LVD 767 FS9 Roll mode problem....

[altstiff]

Hello, I seem to have a problem with the LVD 767 in FS9. I'm using the registered version 3.7.6.6.

It seems that I cannot get the auto pilot to engage a roll mode of any sort (although everything else works fine). It won't track LNAV or any HDG HOLD or HDG SEL even though it is annunciated on the PFD.

I have traced this problem down to using FSUIPC for axis and calibration assignments. If I don't use FSUIPC for these assignments (using the default FS9 joystick settings page for assignments) everything worlks fine.

I posted this problem over at the LVD forums as well.

Any suggestions?

[jp.pilot]

Hello,

I have the same problem as well.

The only way I managed to get it to work was in axis assignment under FSUIPC, set the aileron to the default FS command (not direct to FSUIPC). Then, go into controller assignments in the FS9 menu, and set the aileron to your joystick so FS will control it. Even though its not ideal, it may be the only way.

I got it to work ok but it would sometimes go into a complete roll, but I think that is down to my yokes potentiometers wearing out.

Regards,

JP

[altstiff]

Yup that fixed it, thank you!

[Pete Dowson]

The only way I managed to get it to work was in axis assignment under FSUIPC, set the aileron to the default FS command (not direct to FSUIPC). Then, go into controller assignments in the FS9 menu, and set the aileron to your joystick so FS will control it. Even though its not ideal, it may be the only way.

You should not assign anything in FSUIPC if you also assign it in FS itself. If you do this then every time you move the axis there twill be two copies of exactly the same things being sent through to the Sim engine. Very inefficient. Only assign in FS or in FSUIPC, never both.

The only problem with aircraft like the LevelD is that it assumes the axes are coming from FS controls. If you assign in FSUIPC to go "direct" it bypasses the code in the LevelD aircraft. You can of course make assignments in FSUIPC aircraft-specific, so that LevelD can be treated differently from others quite easily.

Regards

Pete

[altstiff]

Hey Pete. I did not have anything assigned via FS9, all of it was via FSUIPC.

I have kind of figured the LVD problem out now with others help.

Hope you had a great holiday and thanks for the response.

[LOT116]

Hi! I have struggled with this problem lately and literally nothing is helping. FSUIPC does not support the LvlD Ailerons at all. Even if it's set as aircraft specific or not. After many test it was the only solution to make it work for me to reser the aileron axis in FSUIPC settings and let FS9 control it. All other things are controlled by FSUIPC. Just the ailerons are controled by FS9 . This is the way it works.

[Pete Dowson]

Hi! I have struggled with this problem lately and literally nothing is helping. FSUIPC does not support the LvlD Ailerons at all. Even if it's set as aircraft specific or not. After many test it was the only solution to make it work for me to reser the aileron axis in FSUIPC settings and let FS9 control it. All other things are controlled by FSUIPC. Just the ailerons are controled by FS9 . This is the way it works.

You are either using an out-of-date copy of FSUIPC, or you had something set wrong, then. It all works well here and others report good results.

Pete

[jp.pilot]

Pete,

The problem is when under axis assignments, you set the aileron direct to fsuipc callibration. When FS handles the aileron's the problem doesn't occur. I have to set my CH yoke to FSUIPC calibration as the potentiometers within the yoke are wearing out, and under FS control it just rolls to the left (having difficulty trying to get replacement potentiomteres sent to the UK). During cruise I have to roll the aircraft by physically turning the yoke, as the LevelD 767 autopilot will not command any lateral movement; vertical movement is not a problem.

Many have reported the same problem over at the LevelD forums, but you have to be a registered user to gain access. One solution is to delete in fsuipc.ini file which obviously works, but deletes all the assignments. When you set up the assignments to direct to FSUIPC again, the problem reoccurs.

If others have got it fully working with all axis controlled via FSUIPC and not FS, then could they please state what they did as a lot of us would be extremely grateful.

Regards,

James Pearce

[Pete Dowson]

Have you read the earlier messages on this thread? I have, and it seems there are solutions whilst still calibrating in FSUIPC.

Please start at the beginning of the thread and read it all.

Regards

Pete

[Pete Dowson]

The problem is when under axis assignments, you set the aileron direct to fsuipc callibration. When FS handles the aileron's the problem doesn't occur. I have to set my CH yoke to FSUIPC calibration as the potentiometers within the yoke are wearing out, and under FS control it just rolls to the left (having difficulty trying to get replacement potentiomteres sent to the UK). During cruise I have to roll the aircraft by physically turning the yoke, as the LevelD 767 autopilot will not command any lateral movement; vertical movement is not a problem.

Okay. I've been mulling this over in my head the last couple of days, and it occurs to me what might be happening.

Many have reported the same problem over at the LevelD forums

Now so many other reports I really cannot think are due to the same problems you have. I think the earlier parts of this thread do really apply in the majority of cases. But let me discuss your particular case.

You see there is no difference whatsoever in how FSUIPC treats ailerons to any other axis. I'd like you to confirm, or otherwise, the CENTRE positions you have to calibrate in FSUIPC for your yoke (both aileron and elevator) to operate in manual flight to your satisfaction. For if your aileron centre is well off the otherwise expected 0 position then here is what I think is happening:

The LevelD autopilot, unlike real and most other autopilots, is probably assuming that specific values for the aileron control do specific things to the banking/roll of the aircraft. for example, rather than monitor the bank angle and adjust the ailerons left or right to maintain a required bank angle, it assumes, for example, that the value 0 will give zero bank -- i.e. level wings.

If this is indeed the case, and your calibrations in FSUIPC place the wings-level position well off the otherwise FS-defaulted 0 position, then I can see how the A/P would lose control.

In the case of the LevelD aircraft and FSX I had assumed that it was the extra latency, of intercepting axes via SimConnect, which upset LevelD's control, but it may have been a combination of that and the sometimes slightly off-centre calibrations folks had. The problems folks reported in FSX were far greater and more difficult to overcome than those reported earlier in this thread for FS9.

In FSX the solution was to make FSUIPC4 not intercept axes which were being assigned "direct" in FSUIPC4 itself.

Assuming my theories above to be correct, I am going to provide an option in FSUIPC3, which I shall probably have set by default, to avoid intercepting axes which are assigned directly. Please try that and let me know -- it'll be FSUIPC 3.773, probably later today (Tuesday). Check the "Other Downloads" announcement above.

Regards

Pete

[jp.pilot]

Hello Pete,

Many thanks for the quick response!

I've done some testing on FS9 and here are my results:

Ailerons (direct):

-7151 2816 13568

8052

Elevator (direct):

-13262 -3120 15872

4992

(Both axis deleted within the FS9 assignments under options)

With the autopilot set, vertical command is fine, but lateral is not. I have full control in turning the aircraft with my yoke.

When I set the aileron axis within FS9 and delete it via FSUIPC, the lateral command is ok but will open continue to 'tip' over, which is probably due to my yoke (this is the solution I posted earlier in this thread). The default centre for my yoke if 4608.

I tried to hold the yoke to have an input of '0', and try a heading select but the autopilot does not turn the aircraft.

If I move and hold the elevator outside of the '0' input whilst I've set a vertical speed in the autopilot, it will follow my command but try to correct itself.

(Im not to sure about this, but I seem to remember if you try to move the yoke whilst the autopilot is on, the autopilot will disengage and the warning lights/sound comes on. I might be wrong because I believe the PMDG 747 has this, not sure about the LevelD767).

I'm out for the day and will be back this evening. I will do my best to test it out tonight and let you know my results.

Regards,

JP

[Pete Dowson]

H

I've done some testing on FS9 and here are my results:

Ailerons (direct):
-7151   2816   13568
           8052

Elevator (direct):
-13262   -3120   15872
              4992

I think that tends to confirm my theory. The elevator exis centre is okay, operating in your calibration from -3120 to +4992, so including the critical 0 value. The ailerons on the other hand, with that calibration, will be sending a right-aileron signal (-ve value) when 0 is set by the A/P.

It looks like the A/P assumes 0 is wings level (which of course with normally calibrated well-balanced pots it would be).

This being the case I have high hopes for version 3.773, now being uploaded to the Other Downloads announcement.

I tried to hold the yoke to have an input of '0', and try a heading select but the autopilot does not turn the aircraft.

No, the level D is trying to use the same axis controls which you are overriding.

but I seem to remember if you try to move the yoke whilst the autopilot is on, the autopilot will disengage and the warning lights/sound comes on. I might be wrong because I believe the PMDG 747 has this, not sure about the LevelD767).

I think most aircraft do this, but probably not for minor changes -- you probably have to be quite severe in the movement of the axis. I think it is designed that way so you can take manual control quickly in emergencies. I use Project Magenta, not any FS A/P, and it certainly does that but I don't know if any add-on aircraft do it.

I'm out for the day and will be back this evening. I will do my best to test it out tonight and let you know my results.

Okay, thanks.

Regards

Pete

[jp.pilot]

Pete,

Guess what...

It WORKS!!!

In the words of Del Trotter, "Your a 42 carat diamond!".

I'll post a message in the LevelD forums about the new fix.

Your fantastic support is very much appreciated!

Regards,

JP

[Pete Dowson]

It WORKS!!!

Ah, good. Thanks for confirming!

Pete

[jordanal]

I just want to add, since Pete helped be get my Saitek Yoke and Throttles working the other day using direct axis for everything, I have been successfully using FSCUIPC direct axis with the LvL-D 767 on FSX (sp1) with no ill affects. The issue described with the inactive AP on the Airleron axis is only a FS9 issue. I too had run into this exact issue a year ago on FS9 and like others, fixed it by assigning LvL-D axis to Normal. I hope none of these recent changes in FSUIPC will adversly affect the 767 in FSX while using direct axis. Downloading 4.212 now to verify with my current FSUIPC4.ini to validate soonest.

Regards,

AL

[Pete Dowson]

I have been successfully using FSCUIPC direct axis with the LvL-D 767 on FSX (sp1) with no ill affects. The issue described with the inactive AP on the Airleron axis is only a FS9 issue.

The same problem actually did affect FSX + LevelD 767, too, only I put it down to the excess latency -- from intercepting the FS controls used by the Autopilot via SimConnect and relaying them on. I thought that it was this delay causing control to be erratic.

The solution in FSX was simply to not intercept any axes which were being assigned in FSUIPC4 for direct calibration.

It turns out that my understanding of what was really happening was not quite correct, though happily the solution was the right one. The change in FSUIPC 3.773 is effectively the same as that done in FSUIPC4 way back in version 4.16 (August 2007). This is the write-up in the History document about it:

Axis assignments in FSUIPC which are set to go direct to FSUIPC calibration (rather than via FS controls) do not now enable the interception of those controls from FS. This makes this version (again) work correctly with the Level D 767 provided that either FSUIPC is not used for aileron/elevator/throttle calibration, or it is but via its own axis assignments directed to its own calibration.

I hope none of these recent changes in FSUIPC will adversly affect the 767 in FSX while using direct axis.

Of course not. Are you misunderstanding something?

Regards

Pete

[jordanal]

The same problem actually did affect FSX + LevelD 767, too, only I put it down to the excess latency -- from intercepting the FS controls used by the Autopilot via SimConnect and relaying them on. I thought that it was this delay causing control to be erratic.

The solution in FSX was simply to not intercept any axes which were being assigned in FSUIPC4 for direct calibration.

It turns out that my understanding of what was really happening was not quite correct, though happily the solution was the right one. The change in FSUIPC 3.773 is effectively the same as that done in FSUIPC4 way back in version 4.16 (August 2007). This is the write-up in the History document about it:

Ahh, that explains why in works in FSUIPC4 and not in FSUIPC3....

Of course not. Are you misunderstanding something?

Yup, I got it backwards, not realizing the fix was already in FSUIPC4 and you were migrating the fix back to FSUIPC3....

AL

[specialist]

Pete, thanks for version 3.773. I was having similar problems with Carenado's 182 RG II, and this version solved the rolling to the left problem. Sweet!!! 8)