737 VNAV descent debacle

[PENKO]

Originally Posted by A320 Glider

I do think Airbus has a big advantage here over the 737.

Airbus has a speed range for the managed descent.
+/-20kts if I remember from the planned descent speed. It's shown very nicely on the PFD.
.

Curious, why do you see that as an advantage? In light of this debate, the speed would still ‘wander’. Plus as easymxp explains, the new software allows the speed to go up to VMO-5 in order to maintain the profile.

As far as this debate goes, I’ve been told off by ATC and put on a diverging heading to create more separation with the preceding traffic for letting the speed drift +8 knots in managed descent (VNAV in Boeing speak). So you have to use common sense.

[AerocatS2A]

Originally Posted by easymxp

Not anymore. Or better initially yes but as soon as it enters a “capture zone” the profile is latched. This means the descent has priority over the speed and it can accelerate up to vmo-5 in order to keep the profile. (This for aircraft with descent profile optimization)

The latch isn’t part of DPO. Our latch aircraft don’t have DPO and our DPO aircraft don’t latch. It must be common for them to come together but they don’t have to be.

[Uplinker]

Curious, why do you see [the Airbus speed range] as an advantage? In light of this debate, the speed would still ‘wander’.

Why is "wandering" speed a problem in the descent ? (as long as it is "wandering" in order to meet the programmed profile).

It doesn't "wander" as such, it stays slightly above or below the bug until it recaptures the profile.

If ATC or the airspace specifies a speed, then of course you follow it, but if they haven't, then there's no real need - unless you are sequenced close to the one in front, (unlikely until on initial approach).

Varying speed slightly to maintain a profile makes perfect sense, and helps avoid using thrust or drag which is inefficient. (Cannot remember how helpful the Boeing auto-thrust is in doing this).

I don't know about Boeings but if you have programmed a suitable descent constraint, e.g. 8,000' at VWXYZ at 250 kts; the FMGS will meet that for you, possibly calling for more drag if needed.

[Vessbot]

Originally Posted by Uplinker

Why is "wandering" speed a problem in the descent ?

He answered that 2 sentences later: "I’ve been told off by ATC and put on a diverging heading to create more separation with the preceding traffic for letting the speed drift +8 knots in managed descent"

[PENKO]

I don’t think we disagree Uplinker, I only get a bit nervous when the speed does an olympic style ‘wander’ towards VMO-5 to catch the profile or if a very specific controller is on the freq

[FullWings]

I think Airbus and Boeing FMCs work in very similar ways in this particular circumstance, mainly because they have to. They will attempt to stay on a pre-computed path until environmental conditions make it impossible within certain parameters and either call for drag or add power. The exact implementations are probably different but the logic appears the same.

I find the most vulnerable phase is a high CI descent from somewhere close to the tropopause when you’ve been cruising at optimum altitude. A constant angle often does not work well when the speed margins are at their slimmest and wind/temperature can change abruptly with a small change in level.