Plane descends too 10,000 ft 230nm out

[john643]

The IXEG 737 will at times descend too 10,000 at 230 mm out from approach. At 20nm the plane will start descending again for final. I use the same flight plans in other planes with no problems. 

[mmerelles]

I have seen this behavior too. To me this is due to the VNAV fine tuning still required and the developers being working on.

 

If you want to make sure you are on this category and not a special case not considered, you should provide more detail for them being able to reproduce it.

-Entire flight plan (departure airport, runway, sid, route, star, approach, runway, airport)

-Airac cycle you use

-Performance data entered on the FMC, cruise alt, ci, etc.

 

EDIT: forgot to mention, if you are running 1.0.7 the FMC leaves a file on the aircraft root folder that contains all necessary information for the developers. Attach it!

 

 

Edited September 14, 2016 by mmerelles

[Litjan]

Hi john,

mmerelles is correct - we still have some issues to work out with the VNAV part of the descent - it can still be easily confused by multiple and complex descent restrictions. We plan to improve that - as a workaround I suggest going to autopilot modes that give you more control (V/S or FLCHG) so you can adhere to your desired descent path.

Cheers, Jan

 

[lanmancz]

1 hour ago, Litjan said:

Hi john,

mmerelles is correct - we still have some issues to work out with the VNAV part of the descent - it can still be easily confused by multiple and complex descent restrictions. We plan to improve that - as a workaround I suggest going to autopilot modes that give you more control (V/S or FLCHG) so you can adhere to your desired descent path.

Cheers, Jan

 

Hi,

on a similar note I have noticed that the initial descent from cruise altitude with FLCH is quite steep. When I am at 35000 for example and set say 12000 as my initial descent altitude and go into FLCH over my estimated T/D (I use PFPX) I initially get quite a vomit comet - the plane drops at -4000+ fpm, sometimes the needle is even stuck way down at -6000+ fpm so it cannot even indicate any higher descent rate. It then starts leveling out and stabilizes around -2000 fpm when the descent speed gets overshot by a lot. Did you also observe this behavior ? Is this normal or something that will receive some fine-tuning as well, please ?

The workaround is to use V/S mode first, get into stable descent and switch into FLCH later.

Edited September 15, 2016 by lanmancz

[Litjan]

2 hours ago, lanmancz said:

Hi,

on a similar note I have noticed that the initial descent from cruise altitude with FLCH is quite steep. When I am at 35000 for example and set say 12000 as my initial descent altitude and go into FLCH over my estimated T/D (I use PFPX) I initially get quite a vomit comet - the plane drops at -4000+ fpm, sometimes the needle is even stuck way down at -6000+ fpm so it cannot even indicate any higher descent rate. It then starts leveling out and stabilizes around -2000 fpm when the descent speed gets overshot by a lot. Did you also observe this behavior ? Is this normal or something that will receive some fine-tuning as well, please ?

The workaround is to use V/S mode first, get into stable descent and switch into FLCH later.

The initial part of the descent is "constant MACH" - and since the speed of sound gets bigger with warmer temperatures, you will see that the indicated airspeed keeps going up as you descend. So in a way you are not only descending, you are also accelerating!

In addition, the "glide angle" for a constant airspeed is also constant - but you have to remember that at high altitudes the plane is moving a lot faster than indicated. (TAS > IAS). So you will also get a much higher ROD for the same angle.

These effects are portrayed well in X-Plane, and it is not unusual to see 5000+ fpm rate of descent in the real 737 when descending at constant MACH at high altitudes.

The passengers (and pilots) are unable to feel "high ROD" - you could descend at 20.000 feet per minute and no one would know (at least not for a minute or so...)

Jan

 

[lanmancz]

Thanks for explanation! It just seemed odd to me that the plane will lose 10000ft very rapidly during the initial descent from high altitude. Perhaps this is something that is not well modeled in other models that I'm used to flying.

Edited September 15, 2016 by lanmancz

[Litjan]

2 hours ago, lanmancz said:

Thanks for explanation! It just seemed odd to me that the plane will lose 10000ft very rapidly during the initial descent from high altitude. Perhaps this is something that is not well modeled in other models that I'm used to flying.

If you are in VNAV you will see that our plane intially gets a bit "over" the calculated path, then dives down to re-capture it. The real plane plans a smoother initial curve - this is something we skipped to avoid the complexity - VNAV is hard enough to calculate as it is. We may add that at a (much) later stage...

Jan

 

[lanmancz]

5 hours ago, Litjan said:

If you are in VNAV you will see that our plane intially gets a bit "over" the calculated path, then dives down to re-capture it. The real plane plans a smoother initial curve - this is something we skipped to avoid the complexity - VNAV is hard enough to calculate as it is. We may add that at a (much) later stage...

Jan

 

I am not in VNAV, that's what I mean (see screenshot). I've been cruising at FL360, set FL140 as my initial descent altitude and hit FLCHG (not sure if this is correct?perhaps I'm using the FLCHG wrong?). The plane then goes nose down at -6000+ fpm for a while, overshoots the selected speed (M.73 currently) by about 20 kts and then starts leveling out, slows down to the selected speed and stabilizes at about -2000 fpm.

[Litjan]

I will check on my end to see if I can recreate what you are seeing...will take a bit, though, I have to work the next 5 days...

Cheers, Jan