B737 too aggressive VNAV profile intercept

[Fabristunt]

Hello,

since about this summer the VNAV profile intercept is too aggressive. I'll explain the issue using a video

https://youtu.be/BlYeDiRS-ig

As you can see the aircraft goes from 1000fpm to 3600fpm in about 10s, which for me is a bit too much aggressive (there are tons of videos on youtube showing the PFD of the real aircraft and it's smoother. Also the FSX/P3D versions of the PMDG B737 were smoother). Keep also in mind that in the case recorded in the video, the 80-knots headwind helped a bit: if you have 80-knots tailwind it's worse. It's also worse if you are using a high cost index compared to a lower one.

The VNAV aggressive intercept issue not only happens at the TOD or around TOD: it happens also if you level off and then start descend again using VNAV or even during climb in the same situation, so it's a general thing, not specifically related to descend, high altitude, low altitude etc....

In this video, I'm using the BBJ1, which is the aircraft where the issue is much more evident. I also regularly use the B738 and on that type of aircraft the VNAV intercept is much more realistic, even if in some situations it could be better in my opinion.

I'd like to hear opinion from other pilots.

PS: before anyone ask, yes I set the desc forecast page everytime.

[jrw4]

It was my understanding (gained, I think, from a YT video a few years ago by flightdeck2sim) that the real-world procedure for engaging VNAV on descents was to trigger the DES NOW process around 10 nm before TOD. This places the a/c into a 1000 fpm descent which is maintained until the a/c intercepts the VNAV path a few minutes later, at which point a transition to the steeper descent is not so evident.

[Fabristunt]

Hello.

the video clearly shows that the aircraft is descending at 1000 fpm descent rate and the VNAV intercept is still too aggressive.

[jrw4]

Agreed Fabristunt, but the video also demonstrates a Mach number of .800, both in cruise and descent. Have you tried this at a more normal Mach number somewhere in the range .76-.78. I ask this because the only way the aircraft can maintain such a high airspeed in descent, it has to establish a rapid vertical speed. You may have found an interesting edge case which should be reported to PMDG on a support ticket, but it might not appear at slower speeds.

[Fabristunt]

I know M.800 the descent it's pretty fast. The issue is not the rate itself, but the transition between 1000fpm and 3500fpm or whatever is needed to keep the VNAV profile: it shouldn't be so fast to get to 3500fpm even if the CI is high.

[fmadrid]

Someone have the same problem? I have the same and it's very frustrating get -4000 fpm in just 4-5 seconds and, few seconds later, get -800 because desc path is above due to heavy pitch the first time. Sometimes the aircraft get riding between -1000 - 3000 in order to follow the desc path. So frustrating.

[VBHB]

Yes the VNAV doesn't work properly.

[Aeromar]

What I saw on the video looks close enough to what I see in the real airplane. Sometimes it catches it smoothly and sometimes it's more aggressive. You're descending at M.80 so the drag is high. At such a high drag as soon as the thrust levers retard the airplane will fall like a piano like the real one does. Especially before conversion level, when you're descending at MACH instead of CAS, which means you're accelerating as you descend.

For simulation purposes I don't think this is an issue. The airplane is doing what it's supposed to do.

[Fabristunt]

From -1000ft to 3600ft in less than 10s? I think in the real plane it's much smoother ...

[DD_Arthur]

From -1000ft to 3600ft in less than 10s? I think in the real plane it's much smoother ...

And yet the pilot of the real plane posts above you and says it’s its doing what the real plane does…

Who to believe? Hmmm, toughie…..

[Fabristunt]

Funny. I'm not saying it's completely wrong, given that I've done some calculations and the manouvre it's done at a bit more than 1,2G, so nothing dramatic. However I think in the real airplane it's much smoother, there's no need to be a RL pilot, you can check tons of videos on youtube.

[bobsk8]

And yet the pilot of the real plane posts above you and says it’s its doing what the real plane does…

Who to believe? Hmmm, toughie…..

Well who would you trust, a real airline pilot that makes a living flying for an airline, or someone that flies a home simulator, and most likely has never flown anything in real life? 😉

[Fabristunt]

Said the guy who continued to deny the recent engine spool up bug despite PMDG itself admitted there was something not right.

[DD_Arthur]

Thank heavens for YouTube!👍

[Aeromar]

From -1000ft to 3600ft in less than 10s? I think in the real plane it's much smoother ...

It may be but the difference is not big enough for me to notice. You may be absolutely right and maybe the real airplane does it more smoothly.

My point is that even if you are right this isn't a big deviation from the real airplane for it to matter. What you want to happen is happening. The airplane is descending below the path at -1000fpm in VNAV PTH and when it catches the path you get RETARD and an idle descent that matches the VNAV calculated descent.

Boeing does not publish the actual code and the logics behind VNAV and LNAV so all we can do match the real sims and the real airplane as closely as reasonable. Who knows what the real airplane is doing... The real airplane is very organic and the atmosphere is very chaotic. We never see the real airplane behaving exactly the same. Sometimes it's smooth. Sometimes it's aggressive (what you can see in your video is not something that the passengers would feel unless they're paying very close attention).

Try again at a more reasonable descent like .76/270 or whatever ECON speed a common cost index results in (6, 12, 20...). But really I wouldn't bother with this. I'm one to find the smallest differences ad report them. I even sent PMDG a full continuous descent from TOD to GS capture back when they were reprogramming VNAV and things improved a lot after that. If this was a significant deviation from what I see in real life or if it caused negative training I would be the first one to report it.

[Fabristunt]

Thanks for the detailed explanation. I noticed that the "aggressive" VNAV profile intercept is much more evident on the B737 compared to the B738 (any version of them). In the 738, even at high CI like in my video, it's smoother.

[DDowns]

Hey, if you want to see aggressive watch a BBJ at TOD with descent SPD TGT .80/320. There is a serious deck angle! When flying commercial carrier operations I'm usually around .77/280 for the descent and things are quite smooth there.