9 hours ago, Richard McDonald Woods said:

CH PRO PEDALS USB... Rudder axis assigned to Z Axis not reversed. Calibrated as per P3D instructions.

Saitek X52 Pro Flight Control System .... Rudder axis assigned to Rz Axis not reversed; and Throttle axis assigned to Z Axis not reversed....

I’d be willing to bet that this is the exact source of your problem. You have two separate hardware controllers with rudder axis assignments. Delete the rudder axis assignment in the X52 controller, and leave the assignment enabled for the CH Pro pedals.

If you have two individual controllers both assigned to a given axis, they can interact and cause all sorts of unpredictable control responses in the sim.

5 hours ago, pete_auau said:

some  how  I  doubt its  either  of  these  since  if it  was  than there  would  be  heaps  of posts  on this  issues  that Richard  has

Well, I do recall some posts over the years with various ActiveSky settings causing autopilot discos on approach.  Usually had to do with winds and turbulence settings and such. 

But yes, it may well be a controller issue, with the yawing motion and radical turning.

Richard, I just flew this STAR and approach (with an autoland) into stock MRLB.  The AP did not disconnect for me.  I use a CH yoke and proflight rudder pedals, and ActiveSky.  I'm not presently using FSUIPC.  My AS turb setting is currently at 30 I believe.  I had never flown the 777 into Liberia -- I say I will fly into there again with the 777 -- that was pretty fun and the runway isn't as small as I thought.

Gentlemen,

I no longer have the Orbx freeware airports installed, so will have to rely on future payware airports to show any further problems. 

I have now removed FSUIPC5 entirely, as I am sure that I don't need it.

I agree with Henrik about scenery not being the problem, and Pete about the absence of posts from other pilots. And I am glad that Rhett managed MRLB without mishap.

I do believe that Jim has hit the target with the double assignment. I have done exactly as he suggests

I thank you all for your thoughtful inputs to this problem. I owe Kyle an apology for my woeful ignorance about axis calibrations and his valiant attempts to explain where the problem may have lain.

Onto my next flight!

Kind regards, Richard😄

Not entirely on topic, but an observation from previous experience is that quality control on the Orbx freeware airports is all over the place.  I used to have endless trouble with those installed (ditto Orbx Vector).  Orbx seems best for small airports and the overall texture replacement stuff with the base and LC products.  I know that stability in the sim is so vital that I must favor my tubeliner obsession in terms of what I add into the sim.

Given my experience with other things, don't stop suspecting that the  777 can have quirks.  It was PMDG's first (and so far only) fly-by-wire bird and I think the implementation, even 5 years later, isn't quirk-free.

2 hours ago, ahuimanu said:

Given my experience with other things, don't stop suspecting that the  777 can have quirks.  It was PMDG's first (and so far only) fly-by-wire bird and I think the implementation, even 5 years later, isn't quirk-free.

Naturally, something always on my mind, but when it comes to issues like this, you'd see it all over if it were a problem with the product. The fact is, we don't. As such, the most likely avenue is hardware, and now that Richard has noted he isn't super knowledgeable about all the hardware he's stuffed into the sim (not a bad thing, and not something to apologize for, but something that can very negatively impact a sim session if configured appropriately), seems ever more likely.

Kyle,

Yes, the hard disconnect in Richard's case seems acute to some non-777 quirk specific to his environment. However, the idiosyncrasies I speak of are particularly noticeable on approach and landing in the 777. A project I started promptly upon 64-bit availability was overnight long-haul 6 nights per week.  My policy is to hand fly after stabilizing on G/S after around 2,500 feet radar alt, to be stable in the hand-flying stage by 1,500 feet. In my experience, the "quirkiest" between hand-flown approaches in the 777F and the 744F, by far, is the 777F.  I have to chase the FD with more yoke and throttle movement in the 777F than the 744F.  So, from just this simmer's experience, I am less surprised when I encounter, or about hear from others, strange 777 behaviors.

That said, at this point I am not implying that quirk == bug, just that odd behavior on approach is less surprising with the 777F. For me the 777F is not as sophisticated in the hand flying department as its older 4-engined cousin is. Thus, I know a major difference to be the fly-by-wire.  I wonder if I wont fall in love as much as I am hoping with the 748F because of fly-by-wire.  Of course my only experience with fly by wire is in flight simulators.  I personally find the airbus FBW to be less squirrelly on approach than the 777F's . 

I wonder how 2 assigned rudder axes shall make the airplane behave that awkward. the hardware is centered with springs to the 0 value (in contrast to the throttle axis/axes), more or less accurately depending on the quality and age of the controllers, but still to a very low value. now of course they can interact against each other when you're flying manually, using the rudder in crosswind situations and so on but make the airplane completely unflyable? I can't imagine that they are the reason for this behaviour. What would happen is that the rudder could always snap back to zero if you used the feet and then the joystick sends a zero.. The only logical explanation for me is that Richard had twisted the Z axis on the joystick without noticing it when he flew the airplane and actually screwed everything up himself, but that's something only he can know. 

29 minutes ago, Ephedrin said:

I wonder how 2 assigned rudder axes shall make the airplane behave that awkward. the hardware is centered with springs to the 0 value (in contrast to the throttle axis/axes), more or less accurately depending on the quality and age of the controllers, but still to a very low value. now of course they can interact against each other when you're flying manually, using the rudder in crosswind situations and so on but make the airplane completely unflyable? I can't imagine that they are the reason for this behaviour. What would happen is that the rudder could always snap back to zero if you used the feet and then the joystick sends a zero.. The only logical explanation for me is that Richard had twisted the Z axis on the joystick without noticing it when he flew the airplane and actually screwed everything up himself, but that's something only he can know. 

You're assuming the sim somehow logically assigned something to a centered axis, instead of a slider axis. In reality, the sim simply runs down a list. You'll note that button 1 is always assigned to the brakes. Axis X to roll, Axis Y to pitch (note that the Saitek HOTAS combos use X for engine 1 and Y for engine 2 on the throttle side, yet this gets assigned to roll and pitch, so this is one example of an axis that needs a center, but gets assigned to hardware axes that are slider-based...and it's not even the right axes). So, if he has an Rz axis for yaw on one piece of hardware, but a slider listed as Rz somehow got assigned to yaw on the other, it is likely that the Rz axis is set to 0 on an axis that only sends 0-100% (fully left, as centered would be 50% on such an axis), and not the -100-0-100% that a yaw axis needs to function properly. With the slider at 0, it might get ignored for the entirety of the flight until a button is pressed on it, or the controller is moved, or even to the level that it only picks up on the axis out of place if the axis itself is tapped, moved slightly, or used. At that point, the correct axis is showing centered, but the incorrect axis is now sending full left. Now the sim needs to somehow resolve conflicting signals...meanwhile, the plane itself has a yaw damper that is trying to resolve the sim flipping between listening to one controller, and then listening to the other.

Never try to outlogic a computer. It's either going to outlogic you, or you'll apply too much logic to an otherwise dumb process (axis assignment being the latter). Think about it: the sim itself cannot have a list of every possible hardware interface out there, so the people who set it up gave it a basic list to "assist" people in setting things up. This is guaranteed to cause issues unless you actually go in and verify everything yourself. As I always note, even my single button PTT switch that has no business having any function in the sim itself still gets a brakes assignment because the PTT switch is listed as Button 1.

This really isn't all that complex. If people disagree with my logic, I think I've been clear enough in my posts that someone could have look at the concepts I'm explaining on their own to see exactly what I'm saying. Anyone with a HOTAS who loads up in a fresh install of a simulator will have seen this - I can nearly guarantee it. It'll even happen in DCS, too, if you've used that without going through and ripping out the default assignments. The latter is quite a harrowing experience, as I only really fly DCS in VR, which really adds a new element to hurtling at the ground because pitch got assigned to some random slider that's got your elevator jammed down that didn't even get picked up until after you got off the ground.

Edited August 13, 20187 yr by scandinavian13