Sign in to follow this  
Woozie

erratic LNAV turn behavior during approach

Recommended Posts

Hope this havent been brought up already, ran a quick forum search but didnt find anything

This just happened to me yesterday, flying into TNCM using the RWY10 RNAV (GNSS) approach via MNOLO transition. The approach requires a ~90 degree left hand turn at OTMUT entering the base leg followed by another 90 degree left hand turn at AVAKI to line up with the final approach. When going into the first turn at OTMUT the aircaft took it way too tight, then realized its mistake and performed a daredevil style right hand turn to get back onto the magenta line before tackling the second left hand turn into final approach fix. So it basically went full bank left, full bank right, full bank left within 10-15 seconds...

Ignoring the screams coming from the cabin, I ran this approach 5 times to see if i can reproduce the behavior, which happened 2 out of 5 times.

Some more details:

* Weather was relatively calm (AS16 historical weather)

* Aircraft was in LNAV/VNAV mode, speed and alt settings have been verified on LEGS page prior to descent.

* LOC wasnt armed as its not an ILS approach

* appropriate energy management was excerised during descent to OTMUT, no use of speedbrake during the entire descent

* approached OTMUT with ~200 kts / Flaps 1, extended early enough before entering the turn  

 

I've seen a similar behavior when approaching Malaga (LEMG) two weeks ago, AP took turn to tight, performed an overly dramatic correction before diving right back onto the magenta path. 

 

I'm wondering if this is a known issue, a pilot error (me??) or even a quirk of the real airplane (which i hope not)?

Share this post


Link to post
Share on other sites
Help AVSIM continue to serve you!
Please donate today!

45 minutes ago, woozie said:

I'm wondering if this is a known issue, a pilot error (me??) or even a quirk of the real airplane (which i hope not)?

Using EZDok by chance?

Share this post


Link to post
Share on other sites
1 minute ago, woozie said:

nope, but chaseplane

Hmm...that one doesn't mess with the weather for its effects (like EZDok does), so that shouldn't be an issue.

What's the turbulence slider like in AS2016 in your settings?

Share this post


Link to post
Share on other sites

will check when i'm at home, only thing i know is that i turned it down quite a bit (around 30 if i recall correctly) as default turbulence in AS can be quite vicious.

i had about 12 knots tailwind on the downwind leg before i entered base leg turn, no thunderstorm cells in the vicinity and weather radar didn't show any turbulence areas nearby. 

 

Share this post


Link to post
Share on other sites

Just checked my AS16 turbulence settings, i'm on 20 max wind turbulence (default 100) and 10 turbulence effect scale (default 70), so i'm literally in girlie mode in terms of weather ;)

I'll try to reproduce the issue in the next few days using the same weather and aircraft conditions and capture a video if it happens again

Share this post


Link to post
Share on other sites

That's a weird one. What I've noticed with the 744 is that it tends to take turns (both in HDG SEL and LNAV) quite aggressively. The initial roll rate when intercepting a new LNAV course, VOR radial or chaning the heading manually is quite high. The weather doesn't play a factor here, since it's the same with no weather on.  The 737 and 777 were a bit smoother, but even the 777 tended to take things quite harsh at times. The smoothest and most spot on LNAV or FMGS behaviour in general I've seen so far is on the MJC Q400. But that one  has a completely external flight model, which is the main reason I assume.

Share this post


Link to post
Share on other sites

Bogdan, yes thats exactly what i observed as well, it enters the turn too agressively. It wouldnt even be a big deal if the aircraft would try to correct the deviation by slowly drifting back towards the magenta line, but it sometimes seems to go into full panic mode and tries to rejoin the flight path using some pretty radical correction maneuvers

  • Upvote 1

Share this post


Link to post
Share on other sites

This behavior is something we worked with over and over during beta testing.  What we have now is the best that could be accomplished within a reasonable time, and it does perform well in the majority of heading changes we encounter such as 25 deg turns or less.  Yeah, those 90 degree turns murder the LNAV performance and there are lots of examples such as going in to RJAA. One comment I heard is the developer mentioned he could nail it if the actual physics of the turns could be modeled instead of relying on the method used by the platform. This is possible, and desirable, and is done in many other areas such as the drag model for the queen; however, this would require additional development time and I guess it was a decision of waiting some number of months or pushing ahead with the compromise.  This is my view of things and not the opinion of PMDG.

I've learned to accommodate it and do exactly what a pilot should do... over ride the automation and interact with the airplane.  One way is to use HDG mode approach the turn and dial in the new heading then revert back to LNAV after resetting the course to next fix (line select the next fix and EXEC).  Another way is to allow LNAV to start the turn but line select the next fix and at the best time to smooth things out you EXEC the updated course.  There are probably several other ways but the important thing is you don't sit there and watch the automation make you look bad as the pilot.

This behavior is not unlike the 777...., the NGX is not as bad but the physics are the underlying issue.  Maybe PMDG will take turn physics outside the box like they have so many other things.

  • Upvote 1

Share this post


Link to post
Share on other sites

Thanks for shedding some light on this Dan! Glad to hear its a known issue. 

Did quite a few landings already and it only happened 2-3 times, so its not a showstopper for me. I'll be ready and try to accommodate next time I fly an approach with tight turns. 

Actually, i quite like such quirks....pulls you out of your comfort zone and gives you something unexpected to react to... ;)

  • Upvote 1

Share this post


Link to post
Share on other sites

Thanks for the explanation, Dan! I figured It might be something sim-related. Again, a completely external physics calculation / flight model would eliminate this, like is the case in the MJC Q400, as mentioned above.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this