I hope! :-) There are good chances that he will have a look at this issue.

I hope! :-) There are good chances that he will have a look at this issue.

 

That's great to hear! Thanks for all you have done Murmur.

More experiments, and the results are again surprising.

 

So, looks like the tire physics in X-Plane, contrary to my previous apparent results, is fundamentally correct.

 

Or, better said, it's not the culprit for the lack of control authority on crosswind operations.

 

In my previous results, I established that the aircraft, when not moving forward, creeps into the wind for low values of the crosswind. This had led me to believe that the max yaw moment applied by the landing gear was much lower than it should be.

 

But, as it turns out, this creeping motion is a bug that is only present when the aircraft is not moving forward, and is probably due to X-Plane using two different friction models for stopped tires and moving tires.

 

Once the aircraft starts rolling, the landing gear is indeed able to apply all the expected yawing moment. So, the lack of control authority during crosswind takeoffs and landings seems to be caused by excessive yaw stability (weathervaning tendency) or insufficient rudder authority, or a mix of both. So, we're back at step one.

 

So, to recap:

.) the tire model seems to be fundamentally correct when the aircraft is rolling on the runway;

.) there is a bug in that the aircraft creeps into the wind when it is not moving forward and is not braked;

.) this bug causes an unrealistic behaviour when the aircraft is not moving, but seems not to be responsible for the lack of control authority during crosswind takeoffs and landings.

 

Will keep experimenting.

A big step forward.

 

I modified the default C172 so that it has a yaw (weathervaning) stability similar to the real C172 (with reference to the real world engineering data I have), and hence much less than the default X-Plane C172.

 

Now the aircraft can be operated in a significant crosswind (I tried up to 25kts so far) without significant problems. The annoying tire screeching is still there of course, but the handling seems much better.

 

Not only that, but also it doesn't creep anymore when not moving (with the engine off), even for strong crosswinds of 30kts or more.

 

So, all the signs seem to point that the cause of the unrealistic weathervaning tendency is that X-Plane calculates an exaggerated yawing moment due to the tailfin. The cause of this has to be further investigated.

 

I will try to upload the modified C172 in the next hours or days, so that you can try it by yourself.

A big step forward.

 

I modified the default C172 so that it has a yaw (weathervaning) stability similar to the real C172 (with reference to the real world engineering data I have), and hence much less than the default X-Plane C172.

 

Now the aircraft can be operated in a significant crosswind (I tried up to 25kts so far) without significant problems. The annoying tire screeching is still there of course, but the handling seems much better.

 

Not only that, but also it doesn't creep anymore when not moving (with the engine off), even for strong crosswinds of 30kts or more.

 

So, all the signs seem to point that the cause of the unrealistic weathervaning tendency is that X-Plane calculates an exaggerated yawing moment due to the tailfin. The cause of this has to be further investigated.

 

I will try to upload the modified C172 in the next hours or days, so that you can try it by yourself.

 

Awesome ! The community is in debt to you ! Laminar should cut you a check ! Looking fwd to this ! Where are you going to post this ?

Here. It's just a rough modification of the default C172 so that you can try it. The underlying issue is apparently internal to X-Plane flight model, so it should be tackled directly by Austin. :-)

I suggest you pass your results to Austin, I bet that with bug and fix provided he cannot postpone the issue anymore.

Great finding!

 

Did you get it through a significant reduction the the fin/rudder area ?

 

In the past I had tried it with the default 744, by continuously reducing the tail area, but my aim was to find out if fuselage drag did play a role under windy conditions on ground...

Here it is:

 

https://drive.google.com/open?id=0B5fyJwRhjar5Zk1wR0l3ekszUWs

 

You have to replace the original Cessna_172SP.acf file. Remember to make a backup of the original .acf, but in any case, running the X-Plane 11 updater will allow you to restore it.

 

I can takeoff and land with a direct crosswind of 25 kts (without using differential brakes).

It is advisable to keep some weight on the nosewheel (pushing the yoke) on takeoff and landing, to improve directional control.

You'll need crossed controls to keep the wings level.

 

With some difficulties, I also managed to land with a direct crosswind of 30 kts (very tricky), that is more than I could in Aerofly FS or Flight Gear.

Did you get it through a significant reduction the the fin/rudder area ?

 

In the past I had tried it with the default 744, by continuously reducing the tail area, but my aim was to find out if fuselage drag did play a role under windy conditions on ground...

 

I reduced fin area, deleted fin root extension, deleted fuselage, added a hidden wing in front the propeller.

 

With this modifications, I get (in cruise conditions) directional derivatives very close to the real one.

 

Now the problem is, what is the origin of the issue? I suspect excessive sideforce due to fin. But a detailed analysis must be done to understand it better.

Your finding can also help making Austin have some insight over the problem through a wider range, probably explaining why all aircraft also in tghe air are so sensible to "slight shear" and wind intensity variations...

 

In my 744 tests I ended up completely deleting the tail surface, and then the 744 behaved much more plausibly during the takeoff run with a x-wind component, But unfortunately unless we had "hiden" surfaces, the fuselages themselves do not contribute much to directional stability, nor to the correct taking into account of the various prop effects...

 

X-Plane really needs Austin's attention in these área.... It should not continue being best known as the "Simulator of the great night looks" and now day looks too...

 

And again, thatnk you Murmur for your dedication - email it to Austin! and Ben!

Did you remove the  empennage of the 747 ?  now there won't be a yaw moment  when there is X-wind ? how will that big bird be stable without it ?

Did you remove the  empennage of the 747 ?  now there won't be a yaw moment  when there is X-wind ? how will that big bird be stable without it ?

 

​only during takeoff rol, and it wasn't, because of the excessive weathervane stability...

Been tracking this thread and I am confused. Did Austin really focus on tire force modelling when NOT moving as opposed to tire force modelling when moving and in weather? That seems like an absurd prioritization. Or am I comparing apples to oranges here?

 

Here is the thread I am referring to: http://forums.x-plane.org/index.php?/forums/topic/111423-coming-in-xp11-beta3-tire-force-modelling-when-the-plane-is-not-moving/

 

Someone tell me I am missing something because this seems like a much more pressing concern than the one he tackled?

​only during takeoff rol, and it wasn't, because of the excessive weathervane stability...

 

So you mean to say the tail surface is the issue in XP ? meaning when the wind hits it there is a bad yaw movement ? 

Been tracking this thread and I am confused. Did Austin really focus on tire force modelling when NOT moving as opposed to tire force modelling when moving and in weather? That seems like an absurd prioritization. Or am I comparing apples to oranges here?

 

Austin works on what he's interested in at the time. Apparently it has bugged him for a while that running up the engine while standing on the brakes doesn't lower the nose, so he decided to improve that aspect of the sim. 

 

I wonder if maybe Austin hasn't noticed or cared about this weathervaning issue because he never flies tailwheel planes in the sim or real life, where the problem is more noticeable.

 

Anyway, if enough people mention unrealistic weathervaning -- and this is the PERFECT time for it, during a beta before major release -- then maybe it will get addressed. It can take a while when the parameters of the problem aren't known, but it sounds like Murmur has made good progress here.