hi,

 

this effect happens to me with the default baron, king air c172 as well as with my carenado c152. i do not think a fm tweak will help here.

 

the situation should be easy to reproduce, some light wind from variable directions in the metar string. i'm sure enough feedback has been provided. i'm sure we will get some update in one of the next betas ;-)

 

anyway, x-plane itself and the new conversion style here in this forum is just fun!

I am curious, although I dont have any idea about plane maker and the variables required to create a flight model it still has me wondering what you could change in an aircraft to cater for turbulence.

 

I imagine that there are a number of variables that help you model an aircraft. Lets say you get these right and are happy with your flight model, it performs excellently. Then you find that turbulence bounces you around when it shouldnt even though the aircraft is modeled so well.

 

So, what is it you can change that would dumb down the turbulence while not having an impact on the flight model that you know is correct?

I am curious, although I dont have any idea about plane maker and the variables required to create a flight model it still has me wondering what you could change in an aircraft to cater for turbulence.

 

I imagine that there are a number of variables that help you model an aircraft. Lets say you get these right and are happy with your flight model, it performs excellently. Then you find that turbulence bounces you around when it shouldnt even though the aircraft is modeled so well.

 

So, what is it you can change that would dumb down the turbulence while not having an impact on the flight model that you know is correct?

 

I'm really hesitant about posting screenshots and pointing people in the general direction on what can get done to get things working more accurately. The reason is, the last time I did it, my inbox was flooded by people wanting instructions on how to make a flight model. I'll post one screenshot though on one area where the flight model can be adjusted to compensate for turbulence. Like I said, I think the turbulence isn't that bad. It's just that people probably don't know where to adjust the flight model to compensate for the turbulence. And now that X Plane has had its turbulence reduced, it just makes it easier to adjust the flight model. Also, the settings I have set are still being worked on. But it's definitely getting there.

This is a screenshot I took of the Saab flight model I was testing WITH turbulence set at level 3 in X Plane 9. Obviously for X Plane 10, I'll change the settings.

Keep in mind that this is just one area where the flight model can be adjusted.

I'll also point out that yes, this is in sim and I'm adjusting the settings on the fly, but Planemaker has the exact same settings. The difference is, doing it in sim saves me from adjusting it in Planemaker, saving it, then reloading the aircraft in X Plane, taking off and then testing. When it's in sim, I can keep flying and any changes I make are done in real time. Once I'm happy, I make the changes in Planemaker.

Ignore the rather rudimentary panel. I add what I need for testing purposes, and I was testing engine startup and autopilot behaviour. (Yes, the autopilot can be immensely adjusted to perform as accurately as the real aircraft!)

 

EDIT: Just to add, before you ask "Why don't the guys at LR make the same adjustments for the default aircraft??" or "Why don't other developers make the same adjustments for their payware??"...well, I can't speak for other payware developers. Maybe they do make the adjustments, maybe they don't. But as far as LR, I can only assume it's because making these adjustments is extremely tedious and time consuming. It took me about 4-5 hours to make the adjustments I already have, and I consider it half way there. Add that to the 12-18 month time frame it has taken me to build the Saab flight model, and it all gets put into perspective. I've said it a few times before...X Plane adds default values to areas the developer misses. These default values are often wrong. example: Tire friction coefficient does not change unless the developer changes it. Fuselage friction coefficient...same thing.

Remember...garbage in=garbage out.

Here's another screenshot to show a better picture of what is happening. I set turbulence at Level 3. Reset all my adjustments in the flight model that pertains to tolerance of turbulence to default settings, took off, engaged autopilot and let it fly for a minute (notice the "roller coaster" like flight path). Then, paused the sim, re-entered all my settings, unpaused the sim, and let it continue flying while changing heading.

The whole flight, with the exception of the takeoff, was under autopilot control. You can see the differences. Seeing as this is in X Plane 9, and X Plane 10 has had its turbulence effects reduced, the flight path will be even more stable with the same settings in X Plane 10.

Thanks Goran, but I have no clue what settings you are changing in Plane Maker to settle down the turbulence. Again yesterday, while on final in light winds, the Duchess was all over the place and was rolling to about 60 degrees, very frustrating. I noticed the turbulence was showing .080. There is no reference in the Plane Maker manual to turbulence as well, doesn't really help the end user. Is this stuff a secret?

 

Xplane 10 beta 9

Thanks Goran, but I have no clue what settings you are changing in Plane Maker to settle down the turbulence. Again yesterday, while on final in light winds, the Duchess was all over the place and was rolling to about 60 degrees, very frustrating. I noticed the turbulence was showing .080. There is no reference in the Plane Maker manual to turbulence as well, doesn't really help the end user. Is this stuff a secret?

 

 

Honestly I don't think that, unless really out of sync with reality we should touch the radii of gyration because of the weird turbulence effects...

The problem is in the way x-plane is reading/parsing the METAR string and applying what it finds to be the adequate weather effects according to it's weather variables.

 

Problem is it usually defines turbuence, sometimes above 1, when variable wind (which is based on a 10 minute average) is reported, or gusts... (gusts are based on 3'' averages from the reported medium wind (10 minutes))

 

Also, wind variable when in METAR results in way too high a rate of change in the sim. The wind vector is allover the place, and the variations on a reported faint wind should really be a lot less intense....

 

The problem, IMHO, resides in the way X-Plane translates METAR data...and the fact that, among other things, it does not have a concept of variability - just shear, which is certainly NOT the same thing!!!!

 

dddffGf_mf_m followed by KT, MPS or KMH, usually KT can be followed by a d_nd_nd_nVd_xd_xd_x like in

 

31015G27KT 280V350

 

when during the 10' period taken for the average wind reported there has been a marked discontinuity lasting at least 2 minute in which case de mean value should be assessed over the period after this discontinuity.

 

A marked discontinuity occurs when:

 

- there is a direction change of 30 degrees or more with a wind speed of 10kt or more, or a wind speed change of 10kt or more.

 

The discontinuity should last for at least 2 minutes!!!

 

Gusts should be averaged for 3'' when variations from the mean 10' wind are observed....

 

Wind encoded as VRB[P]ff coorespond to situations in which wind speed is less than 3kt, or the wind speed is higher and direction varies by more than 180º and a single direction is impossible to detarmine (i.e. CB/thunderstorm over the aerodrome)

 

Thus, variable wind fields should generate variations lasting at least 2 minutes, instead of continuously varying like we see in the wind vector of i.e. the C90, or in the case of faint winds with a VRB group use turbulence and higher variation rates only when the wind is stronger and then make it vary more tha 180º...

 

Other criteria for setting turbulence values in the range [0..10] should be considered... I'll think about that latter...

To be really honest (and not rude), I could care less how this gets solved, as long as it gets solved, there is a problem here, whether some want to admit it or not. I thought Gorans posts were to shed some light on changes that could be made in Plane Maker to help eliminate some of the turbulence effects in a given aircraft. I can't for the life of me understand or see what changes he might be making to achieve this.

 

I have been going into the weather after it's downloaded and changing the turbulence, would be nice if this wasn't necessary.

 

Glen

I have been going into the weather after it's downloaded and changing the turbulence, would be nice if this wasn't necessary.

 

Not enough, you will also have to change the Shear direction, because it only makes sense when there is shear reported, and only up to 500'.... When used to translate wind variability, the effects will always be too intense.

 

When I take the best aircraft in my hangar - LES dc3 - and it waves like a parrot in a light variable wind, the last thing I would do would be to ruin it's great "airfile" to overcome a weather settings problem, so, until this gets fixed for good in x-plane10, I will simply weep any turbulence or shear values unless thunderstorms or really strong winds are reported!!!

When I take the best aircraft in my hangar - LES dc3 - and it waves like a parrot in a light variable wind, the last thing I would do would be to ruin it's great "airfile" to overcome a weather settings problem, so, until this gets fixed for good in x-plane10, I will simply weep any turbulence or shear values unless thunderstorms or really strong winds are reported!!!

 

Totally agree, this is Austins problem, not Gorans. I'm doing the same, adjusting the real weather. Nine months later and were still trying to get the turbulence sorted out...hmmm.

 

Glen

 

Glen

How about a check box: Turb. full, half or off.

I wouldn't consider it a secret. The reason that I, personally, don't give out the full details of it is because some people might take what I tell them, make their own changes and totally screw up the flight model. Like I said, it's very tedious work and a change in 1 area can have exaggerated effects in another. And just for a bit of clarification, I've left the radii of gyration at default values. Sometimes it's necessary to adjust it, but, with the Saab, I'm trying some different things. I probably will need to adjust it later, at least in the roll, because it's a little too responsive, even after the other adjustments have been made.

The updated flight model for the Duchess has these fixes in place already. However, I have to still run a final check on everything and add a few other details to the 3D mesh.

Also, the Duchess is a 3 year old add on, and I didn't find out about these turbulence adjustments until much later.

Glen, check inbox.