RNAV approach TRK not matching DTK

[Karelpatch 29]

Hi, 

I’m back and kind of losing my mind over 2 separate issues:

- I tried to redo the RNAV approach and now for no apparent reason my autopilot (KFC 200) disconnects every time it reaches the FAF from IAF. And it happens on every airport I tried. I tried staying in NAV mode or APR mode, same issue, it doesn’t catch GS anymore and it disconnects and I am unable to select NAV nor APR afterwards. 

(Just crossed my mind: I guess I should select VLOC manually because I am too close to the FAF. That could be the issue. Will try.)

 

- Second issue, certainly related to the initial topic:

I created a direct route to a VOR. And of course, the course isn’t matching the line! Once more I have something like a 2 degrees offset. Completely different region (Europe). I checked the GTN magvar and it was correct compared to recent real life charts. Could it be that my P3D magnetic variations are not correct? 

Pictures illustrating what I am talking about:

https://imgur.com/gallery/MTZ8Tgp

[RXP 2,338]

Hi,

I'm not sure for the first part and this could be due to the aircraft you're using. Is this happening for example with the default Mooney or Baron 58?

As for the 2nd part, a 1deg difference is normal in my opinion. Please note the value is rounded to the nearest angle for display. So if DTK is 328.01 and TRK is 327.49, the difference is close to half a degree only yet shows 1 deg on the two values.

Now, this makes me wonder if what is happening is related to gyro drift instead. This could be a bug either in our code, using a heading source variable which is subject to drift in FltSim (which I doubt, we're careful about these things but I'll cross check), or, a bug in a recent P3D update which now 'drift' a simvar it is not supposed to. You might try 'resetting' the P3D gyro drift (there is a key shortcut for this) and see if this changes anything?

[Karelpatch 29]

10 hours ago, RXP said:

Now, this makes me wonder if what is happening is related to gyro drift instead. This could be a bug either in our code, using a heading source variable which is subject to drift in FltSim (which I doubt, we're careful about these things but I'll cross check), or, a bug in a recent P3D update which now 'drift' a simvar it is not supposed to. You might try 'resetting' the P3D gyro drift (there is a key shortcut for this) and see if this changes anything?

Will try! 

[Karelpatch 29]

Hi! I’ve posted on A2A forums about my issue with the autopilot disconnecting, and I got an interesting reply from the devs (see last answer): 

https://a2asimulations.com/forum/viewtopic.php?f=139&t=67295&p=511964#p511964

 

Apparently something must have changed between RealityXP gauges, P3D or Garmin’s trainer. 

[RXP 2,338]

Hi,

We're about to release our update either today or tomorrow.

However we've been trying to reproduce your flight and didn't experience the 'TRK drifting' you've reported in a standard P3D4.4 installation. Nor we've experienced any issue with the A/P coupling.

As far as A/P is concerned though there should be nothing in our code disconnecting any armed mode. As a matter of fact, the autopilot related code does not change any autopilot state, it just 'reads' it. Our next update introduces a change in that with AFMS.gps_selected enable, the GTN will automatically switch the A/P from NAV mode to APPR mode when required.

I'll cross check a couple additional things prior releasing the update just in case, but I think the only change which could affect the A2A aircraft might be related to something else: how do they code their A/P and what simvars are they using? v2.5.11 implements more stringent simvar overrides in relation to more precise GTN states, and it could be relying on a particular simvar (say NAV1 something) for the Autopilot code makes it disconnect based on their own A/P code logic?!

NB: these tests were conducted with the upcoming update version during its development, not with the actual release v2.5.11.

 

 

[fppilot 4,365]

9 hours ago, RXP said:

Our next update introduces a change in that with AFMS.gps_selected enable, the GTN will automatically switch the A/P from NAV mode to APPR mode when required.

Hasn't it already been doing that since the previous update?  2.5.11? That is what I have experienced!

Edited February 6, 2019 by fppilot

[outermarker 203]

On 2/1/2019 at 10:49 AM, Karelpatch said:

Hi! I’ve posted on A2A forums about my issue with the autopilot disconnecting, and I got an interesting reply from the devs (see last answer): 

https://a2asimulations.com/forum/viewtopic.php?f=139&t=67295&p=511964#p511964

 

Apparently something must have changed between RealityXP gauges, P3D or Garmin’s trainer. 

I've just switched from F1 to RXP after using their GTN's for the past 4 years. I have the same issue, but it happens with any departure/approach procedure I select. I have my flight plan set, FD set to NAV but as soon as I load a procedure the NAV mode disengages, APPR and CPLD start flashing and eventually turn off...I entered the SID waypoints manually and the flight went well until the same thing happened again when reaching the FAF. In the FSW Falcon 50 NAV won't disengage but the FD roll bar starts shaking out of control so this is not exclusively related to A2A.

[RXP 2,338]

@fppilot Sorry for the confusion: yes it is already in v2.5.11. I should have written "our last update".

@OuterMarker thank you for the additional details. Is this when you load, or load and activate?

Can you tell me how have you set the following options:

Panel Instruments:

- Connect GPS to Autopilot
- Connect GPS to HSI
- Connect HSI/OBS (input)
- Connect GPS to VOR
- Connect CDI Mode to NAV/GPS Switch

Gauge Options | Advanced Settings:

- Auto-Update Simulator GPS Waypoints

AFMS

- GPS Selected

 

ideas:

I'd suspect you'd need maybe 'GPS to VOR' on these aircraft if they use any 'NAV' simvar.

GPS Selected is the settings which sends the command to arm APPR mode when the GTN instructs to do so. Enabled is the default and whenever using a KAP autopilot, it should be set to disabled in the real world. In this case, the GPS prompts the pilot to manually arm APPR mode instead.

It is possible we are arming the APPR mode too quickly for these aircraft which are lagging behind. In this case, prior they get the chance to 'monitor' conditions have changed, we send the command to Flight Simulator and they end up in an unknown state for them. It might be interesting we add a delay prior arming the APPR mode when the condition arises. a 1 sec delay won't change anything to the flying experience but this will let the chance to these A/P to process at least once the new conditions. I'll make sure to review the code triggering this.

Furthermore, without changing any of these settings (make a backup of the RealityXP.GTN.ini file in the aircraft folder just in case) and should you have the Garmin Trainer 6.21 still installed, can you eventually try using the environment variables to tell the RXP GTN to use the older trainer and compare how it goes?
 

[outermarker 203]

@RXP I did some brief testing and "Connect GPS to Autopilot" seems to be the issue here, at least in the A2A Bonanza. These are my settings for now (keep in mind that I'm new to the RXP GTN so let me know if I should change something else).

[RXP 2,338]

@OuterMarker the problem is that you've disabled 'Connect GPS to Autopilot' therefore the GTN won't control the autopilot... However, you've enabled "Auto-Update Simulator GPS Waypoint" which makes the GTN override the default GPS flight plan.

Therefore, instead of having the GTN controlling the flight director / autopilot directly, you have the A2A aircraft autopilot following the default GPS system flight plan made of GTN emitted point to point line segments...

Try this instead:

1. Enable Panel Instruments | Connect GPS to Autopilot
2. Disable Gauge Options | Advanced Settings | Auto-Update Simulator GPS Waypoint
3. Compare with or without Panel Instruments | Connect GPS to VOR
4. Compare with or without AFMS | Main system | GPS Selected

[outermarker 203]

2 hours ago, RXP said:

@OuterMarker the problem is that you've disabled 'Connect GPS to Autopilot' therefore the GTN won't control the autopilot... However, you've enabled "Auto-Update Simulator GPS Waypoint" which makes the GTN override the default GPS flight plan.

Therefore, instead of having the GTN controlling the flight director / autopilot directly, you have the A2A aircraft autopilot following the default GPS system flight plan made of GTN emitted point to point line segments...

Try this instead:

1. Enable Panel Instruments | Connect GPS to Autopilot
2. Disable Gauge Options | Advanced Settings | Auto-Update Simulator GPS Waypoint
3. Compare with or without Panel Instruments | Connect GPS to VOR
4. Compare with or without AFMS | Main system | GPS Selected

That explaines why it followed the departure a little quirky😅sorry.

AFMS | Main system | GPS Selected fixed the APPR CPLD flashing and subsequently disabling the FD NAV mode, the FD/AP had no problems flying the departure and RNAV in NAV (GPS) now. I didn't notice any differences with Connect GPS to VOR on/off so I'll leave it off for now.

Thanks for your help & sorry for hijacking this thread!

Edited February 6, 2019 by OuterMarker