757/767 VNAV descent path angle during deceleration

[firstmrj]

Hi,

I looked into how the FMS of B767 construct the descent profile when simultaneous descent and deceleration was intended by creating some manually entered along-track waypoints at intervals of 1 NM.
In one case it shallowed the descent path to about 1 degree and slow the airspeed at the rate of more than 10 KT per 1 NM.
In other case the descent path angle remained at about 3 degrees and the deceleration rate was about the half of the former case.

In both cases, from my experience it seems we need speed-brake to achieve these trajectories in normal circumstances.

Any idea or thought about these difference?

Thanks in advance,

[oceancrosser]

Well are you going to go down OR slow down? Trying to do both at the same time eats up air distance, more so on the 757. So just use the speedbrake. It is there to be used.

[firstmrj]

oceancrosser,
Thanks for the reply,

I agree that we are totally free to use speedbrake.

I just wonder why the FMS’s algorithm draw such unachievable path without extra drag (speedbrake) path in spite of the below assumptions in the FCOM.

“The path to the first constraint is based on:
• idle thrust
• speedbrakes retracted
• FMC cruise wind
• wind entries on the DESCENT FORECAST page
• predicted use of anti-ice
• applicable target speed”

Besides, in my previous types, i.e. B737CL, B737NG and A320, I could follow VNAV PATH down to FAF in most situations when no short cut or no detour from STAR.

However in case of B767 it seems I have to elect not to use VNAV descent or to use VNAV with help of speedbrakes in the deceleration portion.
​​​​​​​

[oceancrosser]

Originally Posted by firstmrj

oceancrosser,
Thanks for the reply,

I agree that we are totally free to use speedbrake.

I just wonder why the FMS’s algorithm draw such unachievable path without extra drag (speedbrake) path in spite of the below assumptions in the FCOM.

“The path to the first constraint is based on:
• idle thrust
• speedbrakes retracted
• FMC cruise wind
• wind entries on the DESCENT FORECAST page
• predicted use of anti-ice
• applicable target speed”

Besides, in my previous types, i.e. B737CL, B737NG and A320, I could follow VNAV PATH down to FAF in most situations when no short cut or no detour from STAR.

However in case of B767 it seems I have to elect not to use VNAV descent or to use VNAV with help of speedbrakes in the deceleration portion.

That last sentence sums it up. The 757/767 FMC/VNAV is archaic. You need to manage these airplanes accordingly. VNAV works ok during descent until you get into manuevering/speed control in the STAR. So you may/will need different modes. Also keep in mind that the weight span of the 767 is far greater than the other jets you have flown. A very light 767 is a different beast from one at MTOW (or MLW at that).

[firstmrj]

Originally Posted by oceancrosser

That last sentence sums it up. The 757/767 FMC/VNAV is archaic. You need to manage these airplanes accordingly. VNAV works ok during descent until you get into manuevering/speed control in the STAR. So you may/will need different modes. Also keep in mind that the weight span of the 767 is far greater than the other jets you have flown. A very light 767 is a different beast from one at MTOW (or MLW at that).

Thanks for the good advice!
I will be extra careful not to end up too fast and/or too high on low altitude especially when handling a chubby bird.

[firstmrj]

Originally Posted by wondering

The same is true for the 787. It almost always ends up fast/high at lower altitudes when descending in VNAV. Could it be that using speedbrakes at lower altitudes is overall more efficient than descending earlier and not using speedbrakes at lower altitudes? That´s what one of my colleagues is suspecting.

I find it very interesting thought.
As far as I know Boeing/Honeywell doesn't disclose any models including coefficients variables.
I really want to ask about the calculation formula FMS is using to someone from Honeywell or Boeing.
Does someone in this community know anything about the logic VNAV is using?

[Sidestick_n_Rudder]

AFAIK it’s the opposite - descending on a profile, that is too high/too fast and with speed brake uses around twice as much as descending too early and flying a lower speed.

787 VNAV is utterly cr@p. I don’t have any other Boeing experience, but the VNAV (sorry, ‘managed DES’) on A320/330 does a much better job overall - even though It was erring on the conservative side, i.e. normally ended up low/slow.

One thing I’m suspecting is ‘Idle Factor’. At least on the ‘Bus you can change it and it affects the steepness of the calculated path. I actually wonder if it can also be modified on Boeings. Anyone?

[firstmrj]

Originally Posted by Sidestick_n_Rudder

AFAIK it’s the opposite - descending on a profile, that is too high/too fast and with speed brake uses around twice as much as descending too early and flying a lower speed.

787 VNAV is utterly cr@p. I don’t have any other Boeing experience, but the VNAV (sorry, ‘managed DES’) on A320/330 does a much better job overall - even though It was erring on the conservative side, i.e. normally ended up low/slow.

One thing I’m suspecting is ‘Idle Factor’. At least on the ‘Bus you can change it and it affects the steepness of the calculated path. I actually wonder if it can also be modified on Boeings. Anyone?

At least on pilots’ reach, B737 and B757/767 don't have boxes for “IDLE FACTOR”. I don’t think FMS configuration function offers something like that to our mechanics neither.
I’m not sure about much newer 787.

[Beakor]

If it’s a winglet aircraft, VNAV always gets the deceleration wrong. I think that when the FMS software was updated for winglets, the drag reduction was underestimated, hence the aircraft always retains more energy than the FMS predicts. More so for the 767 than the 757. Given the age of the aircraft, an updated software package is unlikely.

[amc890]

There is a cheat available though, just enter a value in the anti ice on line. Your mileage will vary so you will need to experiment a bit with the exact altitude to enter.

[rudestuff]

Originally Posted by amc890

There is a cheat available though, just enter a value in the anti ice on line. Your mileage will vary so you will need to experiment a bit with the exact altitude to enter.

You can let the speed creep up a bit in PATH on the way down, but then you'll definitely need speedbrake for the deceleration point just above 10,000. So TAI at 15,000 will flatten the descent and allow the thrust to go effectively below idle (as long as you haven't actually got the TAI on!!) You'll still need to tweak the levers though to keep the correct speed.

Or just use the brakes when you need them 😉

[hec7or]

Boeing Flight Crew Training Manual says when in VNAV descent, if speed is high use speed-brake and if speed is low use thrust. It couldn't be more simple.
This refers to actual speed v target speed and I have found that the FCTM guidelines work extremely well, where it all goes wrong is when pilots try to avoid using the speed-brake for whatever reason.

[Check Airman]

Nothing to add, except to say Boeings sure seem complicated to this Airbus pilot

[Pugilistic Animus]

The Boeing speed brake is sometimes called " the lever of shame" I don't think that way but some do

[stilton]

As far as fuel efficiency is concerned better be high and use speed brake than low and use power

[amc890]

That’s it and if you get the EAI on alt correct it all comes good at 10000ft. Of course as you pointed out actually having to turn it on does complicate things.

[Sidestick_n_Rudder]

Originally Posted by stilton

As far as fuel efficiency is concerned better be high and use speed brake than low and use power

Our fuel gurus actually claim the opposite - but I must admit that I haven’t seen the numbers, so it is anecdotal.

[Alrosa]

Originally Posted by oceancrosser

That last sentence sums it up. The 757/767 FMC/VNAV is archaic. You need to manage these airplanes accordingly. VNAV works ok during descent until you get into manuevering/speed control in the STAR. So you may/will need different modes. Also keep in mind that the weight span of the 767 is far greater than the other jets you have flown. A very light 767 is a different beast from one at MTOW (or MLW at that).

I’d agree with this more or less, and it’s certainly been my experience when flying the 737 300/400s. There’s a time for VNAV and a time for when another mode is more appropriate.

However, I will say that without interference from ATC on speed and altitude etc, when flying into some very quiet airfields with procedural ILS (no radar vectors due local terrain) VNAV - at least on the 737 classic - actually works well all the way down to loc intercept ...it really depends on the situation.

As for the Speedbrake my view is in line with some others here ....it’s there to be used. “Stick of shame” “Butcher’s knife”......I’ve heard it all. The only thing I would have a problem with, is if the guy next to me is switching between VNAV, LVL CHG etc, and using speedbrake because they hadn’t thought about conditions on arrival, how busy is it, shortcuts and 3x your altitude plus 10 etc....no one’s perfect of course

[firstmrj]

I did small “thought experiment” with very much of simplification.

The comparison of fuel consumption between two cases at descent from FL370 to 3000 ft.
Case (1) start descent at the TOD calculated by FMS using speed-brake to decelerate on lower altitude.
Case (2) start descent 5 NM earlier and decelerate on level flight at 3000 ft.

At cruise: case (1) stays at FL370 for extra 5 NM.
It will take 39 seconds. (assuming Mach.8 TAS 459 kt)
Therefore burn +109LBS (assuming fuel flow of 5000LBS/HR for one engine).

At descent: case (1) need a little longer to reach 3000 ft because of slowed airspeed in the later portion.
I put extra 10 second for this reason. so +6 LBS of fuel. (Fuel flow 1000 LBS/HR for one engine)

At 3000 ft: case (2) will fly level at idel thrust while deceleration at average speed of TAS 214 kt. (deceleration from CAS 240 kt to CAS 170 kt)
So it will take 84 seconds, consequentry burn extra 47 LBS. (assuming FF 1000 LBS/HR per one engine)

Summarizing the above, we can save 68 LBS (case (1) 109+6 LBS versus case (2) 47 LBS ) by starting descent earlier than FMS' TOD by 5 NM and avoiding use of speed-brake.

However, who can tell whether 5 NM is appropriate?
If it is too long, we have to burn some more fuel in the dense air mass of low altitude. It will cost a lot.
Furthermore, we need to think about noise, terrain, traffic flow and etc.


I welcome any comment!

[Pugilistic Animus]

Originally Posted by Check Airman

Nothing to add, except to say Boeings sure seem complicated to this Airbus pilot

Exactly how I feel about Airbus!