RNAV approaches

[Richard McDonald Woods 168]

I am trying to get to grips with using RNAV approaches, so will appreciate any corrections/feedback to my current understanding:

1. Selecting an RNAV approach provides only GPS quality (equivalent to category 1 ILS) LNAV.
2. As opposed to ILS VNAV/LNAV, current PMDG aircraft have no knowledge of WAAS signal receivers. The PMDG  GPS system is 'perfect' enough anyway!
3. RNAV approaches currently do not support vertical guidance, so no LPV (localizer  performance with vertical guidance) approaches are supported.
4. Vertical guidance whilst flying an RNAV approach is given only by setting QNH for the airport. So autoland is not supported.
5. The pilot must use the decision altitiude to determine the aircraft's descent to the runway. So accurate setting of local QNH for the approach is essential.

This leaves me with one main question - what criteria should I use to choose between an ILS and an RNAV approach to an airport? 

 

[berts 97]

First of all, it is normally considered good practise to use all of the available navigational aids at your disposal, even when you are conducting a visual approach. Therefore, unless you are on a training flight, or the approach is offset to the landing runway, or the ILS is U/S or not installed. then in the interests of flight safety I suggest it makes perfect sense to have the correct ILS frequency and inbound course  tuned whenever possible.

Non-ILS approaches, including RNAV approaches, are not as accurate when compared to those flown using an ILS, even allowing for any GPS updating.  In the B744 the maximum RNP error allowed for during FAA certification in order to carry out a RNP/GPS type of approach is 0.3nm or smaller and, because the 744's Autopilot/Flight Director system does not provide any VNAV path warning in the event of a vertical deviation, the autopilot must be disconnected at 360ft AAL (or higher depending on the published minima for that runway).  Provided there is GPS updating then the use of the autopilot with the ND in MAP mode is used; but great care is needed in certain parts of the world where the navigation aids are less reliable and/or accurate GPS updating has not been successful. 

I am not aware of any 744 operators that use anything other than QNH when operating at or below the Transition Altitude; but this has nothing whatsoever to do with being able carrying out an autoland or not.  Rather, the autopilots rely on height information obtained from the radio altimeters after glideslope capture and all the way down to the runway. 

Setting an accurate QNH (or QFE) is essential all of the time (i.e. day or night, VFR or IFR) and not just when on an approach. The vast majority of controlled flight into terrain accidents in the past have occured within 30 miles of the airfield and, almost without exception, the lack of any visible contact with the ground has been a significant factor. 

[iwebber 207]

1 hour ago, Richard McDonald Woods said:

Selecting an RNAV approach provides only GPS quality (equivalent to category 1 ILS) LNAV.

They're difficult to compare, the ILS gets more sensitive as you approach the runway whereas an RNAV has the same accuracy the whole way down. I would suggest they're much of a muchness > 5 miles out.

1 hour ago, Richard McDonald Woods said:

As opposed to ILS VNAV/LNAV, current PMDG aircraft have no knowledge of WAAS signal receivers. The PMDG  GPS system is 'perfect' enough anyway!

Yes but none of the airliners I've flown has any type of Augmentation System on and they were all able to do RNAV approaches to a better than Non Precision minima. RNAV isn't just GPS, it covers a whole gamut of navigational equipment.

1 hour ago, Richard McDonald Woods said:

RNAV approaches currently do not support vertical guidance, so no LPV (localizer  performance with vertical guidance) approaches are supported.

They can encode a 'glideslope' into the box and it's shown in the flight deck in a similar fashion to an ILS glideslope. The autopilot can follow it too.

1 hour ago, Richard McDonald Woods said:

Vertical guidance whilst flying an RNAV approach is given only by setting QNH for the airport. So autoland is not supported.

Yes and yes but they're unrelated. The Capt's baro is used for the alt restrictions (at least on the aircraft I've flown) so you need to make sure it's right. Autoland is not supported because the accuracy of the system isn't sufficient to put the aircraft on the end of the runway, especially in the vertical profile, as you said, it's dependant on the Capts baro (setting and working).

1 hour ago, Richard McDonald Woods said:

The pilot must use the decision altitiude to determine the aircraft's descent to the runway. So accurate setting of local QNH for the approach is essential.

This is true of most approaches, only some peculiar CAT1 approaches, Cat II and Cat III use Decision Height (DH) and thus are independent of baro. Everything else uses Minimum Descent Altitude (MDA). In about 15 years flying commercially I've used a DH about 10 times.

1 hour ago, Richard McDonald Woods said:

This leaves me with one main question - what criteria should I use to choose between an ILS and an RNAV approach to an airport?

Choose the one you're most comfortable with using your knowledge and experience. Generally I'd choose an ILS as the act of flying one is easier, on the jumbos I fly, RNAV approachs have been shoehorned into the box and are a bit clumsey to execute, although very effective.

The main advantage of an RNAV approach is that it doesn't need to be a straight line or a continuous angle descent. You can design and fly an RNAV that guides you around hills and deposits you in the right place, facing the right direction at the right height and speed to take over and land. This can be very advantagous in mountainous areas or areas with a lot of airspace restrictions. In the past these approaches were made up of convoluted radio aids and bearing/distances that were a pain to fly, took up an awful lot of capacity and if you got them wrong, tended to kill everyone.

I've crossed with Bert but what he said too.....

Hope this helps,
Ian Webber

 

[scandinavian13 5,192]

2 hours ago, Richard McDonald Woods said:

This leaves me with one main question - what criteria should I use to choose between an ILS and an RNAV approach to an airport? 

For me?

1. Barring any real reasons otherwise, take what's assigned (ILS is available on 13L at JFK, but they won't use it unless they're really desperate, as it then forces LGA onto 13, which then forces TEB and EWR into airspace sharing and N90 just goes down the drain, traffic-flow-wise - situations like that will mean I pull up an RNAV overlay for a visual approach to 13L, or similar)
2. Is ILS available?
   Yes > Go with that.
   No > Go with RNAV with a fallback to something like a LOC, VOR, etc.

At all times: have some sort of supplemental information displayed as an aid, as Bertie mentioned.

[Richard McDonald Woods 168]

Thanks gentlemen that will help me a lot😉

[skelsey 765]

Not much to add other than to say that most operators will have a policy along the lines of "use the most accurate type of approach available", which generally means ILS, RNAV, VOR, NDB, visual, in that order of preference.

As Kyle says on occasion for noise or other air traffic management reasons ATC may be using an approach other than ILS even when one is installed, and in this case you're obviously going to fly what ATC give you unless there is some overriding reason not to.

Re: vertical guidance: as Ian and Bertie have explained, the issue with RNAV approaches is that the vertical profile (in the 744 at least, and most commercial jets) is synthesised in the box and driven entirely by the barometric altimeter. 

This means that it is essential to have the correct QNH set, and QFE is certainly not an option unless you have a special navigation database with the crossing heights on the procedure coded for QFE.

The other issue with this (in real life) is that temperature will have an effect on the true vertical path flown by the FMC when following a baro-VNAV glidepath. This is because the barometric altimeter is subject to temperature error - it is calibrated to read correctly in ISA conditions. However, when the OAT is colder than standard the altimeter will over-read - that is to say your true altitude will be lower than indicated. The aerodrome QNH value takes temperature in to account so the altimeter will read correctly on the ground, but the higher you are above the aerodrome the greater the error will become.

Geometrically this is a problem because although the 'end' of the glidepath (at the aerodrome) will be in the right place, at the FAF - several thousand feet higher - the aeroplane will actually be a lot lower than indicated. Not only does this compromise the terrain clearance at the FAF itself, it also means that the actual glidepath flown will be shallower than that depicted in the chart which will have implications for obstacle clearance in the final segment as a whole. This is in contrast to an ILS approach where the glideslope is a fixed path in space defined by a radio beam and therefore is unaffected by any of this.

For this reason, RNAV approaches will normally have a minimum temperature associated with them. As long as the temperature at the aerodrome is above the charted minimum temperature you are fine - you can fly the procedure as published, changing nothing in the box, and you will be protected.

However, if the temperature is lower than that on the chart then if you are going to fly the procedure you must apply corrections to all the relevant altitudes and therefore can only fly it to LNAV minima (rather than LNAV/VNAV).

This is because to fly to LNAV/VNAV minima you are not permitted to make any changes to the procedure in the box once you have loaded it - you must load it directly from the navigation database and check that it matches the chart.

In any event, if the procedure is not in the database you cannot 'cook your own' by entering the waypoints from the chart, for instance.

The good news is that temperature error is not currently simulated in FSX or P3D so this is largely academic! 

[JRBarrett 3,070]

6 hours ago, skelsey said:

However, if the temperature is lower than that on the chart then if you are going to fly the procedure you must apply corrections to all the relevant altitudes and therefore can only fly it to LNAV minima (rather than LNAV/VNAV).

I checked a few RNAV approach charts for various US airports, and most have a note specifying the minimum temperature for LNAV/VNAV. At San Diego, the note for the RNAV (GPS) Runway 09 approach states: “For uncompensated Baro-VNAV systems, LNAV/VNAV NA below 6°C (43°F) or above  47°C (117°F).”

Interestingly, at my home airport KELM, in the chilly northeast, the minimum temperature is -20 C (-4 F), I assume due to a more favorable terrain profile. At San Diego, even though the initial part of the approach is over the ocean, the final segment has the aircraft descending over the rapidly rising terrain of Point Loma.

Another curious thing: although temperature may restrict when LNAV/VNAV minima can be used, many RNAV approaches appear to have lower minima for a straight LNAV approach than for LNAV/VNAV in any case.

In the case of the runway 24 RNAV (GPS) at KELM, the LNAV/VNAV DA is 1870 feet with minimum visibility of 2 1/2 sm, while the LNAV MDA permits descent down to 1740 feet, with visibility of 1/2 sm (or RVR of 2400). 

Would this be because in an LNAV approach, the pilot is directly controlling/monitoring the descent path, rather than the autopilot?

[berts 97]

10 hours ago, JRBarrett said:

Would this be because in an LNAV approach, the pilot is directly controlling/monitoring the descent path, rather than the autopilot?

The simple answer is no.  In the case of the B744 it is recommended that all Non-ILS Instrument Approaches (i.e. non-precision, including RNP, VOR, NDB, RNAV visual) are flown using the autopilot. On every approach the pilots should continuously monitor the horizontal as well as the vertical descent profile and not simply rely on the autopilot(s) to control the descent path.  It is good airmanship to be prepared for the unexpected - just in case something does go wrong!

Simon has already mentioned the Final Approach Fix (FAF) in his earlier post. This position is particularly important on a non-ILS (i.e. non-precision) approach, because it is often the last opportunity a pilot has of ensuring their altimeter subscale reading is set correctly.  Assuming it is set correctly then the aircraft's altitude should agree with that shown on the published procedure.  However, if the altimeter reading disagrees by a significant amount then it could be potentially very dangerous to continue the approach, because you could be starting the final descent at a much lower (or higher) altitude than you think.  If you are still in cloud when this happens then there is a real risk of early ground contact, so the safest course of action will inevitably be to discontinue the approach and climb to at least the airfield safety altitiude before trying to determine the cause for the false altimeter reading.

Assuming the aircraft passes over the FAF at the correct height, then the stopwatch should be started and the appropriate ROD for the aircraft's groundspeed initiated.  The timings can then be used to adjust the ROD and determine when the Minimum Descent Altitude (MDA) or Decision Altitude (DA) is reached, so that at this point a safe landing can be accomplished visually with the aircraft correctly aligned horzontally to the runway and descending on the correct profile; otherwise a missed approach should be flown.

[downscc 1,327]

11 hours ago, JRBarrett said:

In the case of the runway 24 RNAV (GPS) at KELM, the LNAV/VNAV DA is 1870 feet with minimum visibility of 2 1/2 sm, while the LNAV MDA permits descent down to 1740 feet, with visibility of 1/2 sm (or RVR of 2400). 

Would this be because in an LNAV approach, the pilot is directly controlling/monitoring the descent path, rather than the autopilot?

Excellent question, and one that I had to have a look at the KELM RNAV 24 plate and do some searching.  An article that explains this well is https://www.boldmethod.com/learn-to-fly/navigation/when-lnav-minimums-are-lower-than-vnav-which-should-you-use/

The TERPS experts can explain this better perhaps, but it boils down to the differences between determining a DH vs MDA and in this case how obstacles will influence the outcome.  As mentioned by Bertie and discussed in the article the choice of which one to choose is part of good piloting, and one must consider the weather, equipment available and ones own piloting experience.  The LNAV MDA will have more pilot workload and if the weather is close to minimums I probably would not select the non precision LNAV over the LNAV/VNAV to get that extra 130 foot of ceiling if there were other options, unless I was familiar with that approach and the area.

Edited February 17, 2019 by downscc

[JRBarrett 3,070]

1 hour ago, downscc said:

Excellent question, and one that I had to have a look at the KELM RNAV 24 plate and do some searching.  An article that explains this well is https://www.boldmethod.com/learn-to-fly/navigation/when-lnav-minimums-are-lower-than-vnav-which-should-you-use/

The TERPS experts can explain this better perhaps, but it boils down to the differences between determining a DH vs MDA and in this case how obstacles will influence the outcome.  As mentioned by Bertie and discussed in the article the choice of which one to choose is part of good piloting, and one must consider the weather, equipment available and ones own piloting experience.  The LNAV MDA will have more pilot workload and if the weather is close to minimums I probably would not select the non precision LNAV over the LNAV/VNAV to get that extra 130 foot of ceiling if there were other options, unless I was familiar with that approach and the area.

 

[JRBarrett 3,070]

Great article Dan, thanks.

[berts 97]

4 hours ago, downscc said:

if the weather is close to minimums I probably would not select the non precision LNAV over the LNAV/VNAV to get that extra 130 foot of ceiling if there were other options, unless I was familiar with that approach and the area.

Good article and I tend to agree with you, Dan, although I think your Chief Pilot wouldn't be too happy if you had to divert because you didn't fly to the lowest published minima for a particular runway.  

Professional pilots are trained to operate to the lowest published minima for their aircraft type at any airport they fly into and this includes a sound knowledge of the airline's route structure as well as all normal and alternate airfields. Some airfields and approaches are more demanding than others and this can sometimes result in a pilot having to undergo a special briefing, a simulator detail and even a visit under supervision with a Training Captain before being allowed to operate there on their own.

[skelsey 765]

Agreed -- a very good article!

One thing which is worth noting is that the use of "LNAV/VNAV" in the minimums section of an approach chart is slightly misleading in that it refers to the "3D" aspect of the approach design, which is distinct from the autopilot modes LNAV and VNAV; in theory, if permitted by your airline, you could temperature-correct the altitudes for the FAF and MDA and fly the profile using LNAV and VNAV modes to the LNAV (2D) minimums. However, most airlines do not permit changing anything (including altitudes) in the FMC from the FAF onward and so this effectively rules that out.

In practice, in the 744 VNAV is only permitted where there is a published glidepath angle. If the temperature is below the minimum published on the chart AND LNAV minima are available then the approach may be flown in LNAV and V/S using temperature-corrected minima and a temperature-corrected vertical profile.

One other aspect I was reminded of when checking the books -- in the 747-400, the FMC RNP value defaults to RNP 0.5; therefore all RNAV approach procedures will require the RNP value to be manually amended to 0.3 on POS REF page 2 prior to the approach.

[downscc 1,327]

13 minutes ago, berts said:

Dan, although I think your Chief Pilot wouldn't be too happy if you had to divert because you didn't fly to the lowest published minima for a particular runway.  

I don't pretend to represent myself as a professional pilot thoroughly trained on a route before allowed to Captain that route.  My comments regarding familiarization with the approach and airport environment were based on my own experience.  In my four decades of flying in the GA environment, I've had many many experiences with approaches I've never seen before and airspace that is totally new and in this environment the familiar saying, 'There are old pilots and bold pilots but no old bold pilots,' has served me well.  To be sure I've flown many approaches to mins (LOL my first instrument cross country in a PA-28R Arrow ended as an ILS into Keesler AFB at minimums, coming in over the back bay it was the blackest darkness I've ever seen until over the approach lights, whew). As a rule, I tend to be conservative when assessing my own skills, and prefer to rely on those skills rather than luck.

[berts 97]

22 hours ago, downscc said:

As a rule, I tend to be conservative when assessing my own skills, and prefer to rely on those skills rather than luck.

That's great advice Dan - and having read a little of your aviation background in your other posts my comment about your Chief Pilot was meant as a joke.  Anyway, I agree we shouldn't take things for granted or rely on luck when we are flying.

I remember seeing a rather clever message on this subject some years ago.  It was in the form of a simple cartoon drawing with two single engine light aircraft flying directly towards each other,  They were at the same height and entering into the opposite edges of the same large cumulus cloud, so obviously the two pilots couldn't see each other's aircraft.  The caption underneath had them both saying to their passengers "It's alright, I've got an IMC rating!"  

Those two pilots might have had the skills to fly in cloud, but I think you will agree that if they had any hope of avoiding a collision they would definitiely have needed a lot of luck!  Reading about your PA-28R Arrow experience reminded me of that cartoon drawing, because it was another example of  'I learnt about flying from that' - so thanks!