News:

Precision Simulator update 10.173 (24 February 2024) is now available.
Navburo update 13 (23 November 2022) is now available.
NG FMC and More is released.

Main Menu

"On-Approach" Logic on STAR?

Started by emerydc8, Sat, 1 Aug 2015 10:56

emerydc8

Hi Hardy,

I didn't mean to imply that you believed the airplane would level off for the SPD RESTR. I just wanted to add that caveat:

QuoteSPD RESTR: Manually entered. Caution: The speed bug will automatically set to the target speed at the designated altitude, but the aircraft will not level off to obtain this speed as it does at the SPD TRANS constraint (normally 10,000'). It is therefore best to create a fix on the LEGS page and set in a speed/altitude restriction and use the SPD RESTR method as a backup. For example, your STAR requires 220 knots below 13,000'. You could create a fix 40 miles prior to destination or find a fix currently on the LEGS page that shows the crossing altitude approximately 13,000'. Insert the restriction (220/130) as a speed/altitude constraint next to the fix. It is a good idea to backup the restriction in the VNAV DES page, since radar vectoring by ATC, or, a CDU direct to operation, may eliminate the LEGS page restriction. Entries that conflict with the SPD TRANS will blank the SPD
TRANS parameters.


Britjet

This is all a bit puzzling to me. I note the quotes re the JAL incident, but I have never seen the aircraft get so close to VMO as -11 or whatever, although perhaps the many examples of a near over speed at TOD that I have seen are exactly that, because the aircraft is above the first speed constraint. I was assuming it was just poor pitch control.

At lower altitudes I have never seen it try to go faster that + 15 or so in PTH, but again maybe that is a result of the increased drag as Hardy suggests.

I am starting to lose the will to live on this one!

Peter

emerydc8

Quote from: Britjet on Sun,  4 Oct 2015 09:44I am starting to lose the will to live on this one!
No! We've come so far on this one and we are so close to figuring it out.

Like Hardy said, when he lifted the target+15 restriction on descent, he didn't see it go over target +20 to stay on path anyway. I think that's probably the most we are going to see unless someone can figure out how to program a 30 knot loss of tailwind resulting in a 20 knot increase in speed during a six second period at FL400.

I think Boeing should have been more explicit with their use of the words "unknown winds." The FCOM left me with the impression that a speed of VMO-11 was not very unusual when in fact even the JAL report states that it is very unusual.

The memo at the end of the JAL report gave me some good take-away items. Specifically, if your speed rapidly increases to BP on descent like this, leave the A/P on and switch to V/S or ALT mode.

I have to admit, though, that it is getting to be a long string.

Hardy Heinlin

The +150 ft headroom is so tiny; if you encounter a sudden headwind, the sudden lift will make you exceed +150 ft before you exceed +15 knots. That +150 ft exceedance will engage VNAV SPD. So there's nearly no chance of exceeding +15 kt with VNAV PTH still engaged.

And if the headwind increase is very slow and smooth, the chance is high that it's a forecast wind anyway, with the FMC taking it into account in its predicted path profile. So in that case there's nearly no chance of an exceedance either.

I just learned from another, more recent LH manual that this +15 kt limit is applied only below the first constraint; above this constraint it may accelerate up to VMO-11 indeed.


Cheers,

|-|ardy

Britjet

interesting....

Hardy Heinlin

#85
According to the LH OM, the +150 ft limit is not the only condition for the VNAV SPD engagement. There are two conditions:


1) impossible to maintain +150 ft
2) without further acceleration

(Vice versa for -150 ft with deceleration.)


I may be above the command speed and beyond 150 ft. But that doesn't necessarily mean that I cannot accelerate further to VMO-11 (or -16 with ECON) which might bring the aircraft back onto the path. Thus, only when my speed is at the outer limit and the path deviation isn't getting better, I can tell the system: "impossible to maintain +150 ft".

That makes sense, and works much better. Otherwise you will get speed reversion for every little fart beyond that tiny 150 ft corridor, which would be completely useless anyway because VNAV PTH can do the job just as well as VNAV SPD as long as the deviation trend is not hopeless.

For over a week now, I've been making countless re-designs of all the pitch and thrust control algorithms for smooth transits etc. pp., mode change conditions, endless test runs and re-tunings for every configuration and situation etc., and still no end in sight ... I need some more days before I can release beta7. But it's looking very good now.


Cheers,

|-|ardy



emerydc8

Thanks, Hardy. Somehow I knew you were burning the midnight oil for many days on this one.

LH must have researched and cared enough to actually publish this information. The LH manual sounds like a good one to have.

Have you run into situations where not even VMO-11 would maintain the path with the expected/forecast winds? Or is it that the pitch can't change fast enough to stay within the narrow +/- 150' band but it can always accelerate to VMO-11 and get back on path? Thanks again.

Jon D.


Hardy Heinlin

In most cases, with good forecast data, it remains within +/- 50 ft, and even returns after a disturbance by an unforecast wind change.

When the wind change is significant, it may exceed +/- 150 ft, but will return by adding less than +/- 20 knots. It then may remain, for example, at +80 ft until the first constraint, with a constant additive of +10 knots, for instance, flying a parallel path. A hair more and it may re-intersect the path shortly before the first constraint.

High speed descents are bit harder to predict precisely, because the drag rises with the square of the airspeed. A small speed fluctuation causes a large drag fluctuation. In such descents the deviation may exceed +/- 150 ft more often.

To test speed reversion I really have to add thrust or speedbrakes to get intentionally away from the clean idle configuration, or have to vary the wind speed by 50+ knots or so.


|-|ardy

emerydc8

Quote from: Hardy Heinlin on Thu,  8 Oct 2015 06:52
According to the LH OM, the +150 ft limit is not the only condition for the VNAV SPD engagement. There are two conditions:


1) impossible to maintain +150 ft
2) without further acceleration

(Vice versa for -150 ft with deceleration.)

How would one get his hands on an LH OM?

Hardy Heinlin

I can't help, I'm sorry. It's paper print anyway.