News:

Precision Simulator update 10.181 (1 February 2025) 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 09:56

emerydc8

I agree -- you will never get low if you are descending in VNAV PTH because thrust will be added if airspeed isn't enough. I should have been clearer. I was looking at it more from a situation where you were descending on the path and ATC assigned you a speed that was higher than your FMC-target speed. If you speed intervened with that speed, it would take you below the path.

If they subsequently told you to resume normal speed and you closed the speed window, you would be below the path and the path capture feature would activate. If you were really far below the path it would take power to reduce your descent rate to 1250'/min and get back on the path (THR/HOLD||VNAV SPD). But if you were close to the path but still more than 150' low, you might get away with just closing the speed window and not having it add any power. It will pitch up for the FMC target speed and in doing so will capture the path in idle thrust.

Hardy Heinlin

Yes, understood, agreed, that's what I meant by the difference between "capture mode" and "maintain mode".

emerydc8

Got it. Now I see. Thanks, Hardy.

Hardy Heinlin

#63
Quote from: emerydc8 on Sun, 20 Sep 2015 21:09
With regard to getting high on the path, prior to the first speed constraint, based on all my research, it would be very unusual to lose the path because the aircraft will accelerate up to VMO-11 (354 knots) to maintain it. If you did lose the path, it would shift to HOLD||VNAV SPD (speed reversion) when it gets more than 150' high.. I have observed PSX correctly shifting into VNAV SPD when it was high, but it usually occurs when much higher than 150'. If you could change that parameter, it would be helpful.

Hi Jon and all,

in the current PSX versions the upper limit is 350 ft instead of 150 ft (350 is easier to maintain without too much mode flip-flopping). I'll try to change it to 150.

But I'm not convinced yet that the FMC should be able to exceed the command speed by more than 15 kt, respectively by more than 10 kt when below FL100. What would ATC say about this?

If there is a "whichever-is-higher/lower" logic in there, then I would say:

This one makes no sense (because VMO must never be exceeded):
The max target is VMO-11 or command+15, whichever is higher.

This makes sense (because VMO is never exceeded):
The max target is VMO-11 or command+15, whichever is lower.

One could also remove the +10/+15 stuff completely. But why is it mentioned in the manuals then?


Regards,

|-|ardy



P.S.: One thing should be considered re VNAV SPD engagement at +150 ft: You expect that the airspeed additive will increase the descent rate in order to stay below +150 ft. But sometimes the +150 ft is exceeded before the full airspeed additive is reached; that is, you may see +300 ft and more without getting VNAV SPD because the full airspeed additive is not reached yet. Only when the airspeed additive too is at the limit, VNAV SPD will engage. So don't be surprised when you see higher values than +150 ft while VNAV PTH is still engaged :-)

emerydc8

Hi Hardy,

I did some digging around on this and I got two interesting (but not entirely consistent) answers. If you don't like the first approach (e.g., it would be really difficult to program up to VMO-11 into PSX), then there is a second, fallback answer that may simplify this to just using the figures given when below the first speed constraint (generally FMC target +15), even while you are technically above the first speed constraint.

First, in the real airplane, you will almost never be in a position where you have to accelerate close to VMO in order to maintain the path because your winds and descent speeds are already entered and accounted for and the FMC is able to accurately calculate the ToD to keep it very close to +/- 150' of path.

When you descend in VNAV PTH you are always within 150' of the path. This isn't really that much for the airplane to handle with a small airspeed increase. If you are more than 150' high, then you are already in VNAV SPD and the airplane will not be pitching to stay on the path anyway (speed reversion occurs).  It would have to be a significant unforecast wind in order to cause a speed close to VMO-11.

One exception, I am told, actually happened to a JAL 744 crew a few years back. Their Econ speed was very high at Mach .86 and during descent they entered a jet stream leading to a rapid speed increase. At first the automatics went to VNAV SPD at MMO-11 knots but the speed still increased to BP.

It became a reportable incident due to the captain pulling on the controls with the autopilot still engaged and when the autopilot became disengaged there was one of those positive followed by negative g things that injured people.

Another exception where you could get close to VMO/MMO on descent might be where you have a high Econ Cruise like M.86 or more and you are cruising at a high altitude. In this case, some guys will actually speed intervene and roll it back a few knots just prior to the ToD to prevent any possibility of overspeeding when it initially starts downhill.

The VMO -11 seems high to most pilots because they are not used to seeing the speed get that high to stay in VNAV PTH. But as far as ATC is concerned, unless they assign you a speed or there is a published speed, you are free to go as fast as you want above 10,000. Even without VNAV (747 classic, DC-8, DC-10) where normal descent speed is about 310 knots, if you were getting high you could just pitch the V/S wheel over and run it up to 340 knots with no problem.

The second answer I got raised an interesting point that I had not considered and I'm not even sure I agree with; but it does help to explain some questions. I almost always enter a descent mach/spd (e.g., M.84/310) on VNAV 3/3 prior to ToD, because I like to descend at 310 knots and the low cost index will usually put us well below this speed if we don't change it. Someone proposed that the mere entry of a mach and speed on the VNAV descent page creates a speed constraint; so the second set of figures (generally up to FMC target +15) would apply. I'm not sure this is the case, but it does sound like something Peter mentioned was the reason the DRAG REQUIRED message would appear at speeds much lower than VMO-16. Although I would be interested in his input on this proposal.

With regard to your PS, I think the airplane will always increase speed first before going above path more than a small amount, if any. So the elevator is going to pitch to keep it on path until it reaches VMO/MMO -11, then it would allow a deviation up to 150' above the path before reversion finally occurs and it gives up.




Hardy Heinlin

#65
Hi Jon,

re +15 versus VMO-11: I can easily remove the +15 limit and let VMO-11 be the only limiter. But I'm not convinced that this will be correct because various manuals (e.g. Bulfer page 199 under "Normal descent") say the target speed will go to +15 (+10 when below FL100). I think you are right if you say it can go to VMO-11. But I say it can go to VMO-11 only if that lies within command speed+15 (or +10). If you use SEL SPD 310 or(OK, not with M.84) a high cost index, it's pretty normal that your command speed +/-0 is already close to VMO, maybe even closer than 15 kt.

Re "in the real airplane, you will almost never be ... close to VMO": I understand. But in the simulator you will, if you will :-) And I have to design it correctly :-)

Re my last P.S.: I just removed the speed condition for the VNAV SPD engagement; it now engages at +150 ft, no matter what the airspeed is. It works very well. I expected some ugly side effects, but there are none. Another problem solved.


Regards,

|-|ardy

emerydc8

Hi Hardy. I haven't been able to find much to help explain whether it is VMO-11 regardless of FMC target or only if it lies within 10 or 15 knots of your FMC target range. The Boeing FCOM does say, "the airplane may accelerate up to 15 knots above target speed to maintain the path," when below the first speed constraint subject to 5 knots above the transition speed (below the trans alt) or 5 knots below the flap placard speed.

I did find something that may shed some light on this but I haven't had a chance to read through it yet. This should put you to sleep for sure. http://www.mlit.go.jp/jtsb/eng-air_report/JA8903.pdf

I'm hoping there will be some good information in there. Cheers.

PS - Great news about changing the speed reversion with minimal issues. Thanks!


Hardy Heinlin

I'm flying beta 7 (not uploaded yet) in the EHAM descent situ with your .84/310 SEL SPD figure. It's fantastic now. Max deviation +30 / -40 feet. Every now and then, for a minute or two, it's even staying at 0 feet. That's magic.

emerydc8

Very cool! I'm looking forward to trying it. BTW, Beta 6 is great -- V1 cuts are much better; You fixed the airway intercept issue and the ALT capture issue, among others. The direct-to behavior is nicer too. Thanks!

Hardy Heinlin

Quote from: emerydc8 on Sat,  3 Oct 2015 06:14The Boeing FCOM does say, "the airplane may accelerate up to 15 knots above target speed to maintain the path," when below the first speed constraint ...

OK, I'll try it ...

emerydc8

#70
Check out the JAL report pdf page 40:

QuoteDuring descent path tracking in the VNAV mode, the autopilot will transition from VNAV PTH mode to the VNAV SPD mode as the airspeed approaches 11kt below Vmo/Mmo. When the VNAV SPD mode becomes active for speed protection, the Autopilot controls speed to 16kt below Vmo/Mmo as a target speed.

pdf page 69:

QuoteDuring the descent in which the accident occurred, the aircraft was flying at the ECON speed of around Mach 0.865. As described in section 2.12.1(2), JAL's Operation Information bulletin stated that "During flight, it is possible for speed to approach VMO/MMO when the headwind increases due to a change in the wind at heavy weights, or if the tailwind decreases. If these conditions are anticipated, it is essential not to stick to ECON speed but to operate at an appropriate speed with a margin for VMO/MMO in advance."

Hardy Heinlin

So there are two conditions?

Change to VNAV SPD if ...

... path > +150 ft

OR

... speed > VMO-13





By the way, I just removed the +15 limit for the segment above the first speed restriction. It works. The algorithm is so self-controlled that it will never exceed +20 anyway, as you said. The trend back to the path already starts at +15 knots anyway. Unless there is an unexpectedly increased tailwind.

emerydc8

That makes sense. So +15 was just restrictive enough to throw the descent path off, but it doesn't really need much more than +20 to hold it?

Hardy Heinlin

Yes. The drag increase by the higher airspeed is just enough.

Britjet

Just be aware that the rules should still work satisfactorily if the aircraft is artificially made high on the path by changing a downroute constraint of a shorter arrival. Personally I am not a fan of the VMO-XXX idea - I have never seen that.

Jon is quite right about the pitch over from TOD at high cost index etc - it is quite alarming on the real aircraft and can lead to the aircraft overspeeding, particularly if shear occurs in the descent. As Jon says the guys will sometimes put in a "fudge", either an intervene approaching TOD, or in my case, a lower descent Mach than the cruise Mach (eg .84 from cruise of .86).

Peter

emerydc8

Hi Peter,

I agree that it would be unusual to see it get that fast to stay on path. You would have to have something like the JAL situation where they entered the core of a jetstream while descending from a high altitude at M.86.

This is from the Big Boeing Book under Speed Reversion 16.12:

QuoteIf above the first altitude constraint or SPD TRANS (not in Speed Intervene), Speed Reversion occurs (on the high side):

11 kts from Mmo/Vmo.
(This is called Vgmax.)
Command speed reverts Mmo/Vmo minus 5.

I think the last sentence was a typo. It should say Command speed reverts Mgmax/Vgmax minus 5, which would be Mmo/Vmo minus 16.


Hardy Heinlin

In this context, the manuals always mention the first speed/altitude constraint as a reference.

I think the SPD RESTR on the VNAV DES page may also be the reference -- if it's higher than the SPD TRANS and higher than the first speed/altitude constraint.

When the SPD RESTR is not dashed, and if the first speed/altitude constraint on the LEGS is below SPD TRANS and below SPD RESTR, there will be two deceleration segments ("knees") on the idle path. These two knees are also shown on the predicted idle path graph on Instructor > Analysis > Misc.


|-|ardy

emerydc8

I haven't experimented much with the SPD RESTR feature on descents, but I do know that it will not level off at the SPD RESTR altitude to keep you from exceeding that speed before descending further, like it does at the SPD TRANS altitude. The target speed will ratchet down, but it will keep descending, albeit at a lower rate, until it gets to the entered RSTR speed.

emerydc8

Quote from: Britjet on Sat,  3 Oct 2015 09:04
Just be aware that the rules should still work satisfactorily if the aircraft is artificially made high on the path by changing a downroute constraint of a shorter arrival.

Hi Peter,

I've been thinking about this and I wanted to get your input. My understanding is that if you did make it artificially high by changing a downroute constraint (speed or altitude), the FMC would calculate a new path and kick you into VNAV SPD if you were more than 150' above that new path. In that case, VNAV SPD would not be pitching to try to get back to the path anyway. So, you can make yourself artificially high and stay in VNAV PTH, but anything more than 150' high and you are in VNAV SPD.

Just like when you enter a new descent speed on VNAV 3/3, as soon as you execute, the FMC automatically computes a new path based on this new speed. If you are within 150' of that new path, I can see where it will stay in VNAV PTH and attempt to capture the new path; but if you are outside that envelope (>150' high) because of the new restriction that you entered, VNAV will revert to VNAV SPD.

So in VNAV PTH you are always within 150' of the path. That's the worst-case scenario you would see. If you are higher than that, you would not be in VNAV PTH, so the rules (VMO-11 or FMC target + 15 -- whichever you like) would not apply.


Hardy Heinlin

Quote from: emerydc8 on Sat,  3 Oct 2015 19:18
I haven't experimented much with the SPD RESTR feature on descents, but I do know that it will not level off at the SPD RESTR altitude to keep you from exceeding that speed before descending further, like it does at the SPD TRANS altitude. The target speed will ratchet down, but it will keep descending, albeit at a lower rate, until it gets to the entered RSTR speed.

I didn't mean "level off" but "deceleration" (using -500 fpm or so) -- as you say "lower rate".


|-|ardy