Old 744 Forum

Archived posts

"On-Approach" Logic on STAR?

Page: 1 2 >

Author Post
Member
Registered: Jun 2015
Posts: 103
Location: Tucson, AZ USA
Hello,

I was playing around with the SIERA 6C arrival for the ILS 07L into HKG today and noticed that on the descent, the aircraft is going into on-approach logic at SIERA, which is almost 30 miles from HKG and is the initial waypoint on the STAR. It should not be going into on-approach logic until sequencing the course fix for 07L (CI07L).

I can tell that it's going into on-approach logic on the descent right at SIERA because I kept speed intervening and watching it go from | HOLD | LNAV | VNAV PTH | to | HOLD | LNAV | VNAV SPD | when I speed intervened. As soon as it cycled past SIERA, it stayed in VNAV PTH when I speed intervened and the autothrottle went straight to | SPD | even though it was on the path. This is consistent with on-approach logic.

As far as the descent profile itself goes, it made the altitudes and speed restrictions perfectly, but the FMA's were not accurate here:

Regardless of whether you are in on-approach logic or not, if you are doing a VNAV PTH descent and you get more than 150' away from the path, it will drop out of VNAV PTH and into VNAV SPD. If the autothrottles are active, VNAV should not let you get more than 150' below the path. It will add power to keep you on the path.

But if you are descending in VNAV, in on-approach logic, and you get above the path by more than 150' (as happened on this arrival), you should see | IDLE/HOLD | VNAV SPD | as soon as the aircraft gets more than 150' high. It will not pitch down and increase speed to get back to the path once in VNAV SPD. If you catch the path at idle or use speed brakes, that's fine, but once the airplane gets kicked into VNAV SPD (more than 150' high), it's not going to increase airspeed to catch the path. If you do manage to get it back down to the path, it will shift from VNAV SPD back to VNAV PTH when within 150' of the path. Note that if you are not in on-approach logic, the speed window must be closed for this to happen (you cannot descend in VNAV PTH with the speed window open unless you are in on-approach logic).


This is a strange arrival and I can see it would be difficult for the FMC to get confused. At one point, between SIERA and BORDA, you are actually flying away from the airport. During this situ, I kept PROGS 2/3 open to get a digital reading of VTK Error. There were a number of times that it got well past 200' high, but never dropped into VNAV SPD.

So, one issue is that it's going into on-approach logic early. The other is that it is not shifting to VNAV SPD when more than 150' above the path.

I made a situ file for anyone interested in dropping it into your situation file folder and flying it for yourself. It starts with the motion frozen so you can assess everything before cutting it loose.

https://drive.google.com/file/d/0B-WRh0Hf7VdZYkhUcU1QTkxjQ3c/view?usp=sharing

I have done this arrival quite a few times in the real airplane and it never went into on-approach logic on the STAR. I don't think the -400 would go to on-approach logic even if the STAR was coded with a gradient path the whole way down, but this is not the case on the SIERA 6C arrival anyway.

I would be interested in any thoughts on these issues.

BTW, here's a copy of the SIERA 7C arrival which is essentially the same as the SIERA 6C.

https://drive.google.com/file/d/0B-WRh0Hf7VdZSGZHdjk2WTZsUUU/view?usp=sharing

Jon D.
Moderator
Registered: May 2009
Posts: 5140
Hello,

different pilots, different opinions. Unfortunately, I can't make everyone equally happy:

http://aerowinx.com/forum/topic.php?id=2788


Regards,

|-|ardy
Member
Registered: Dec 2011
Posts: 497
Location: Portugal
I confess I get completely lost reading this threads, but tonight I'll make my best to discuss it with my hosts, Jeroen and Cristina, inspired, so I hope, by good Portuguese food and drink, but..., allow me a question gentlemen...

Can this mode transitions be a function of the programmed STARS / SIDs, or is there a kernel of the Logic that is only up to the FMS to control ?

I ask based on the OP, and on HH's answer, pointing into a thread where another pilot ( ? ) describes apparently the opposite, and basis his observation on his company SOPs...

So, while I try to understand every bit of it ( good red wine will help latter today... ) I would really like to know if:

- Can this sort of Logic depend on the way the procedures are programmed ?

..or...

- it depends only on the FMS, and is invariable for the way STARS are programmed ?
_______________
Jose Monteiro
Member
Registered: Jun 2015
Posts: 103
Location: Tucson, AZ USA
« Last edit by emerydc8 on Sat, 01 Aug 2015 17:21:55 +0000. »
Moderator
Registered: May 2009
Posts: 5140
Then perhaps we're talking about two different things.

You call the usage of VNAV PTH at +150 feet above the path "on-approach logic"? I don't. It may be a function of the idle path descent mode. When the aircraft is high and the trend is such that it will recapture the path, PSX will keep VNAV PTH. Only when it tends to leave the path and cannot keep the target speed+15, PSX will change to VNAV SPD.

I never heard of a fixed 150 feet value for the mode change-over trigger. Is this a value you observed on the big sims?

PSX will switch to on-approach logic only when the active leg is coded as an approach leg. The code is different to that of a STAR leg.


|-|ardy
Member
Registered: Jun 2015
Posts: 103
Location: Tucson, AZ USA
« Last edit by emerydc8 on Sat, 01 Aug 2015 17:40:08 +0000. »
Moderator
Registered: May 2009
Posts: 5140
Member
Registered: Jun 2015
Posts: 103
Location: Tucson, AZ USA
On the descent in VNAV (with the speed window closed), you should see IDLE/HOLD | | VNAV PTH if you are on path.

If you start getting low, you should see THR as the throttles come up, but it should still remain in VNAV PTH. Once it recovers its speed (it pitched up and slowed down to try to stay on path), it should go back to ILDE/HOLD. It should not go into SPD || VNAV PTH if power is needed to maintain the path on a descent.

If you get high (more than 150' above the path), not only should the autothrottle FMA indicate IDLE/HOLD, but the pitch FMA should go from VNAV PTH to VNAV SPD.

In any case, you should never see SPD || VNAV PTH while descending on a STAR. I have never heard of STAR logic. Is this published somewhere?
« Last edit by emerydc8 on Sat, 01 Aug 2015 18:28:30 +0000. »
Moderator
Registered: May 2009
Posts: 5140
emerydc8 wrote
In any case, you should never see SPD || VNAV PTH while descending on a STAR.

This is new to me. But I believe you until someone proves it may happen on a STAR as well :-)


emerydc8 wrote
I have never heard of STAR logic.

There's no such term. I just used this wording above to emphasize that this is not "on-approach" logic, and that PSX is not "on-approach" logic when on a STAR.
Member
Registered: Jun 2015
Posts: 103
Location: Tucson, AZ USA
Okay, I hope to get some input on this from others. While it may not be in full-blown on-approach logic during that arrival, it is about half way there because you can speed intervene and it will stay in VNAV PTH. That's not supposed to happen unless you are in on-approach logic.
Moderator
Registered: May 2009
Posts: 5140
What if you have a level-off segment on the STAR, e.g. two consecutive "AT" constraints at FL150? Will you maintain FL150 with THR || VNAV PTH?
Member
Registered: Jun 2015
Posts: 103
Location: Tucson, AZ USA
If you are level and flying the level segment, then you would see SPD || VNAV PTH. But you would not see that while "descending" on a STAR.

Let me do some more research on this today.
Member
Registered: Jun 2015
Posts: 103
Location: Tucson, AZ USA
Okay, I think I was wrong about never seeing SPD || VNAV PTH on a descent. In the case of a flight plan modification or unknown winds, if you get 15 knots below the target speed while attempting to maintain the path the autothrottles will go to SPD in order to accelerate back to the FMC target speed, then back to IDLE/HOLD once it reaches that speed. It should not sit there in SPD || VNAV PTH once it regains the target speed. If the autothrottle is not active you will get a "THRUST REQUIRED" on the scratchpad.

In the case of this arrival, there were no unforecast winds (I uplinked the descent winds) and no flight plan modifications; and you are never close to being 15 knots below the FMC target speed, so you should not be seeing SPD in the autothrottle FMA. It should be descending at IDLE/HOLD. The reason I felt something was wrong here is that you should not be staring at SPD || VNAV PTH during the entire descent. That should just be transitory until you regain your speed, then back to IDLE/HOLD.

The only at-or-below restriction is at MURRY (FL130-FL110). All the rest of the altitdue restrictions are at-or-above FL130. Because the database did not have the SLP speeds in it, I had to enter at-or-above altitudes in order to assign the speed limits at SIERA (280) and BORDA (250). So, it should have been able to descend at idle the whole way. There were no level or shallow segments.

It would be appreciated if any -400 guys could try flying this situ and giving some input.
Moderator
Registered: May 2009
Posts: 5140
If it can't maintain the idle descent path at -10 knots with idle thrust, I think the mode should change to THR, not SPD.
Member
Registered: Jun 2015
Posts: 103
Location: Tucson, AZ USA
That's what I initially thought too -- and maybe you are right. But from doing some research on it, I'm now leaning towards it going to SPD to accelerate back to the FMC target speed, then back to IDLE/HOLD.

I'm not sure why it can't do an idle descent all the way to MURRY on this. All the other altitudes are at-or-above. Crossing BORDA, ROCCA and CANTO at-or-above FL130 should not be restrictive.
Moderator
Registered: May 2009
Posts: 5140
Set all constraints before MURRY to an altitude lower than that at MURRY. MURRY should have the highest constraint AND it shouldn't be an altitude AB window but an exact AT constraint. Maybe this helps. I haven't tried it.
Member
Registered: Jun 2015
Posts: 103
Location: Tucson, AZ USA
Okay. I'll try it.
Moderator
Registered: May 2009
Posts: 5140
Just tested. It won't help.
Member
Registered: Jun 2015
Posts: 103
Location: Tucson, AZ USA
Me too. It initially starts out approaching SIERA and says it's 7000 high. When I extend the speedbrakes and try to get down, I eventually do, but it's not until the segment between BORDA and ROCCA.

During this descent, the initial descent FMA is correctly in | HOLD | | VNAV SPD | because it's indicating more than 150' above the path. Then, when I finally do get down to the path, it goes back to | SPD | | VNAV PTH | the rest of the way and the power is sitting at about 50% N1. So, it probably wasn't really 7000' high since it's taking a good amount of thrust once I do get on the FMC path to fly the rest of the arrival.

Ideally, it should go to | IDLE/HOLD || VNAV PTH | once it finally does capture the path and the airspeed should decrease below the FMC target speed to try and maintain the path before power is ever added.

When the manual refers to passing the first "altitude constraint waypoint," would that be an at-or-above waypoint like SIERA (280/FL130A), or is it looking for a hard or at-or-below altitude like MURRY (FL130BFL110A)? Since the database does not have the speed limit points coded in, I had to assign an at-or-above altitude in order to assign a speed at those waypoints. It's either that or speed intervene.
Moderator
Registered: May 2009
Posts: 5140
Member
Registered: Jun 2015
Posts: 103
Location: Tucson, AZ USA
It's complicated enough when you are descending straight towards the crossing fix. But when you are flying away from it in some parts of the STAR, I suspect it makes it even more complicated. Maybe this was a bad arrival to use to test this.

So, if it was looking at MURRY at FL130 (or between FL110 and FL130) and projecting back towards towards the top of descent, shouldn't it be able to do a 3 degree path without any knees getting in the way? All of the waypoints going backwards from MURRY to the top of descent are FL130A, so they are really no more restrictive than MURRY at FL130.

This is so confusing! I wish Peter or others could give us their input here. I have to step out for a few hours, but I should be back at it later today. I want to try some different arrivals that aren't as circuitous!
Member
Registered: Aug 2014
Posts: 375
Location: Camberley, UK
emerydc8 wrote
.......you can speed intervene and it will stay in VNAV PTH. That's not supposed to happen unless you are in on-approach logic.

I agree.
I am pretty sure that you can only get SPD and VNAV PTH when flying level at an FMC altitude constraint, or when on approach logic.

Otherwise, "intervening" should give VNAV SPD...

Maybe something amiss here?

Peter
Member
Registered: Jun 2015
Posts: 103
Location: Tucson, AZ USA
I apologize if I'm adding more confusion with this post, but I did more research and wanted to get it out here so that we can have a better discussion of what the FMS should be doing.

First, I wanted to distinguish situations where you can speed intervene and still remain in VNAV PTH and situations where you don't speed intervene (speed window is closed) but you see |SPD|LNAV|VNAV PTH| on the FMA during a VNAV descent.

I agree with Peter that the only time you can speed intervene and stay in VNAV PTH is when flying level at an FMC altitude constraint, or when in on-approach logic.

I did, however, find three situations on a descent where you can see |SPD|LNAV|VNAV PTH| with the speed window closed:

1) If after the first altitude constraint waypoint, the second VNAV path segment is too shallow to be flown in IDLE, then you will see |SPD|LNAV|VNAV PTH| while descending on that segment.

2 & 3) If there is a flight plan modification or unknown winds when above the first speed constraint, VNAV will slow up to 15 knots below the FMC target speed to stay on path. If that's not enough, the autothrottle will change from IDLE/HOLD to SPD. Here's a link to the two pages I'm using for reference:

https://drive.google.com/file/d/0B-WRh0Hf7VdZM3Jud19VOXA3Mms/view?usp=sharing

I tried a new situ today going into JFK on the LENDY 6 Arrival. https://drive.google.com/file/d/0B-WRh0Hf7VdZM3Jud19VOXA3Mms/view?usp=sharing

To make it simple, I didn't insert the FL230 at JENNO. I just set up one crossing restriction at LENDY (FL190 and 250 knots). You can see for yourself how it does. Actually, if you do nothing, it comes pretty close to making the altitude and speed. It crosses LENDY at about FL196 and 265 knots. It does get 1500' high when slowing to 250 about 12 miles from LENDY, but it seems to close up the gap by the time it gets there.

As for the FMA, at top of descent it correctly goes to |IDLE/HOLD|LNAV|VNAV PTH|. And if you speed intervene, it correctly goes to |HOLD|LNAV|VNAV SPD|.

On a VNAV PTH descent, there are two ways that I know of to get into VNAV SPD -- speed intervention or getting more than 150' away from the path. During the descent in the situ above, it correctly went to VNAV SPD when I speed intervened; but when the aircraft was more than 150' from the path (above or below), it incorrectly displayed VNAV PTH instead of VNAV SPD.

Since it went above the path on its own in my situ, I purposely put it in a position below the path to test the FMA here as well. As soon as it starts descending out of FL370, speed intervene and select M.90. It will correctly go to |HOLD|LNAV|VNAV SPD| and start diving below the path (keep PROGS 2/3 open to see how far below). When it gets at least 1000' below the path, close the speed window. The displayed FMA -- |HOLD|LNAV|VNAV PTH| -- is incorrect. It should display |HOLD|LNAV|VNAV SPD| until getting back within 150' of path.

Once you close the speed window, it will revert back to the FMC target descent speed and in doing so it will reduce the rate of descent to 1200 FPM. This is the path capture feature of VNAV which is basically the same thing it does if you start an early descent. If power is also needed to maintain the 1200 FPM descent, it should display |THR/HOLD|LNAV|VNAV SPD| until within 150', then it should go to |IDLE/HOLD|LNAV|VNAV PTH|.

So, while the actual descent profile seems to be pretty close when using just one crossing altitude and speed restriction (HKG was probably not a good place to start), the FMA's do not seem to be indicating correctly when more than 150' from the path.

I also noticed that the speed variations available to capture or stay on path were limited to only about 8 knots. According to the attached material, prior to reaching the first speed constraint (in this case LENDY), the FMC should have accelerated up to Vmo - 16 (349 knots) before the DRAG REQUIRED message was displayed on the scratchpad. Likewise, if below the path, the aircraft should have slowed to FMC target speed -15 before adding thrust.

When I put the airplane 1000' below the path, it added thrust in an attempt to re-acquire the path before it slowed to this speed . When thrust was added, the FMA incorrectly indicated |THR|LNAV|VNAV PTH|. It should have indicated |THR|LNAV|VNAV SPD| because it was more than 150' below the path.
Member
Registered: Aug 2014
Posts: 375
Location: Camberley, UK
That is a very comprehensive manual that you have, Jon. I have never seen such detail and was certainly not aware of many of the nuances...

There seems to be two aspects to this, apart from the approach logic discussion, which I think we all agree on. They are whether we should see THR or SPD, and whether we should see VNAV SPD or VNAV PTH. One caveat is we should never see SPD-VNAV SPD!

Peter
« Last edit by Britjet on Sun, 02 Aug 2015 10:14:43 +0000. »
Member
Registered: Jun 2015
Posts: 103
Location: Tucson, AZ USA
Yes, but SPD-V/S will not cause the wings to fall off the airplane like some guys fear!

Maybe another aspect would be the amount of airspeed deviation from FMC target speed to stay on path.

Page: 1 2 >

Old 744 Forum is powered by UseBB 1 Forum Software