RNAV GNSS approaches

[jpwelton]

Hi all,

I understand that standard procedure for an RNAV approach is to be 'own nav'd' to the IAF, and that a period of level flight of 2nm is required. My questions are:
does this level flight have to be straight, or could it be in the turn?
Could you be sent to the final approach fix instead of the initial, or would the a/c FMS (or crew!) not like that?

Cheers

John
(curious ATCO!)

[Right Way Up]

Hi John,

I will look for the reference but I am pretty sure that EU-Ops predicates that you must never be sent direct to the final descent point on an RNAV approach. Therefore you will be given direct to the IAF or given vectors to a point at least 2nm from the final descent point.

[eckhard]

Hi John,

Here is an extract from an EU-OPS Airline's Operations Manual:

P-RNAV/RNP STARs and RNAV Approaches may utilise specific phrases which identify the method of intercept, navigation, climb or descent to be used. Details will be published on relevant approach charts. ATC must be advised of any deterioration or failure of navigation equipment below that required for entry into a scheduled navigation area or continued operation in such airspace.

Vectoring and Positioning
ATC tactical interventions in the terminal area may include radar headings, ‘direct to’ clearances which bypass the initial legs of an approach, interceptions of an initial or intermediate segments of an approach or the insertion of additional waypoints loaded from the data base. In complying with ATC instructions, the flight crew should be aware of the implications for the navigation system.

‘Direct to’ clearances may be accepted to the Intermediate Fix (IF) provided that it is clear to the crew that the aircraft will be established on the final approach track at least 2 miles from the FAF.

‘Direct to’ clearance to FAF is not acceptable. Modifying the procedure to intercept the final approach course prior to the FAF is acceptable for radar vectored arrivals or at other times with ATC approval.

The final approach trajectory should be intercepted no later than the FAF in order for the aircraft to be correctly established on the final approach course before starting the descent (to ensure terrain and obstacle clearance).

‘Direct to’ clearances to a fix that immediately precede an RF leg are not permitted.

For parallel offset operations enroute (in RNP 4), transitions to and from the offset track should maintain an intercept angle of between 30 and 45° unless specified otherwise by ATC.



So it appears that the 2nm segment that you refer to does not necessarily have to be level flight but should be aligned with the FAT.

As Right Way Up said, vectors should be given either to the IAF or to a point at least 2nm prior to the FAF.
Another option that I have seen is to be vectored onto an intercept heading and then told to intercept the final approach course. Again, this intercept point is at least 2nm prior to the FAF.

[underfire]

What if the entire procedure is coded from DEP to ARR? With these procedures, I am always curious how the clearances work.
or this one? (notice where the IAF's are located?)

http://i40.tinypic.com/14mys7k.jpg

[divinehover]

On the above procedure, you would be cleared to the IAF with a 'descend via' clearance. Follow the whole procedure, beginning to end. You can't be cleared to a waypoint which is the beginning of a RF Leg, which in this approach is just about the whole procedure.

[aterpster]

That approach is issued only to China Airlines. Also, there is very little traffic at that airport so there is not much in the way of ATC involvement once cleared for the approach. The crews also receive a lot of training before they are authorized that airport and approach.

[underfire]

Of course terpster, you are aware that I designed this?

[Capn Bloggs]

I designed this?
Arr, that explains it! :p

[aterpster]

underfire:

Of course terpster, you are aware that I designed this?

No, because I don't know who you are. Are you Steve Fulton?

[underfire]

Sorry, thought you were aware I had been with Naverus....

[aterpster]

underfire:

Sorry, thought you were aware I had been with Naverus....

I did not know that. It was a good company with a great CEO, pre GE. I don't know anything about the GE operation.

Steve did a great job for Alaska Airlines. I believe Alaska Airlines is/was anti-competitive making them all specials. But, that wasn't Steve's decision.

[B-HKD]

That approach is issued only to China Airlines. Also, there is very little traffic at that airport so there is not much in the way of ATC involvement once cleared for the approach. The crews also receive a lot of training before they are authorized that airport and approach.

Air China,

China Airlines is a Taiwan based operator.

[mi68guel]

FINAL APP MODES (A330 FCOM DSC-22_30-80-30-20 P 1/8)

The FINAL APP mode includes the following modes:
• APP NAV mode for lateral guidance
• FINAL mode for vertical guidance.

FINAL APP ARMING CONDITIONS

The flight crew arms APP NAV and FINAL modes by pressing the APPR pb on the FCU, provided all
of the following conditions are met:
• The aircraft is above 400 ft AGL
• The active flight plan is valid (lateral and vertical profile)
• A non-precision approach has been selected in the active flight plan
• GA mode is not engaged.
The FMA displays "FINAL" and "APP NAV" in blue.
If NAV mode was already engaged, APP NAV engages immediately.

FINAL APP ENGAGEMENT CONDITIONS

APP NAV and NAV modes engage under the same conditions:
If NAV mode was engaged, APP NAV engages immediately. If HDG/TRK is engaged, APP NAV
engages when the intercept conditions are met (the aircraft’s heading or track intercepts the flight
plan active leg).
APP NAV will engage if the "TO" waypoint is displayed in white on the ND and MCDU.
FINAL mode engages, if:
• FINAL is armed (by pressing APPR on the FCU), and
• The aircraft intercepts a descending leg of the vertical flight path, or
• In V/S (FPA) or OPEN DES, the aircraft intercepts a level-off segment of the vertical flight
• profile, with an FCU selected altitude different from this level-off segment.
• APP NAV is engaged,
• Predictions are available,
• The APPR phase is active, and the deceleration point has been sequenced, and
• Note: APPROACH phase may have to be manually activated on MCDU PERF page if the
• approach starts at high altitude above aerodrome level (i.e. RNAV(RNP) approaches).
• The aircraft is within the capture area of the vertical profile
Laterally: ±1.5 NM from the Final Descent Point.
Vertically: +150 ft above the Final Descent Point.
• A blue arrow is displayed on NDs to indicate where the FINAL APP engagement conditions are met
• and where the final descent will begin automatically.
• If the same arrow is shown in white, at least one engagement condition is not fulfilled, FINAL APP
• will not engage and the aircraft will not descend automatically.

Definition of the Final Descent Point (also called Vertical Intersection Point "VIP" for
RNAV(RNP) approaches)

The Final Descent Point is the capture point of the final descent segment coded in the Navigation
Database. It is the latest point where FINAL APP is supposed to be engaged. For RNAV(RNP)
approaches, this point may be indicated on the chart as "VIP".
This point is defined in the Navigation Database by:
• A constant vertical flight path beyond this point,
• A coded altitude constraint that may be "at" or "at or above" (e.g. +3 000 ft). This Constraint
is displayed on ND (in magenta), next to the corresponding waypoint, when the CSTR key is
selected on the EFIS Control Panel. It is also shown on the F-PLN page at this WPT.
Note: The Final Approach Fix (FAF) is the position from where the obstacle clearance is defined
relative to an Obstacle Clearance Surface (OCS). Obstacle clearance is only ensured if
the aircraft is flying on the defined vertical flight path. Before the FAF, minimum altitudes
are published with fixed minimum Obstacle Clearance (MOC). The Final Descent Point is
the point in the procedure at or before the FAF from which a constant vertical flight path is
defined.

[aterpster]

B-HKD:

Whatever the state airline is of mainland China. That wasn't really my point in any case. :)

[underfire]

miguel, ahhh yes, the infamous VIP....this was user preference (and seemed to cause more confusion than it was worth)

http://i.imgur.com/H6ikMnL.jpg

Yes, terpster, things were certainly much different pre-GE. I enjoyed the challenges with real RNP-AR, stuff with WestJet, PSP, SCC, all of the AUS procedures (with goofy waypoints) and especially the procedures in China, with the extreme altitudes.

In procedure design, just dabble with the real crazy stuff, like VNKT, CAT III autoland GBAS, offshore platform RNP, and RNP transition to GBAS final....

Currently working on several airports for Time Based Operations and wake turbulence mitigations.

What are you up to these days?

[aterpster]

underfire:

Yes, terpster, things were certainly much different pre-GE. I enjoyed the challenges with real RNP-AR, stuff with WestJet, PSP, SCC, all of the AUS procedures (with goofy waypoints) and especially the procedures in China, with the extreme altitudes.

In procedure design, just dabble with the real crazy stuff, like VNKT, CAT III autoland GBAS, offshore platform RNP, and RNP transition to GBAS final....

Currently working on several airports for Time Based Operations and wake turbulence mitigations.

Sounds great. I am not/was not a procedures designer. Rather an airline pilot, then retired airline pilot, who dug deep into TERPs from the beginning of my career.

Four years as a member of the PARC really helped my understanding of RNP AR. I am very familiar with VNKT. I thought that RNP AR was designed by Quovadis.

[chevvron]

Hi all,

I understand that standard procedure for an RNAV approach is to be 'own nav'd' to the IAF, and that a period of level flight of 2nm is required. My questions are:
does this level flight have to be straight, or could it be in the turn?
Could you be sent to the final approach fix instead of the initial, or would the a/c FMS (or crew!) not like that?

Cheers

John
(curious ATCO!)
Similar to an ILS to allow the aircraft to be stabilised on final approach track before commencing descent.

[underfire]

John (OP)

You have not commented on the various replies, so I am not certain if your question has been answered.

I am not aware of any constraint where there has to be a 2nm level section prior to the IAF. Some procedures are on the STAR, and the TOD is in the procedure. Turns before/after IAF can be included, as the IAF is simply a coded waypoint, but the procedure does not have to begin with an IAF, and there can be turns before the IAF, especially when coding from the STAR.

Its a computer, make the damn thing do what you want,

[Ian W]

Similar to an ILS to allow the aircraft to be stabilised on final approach track before commencing descent.

Not really true any more.
One of the problems with implementation of 'RNAV approaches' is that the simple thing for regulators was to overlay them on ILS procedures. Beancounters then asked what are we doing all this extra training for? At which point some airlines ceased equipping for RNAV approaches.

The future is more likely to be curved approaches (sometimes multiply curved approaches) where the aircraft is 'established' once it starts the procedure regardless of whether the aircraft is flying a straight track on the ground.

See https://www.faa.gov/nextgen/snapshots/stories/?slide=6

When 4DTBO eventually becomes accepted the trajectory is likely to include 'Greener Skies' type approach paths as part of a continuous descent.

[underfire]

The criteria simply states 30 seconds of stabilised flight, and that can include turns.
FROP can be at 500 feet, with a FAF at the beginning of the turn.

[aterpster]

Underfire:

The criteria simply states 30 seconds of stabilised flight, and that can include turns.
FROP can be at 500 feet, with a FAF at the beginning of the turn.

I can only speak to FAA public criteria for RNP AR. (I know Alaska Airlines has their own world of special, "fleet specific" performance-based criteria):

1. DA cannot be in a turn.
2. At least 15 seconds of flight from the FROP to the DA with a non-RNP missed approach, at least 50 seconds with a missed approach of less than RNP 1.0.

[underfire]

yes, the FAF can be at the beginning of a turn, with 30 seconds to the DA. The 30 seconds was for when the FAF was at the beginning of the turn. 50 seconds of stabilised flight before the DA, jeez.

We use 15 seconds of stabilised flight from FROP to DA, with the RNP missed. I do not subscribe to the common concept that on a 0.1/0.3 RNP, you go missed and the world falls apart to a 1.0 RNP (which half of the time gets the tower as an obstacle and the DA is unusable).

Many variables and if/then scenarios to consider.

[Goldenrivett]

Hi underfire,
I do not subscribe to the common concept that on a 0.1/0.3 RNP, you go missed and the world falls apart to a 1.0 RNP (which half of the time gets the tower as an obstacle and the DA is unusable).

How did we all cope when doing a GA from an ILS APP using only heading initially?

Unless your tower is outside the airfield boundary and close to the approach path, it won't be a problem. You commence the GA from a DA which is on the final approach and beyond the airfield boundary.

[Ian W]

yes, the FAF can be at the beginning of a turn, with 30 seconds to the DA. The 30 seconds was for when the FAF was at the beginning of the turn. 50 seconds of stabilised flight before the DA, jeez.

We use 15 seconds of stabilised flight from FROP to DA, with the RNP missed. I do not subscribe to the common concept that on a 0.1/0.3 RNP, you go missed and the world falls apart to a 1.0 RNP (which half of the time gets the tower as an obstacle and the DA is unusable).

Many variables and if/then scenarios to consider.
It would be nice to see the genesis of some of these grandfathered in restrictions. Just looking at the calculations made for RNP accuracy and comparing that to the calculations made when building procedures and the mismatch in assumptions is enough to raise significant concern about lack of joined up thought. Taking 'finger in the air' assumptions then using them in complex formulae with false precision inside tools like TARGETS doesn't help. Logic seems to have been abandoned.

[aterpster]

underfire:

yes, the FAF can be at the beginning of a turn, with 30 seconds to the DA. The 30 seconds was for when the FAF was at the beginning of the turn. 50 seconds of stabilised flight before the DA, jeez.

As you know, with an RNP missed approach (less than 1.0) the LNAV needs to be engaged. My FAA friends tell me the 50 seconds is required to accommodate some models of the AirBus. Beats me, I have no first-hand knowledge.

We use 15 seconds of stabilised flight from FROP to DA, with the RNP missed. I do not subscribe to the common concept that on a 0.1/0.3 RNP, you go missed and the world falls apart to a 1.0 RNP (which half of the time gets the tower as an obstacle and the DA is unusable).

I don't follow you. We have lots of RNP AR approaches with minimums of less than 0.30 that do not have RNP missed approaches. One example is KGUC RNP Rwy 24, where only 0.10 would work in the final segment, but then the terrain opens up for a conventional TERPs missed approach.

http://aeronav.faa.gov/d-tpp/1607/00517rr24.pdf

If a tower creates an issue, with FAA policy, then an RNP missed approach is used to avoid it. Or a climb gradient, or a combination of both.

[underfire]

I know you can have RNP approach without RNP missed, especially with the 1.0 RNP missed gathering so many obstacles, and a turn, well, just forget about that containment area...(Imagine Cuzco designed that way!)

I am not aware of any AB variants that require 50 seconds for the missed approach to engage. That was certainly not the case in any of the flight vals on the A320/A380 that I was on...

I was referencing the tailored RNP approaches that are for an airline and not public. In those, part of the service was to design the missed and the EO missed. In these, the same RNP level carries through the missed.
As you are aware, there is no FAA or ICAO criteria for EO procedures, so those are all custom.

I see that the RNP procedure into JFK has a DA in a turn! (FROP is 0.8nm from the TCH, 250HAT!)

http://i.imgur.com/BymPHg7.jpg

[aterpster]

underfire:

I know you can have RNP approach without RNP missed, especially with the 1.0 RNP missed gathering so many obstacles, and a turn, well, just forget about that containment area...(Imagine Cuzco designed that way!)
Cusco should have a line of minimums with a lower DA for the folks that have the performance. The density altitude is a problem for many, but not all, airframes.

I am not aware of any AB variants that require 50 seconds for the missed approach to engage. That was certainly not the case in any of the flight vals on the A320/A380 that I was on...
The way it was explained to me, there had to be at least 50 seconds of flight on a TF leg from the FROP to the MAP, or there was some type of issue reengaging LNAV for the miss. As I said before, I don't have actual knowledge of the issue. I do know it almost prevented a good design at KGUC Rwy 24.

I was referencing the tailored RNP approaches that are for an airline and not public. In those, part of the service was to design the missed and the EO missed. In these, the same RNP level carries through the missed.
As you are aware, there is no FAA or ICAO criteria for EO procedures, so those are all custom.
Lots of variants outside the U.S. I see Australia has EO procedures associated with their RNP AR approaches.

I see that the RNP procedure into JFK has a DA in a turn! (FROP is 0.8nm from the TCH, 250HAT!)
Indeed that one does, but it is a Jet Blue special, not a public procedure. Jet Blue had to do some demonstration flights to convince FAA's Flight Standards that Jet Blue's airframes and crews could safely perform that maneuver. Hopefully, someday the FAA will accept DA in the turn as public criteria. That Jet Blue procedure would be much more useful if everyone flying into JFK could do it. Rolling out at 250 HAT is something that did not make some folks at the FAA very happy. But, Jet Blue prevailed with the people that held the authority.
:ooh:

[underfire]

Lots of variants outside the U.S. I see Australia has EO procedures associated with their RNP AR approaches.

Of course. AUS was great to design for, instead of studying, the procedures just happened...Those started as the tailored designs for QF, then were made multi-variant to some extent. Most of the good EO missed went away, especially with the temp limitations, because of public outcry with "aircraft on fire" going over their house.

[OK465]

The JB special was accepted because JB was the first to get TOGA to Managed Nav capability on their 320s (equivalent to TOGA to LNAV on the NG).

You've got to have this going missed off an RF turn. Without it, when you TOGA (previously to heading or track) the FMS arrived at an average heading over some previous time increment on the RF which resulted in a turn reversal off the RF.

Did 100s of hours of DA in a Turn simulations directed at making the option available for public use. Happily I don't know where that stands now. :}

[aterpster]

OK465:

The JB special was accepted because JB was the first to get TOGA to Managed Nav capability on their 320s (equivalent to TOGA to LNAV on the NG).

Knowing virtually nothing about a 320, what is its behavior in this respect if it does not have Managed Nav?

[OK465]

The older software resulted in a lateral straight track mode becoming active for both FD & AP when the TOGA detent was selected in the 320/330 or 340 on the RF. The requirement with the DA in a Turn is to stay on the RF during the missed, the Nav selector had to be manually re-engaged (pushed in) to get correct missed steering commands for the FD/AP, by which time the FMGEC had computed the initial straight track in the opposite direction off the RF based on 'an average' of the previous continually changing tracks on the RF itself and supplied this to FD/AP. Unacceptable cross track errors were occurring prior to manually engaging the Nav mode for continuity on the RF in Managed Nav. Depending on where the DA is placed, I have seen the initial lateral steering commands on the 13R transition, if followed, point directly at JFK Tower on the missed. Managed Nav is LNAV equivalent.

Now the FD/AP commands in TOGA to Managed Nav are for staying on the RF to the MApp profile and tracking in Managed Nav on the missed. Coded track continuity maintained.

In the 738, the older software resulted in TOGA to heading, and the FMS provided to the FD/AP an 'average heading' over some previous time increment on the RF, resulting in an opposite direction turn off the RF. Hence updated software with TOGA to LNAV to avoid delay with re-engaging LNAV.

First time I saw this was way back in the 738 with a TOGA actuation on the left RF prior to the FROP performing the PSP procedure. On AP, aircraft reversed the turn direction to the right and headed for the Dino Martin memorial mountains. :eek:

[underfire]

Concur, when first looking at these, the B models wanted to go straight to the endpoint of the rf turn, rather than follow the turn. This was problematic at places, especially when the included angle was more than 90 degrees. Was a real issue at Kelowna. (did not have these issues with the A models)

Was that the Naverus PSP procedure or the later public procedure? I think it must have been the later, because with the original AA procedure, the 738 was too slick and would bust the final turn..

[underfire]

ahhh. okay...that makes sense.

It was fun, well, sort of. Dont remember a turn intercept, except in the sim when flight validating, not wanting to run the whole procedure again...

Tried for a few weeks to get the Smith box to take a RWY over 10K, PIN, offsets, anything, just would not do it. Then some variants of the 320 would not keep the descent profile, and would dive to the next waypoint altitude.

The really complicated procedures are still out there, especially in China (as shown by the Lhasa/Lhize type procedures.)

As shown by some of the threads on pprune, the drivers come up with some combinations/events that would be difficult to think of in the validations.

Currently, I am working on the RNP transition to GBAS final procedures, which brings back the old days with a bit of fun in the mix.

[aterpster]

ok465:
First time I saw this was way back in the 738 with a TOGA actuation on the left RF prior to the FROP performing the PSP procedure. On AP, aircraft reversed the turn direction to the right and headed for the Dino Martin memorial mountains.

AKA, Mother Sinatra memorial mountains.

Having said that, part of RNP AR training is not to go to TOGA until the MAP in the event of an early missed approach. Really no different than in any missed approach. Climb, but maintain final segment track to the MAP.

Some foreign countries are quite specific on the chart. At WAPP, for example, the chart states: "Missed approach transition to missed approach RNP for lateral guidance must not be initiated prior to the along track position of DA/H."

[underfire]

I looked at doing WAPP and WAMM, they balked at the cost of the survey for the ICAO obstacle area.

Who ended up doing these?

[aterpster]

underfire:

I looked at doing WAPP and WAMM, they balked at the cost of the survey for the ICAO obstacle area.

Who ended up doing these?

Don't know. But, they are public, both as to charting and database.