PMDG 747-400 trouble with turbulence with 4.28

[ahaka]

Hello,

I've posted about this in PMDG support forum too here: http://forums.avsim.net/dcboard.php?az=9015&page=

It seems to be that after installing FSUIPC 4.28 the aircraft banks steeply to left or right, with or without autopilot, in turbulent conditions. I'm talking about more than 60 degrees bank here, and then the plane comes crashing down, or at least gets totally off course.

So far it seems the worst banking has appeared in cloud turbulences, but some banking at least also happens in wind turbulence.

Could it be that theres something new in 4.28 causing this as I dont recall having it with 4.26? could it be the vertical air added in 4.28?

Disabling all turbulences in FSUIPC solves the problem, so at least it seems very clear that the cause for the unstable behavior of the plane is the turbulence emulations.

Here's my FSUIPC4.INI file, in case there's any help from it:

[General]

CustomWeatherRewrite=No

History=X7GAGJQJXY0EVGE5381M9

TCASid=Flight

TCASrange=40

AxisCalibration=No

DirectAxesToCalibs=No

ShowMultilineWindow=Yes

SuppressSingleline=No

SuppressMultilineFS=No

AxisIntercepts=No

WeatherReadFactor=2

WeatherRewriteSeconds=1

SimConnectStallTime=1

MouseWheelTrim=No

FixControlAccel=No

VisibilityOptions=No

OneCloudLayer=No

CloudTurbulence=No

CloudIcing=No

GenerateCirrus=No

SuppressCloudTurbulence=Yes

MaxIce=-4

MinIce=-4

UpperWindGusts=No

SuppressWindTurbulence=Yes

SuppressWindVariance=No

WindTurbulence=No

TurbulenceRate=1.0,5.0

TurbulenceDivisor=20,20,40,40

SuppressAllGusts=No

MaxSurfaceWind=0

WindLimitLevel=200

WindDiscardLevel=400

WindAjustAltitude=No

WindAjustAltitudeBy=2000

WindSmoothing=Yes

WindSmoothness=1

WindSmoothAirborneOnly=No

PressureSmoothness=0

TemperatureSmoothness=0

DisconnTrimForAP=No

ZeroElevForAPAlt=No

ThrottleSyncAll=No

WhiteMessages=No

ShowPMcontrols=No

SpoilerIncrement=512

MagicBattery=No

RudderSpikeRemoval=No

ElevatorSpikeRemoval=No

AileronSpikeRemoval=No

ReversedElevatorTrim=No

ClockSync=No

ClockSyncMins=5

ClearWeatherDynamics=No

OwnWeatherChanges=Yes

TimeForSelect=4

LoadFlightMenu=No

LoadPlanMenu=No

PauseAfterCrash=No

ZapSound=firework

ShortAircraftNameOk=No

MouseWheelTrimSpeed=1

FixMachSpeedBug=No

[WideServer]

WideFSenabled=Yes

[GPSout]

Port=

Speed=4800

Interval=1000

PosTo6Decimal=Yes

Sentences=

GPSoutEnabled=No

[AutoSave]

AutoSaveEnabled=No

[JoystickCalibration]

ExcludeThrottleSet=Yes

ExcludeMixtureSet=Yes

ExcludePropPitchSet=Yes

SepRevsJetsOnly=No

ApplyHeloTrim=No

FlapsSetControl=0

FlapDetents=No

ReverserControl=66292

Reverser1Control=66422

Reverser2Control=66425

Reverser3Control=66428

Reverser4Control=66431

MaxThrottleForReverser=256

AileronTrimControl=66731

RudderTrimControl=66732

CowlFlaps1Control=66162

CowlFlaps2Control=66163

CowlFlaps3Control=66164

CowlFlaps4Control=66165

SteeringTillerControl=0

MaxSteerSpeed=60

Aileron=-16383,0,0,16383

Elevator=-16383,0,0,16383

Rudder=-16383,0,0,16383

Throttle=-16256,16256

LeftBrake=1203,7899

RightBrake=1203,7834

[Pete Dowson]

It seems to be that after installing FSUIPC 4.28 the aircraft banks steeply to left or right, with or without autopilot, in turbulent conditions. I'm talking about more than 60 degrees bank here, and then the plane comes crashing down, or at least gets totally off course.

So far it seems the worst banking has appeared in cloud turbulences, but some banking at least also happens in wind turbulence.

The turbulence emulation has been thoroughly tested and approved by PMDG's own test pilot. There's been no changes at all in this since then. Obviously extreme turbulence is to be avoided. What level did you see?

Could it be that theres something new in 4.28 causing this as I dont recall having it with 4.26? could it be the vertical air added in 4.28?

What vertical air added? You mean the the fix to the bug which prevents thermals and so on from operating? FSUIPC doesn't add any vertical component EXCEPT in turbulence, so for tubulence that changed nothing.

Disabling all turbulences in FSUIPC solves the problem, so at least it seems very clear that the cause for the unstable behavior of the plane is the turbulence emulations.

In all levels of turbulence? I would expect severe turbulence to be unflyable, but that is very rare. Have you checked settings in your weather -- are you using ASX?

Here's my FSUIPC4.INI file, in case there's any help from it:

That shows all turbulence suppressed. You can play with the turbulence emulation via the parameters

TurbulenceRate=1.0,5.0

TurbulenceDivisor=20,20,40,40

Regards

Pete

[ahaka]

Pete,

It does happen even in light to moderate turbulence.

I'm using ASX with SP3, but it shouldnt override FSUIPC turbulence emulation right? I've also checked the wind values in turbulence and its going just like in 4.26, just that the plane is really banking like crazy. Even the autopilot wont do anything to it.

In my INI file the supression is on since the last test I did was disabling all turbulence.

I admit that this is a really strange problem, but I wish it can be solved since I cant really fly the PMDG plane (unless I disable all turbulence, but I would like it to work with turbulence on)

[Pete Dowson]

It does happen even in light to moderate turbulence.

Strange, as that's been perfect for weeks. However, it seems a lot of new stuff is happening since ASX SP3 was released.

I'm using ASX with SP3, but it shouldnt override FSUIPC turbulence emulation right?

It instigates it. FSUIPC doesn't create any by itself unless you enable its random turbulence.

If you are using ASX SP3 you may want to add the line "CustomWeatherRewrite=No" to the [General] section of the FSUIPC4.INI file, as otherwise there seems to be a lot more unwanted interaction going on.

I've also checked the wind values in turbulence and its going just like in 4.26

Sorry, what does that mean? Where are you looking?

Unfortunately it seems folks are changing to 4.28 and ASX SP3 at the same time, even though 4.28 has been out there for 4 weeks with no weather problems at all reported until after ASX SP3 -- now suddenly I am getting "super fog" and "unmanageable turbulence" reported despite none of those areas being changed since 4.26. From what I've seen so far, and what I know about the 4.28 changes, I really must assume that the same things might arise with 4.26 + ASX SP3.

Please do try reducing the turbulence parameters, to see if those help. If not I suspect something else is going on.

Regards

Pete

[ahaka]

Hello Pete,

Changing the turbulence rate values did not help.

I guess all this is really due to ASX SP3 then. What a "nice" service pack... :? I'll try to change back to SP2 if possible and see if theres any change.

If you have any other ideas though, let me know.

[Pete Dowson]

Changing the turbulence rate values did not help.

What values did you try?

If you have any other ideas though, let me know.

Did you try adding that parameter to the INI file as I suggested?

Regards

Pete

[ahaka]

Pete!

I've discovered something while testing!

First of all, I tried everything without ASX. Well, the problem persisted.

Trying different scenarios, I found out that the severe banking and loss of control happens ONLY with CLOUD TURBULENCE. Checking the FSX wind values, in WIND TURBULENCE the actual wind values are not moving, just the indicate airspeed moves around and the airplane bounces a bit but stays of course. Well, with CLOUD TURBULENCE, the FSX wind starts shifting rapidly both in speed and direction, and no matter if its light to moderate or severe turbulence, at some point the airplane banks steeply to one direction and gets off course.

Both wind and cloud turbulence was ENABLED from FSUIPC when I was testing this.

Does this give you any ideas of what might be going on? Does the cloud and wind turbulence have different behavior written in FSUIPC code?

Regards,

Antti Haka

[ahaka]

Changing the turbulence rate values did not help.

What values did you try?

I adjusted the Turbulence rate values. I tried lowering to 2.5,0.5

If you have any other ideas though, let me know.

Did you try adding that parameter to the INI file as I suggested?

Yes I have added that parameter.

Regards

Pete

[Pete Dowson]

I adjusted the Turbulence rate values. I tried lowering to 2.5,0.5

I hope you mean the other way around, "0.5,2.5", else you would have increased the first one considerably!

Yes I have added that parameter.

Hmmm. I'm stuck then.

Before trying to get back to ASX SP2, why not try with FSX downloaded weather instead of ASX first. Might be easier. Try to find some with turbulence, or re-enable the random facility FSUIPC provides (but if you do that, you'd need to check the "change FS own weather" option before it will add the turbulence).

Regards

Pete

[Pete Dowson]

I've discovered something while testing!

First of all, I tried everything without ASX. Well, the problem persisted.

Ah, I just asked thatyour message crossed with mine.

Trying different scenarios, I found out that the severe banking and loss of control happens ONLY with CLOUD TURBULENCE.

Strange, as they are both emulated the same way, with the same code and same values. Only the test is different -- with wind turbulence it gets enabled by the turbulence value in the current wind layer, with cloud turbulence it's the same test but in the cloud layer.

Checking the FSX wind values, in WIND TURBULENCE the actual wind values are not moving

Sorry, wehat do you mean by "moving"? Where are you reading these "actual wind values"? And what is the difference with Cloud Turbulence? As I say, they are the same inside FSUIPC.

Well, with CLOUD TURBULENCE, the FSX wind starts shifting rapidly both in speed and direction

It's supposed to do that with wind turbulence too. They are (or should be) the same. Even for mild turbulence you should seem some small variations in both direction and speed.

I'll see what I can find. Probably not till next week some time now.

Pete

[ahaka]

Sorry, wehat do you mean by "moving"? Where are you reading these "actual wind values"? And what is the difference with Cloud Turbulence? As I say, they are the same inside FSUIPC.

What I mean by that is when you SHIFT + Z key the FSX displays its wind values. In wind turbulence, the values were not changing, not even a bit, but in cloud turbulence both the speed and direction were shifting/varying.

It's supposed to do that with wind turbulence too. They are (or should be) the same. Even for mild turbulence you should seem some small variations in both direction and speed.

I guess thats the strange thing here. It seems that the wind turbulence does add turbulence, but doesnt variate the wind, making it possible to stay on course with the PMDG plane. Hoever, when the cloud turbulence does indeed variate the wind, the problems start.

[Pete Dowson]

What I mean by that is when you SHIFT + Z key the FSX displays its wind values. In wind turbulence, the values were not changing, not even a bit, but in cloud turbulence both the speed and direction were shifting/varying.

Sounds like wind turbulence wasn't working thenthere should be some changes.

I've just checked the code. There's no separate code for wind and cloud turbulence. If gets the wind turbulence value (0-4), then checks for cloud turbulence (also 0-4) and takes the maximum of the two. After that there is absolutely no difference -- the number (0-4) is used as a multiplier for the maximum changes to be simulated.

I guess thats the strange thing here. It seems that the wind turbulence does add turbulence, but doesnt variate the wind

If it doesn't vary the wind it isn't doing anything! How can you tell you have turbulence if there's no changes?

Pete

[Pete Dowson]

Maybe you could save a flight and ZIP up the FLT+WX+FSSAVE files, send to petedowson@btconnect.com. Save one which definitely shows the problem, in case I cannot reproduce it.

Pete

[ahaka]

If it doesn't vary the wind it isn't doing anything! How can you tell you have turbulence if there's no changes?

I can tell theres turbulence because the plane is still wobbling around just like in normal turbulence, and the airspeed is variating rapidly just like in turbulence, so I'm 100% sure there is turbulence.

An update of the testing is that the "supress wind turbulence" did not remove this turbulence. Seems like the FSUIPC is not doing anything to the wind turbulenc. It just smooths the wind, but the turbulence is like its coming from FSX weather engine!

[Pete Dowson]

An update of the testing is that the "supress wind turbulence" did not remove this turbulence. Seems like the FSUIPC is not doing anything to the wind turbulenc. It just smooths the wind, but the turbulence is like its coming from FSX weather engine!

Hmm. In that case it sounds like the wind smoothing isn't working either, then, because the only reason I added turbulence emulation was to generate turbulence to replace the smoothed out air which smoothing produced. Seems that somehow what you've got is the reverse of everything I strove for between Christmas and April, spending many hundreds of hours, to achieve something like success. I really can't understand what has happened -- nor why it seems to be only happening to you, and four weeks after the last FSUIPC change?

weird.

Pete

[ahaka]

An update of the testing is that the "supress wind turbulence" did not remove this turbulence. Seems like the FSUIPC is not doing anything to the wind turbulenc. It just smooths the wind, but the turbulence is like its coming from FSX weather engine!

Hmm. In that case it sounds like the wind smoothing isn't working either, then, because the only reason I added turbulence emulation was to generate turbulence to replace the smoothed out air which smoothing produced. Seems that somehow what you've got is the reverse of everything I strove for between Christmas and April, spending many hundreds of hours, to achieve something like success. I really can't understand what has happened -- nor why it seems to be only happening to you, and four weeks after the last FSUIPC change?

weird.

Pete

Pete,

I only changed to version 4.28 recently. I had other things to do so I missed the exact day it was released.

I'm also puzzled and it seems like nothing is making sense. The wind smoothing IS working perfectly all the time no matter what is going on with the turbulence.

I am sending you the files and the PMDG panel state (the panel state file needs to be copied to the PMDG panel state directory) of a situation where the cloud turbulence is causing that problem for me. I am praying you would have the same situation so you could see whats happening.

[Pete Dowson]

I am sending you the files and the PMDG panel state (the panel state file needs to be copied to the PMDG panel state directory) of a situation where the cloud turbulence is causing that problem for me. I am praying you would have the same situation so you could see whats happening.

Okay. Thanks. I won't be able to get to this immediately -- I'm tied up till Monday or so. Watch this space.

Pete

[ahaka]

Okay. Thank you Pete.

[ahaka]

Small update:

I re-installed 4.26 and have been testing with the same scenario and so far I have not had any problems. Even in severe turbulence the plane keeps its course and does not bank more than 10 degrees.

It starts to seem like there really was something in the 4.28 that, if not to others, but to me at least caused some strange behavior in the PMDG aircraft.

[Meela]

I have experienced this same problem. I can confirm that it only happens while flying through clouds and having turbulence. Yesterday I did a re-install of FSX (w/Acceleration) and and added FSUIPC 4.28. Presto! my PMDG 747 goes out of control whenever flying through cloud turbulence. I was not using any FSUIPC with my FSX prior to this install yesterday.

I can also confirm this happens both with, and without ASX sp3.

My first post here - I came to these forums hoping to find a solution to exactly this subject. I hope a solution is found soon. I'm developing an unhealthy complex about clouds. It's not easy driving my jumbo jet through the sky like a sports car trying to avoid the clouds!

Steven

[Pete Dowson]

It's not easy driving my jumbo jet through the sky like a sports car trying to avoid the clouds!

Why not simply turn off cloud turbulence for the time being?

Pete

[ahaka]

It's not easy driving my jumbo jet through the sky like a sports car trying to avoid the clouds!

Why not simply turn off cloud turbulence for the time being?

Pete

Or go back to 4.26. In my tests yesterday, this really seemed to work for me. I had no problems with that version.

[Pete Dowson]

Or go back to 4.26. In my tests yesterday, this really seemed to work for me. I had no problems with that version.

That's all very well, but you'll be stuck on that version forever, despite any new facilities and fixes, unless we can sort out why 4.28 doesn't appear to work. And I am not going to support old versions, ever.

Best to stick to the latest and temporarily turn off the cloud turbulence, as I said.

Pete

[ahaka]

Or go back to 4.26. In my tests yesterday, this really seemed to work for me. I had no problems with that version.

That's all very well, but you'll be stuck on that version forever, despite any new facilities and fixes, unless we can sort out why 4.28 doesn't appear to work. And I am not going to support old versions, ever.

Best to stick to the latest and temporarily turn off the cloud turbulence, as I said.

Pete

Well, I meant as a temporary solution ;) When the problem is solved, then of course will go permanently to the latest version.

[Pete Dowson]

Well, I meant as a temporary solution ;) When the problem is solved, then of course will go permanently to the latest version.

Well, I've tried your saved flight (which had user-set weather with a cloud layer featuring very high turbulence (level 4). I deliberately re-installed the release 4.28 (I'm a bit later here with some other changes), and the wind changes are small and smooth, as programmed. The wind speed changes up and down a little, by 1-3 knots and the wind direction likewise. nothing drastic and certainly nothing not already tested and found to be okay / good by the PMDG 747 test pilot. The loss of control problems in the earlier turbulence effects, before 4.26, were as you describe and were due to sudden changes as opposed to smooth up and down changes which I added.

I've also brought up my Source Control and examined al of the changes made between 4.26 and 4.28, and there is only one, small one, which has anything to do with weather, and that it the one to stop the vertical wind being changed all the time (nothing to do with turbulence or any of those facilities, just to do with thermals for gliding).

Interestingly, and perhaps meaningfully, the symptoms you describe, with cloud turbulence (only) being a problem, are exactly those which were apparent in a much earlier release, 4.25 or before, before the finalisation of the wind smoothing and turbulence emulation in 4.26. If you are genuinely using 4.28, then this is a huge puzzle -- I don't suppose you have any Logs left from when you were having these problems, have you -- the version number would show at the top of the Log?

Is it possible that you installed something which overwrote your 4.28 with some other, older, version? Please, to be really sure, can you download 4.28 again, from the Schiratti site, and re-install and check again, as I have just done? I'm beginning to think this is all one big mistake and I don't want to waste my time further.

I cannot emphasise enough that the treatment of wind smoothing and turbulence is absolutely identical in 4.26 and 4.28.

Regards

Pete