one of the most common problems while learning not to overcompensate in the simulation. I still insist and will continue to do so, that it's absolutely possible for an experienced pilot with a carefully configured controller to direct the plane straight down the runway through a take off roll even with significant wind.

Notice how a child in a car seat with a steering wheel tries to mimic the parents by swerving the wheel all over the place. It makes us smile. Now, when you as a driving trainee first gripped the wheel to learn to drive, the reality set in on how much you misunderstood how the wheel really effects the car and you had to re-train yourself not to use your childish understanding and stop over-compensating. "Driving" aircraft is the NEXT level up from CARS in understanding inertial forces and momentum. Allow yourself to be re-trained instead of fighting it.

 

Here is a video I made of the Concorde (concall2012) taking off from Honolulu in wind. Notice the slight yaw as the throttle for engine four moves up and down. Watch the video several times by downloading it to your computer. Unlike most planes, there is a control position gauge on the panel so you can watch me control the plane. Watch this gauge carefully.

 

https://www.hightail.com/download/UlRTNWNsT010Njk4SjhUQw

 

-Pv-

As I said above

Let's just forget it. We'll just need to agree to disagree...

 

You do not understand the model and you did not read the documentation.

You're absolutely correct! I didn't know I needed to "understand the model" and I didn't know I'd have to "read the documentation" to get a little speed and try a turn to see the friction. I don't need the aggravation of supposedly complex airplanes that won't behave in a real world fashion. Sure, I don't know the systems, but I should be able to fly it, stick throttle and rudder. And how was I to know that you would AUTOMATICALLY reduce power on an engine. I have two hours in a 757/767 sim (United, full motion), which was enough to show me that they behave just like other aircraft, on the ground and in flight. I have flown over 60 types of real world aircraft. And they all have certain behaviors in common with the 767, as well as with each other, even though they may have complex SYSTEMS on board.

 

Fuel loading, speeds and throttle settings are all critical at various altitudes and air speeds.

 

I didn't even get airborne, and in any case those items have nothing to do with surface friction. I asked for something to display your idea of realistic friction and you give me something that, according to you, takes a lot of study just to get it to roll straight down the runway. How about a Cessna, or a Beechcraft, or even a 767?

 

Again, I say

Let's just forget it. We'll just need to agree to disagree...

 

I'm done!

"And how was I to know that you would AUTOMATICALLY reduce power on an engine"

 

That's what documentation is for and why I went through the effort to provide it so you would know.

 

"I have two hours in a 757/767 sim (United, full motion), which was enough to show me that they behave just like other aircraft, on the ground and in flight. I have flown over 60 types of real world aircraft."

 

The Concorde is unlike every other aircraft ever built. None of the aircraft you mentioned experience in have a delta wing and take off, land, and fly mach 2 without flaps. None of the airliners you mentioned have afterburners. None of the aircraft you mentioned have had to modify the engine controls to keep engines from flaming out on take off because a wing vortex is IN FRONT OF AN ENGINE. The Concorde takes off around 200+ knots and lands around 160 knots. The 767 does not normally operate this way. These are only a few characteristics that make the Concorde unique among ALL airliners.

 

Not every airline pilot could fly the Concorde.

 

-Pv-

The Concorde is unlike every other aircraft ever built. None of the aircraft you mentioned experience in have a delta wing and take off, land, and fly mach 2 without flaps.

 

And what does that have to do with surface friction? I didn't want to fly a Concorde, I just wanted to see the realistic surface friction you claimed.

 

Bye!

Hey aeroone,

In the real world C172 there is no connection between the rudder control and the nose wheel. The nose wheel casters in response to other forces. At low speeds the brakes and then as speed increases, the rudder.

As a sidenote that may be of interest I'll relate my experiences of approximately 2 years ago flying three different light aircraft. I took control shortly after liftoff until the threshold when landing. I try to simulate the real world as accurately as possible in FSX and felt vindicated that everything felt very similar when actually flying. The biggest problem I had was taxiing. I had a heck of a time trying to stay on the yellow line and often would actually turn in the opposite direction I wanted. My focus was on a lot of different things and I didn't have a lot of time to analyze why I was having this problem. I did not have rudder pedals prior to flying but got some shortly thereafter. When I would taxi in FSX I found I was having the exact same problem that I did when at the aircraft's controls. But now I had the time to figure out what the problem was. I was used to downhill skiing and in it you apply pressure to the right foot in order to turn left; just the opposite when using the brakes in an aircraft. So if your a skier be warned.

Jim F.

 

Edit: How about "if you're a skier"? That seems better.

"And what does that have to do with surface friction? I didn't want to fly a Concorde, I just wanted to see the realistic surface friction you claimed."

 

The realism I claim is only possible if the model is operated to specification. One test to verify ground steering actually WORKS is to just taxi around without trying to take off. Do this with no weather so there are no unpredictable side forces as you are getting used to the feel of the controls. This way you do not have "fly" the Concorde to experience tire/surface interaction.

 

The things which should be noted is:

1) Unlike a single engine prop plane or MOST airliners, the nosewheel is 20 feet BEHIND the pilot. This means you have to turn to follow a taxiway line 20 feet AFTER your viewpoint passes over it.

2) The slower you go, the more responsive the steering because it's primarily speed which causes the wheels to lose contact with the surface to cause skidding. If you want even more responsive steering, reduce the fuel weight but keep the CG within specified limit. The Concorde has a 3400 miles range. You do not need all that fuel for a test flight around the block.

 

Watch my video which should prove it's possible to direct a FSX model down the runway without losing control even in a slight crosswind and an engine that partially fails during take off roll ;)

 

-Pv-