xkoote

B748 VNAV descent

Recommended Posts

Hello,

maybe someone can clarify this for me. I have flown two different aircraft types in my life that are fully VNAV capable, and countless in the FS world. Is it just me, or does the B748 remain endlessly in SPD | xxx | VNAV PTH when initiating a descent? Maybe this is specific to the B748? From what I know, the Boeing VNAV will;

- Calculate a ToD point using as much information it has at its disposal ( e.g. descent winds, gross weight, descent speed and crossing restrictions )

- Once that point is reached, VNAV will initiate a sequence that without variation will reduce the thrust to idle. IDLE is displayed during this time. When the throttle reaches idle, thrust mode changes to HOLD.

- ONLY when current airspeed drops below a preset value, does thrust mode change back to SPD to re capture the speed before going back to idle.

As I am experiencing for a few flights now, VNAV will stay in SPD | xxx | VNAV PTH at top of descent, and thrust will keep managing speed during the descent. I kept it in VNAV for about 2 minutes to see if it finally changes to idle. It never does. Am I missing something? Or is this maybe a known issue?

 

Cheers,

Share this post


Link to post
Share on other sites
Help AVSIM continue to serve you!
Please donate today!

The -8 FMC calculates an "off-idle" VNAV descent, so it's normal behavior AFAIK.

Share this post


Link to post
Share on other sites

Interesting ! Thanks. I’ll go find some reading material ...

Share this post


Link to post
Share on other sites

But doesn't the "off-idle" descent mean that the engines are just not completely in idle but a few % above it (flight idle) in order to reduce the drag? What I've seen so far is however that it is constantly managing the speed, so it's always increasing and decreasing the thrust.

Share this post


Link to post
Share on other sites
1 hour ago, Skyrock said:

But doesn't the "off-idle" descent mean that the engines are just not completely in idle but a few % above it (flight idle) in order to reduce the drag? What I've seen so far is however that it is constantly managing the speed, so it's always increasing and decreasing the thrust.

Drag is actually what you want when you move your throttles back. It‘s hard enough to slow and bring an airliner down. AFAIK the only reason for the engines not being in full idle during flare is the fact that it takes forever (and definitely too long) to bring them out of the N1 cellar if you have to go around. But in decent you actually appreciate any additional drag that „prevents“ you from using the spoilers/speed brakes. But unfortunately jet engines don‘t produce any significant drag when idling, „idle“ is where the air stream let‘s them slow down to. That‘s why you see them at around 40% at „decent idle“ in opposite to 25 or so on the ground. 

Share this post


Link to post
Share on other sites
1 hour ago, Ephedrin said:

Drag is actually what you want when you move your throttles back. It‘s hard enough to slow and bring an airliner down. AFAIK the only reason for the engines not being in full idle during flare is the fact that it takes forever (and definitely too long) to bring them out of the N1 cellar if you have to go around. But in decent you actually appreciate any additional drag that „prevents“ you from using the spoilers/speed brakes. But unfortunately jet engines don‘t produce any significant drag when idling, „idle“ is where the air stream let‘s them slow down to. That‘s why you see them at around 40% at „decent idle“ in opposite to 25 or so on the ground. 

Still this is not the point here. The point is that the engines don't remain in idle (regardless of which idle that means). It's constantly spooling up and down because it would slow down if it was kept in idle. The winds are completely entered into the des forecast page, but still. I think the -400 handles the descent better and also the 777 does.

Share this post


Link to post
Share on other sites
1 hour ago, Skyrock said:

Still this is not the point here. The point is that the engines don't remain in idle (regardless of which idle that means). It's constantly spooling up and down because it would slow down if it was kept in idle. The winds are completely entered into the des forecast page, but still. I think the -400 handles the descent better and also the 777 does.

Yes I thought about adding that this doesn‘t explain the behaviour you see but was too lazy as to me it was obvious that I was talking about just the idling itself 😄 nevermind, I noticed the throttles going up and down a bit too. Sometimes they spool up way too much so that I have to retard them manually. But when you watch YT vids of boeings on approach you often see pilots pulling them back. There‘s a russian 737 captain (Denis Okan) who in one of his videos engages the A/T after a go around (I believe) but as it gives far too much thrust he immediately turns it off again.. So I’m not sure if the Boeings are always that perfectly adjusted as one would wish. The 737 isn‘t the best VNAV follower at all, so although the 748 is newer I wouldn‘t expect a perfectly balanced system (just because I would like it to be :D) 

Sounds a bit strange but I hope you get what I want to say lol

Share this post


Link to post
Share on other sites
2 hours ago, Ephedrin said:

Yes I thought about adding that this doesn‘t explain the behaviour you see but was too lazy as to me it was obvious that I was talking about just the idling itself 😄 nevermind, I noticed the throttles going up and down a bit too. Sometimes they spool up way too much so that I have to retard them manually. But when you watch YT vids of boeings on approach you often see pilots pulling them back. There‘s a russian 737 captain (Denis Okan) who in one of his videos engages the A/T after a go around (I believe) but as it gives far too much thrust he immediately turns it off again.. So I’m not sure if the Boeings are always that perfectly adjusted as one would wish. The 737 isn‘t the best VNAV follower at all, so although the 748 is newer I wouldn‘t expect a perfectly balanced system (just because I would like it to be :D) 

Sounds a bit strange but I hope you get what I want to say lol

There are certainly some flaws when it comes to VNAV descents, but it seems to be happening everytime and from what I saw, the -400 does a better job here despite it being older. Only when the -8 is forced to go into idle (directs or rapid wind changes), it seems to work, but in normal conditions, it's always spooling up and down.

Share this post


Link to post
Share on other sites
2 hours ago, Skyrock said:

There are certainly some flaws when it comes to VNAV descents, but it seems to be happening everytime and from what I saw, the -400 does a better job here despite it being older. Only when the -8 is forced to go into idle (directs or rapid wind changes), it seems to work, but in normal conditions, it's always spooling up and down.

I strongly disagree with the assertion that the -400 does a better job than the -8 on VNAV descent.  You are comparing two different airplanes, two different wings, very different engines and much newer AFCS that resembles the 787 more than the 747.

I also disagree that off idle descent is a flaw.  I cannot even imagine what kind of logic comes to that conclusion.  The NG FMS has a default descent profile that is the economy descent to 10,000.  You will see much less drag required messages too.

  • Like 1

Share this post


Link to post
Share on other sites
1 minute ago, downscc said:

I strongly disagree with the assertion that the -400 does a better job than the -8 on VNAV descent.  You are comparing two different airplanes, two different wings, very different engines and much newer AFCS that resembles the 787 more than the 747.

I also disagree that off idle descent is a flaw.  I cannot even imagine what kind of logic comes to that conclusion.  The NG FMS has a default descent profile that is the economy descent to 10,000.  You will see much less drag required messages too.

Of course I don't see as much drag required messages because the A/T has to apply thrust the whole time.

I did not say that the off idle descent is a flaw, I said that VNAV descents in general have some flaws, because they are not perfect and will never be.

Share this post


Link to post
Share on other sites
33 minutes ago, Skyrock said:

because they are not perfect and will never be.

Pure curiosity, what is your definition of a prefect descent?  Are we referring to the actual aircraft or the simulation of the aircraft?

Edited by downscc

Share this post


Link to post
Share on other sites
3 minutes ago, downscc said:

Pure curiosity, what is your definition of a prefect descent?  Are we referring to the actual aircraft or the simulation of the aircraft?

An idle descent from the T/D to the approach. If there are constraints or ATC interventions along the way which have to be met, then it can not necessarily be an idle descent, but so far my descents were without constraints...

Even in reality, VNAV sometimes screws up. It most probably is more reliable than the simulation, but still not always.

Share this post


Link to post
Share on other sites

Take a read of Chapter 11 in the 747-8 FCOM.  There are some notable differences between the -8 and the 747-400.

The expected modes are explained pretty well, along with “off-idle” descents and the “VNAV Speed Band”.

Share this post


Link to post
Share on other sites
6 minutes ago, Skyrock said:

An idle descent from the T/D to the approach. If there are constraints or ATC interventions along the way which have to be met, then it can not necessarily be an idle descent, but so far my descents were without constraints...

Even in reality, VNAV sometimes screws up. It most probably is more reliable than the simulation, but still not always.

Okay that explains it.  You would consider the use of FLCH with idle thrust a perfect descent.  However, the VNAV profile uses the most economical descent that is determined primarily by the CI value.  If time has not value then you would use a CI = 0 and I suspect there the descent would use much less thrust and a lower descent speed.  Of course time is very expensive in a -8 (labor, insurance, schedules, etc) so you'll probably find most carriers using a CI closer to 85 than 0.  The VNAV profile is almost perfect in matching the economical constraints.  The implementation in the PMDG simulation is imperfect and at times the vertical path will have discontinuities at waypoints but this is going to take new FMS code from them to fix.

Share this post


Link to post
Share on other sites
30 minutes ago, rcoultas said:

Take a read of Chapter 11 in the 747-8 FCOM.  There are some notable differences between the -8 and the 747-400.

The expected modes are explained pretty well, along with “off-idle” descents and the “VNAV Speed Band”.

I did. In chapter 11, it reads:

Quote

If the VNAV path segment is too shallow to be flown satisfactorily at IDLE thrust, the FMC commands speed on thrust levers (SPD).

For me, this sounds like applying thrust is the exception rather than the usual procedure.

Again: my "problem" here is not the off-idle. Thats completely fine for me. My problem is the A/T spooling up and down in the descent due to a too shallow descent.

Quote

Okay that explains it.  You would consider the use of FLCH with idle thrust a perfect descent.  However, the VNAV profile uses the most economical descent that is determined primarily by the CI value.  If time has not value then you would use a CI = 0 and I suspect there the descent would use much less thrust and a lower descent speed.  Of course time is very expensive in a -8 (labor, insurance, schedules, etc) so you'll probably find most carriers using a CI closer to 85 than 0.  The VNAV profile is almost perfect in matching the economical constraints.  The implementation in the PMDG simulation is imperfect and at times the vertical path will have discontinuities at waypoints but this is going to take new FMS code from them to fix.

A higher CI would result in a higher speed which can be easily achieved by delaying the descent and flying a steeper profile, in particular for a 747 which has more drag than a 777 this should be working just fine. It just doesn't make sense to me to have a rather shallow descent path where you'll always have to apply thrust which burns more fuel (besides being slower than at CRZ level given there are not better winds down there) instead of remaining at optimum CRZ level and having an idle descent. I was using CI70 all the time, so I would not consider this as particularly fast and still it had to keep the speed up (with a rather shallow V/S of around 1800-2200ft/min)

Edited by Skyrock

Share this post


Link to post
Share on other sites
1 hour ago, Skyrock said:

A higher CI would result in a higher speed which can be easily achieved by delaying the descent and flying a steeper profile, in particular for a 747 which has more drag than a 777 this should be working just fine. It just doesn't make sense to me to have a rather shallow descent path where you'll always have to apply thrust which burns more fuel (besides being slower than at CRZ level given there are not better winds down there) instead of remaining at optimum CRZ level and having an idle descent. I was using CI70 all the time, so I would not consider this as particularly fast and still it had to keep the speed up (with a rather shallow V/S of around 1800-2200ft/min)

You make a common mistake of thinking that fuel burn is the only economic determinant.  Also, comparing the drag of a 777 to a 747 is comparing apples and oranges and actually it turns out that the constant body width form that the Queen affords by having a equivalent to a body pinch at the wing roots thanks to the upper deck significantly lowers the Queens drag below that of a straight tube.  She is bigger but sleeker. I doubt that Boeing has made a big error in designing what an economic descent profile looks like,, I suggest you open your mind to seek an understanding of why it works rather than sticking to your definition of an perfect descent profile.

Share this post


Link to post
Share on other sites
16 minutes ago, downscc said:

You make a common mistake of thinking that fuel burn is the only economic determinant.  Also, comparing the drag of a 777 to a 747 is comparing apples and oranges and actually it turns out that the constant body width form that the Queen affords by having a equivalent to a body pinch at the wing roots thanks to the upper deck significantly lowers the Queens drag below that of a straight tube.  She is bigger but sleeker. I doubt that Boeing has made a big error in designing what an economic descent profile looks like,, I suggest you open your mind to seek an understanding of why it works rather than sticking to your definition of an perfect descent profile.

I did not say fuel burn is the only economic determinant, but I have already learnt in the topic about ISA dev. that you tend to pick those parts in my post which suit your argumentation, ignoring everything else, so I'll leave that here. Please stop turning my statements into something I've never said nor meant..

Hopefully someone from PMDG might be able to give a statement regarding this topic.

Edited by Skyrock

Share this post


Link to post
Share on other sites
39 minutes ago, Skyrock said:

I did not say fuel burn is the only economic determinant, but I have already learnt in the topic about ISA dev. that you tend to pick those parts in my post which suit your argumentation, ignoring everything else, so I'll leave that here. Please stop turning my statements into something I've never said nor meant..

Hopefully someone from PMDG might be able to give a statement regarding this topic.

The only way to communicate directly with PMDG is via their support portal.

Let's use your CI of 70 to set up an example of an ECON VNAV descent path.  Trying to find something that you can relate to so using your CI that means the aircraft costs 70 times as much as the fuel per hour.  Putting some numbers to that and assuming an airline has to pay much less than you or I would have to at Signature say $4.25 /gal Jet A.  We are going to convert to cents per lbs so a gallon is roughly 6.72 lb on a good day, which gives us a fuel cost of 0.63/pound.   This means the CI = 70 is set by the ration aircraft cost per hour to the fuel cost cents/pound or (70 * 63) / 63 which is the ration 4,410 / 63.  The aircraft is costing $4,410 /hr given this scenario.   Flying the economic profile will result in the least cost where the cost is a combination of aircraft cost and fuel cost.  The aircraft cost includes insurance, labor, cost of ownership (be it lease recovery or depreciation).  The ECON descent speed is the speed that minimizes the total cost.  It will not work out to be the zero thrust descent unless your CI is closer to zero.

See also: https://www.google.com/search?q=boing+cost+index&ie=utf-8&oe=utf-8&client=firefox-b-1

Edited by downscc

Share this post


Link to post
Share on other sites

Unless something drastic has change on the -8, I think the OP is asking why the throttles aren’t commanded to idle at top of descent like they are in the -400

Edited by VHOJT

Share this post


Link to post
Share on other sites
16 hours ago, downscc said:

The only way to communicate directly with PMDG is via their support portal.

Let's use your CI of 70 to set up an example of an ECON VNAV descent path.  Trying to find something that you can relate to so using your CI that means the aircraft costs 70 times as much as the fuel per hour.  Putting some numbers to that and assuming an airline has to pay much less than you or I would have to at Signature say $4.25 /gal Jet A.  We are going to convert to cents per lbs so a gallon is roughly 6.72 lb on a good day, which gives us a fuel cost of 0.63/pound.   This means the CI = 70 is set by the ration aircraft cost per hour to the fuel cost cents/pound or (70 * 63) / 63 which is the ration 4,410 / 63.  The aircraft is costing $4,410 /hr given this scenario.   Flying the economic profile will result in the least cost where the cost is a combination of aircraft cost and fuel cost.  The aircraft cost includes insurance, labor, cost of ownership (be it lease recovery or depreciation).  The ECON descent speed is the speed that minimizes the total cost.  It will not work out to be the zero thrust descent unless your CI is closer to zero.

See also: https://www.google.com/search?q=boing+cost+index&ie=utf-8&oe=utf-8&client=firefox-b-1

Congratulations, you have come up with a lengthy calculation, yet failed to provide a proof that an idle descent is less economic than having to apply thrust during the whole descent. Even Airbus states in their "getting to grips" manual (see here) that the higher the cost index

- the steeper the descend path

- the shorter the descent distance

- the later the top of descent (TOD)

But I'm sure you won't accept this since it's Airbus and not Boeing. They haven't re-invented the CI and it's not a rocket science, so this can be applied to Boeing, too. Even in our own FCOM provided by PMDG it implies that the usual descent is flown in IDLE thrust. The NGX, the 777 as well as the 747-400 prove that the above mechanic works there the exact same way, so I don't accept the "it's not a bug, it's a feature" attitude towards this VNAV behavior.

I'll consider opening a ticket since this discussion here leads to nowhere.

Edited by Skyrock

Share this post


Link to post
Share on other sites
2 hours ago, Skyrock said:

yet failed to provide a proof that an idle descent is less economic than having to apply thrust during the whole descent.

You are correct in that less fuel is used.  Boeings do the same thing with higher CI the descent speed is higher.  I get it that you don't like the off idle descent.  I do not have the knowledge of the specifics as to why the -8 does this but it's not a bug.  It is discussed in the FCOM so someone that is a lot smarter than either of us has engineering the aircraft to do what you see it doing.  Over at airliners.net there are posts on this topic that support the notion that going down faster is not the most economical descent but no factual basis.  Another comment on the -8 from there is that there is less engine drag with a low level of thrust (off idle).  I've exhausted my knowledge.

Share this post


Link to post
Share on other sites

The off-idle is mainly because the -8 is a more “clean” airplane than the -400. It can be difficult to slow down, especially at high weights. The off-idle allows for a bit of maneuvering room as to avoid an unstable descent. I’m not sure about the sim, but the real aircraft has descent profile information set up in the AMI (Airline Modifiable Database) that allows carriers to customize the descent profiles and whether or not they use the off-idle. 

Share this post


Link to post
Share on other sites

From the real world -8 FCOM Vol 2 page 11.31.29: “off-idle thrust is selectable in the AMI; it allows an off-idle thrust setting to stabilize descent path and minimize maneuvering at low altitudes” 

Share this post


Link to post
Share on other sites

Gents,

 

I thought I'd quickly insert a couple of paragraphs of the VNAV Differences manual regarding the -8 to clarify the issue (my underlining).

(NG FMC) The descent gives path control priority and is constructed as a flyable, non-idle descent path, to allow VNAV to control speed and maintain the path with minimal flight crew input. To support path control priority, when on path in descent, the VNAV mode is VNAV PTH and autothrottle mode is SPD.
 
(NG FMC) The offidle descent path is constructed slightly shallower than a traditional idle path descent to allow a margin of speed control during the descent. The VNAV descent mode is VNAV PTH and autothrottle SPD for the entire descent profile, similar to today’s 777 or 747-400 on-approach mode logic. This enables the autothrottle to remain active during the descent and automatically make small thrust adjustments to maintain the descent speed while remaining on path.
 
(NG FMC) With thrust levers at the idle stop in an accurately forecast VNAV PTH descent, the aircraft decelerates approximately 5 knots over 3000 feet altitude. This is heavily dependent on the accuracy of forecast winds. The more accurate the forecast, the more accurate the FMC calculates predictions and generates the descent path. Speed brakes may be required.
 
This is exactly how we have modelled it. I hope this is the definite answer to questions as it comes directly from the manufacturer.
 
 
Best,
 
Vangelis
Edited by emvaos

Share this post


Link to post
Share on other sites
4 minutes ago, emvaos said:

This is heavily dependent on the accuracy of forecast winds. The more accurate the forecast, the more accurate the FMC calculates predictions and generates the descent path. Speed brakes may be required.

This is why on long 8+ hour flights with ASP4 I'll go request an updated wind data download for the descent forecasted winds in the VNAV DES page. I typically see the T/D point adjust on the ND. 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now