Auto flight Navigation and Auto Flight.

[berts 97]

On 5/22/2018 at 9:58 PM, Oskar Speer said:

Yes, you are right. I going to create a flight from LFMN to morrow when I get time for the simulator and activate a direct intercept to the next waypoint. I believe that the PMDG B744 is not a very advance simulator but still, I do not know how to "activate a direct" in this case. 

As I said earlier, the first three entries on your LEGS page (   420); (NIZ-8) and (NIZ-11) are all conditional waypoints.  They are NOT in a fixed geographical position and only indicate when an event will occur, so they are not fixed in space and neither is the LNAV track to them. The QOTS v3 is actually one of the most advanced simulator addons you can buy, especially if you have the 64bit verson of P3D (which is why most of us can't wait for the latest PMDG 744-8 to be released).  There are a few well documented issues to do with LNAV tracking which might date back to the original version, but they are relatively minor and certainly do not spoil the sense of realism when you engage the autopilot and 'fly' using LNAV and VNAV.

The main problem as I see it is in the way these videos show the aircraft being operated. For example in one takeoff LNAV and VNAV are armed and then shortly after getting airborne HDG is selected (why?); shortly followed by speed intervene (why?); and then deleting the manually tuned NAV radios (why?).  Try leaving the aircraft in LNAV and VNAV the next time and sit on your hands more.  Let the aircraft autotune the NAV radios for you on the departure too (because that is what it is meant to do) and see what happens.  Oh - and start the elapsed timer and not the stopwatch at the start of the takeoff roll (the stopwatch is used if you have an engine failure and need to time the discharge interval for the fire bottles).  The chances are you will then be able to sit back, relax a bit and even look out of the virtual cockpit for other AI traffic! 

Bertie

          

Edited May 23, 2018 by berts
addition

[Copper. 78]

6 hours ago, berts said:

The main problem as I see it is in the way these videos show the aircraft being operated. For example in one takeoff LNAV and VNAV are armed and then shortly after getting airborne HDG is selected (why?); shortly followed by speed intervene (why?); and then deleting the manually tuned NAV radios (why?).  Try leaving the aircraft in LNAV and VNAV the next time and sit on your hands more.  Let the aircraft autotune the NAV radios for you on the departure too (because that is what it is meant to do) and see what happens.  Oh - and start the elapsed timer and not the stopwatch at the start of the takeoff roll (the stopwatch is used if you have an engine failure and need to time the discharge interval for the fire bottles).  The chances are you will then be able to sit back, relax a bit and even look out of the virtual cockpit for other AI traffic!

          

The intent of videos is to test lnav badod6e depiction as mentioned; not to fly procedurally.

Even with modification to the SID, the aircraft didn’t want to track 138deg hdg to intercept NIZ 158 radial at 8nm (knew it wouldn’t), hence hdg select as a matter of experimentation, remembering the intent here is not not to fly procedurally correct, but to experiment. I would’ve thought the flight number set during the quick preflt would’ve given that away

Nav tuning - with this NDB procedure (not RNAV) depending on nav by what I’d describe as raw data, why wouldn’t you manually tune in NIZ and NC? Additionally, why wouldn’t you delete them when they’re no longer needed?

My reasoning to manually tune is to ENSURE NIZ wasn’t lost due to auto tune. With a manual entry, it’ll remain as long as I leave it tuned. I delete the entry when I no longer depend on it. Same goes for NC, except unlikely to need to nav by ndb again so I left the entry.

as for the vertical mode selections in both videos... experimenting! I don’t leave VNAV during normal procedures often. It’s good to let loose and do weird stuff during these sessions to revise behaviour. How boring a life if all you ever do is LNAV/VNAV autopilot x to command...... takes the fun right out of things keeping in mind we are in a desktop sim and I DONT FLY WITH TRAFFIC as previously stated.

back to the problem. I refute that the 2nd and 3rd cond wpt’s are not fixed.... they are! Or at least they should be. They are defined by the chart as being on the 158 radial from niz at 8nm and 11nm. Those points are fixed, laterally. After passing 420’ we’re meant to turn trk 138 to intercept that radial at that defined point.

so then why is fmc drawing the lnav path lower to those 2 DEFINED and FIXED points in space? You can see the difference in the first video.

I surmise it’s because either nav data or the fmc, or both... and am happy to keep believing it, just as happy to admit I’m wrong if that’s not the case.

 

Snippet of sid proc https://ibb.co/hFV8MT

nd lnav track compared to fix page radial line and range rings https://ibb.co/i5EUT8

as you can see, lnav niz8 and niz11 SHOULD be laterally placed where the 8nm and 11nm rings intersect with the niz 158 radial line..... they do not.

Edited May 24, 2018 by Copper.

[Johannvr 3]

Brian, as Bertie suggested in his first reply, if you select a non-RNAV procedure with conditional waypoints very close to the departing rwy, like this BADOD6E departure at LFMN, the chances are you might miss some and LNAV mode wants to turn back to the active waypoint that was "missed", probably the first or second conditional one. Your first video clearly shows it. On the ND you can see flying Hdg mode to "correct" the procedure by intercepting 8dme R158 NIZ. (The official procedure states when at 400ft, RT on hdg 138 till 8dme NIZ, then intercept NIZ R158 till 11dme then RT dct NCNB) and in the process missed the second conditional wpt. The FMC Legs page shows this wpt as still active and it's confirmed on the ND by being magenta. Selecting LNAV will bring you back to this active wpt, You have to clear this by selecting the next wpt and making active in the FMC, which will then show up on the ND as magenta and not white.

I too have noticed the ND does not display the conditional wpts correctly at first, but I agree with Bertie here. It's a minor issue, just clear the missed wpts and get the next wpt active and LNAV will function correctly. I have flown this departure a couple time and noticed sometimes the first conditional wpt "shifts" as you progess and sometimes not and then gets missed by going wide past it. I just correct this by selecting the next wpt and activate it in the FMC. No problem, I am happy and even ATC (if online).

[Copper. 78]

That’s the point. Lnav path is drawn incorrectly. The result is, if the procedure is flown as charted, that wpt niz8 & niz11 are missed and are not sequenced. That is not normal. The “why” is what needs evaluating. 

Yes, the current fmc sid wpt’s are too near, and at an acute angle to rwy04 l/r der. Sound airmanship dictates the solution you and berts proffer but does nothing to evaluate the cause of the situation Oskar reported, which is what I’m trying to do as critically and objectively as I can. My aircraft handling aside I believe there is an issue with nav data or fmc, or both, and in previous posts I’ve provided the reasons why.

Others suggest this is due to to the conditional nature of the wpt’s. Ok, let’s explore that. My questions is - why is niz8 and niz11 conditional, when they are charted as defined points in space? If you can answer that critically, without reference to my aircraft handling which shouldn’t impact on niz8 and niz11 placement (due to them being defined points in space) I’d be appreciative.

note that I’m trying to do the evaluating because *IF* there is something amiss, then we would all benefit if it gets fixed *IF* there is something wrong which I believe there is. The root cause of this problem could impact on other procedures.

 

 

[Copper. 78]

Having tried and confirmed in another product, and having reviewed a video of another independent product I believe it is the nav data as lateral path depiction appears the same confirmed on 2 of the 3 separate (different companies) products. The 3rd I can't confidently say but looks like it.

NOTHING TO DO WITH FMC

 Here endeth the lesson :) Off to navigraph I go.

Edited May 24, 2018 by Copper.

[Johannvr 3]

Brian, I will try and answer your question. It seems the AIRAC data is the issue. If you look at the Navigraph data you quoted, the section of concern in bold:

SID BADO6E RNW 04R RNW 04L TRK 043 UNTIL 420 TRK 138 UNTIL 8.0 FROM FIX NIZ TRK 158 UNTIL 11.0 FROM FIX NIZ TURN RIGHT DIRECT FIX NCNB AT OR ABOVE 6000 FIX AMIRO FIX BADOD 

NIZ8, as you call it, is not chartered as a defined point in space or a fix, but "arbitrary", hence conditional on a hdg of 138 from the previous condition till 8dme NIZ, not fixed. NIZ11 is fixed (named MN044 on the RNAV SID BADOD6A) as it is 11dme on NIZ radial 158. According to the highlighted section above an incorrect wpt will be calculated on 158 deg from the previous "arbitrary" wpt instead of on NIZ radial 158. This is clearly what we see on the ND with the third wpt incorrectly positioned.

I tested this today with A/P and LNAV selected just passing 400ft. The aircraft could not reach the second wpt which was "held fixed" by the incorrect third fixed wpt. On my second attempt I selected BADOD6A with the third wpt now a fix MN044, the 2nd wpt "moves" on the niz8 arc and could be reached with the aircraft following track through MN044 without any issue.

I hope this explains it good enough.

 

(you just beat me to it, yes Navigraph will have to fix it)

 

Edited May 24, 2018 by Johannvr

[Copper. 78]

Johann. That is a perfect explanation. Thank you for looking at this problem critically and objectively with me! You’ve expressed the anomaly with the nav data better than I ever could, particularly your comparison between the conditional (NIZ-11) and fixed (MN044) waypoints and their interaction with the preceding waypoints and the impact on visual representation of the lateral path.

The problem is with Navigraph. The question is whether BADOD6E and other procedures like it are impacted in the same way, and if the nav data can be written better to reconcile with charted requirements.

Further discussion is not required on PMDG forum as it does not relate to them in any way shape or form and I do not wish to further implicate them more than I might already have.

🙂 thank you all for playing

 

[downscc 1,327]

23 hours ago, Johannvr said:

TRK 158 UNTIL 11.0 FROM FIX NIZ

 

10 hours ago, Copper. said:

The problem is with Navigraph.

I've done a lot of terminal procedure encoding, and while not the expert that Terry Yingling was I do have a basic knowlege.  The procedure encoding is using the correct syntax and one has to stare at the chart awhile to see the problem.  When posting to the Navigraph bug report forum, recommend to them that they use the NIZr158 as published for then the procedure becomes:

SID BADO6E RNW 04R RNW 04L TRK 043 UNTIL 420 TRK 138 INTERCEPT RADIAL 158 FROM FIX NIZ TRACK 158 UNTIL 11.0 FROM FIX NIZ TURN RIGHT DIRECT FIX NCNB AT OR ABOVE 6000 FIX AMIRO FIX BADOD 

The tricky part about this procedure is that it is poorly laid out and relies on the minimum climb gradient to reach 420 and the turn to 138 such that the NIZr158 is reached before NIZd11, and ideally about NIZd8.  However, flying in the cockpit in IMC conditions there is no practical way to ensure you intercept radial 158 exactly at NIZd8.  A better way for the authority to design this would be to say something like:  TRK 043 UNTIL 420 TURN RIGHT DIRECT FIX NIZ158/008 TRACK 158 UNTIL 11 FROM FIX NIZ.....; but that's not how it's drawn.

[Johannvr 3]

That is a good suggestion to Navigraph, Dan. It seems the idea of rwy04L/R procedures are to turn to hdg138 when at 420ft and then till NIZ 8d to avoid the buildup hill to the east. Only then continue to intercept NIZ R158 or the fixes applicable to the different SID's

LFMN is a NICE airport to do those non-RNAV rwy04L/R departing procedures and the approaches to rwy22L/R

 

[berts 97]

Maybe we can all agree then that the PMDG B744 FMC isn't the problem here after all, because for all practical purposes it models Conditional Waypoints very well, doesn't it? 

As I said earlier, as far as the FMC is concerned these Conditional Waypoints are NOT in a fixed geographical position and only indicate when an event will occur.  Therefore the conditions for crossing each one will inevitably be met in a different geographical position on every departure and will depend on all sorts of variables, including the effects of wind, aircraft weight etc.pilot handling, speed and bank angle in the turn etc., etc. 

In the case of the "TRK 158 UNTIL 11.0 FROM FIX NIZ" where you end up geographically will be somewhere on the 11.0 DME Arc from NIZ and that will depend on the position where you started tracking 158 degrees.

The Aeronautical charts can normally only be drawn to depict one fixed position for each Conditional Waypoint, so they have to be interpreted by the pilot in a practical sense and flown as accurately as possible.  If the aircraft isn't capable of complying with a particualr SID or STAR for performance reasons then there will normally be a note to that effect in the aircraft's performance manual.

Bertie G