LPV DA NA - Can I fly this approach and to what minimums?

[azure]

Title says it all. The approach in question is the RNAV 17 at KMPV. The approach has an LPV DA, an LNAV/VNAV DA that is about 100 feet higher, and an LNAV MDA that is higher still. The LPV DA is NA by NOTAM due to trees close to the airport perimeter, evidently affecting clearance on the missed. Therefore, when flying this approach with a WAAS GPS (in my case, a CNX-80 (aka GNS-480)), the unit will annunciate LPV.

According to one opinion, claimed to be based on 91.175 (b), you are legally flying whatever the unit annunciates. Since the LPV DA is NA, you cannot fly the approach to that DA, but also, you cannot fly to the LNAV/VNAV DA since the unit will not annunciate LNAV/VNAV under these circumstances (WAAS GPS and no Baro-VNAV). In one version of this opinion, you can still fly to the LNAV MDA even with the unit annunciating LPV. In another, you cannot even do that unless you turn WAAS off to force annunciation of LNAV.

Another opinion says that you can fly to any higher DA or MDA that is still authorized.

I have no idea which opinion is correct. As a matter of safety, I should think the last opinion is correct unless the LNAV/VNAV glide path is different from that of the LPV approach.

In particular, I'm hoping that @John Collins will weigh in on this.

 

[Brad Z]

The navigator annunciates for whatever service it has available, based on integrity. You can fly a on lnav/vnav approach using WAAS integrity, just as you can opt to fly an approach as an LNAV approach even when LPV is an annunciated.

See page 6:
https://www.faa.gov/documentlibrary/media/advisory_circular/ac 90-107.pdf

 

[Captain Bubba]

You can fly LNAV/VNAV minimums with WAAS. Some where I have read the FAA's official statement on this, but I don't remember where. I think it's in the latest version of the AC that covers RNAV. It's probably on Bruce's website. Some 142 schools teach you to switch to baro, but the FAA doesn't require it.

 

[benyflyguy]

The glideslope isn’t inherently different it is just that the LPV is to much tighter tolerances. You can fly with LPV annunciated to the LNAV/VNAV DA. Apparently that 100 feet lower puts u too close to maintain obstacle clearance hence three notam.
The LPV glideslope just gets tighter the closer you get.I try to think of LPV flying into a cylinder and LNAV/VNAV flying into a long cylinder

 

[Captain Bubba]

AC 90-107, Para 4(n), and Para 6(b)

 

[Clip4]

Look in your manual for the GPS there should be a method to defeat the WAAS if you desire to for the approach.

 

[RussR]

Look in your manual for the GPS there should be a method to defeat the WAAS if you desire to for the approach.

If you defeat WAAS, which I have done on occasion as a training scenario, the popular GPS units will no longer annunciate LNAV/VNAV either, defaulting to LNAV.

 

[RussR]

Title says it all. The approach in question is the RNAV 17 at KMPV. The approach has an LPV DA, an LNAV/VNAV DA that is about 100 feet higher, and an LNAV MDA that is higher still. The LPV DA is NA by NOTAM due to trees close to the airport perimeter, evidently affecting clearance on the missed. Therefore, when flying this approach with a WAAS GPS (in my case, a CNX-80 (aka GNS-480)), the unit will annunciate LPV.

According to one opinion, claimed to be based on 91.175 (b), you are legally flying whatever the unit annunciates. Since the LPV DA is NA, you cannot fly the approach to that DA, but also, you cannot fly to the LNAV/VNAV DA since the unit will not annunciate LNAV/VNAV under these circumstances (WAAS GPS and no Baro-VNAV). In one version of this opinion, you can still fly to the LNAV MDA even with the unit annunciating LPV. In another, you cannot even do that unless you turn WAAS off to force annunciation of LNAV.

Another opinion says that you can fly to any higher DA or MDA that is still authorized.

I have no idea which opinion is correct. As a matter of safety, I should think the last opinion is correct unless the LNAV/VNAV glide path is different from that of the LPV approach.

In particular, I'm hoping that @John Collins will weigh in on this.

We really do try hard to cripple our use of aviation, don't we? Please nobody submit this to the Chief Counsel, nothing good can come of that.

I don't believe you will find any definitive guidance on this. So we are left to determine what is "safe" to do.

From a TERPS perspective, a glideslope is a glideslope. The underlying surfaces are different between LPV and LNAV/VNAV, but that's because LNAV/VNAV was designed for Baro-VNAV systems, not WAAS systems. We are allowed to fly L/V using WAAS because it's not affected by those temperature and pressure errors inherent in Baro-VNAV systems. In other words, WAAS is better than Baro-VNAV, and does a better job of keeping us on course and glideslope.

If the obstacle in this case only affects the LPV, making it NA, then there is absolutely no safety reason to not fly the LNAV/VNAV line. You are flying the same glideslope, just going missed at a higher altitude. If the obstacle affected the LNAV/VNAV too, the DA would be adjusted or the LNAV/VNAV line made NA too.

If I came across this situation, I know what I'd do, without hesitation. I'd fly the LNAV/VNAV, land and go about my life.

 

[dtuuri]

If the obstacle in this case only affects the LPV, making it NA, then there is absolutely no safety reason to not fly the LNAV/VNAV line. You are flying the same glideslope, just going missed at a higher altitude. If the obstacle affected the LNAV/VNAV too, the DA would be adjusted or the LNAV/VNAV line made NA too.

Isn't LNAV/VNAV linear deviation, i.e., a different obstacle surface? Could navigating laterally via angular LPV possibly place you outside of LNAV/VNAV protection?

 

[azure]

So far the answers all echo my own thinking about this... but I have a very firm answer to the contrary from someone who claims to be very confident (a respected CFII, whom I won't name because he no longer posts here), and who discussed this with Garmin earlier today and came away even more convinced.

@Captain Bubba, thanks for the AC reference. But this doesn't answer the question of whether it's legal to fly to L/V minimums when L/V is not annunciated on the instrument.

@Clip4 and @RussR, the purpose of defeating WAAS would be to force an LNAV annunciation. Of course the unit would not annunciate L/V.

@RussR, what you're saying is exactly my thinking... assuming there is no difference in glidepath *angle* between the LPV glidepath and the L/V one. Assuming that (and that's the main thing I was unsure of), I'd have no concerns as regards safety with breaking off at the L/V DA.

Whether it would be legal or not is another matter.

 

[azure]

Isn't LNAV/VNAV linear deviation, i.e., a different obstacle surface? Could navigating laterally via angular LPV possibly place you outside of LNAV/VNAV protection?

Well now, again, that's what I'm not sure of. I *thought* the two glidepaths had the same angle but different vertical thicknesses - the L/V being thicker and thus less precise. But I'm not certain of that.

 

[Clip4]

There no restriction against using the LPV angle of descent to LNAV minimums. That’s the best you can do.

 

[azure]

There no restriction against using the LPV angle of descent to LNAV minimums. That’s the best you can do.

So you cannot follow the LPV GS to the LNAV/VNAV DA? Why not?

 

[Clip4]

I will respectfully point out that MDA is not the only factors in this approach. The RNAV has a flight visibility of 3/4 mile and the LNAV VNAV 1-1/4 and you are only getting 56 ft lower. In the big scheme of things it not much.

The LPV also has a lower flight visibility (3/4) than the LNAV VNAV.

 

[azure]

I will respectfully point out that MDA is not the only factors in this approach. The RNAV has a flight visibility of 3/4 mile and the LNAV VNAV 1-1/4 and you are only getting 56 ft lower. In the big scheme of things it not much.

The LPV also has a lower flight visibility (3/4) than the LNAV VNAV.

I know - but it is still something that could make the difference between getting in and not. And I'm interested in the principle here as well - there seems to be a rash of such new NOTAMs in the area lately, and with some of these approaches LPV may confer a greater advantage.

And tbh, personally, if visibility is the limiting factor rather than ceiling, I'm not likely to be out flying.

 

[midlifeflyer]

I'm getting a sense of deja vu. @RussR, this question on the other forum was the basis for my field test question on FB. The nay-saying CFI was also convinced ADC was required until calling Garmin.

 

[RussR]

So far the answers all echo my own thinking about this... but I have a very firm answer to the contrary from someone who claims to be very confident (a respected CFII, whom I won't name because he no longer posts here), and who discussed this with Garmin earlier today and came away even more convinced.

I will assume by the tone and content of your comment that you are talking about Ron Levy. In which case realize you will always get a firm answer, regardless of the correctness of his answer. And "claims to be very confident" is exactly correct. Also, I will add that Garmin hardly sets national policy or regulation. It makes no more sense to call them, than to call Piper to see if you need a CFI along to get night current.

@RussR, what you're saying is exactly my thinking... assuming there is no difference in glidepath *angle* between the LPV glidepath and the L/V one. Assuming that (and that's the main thing I was unsure of), I'd have no concerns as regards safety with breaking off at the L/V DA.

There is no difference. 3 degrees is 3 degrees. The formula for plotting this glidepath in space is the same. Note that this was not always the case, about 15 years ago there were as many as 3 different formulas, although the differences were inconsequentially small.

Isn't LNAV/VNAV linear deviation, i.e., a different obstacle surface?

The obstacle clearance surfaces ARE different, but LNAV/VNAV offers MORE protection than LPV. Which means, using LPV to fly LNAV/VNAV, you are safer.

Could navigating laterally via angular LPV possibly place you outside of LNAV/VNAV protection?

No. Laterally, the LNAV/VNAV protection is much wider than LPV. LNAV/VNAV primary area width is a (mostly) constant 0.6 nm either side of centerline. LPV is much tighter, exactly like an ILS.

 

[dtuuri]

Laterally, the LNAV/VNAV protection is much wider than LPV. LNAV/VNAV primary area width is a (mostly) constant 0.6 nm either side of centerline. LPV is much tighter, exactly like an ILS.

Ah, I was thinking in terms of .3 nm CDI sensitivity. Got it now.

 

[azure]

I will assume by the tone and content of your comment that you are talking about Ron Levy. In which case realize you will always get a firm answer, regardless of the correctness of his answer. And "claims to be very confident" is exactly correct. Also, I will add that Garmin hardly sets national policy or regulation. It makes no more sense to call them, than to call Piper to see if you need a CFI along to get night current.

Since you have mentioned the name, I will simply confirm your assumption.

There is no difference. 3 degrees is 3 degrees. The formula for plotting this glidepath in space is the same. Note that this was not always the case, about 15 years ago there were as many as 3 different formulas, although the differences were inconsequentially small.

Thank you, that answers my question as regards safety...

The obstacle clearance surfaces ARE different, but LNAV/VNAV offers MORE protection than LPV. Which means, using LPV to fly LNAV/VNAV, you are safer.

... and this confirms my understanding of the relationship between the two glidepaths. Identical angles, tighter guidance on the LPV.

It seems the only question remaining is the legality. I posed this question also on the GNS-480 users group, and received a private communication from a member who pointed me to AC 90-107. I'll copy and paste from a post I made on the other forum Mark referred to:

From paragraph 4 ( n ) : "Lateral Navigation/Vertical Navigation (LNAV/VNAV). An RNAV function that computes, displays, and provides both horizontal and approved vertical approach navigation. Both WAAS vertical guidance and baro-VNAV support approaches to LNAV/VNAV lines of minima." (bolding mine)

From paragraph 6 (b): "WAAS Overview. WAAS improves the accuracy, integrity, availability and continuity of GPS signals. Additionally, the WAAS geostationary satellites provide ranging sources to supplement the GPS signals. If there are no airworthiness limitations on other installed navigation equipment, WAAS avionics enable aircraft navigation during all phases of flight from takeoff through vertically guided approaches and guided missed approaches. WAAS avionics with an appropriate airworthiness approval can enable aircraft to fly to the LPV, LP, LNAV/VNAV and LNAV lines of minima on RNAV (GPS) approaches. One of the major improvements WAAS provides is the ability to generate glidepath guidance independent of ground equipment. Temperature and pressure extremes do not affect WAAS vertical guidance unlike when baro-VNAV is used to fly to LNAV/VNAV line of minima. However, like most other navigation services, the WAAS network has service volume limits, and some airports on the fringe of WAAS coverage may experience reduced availability of WAAS vertical guidance. When a pilot selects an approach procedure, WAAS avionics display the best level of service supported by the combination of the WAAS signal-in-space, the aircraft avionics, and the selected RNAV (GPS) instrument approach.

NOTE:

This AC does not preclude the use of WAAS receivers to fly to LNAV and LNAV/VNAV lines of minima." (This last note is even bolded in the AC.)

Given that all this is from the FAA's own publication, I'm having a really hard time understanding the legal objection as well.

 

[RussR]

Ah, I was thinking in terms of .3 nm CDI sensitivity. Got it now.

The width is +/-0.6 nm (2x the CDI sensitivity) BECAUSE the CDI width is +/- 0.3 nm. Or may vice-versa (chicken vs egg). But they are directly related.

 

[sarangan]

Title says it all. The approach in question is the RNAV 17 at KMPV. The approach has an LPV DA, an LNAV/VNAV DA that is about 100 feet higher, and an LNAV MDA that is higher still. The LPV DA is NA by NOTAM due to trees close to the airport perimeter, evidently affecting clearance on the missed. Therefore, when flying this approach with a WAAS GPS (in my case, a CNX-80 (aka GNS-480)), the unit will annunciate LPV.

According to one opinion, claimed to be based on 91.175 (b), you are legally flying whatever the unit annunciates. Since the LPV DA is NA, you cannot fly the approach to that DA, but also, you cannot fly to the LNAV/VNAV DA since the unit will not annunciate LNAV/VNAV under these circumstances (WAAS GPS and no Baro-VNAV). In one version of this opinion, you can still fly to the LNAV MDA even with the unit annunciating LPV. In another, you cannot even do that unless you turn WAAS off to force annunciation of LNAV.

Another opinion says that you can fly to any higher DA or MDA that is still authorized.

I have no idea which opinion is correct. As a matter of safety, I should think the last opinion is correct unless the LNAV/VNAV glide path is different from that of the LPV approach.

In particular, I'm hoping that @John Collins will weigh in on this.

Sounds to me like the NOTAM should have raised the DA instead of making it NA.

 

[RussR]

Sounds to me like the NOTAM should have raised the DA instead of making it NA.

That is the usual approach for new obstacles, of course.

However, the most common reason for a vertically-guided line of minima being NOTAMed NA is due to a penetration of what is called the "VGS", Vertical Guidance Surface. It is a surface that extends from the runway to the DA point. If this surface is penetrated, a vertically-guided line of minima is not allowed. The most common penetrators of this surface are trees as they grow.

Now, typically, this affects both the LPV and LNAV/VNAV lines. So you would normally expect to see them both NA.

However, due to the particulars of the shape of this surface, it is possible that one line is affected but not the other. Since the VGS extends out to the DA point, a line of minima with a higher DA kind of "stretches" the surface out - it gets longer but actually narrower at equivalent distances from the runway. There is a difference of just a few feet in width typically between the shape of one with a lower DA (wider) and one with a higher DA (narrower). If an obstacle (like a tree) happens to fall within this narrow band, it could affect the LPV (making it NA) but not the LNAV/VNAV.

I suspect that's the case here.

 

[John Collins]

If LPV is annunciated for this approach, it means that the lateral integrity is 40 meters or better and the vertical integrity is 50 meters or better. The LNAV/VNAV requires the lateral integrity to be 556 meters or better and the vertical integrity to be the same as an LPV 250 ft or higher DH, IOW 50 meters or better. An LNAV only requires that the lateral integrity be 556 meters or better. So if LPV is annunciated on the GPS for this approach, it authorizes any procedure on the chart that is not NOTAM'ed as being NA. The annunciation of LPV authorizes LPV or LNAV/VNAV or LNAV. There is no logic for it to mean anything else. If the position that is being proposed is that if the annunciation is LPV, you are only authorized to use the LPV minimums, and that it must annunciate L/VNAV in order to fly the procedure to the LNAV/VNAV DA, and that you must have it annunciate LNAV to use any of the LNAV MDA, then this procedure can not be used for circling or LNAV straight in or LNAV/VNAV if LPV is annunciated. This is patently ridiculous. All the annunciation indicates is the "most accurate approach where the alert limit(s) are predicted to be supported" or IOW integrity service level. The integrity service level is a hierarchy from most accurate to least accurate in the following order [LPV then LNAV/VNAV] for vertically guided procedures and [LP then LNAV] for procedures without vertical guidance. Regardless of the annunciation, if WAAS is providing the vertical guidance, the path is the same. If the database contains support for an LPV, the FAS data block is used to establish both the lateral and vertical final approach segment guidance.


From RTCA DO 229D, the WAAS GPS TSO specification, this is what is specified, emphasis is mine:

2.2.5.2.4 Selection of the Approach Type

The equipment shall provide a means to select which type of approach will be conducted (LPV or LP, LNAV/VNAV or LNAV). This selection may be manual or automatic.

For automatic selection, the equipment shall select either Approach (LPV or LP), Approach (LNAV/VNAV), or Approach (LNAV) when entering approach mode. The automatically-selected approach type shall be the most accurate approach where the alert limit(s) are predicted to be supported, and where a minimum is published for the selected procedure. The order of precedence is LPV or LP, LNAV/VNAV, then LNAV. If LPV or LP is both published and predicted to be available, the equipment shall indicate that it is available. If LPV or LP is published and is not predicted to be available, the equipment shall indicate that it is not available and shall indicate the approach type that is available (e.g., “LPV not available – Use LNAV/VNAV minima”). A prediction for Approach (LNAV) is not necessary. Once annunciated, the equipment shall not change from Approach (LPV) to Approach (LNAV/VNAV) or from Approach (LP) to Approach (LNAV) unless the approach is reselected or the pilot selects a different approach type.

 

[azure]

If LPV is annunciated for this approach, it means that the lateral integrity is 40 meters or better and the vertical integrity is 50 meters or better. The LNAV/VNAV requires the lateral integrity to be 556 meters or better and the vertical integrity to be the same as an LPV 250 ft or higher DH, IOW 50 meters or better. An LNAV only requires that the lateral integrity be 556 meters or better. So if LPV is annunciated on the GPS for this approach, it authorizes any procedure on the chart that is not NOTAM'ed as being NA. The annunciation of LPV authorizes LPV or LNAV/VNAV or LNAV. There is no logic for it to mean anything else. If the position that is being proposed is that if the annunciation is LPV, you are only authorized to use the LPV minimums, and that it must annunciate L/VNAV in order to fly the procedure to the LNAV/VNAV DA, and that you must have it annunciate LNAV to use any of the LNAV MDA, then this procedure can not be used for circling or LNAV straight in or LNAV/VNAV if LPV is annunciated. This is patently ridiculous. All the annunciation indicates is the "most accurate approach where the alert limit(s) are predicted to be supported" or IOW integrity service level. The integrity service level is a hierarchy from most accurate to least accurate in the following order [LPV then LNAV/VNAV] for vertically guided procedures and [LP then LNAV] for procedures without vertical guidance. Regardless of the annunciation, if WAAS is providing the vertical guidance, the path is the same. If the database contains support for an LPV, the FAS data block is used to establish both the lateral and vertical final approach segment guidance.


From RTCA DO 229D, the WAAS GPS TSO specification, this is what is specified, emphasis is mine:

Thank you. I believe this settles the question. Given (as you confirmed) that the path is the same for LPV and LNAV/VNAV under WAAS, there is no basis in safety for any more restrictive interpretation.

Just as an aside, I'm curious that this TSO seems to allow the GPS to downgrade the annunciated approach from LPV to LNAV/VNAV. I'm fairly certain that I've read (in the Pilot's Guide?) that if forced to default to a lower service level because of integrity issues, the CNX-80 / GNS-480 will always downgrade to LNAV. I don't have time at the moment to find the specific reference, but if this is true, I wonder why the unit was designed this way? Did the design predate that particular TSO? (I *think* this unit is certified to conform to the C146a TSO, not sure if that is different from RTCA DO 229D or the same thing.)

 

[John Collins]

The CNX-80/GNS 480 were approved to an earlier version of the TSO, TSO C146a RTCA DO-229C. I believe TSO C146e, RTCA DO 229-E are the most current versions. There is not that much difference to the various TSO versions.

The GNX 175, GNC 355, and GPS 175 are the latest versions of the TSO. The TSO allows for fail down to LNAV/VNAV but as a practical matter, this does not make much sense because of geometry. The vertical integrity is generally worse than the horizontal integrity, usually by a factor of an additional 50% due to the fact that GPS satellites are available over the entire sky to determine for best geometry for a lateral position, but only over half of the sky for the vertical since those satellites that could be used for the vertical are low or below the horizon and not in view (best scenario is a satellite low to the horizon vs those overhead or high above the horizon to obtain a good vertical position). So it is way way more likely that the vertical integrity for LPV will fail to meet requirement than the lateral integrity will fail to meet requirements yet the vertical still meet requirements. Lateral is 40 meters, vertical is 50 (or 35 meters LPV200) for LPV, so if lateral is > 40 meters, vertical is likely to be 60 meters or so. With LNAV/VNAV, lateral integrity allows for an enormous at 556 meters, but vertical is still 50 meters, so LNAV/VNAV is not very likely to be an option. On the other hand, LNAV only requires 556 meters without any vertical requirement, so the designers decided not to waste their effort on programming for a rare if ever possibility and just programmed for the straight forward downgrade to LNAV without vertical guidance.

 

[midlifeflyer]

I'm going to say this about Ron. I like and respect him even though we disagree from time to time. He is very knowledgeable. I don't like to see him insulted or trashed (especially in absentia). But I think we have one basic differences. I tend to look at a problem and think, "how can comfortably I say "yes" (safely and within the regulatory world). He tends looks for the "no."

 

[MauleSkinner]

The CNX-80/GNS 480 were approved to an earlier version of the TSO, TSO C146a RTCA DO-229C. I believe TSO C146e, RTCA DO 229-E are the most current versions. There is not that much difference to the various TSO versions.

The GNX 175, GNC 355, and GPS 175 are the latest versions of the TSO. The TSO allows for fail down to LNAV/VNAV but as a practical matter, this does not make much sense because of geometry. The vertical integrity is generally worse than the horizontal integrity, usually by a factor of an additional 50% due to the fact that GPS satellites are available over the entire sky to determine for best geometry for a lateral position, but only over half of the sky for the vertical since those satellites that could be used for the vertical are low or below the horizon and not in view (best scenario is a satellite low to the horizon vs those overhead or high above the horizon to obtain a good vertical position). So it is way way more likely that the vertical integrity for LPV will fail to meet requirement than the lateral integrity will fail to meet requirements yet the vertical still meet requirements. Lateral is 40 meters, vertical is 50 (or 35 meters LPV200) for LPV, so if lateral is > 40 meters, vertical is likely to be 60 meters or so. With LNAV/VNAV, lateral integrity allows for an enormous at 556 meters, but vertical is still 50 meters, so LNAV/VNAV is not very likely to be an option. On the other hand, LNAV only requires 556 meters without any vertical requirement, so the designers decided not to waste their effort on programming for a rare if ever possibility and just programmed for the straight forward downgrade to LNAV without vertical guidance.

Is there any way for a system that requires WAAS for LNAV/VNAV to downgrade itself from LPV to LNAV/VNAV? Seems like that would be primarily for WAAS signal issues, in which case the system would have to be capable of Baro VNAV to downgrade to LNAV/VNAV instead of LNAV.

 

[John Collins]

I don't think there is much use for downgrade from LPV to LNAV/VNAV using WAAS for vertical guidance. There are some LNAV/VNAV that are charted with the note "WAAS VNAV not authorized", example TJSJ RNAV (GPS) Rwy 10 in Puerto Rico. I would assume, but don't know for sure that some high end FMS systems that support LPV and Baro/VNAV for the vertical on an LNAV/VNAV could downgrade from LPV using WAAS vertical to LNAV/VNAV using Baro/VNAV for the vertical. Any approach that charts both LPV and LNAV/VNAV may be flown with WAAS based vertical guidance if the GPS system supports these approach types and integrity is satisfactory. The WAAS vertical guidance is based on the GP angle and the vertical CDI FSD is +/- 25% of the GP angle. At distances greater than 150 meters/tan(GP angle/4) the FSD is fixed at +/- 150 meters. This occurs at a point rough
ly 6 NM from the threshold.

 

[Rcmutz]

There no restriction against using the LPV angle of descent to LNAV minimums. That’s the best you can do.

thats what I would do. Fly the LPV, but to the LNAV/VNAV minimums.

 

[azure]

[...] The TSO allows for fail down to LNAV/VNAV but as a practical matter, this does not make much sense because of geometry. The vertical integrity is generally worse than the horizontal integrity, usually by a factor of an additional 50% due to the fact that GPS satellites are available over the entire sky to determine for best geometry for a lateral position, but only over half of the sky for the vertical since those satellites that could be used for the vertical are low or below the horizon and not in view (best scenario is a satellite low to the horizon vs those overhead or high above the horizon to obtain a good vertical position). So it is way way more likely that the vertical integrity for LPV will fail to meet requirement than the lateral integrity will fail to meet requirements yet the vertical still meet requirements. Lateral is 40 meters, vertical is 50 (or 35 meters LPV200) for LPV, so if lateral is > 40 meters, vertical is likely to be 60 meters or so. With LNAV/VNAV, lateral integrity allows for an enormous at 556 meters, but vertical is still 50 meters, so LNAV/VNAV is not very likely to be an option. On the other hand, LNAV only requires 556 meters without any vertical requirement, so the designers decided not to waste their effort on programming for a rare if ever possibility and just programmed for the straight forward downgrade to LNAV without vertical guidance.

This makes perfect sense, thank you again.

 

[aterpster]

KISS principle:

NOTAM:

!FDC 1/5062 MPV IAP EDWARD F KNAPP STATE, BARRE/MONTPELIER, VT. RNAV (GPS) RWY 17, AMDT 1A... LPV DA NA ALL CATS. 2108111311-2308111311EST

The NOTAM speaks only to the LPV DA, not the LPV GS.

 

[midlifeflyer]

KISS principle:

NOTAM:

!FDC 1/5062 MPV IAP EDWARD F KNAPP STATE, BARRE/MONTPELIER, VT. RNAV (GPS) RWY 17, AMDT 1A... LPV DA NA ALL CATS. 2108111311-2308111311EST

The NOTAM speaks only to the LPV DA, not the LPV GS.

BINGO.

 

[midlifeflyer]

Can I toss in a twist? Here's the minimums box for the RNAV 21 at my home base, KTTA. I will add that whether Garmin or Avidyne, the GPS annunciates LPV, not LNAV/VNAV.
@aterpster or @RussR, the "can you fly to the lower minimums" question aside, can you explain how the differences in design requirements lead to this? It's way above my pay grade.

 

[azure]

On an approach with both LPV and LNAV/VNAV sets of minimums, why might the LNAV/VNAV DA be lower than the LPV DA? Ron says he is aware of such an approach, but I have no more details.

I know there are cases where the LNAV MDA can be lower than the corresponding APV DA, because the obstacle clearance requirements are different for a non-precision approach vs. a vertically guided one. But LNAV/VNAV and LPV are both vertically guided approach types. I know the OCSes are different for the two types, but I'm trying to imagine a geometry where the same procedure for defining the DA gets you lower when the APV's OCS is lower, as I think it is for the LNAV/VNAV as compared with LPV (i.e., the vertical guidance is looser for LNAV/VNAV and thus the OCS is more easily penetrated) and I'm coming up short. An obstruction that penetrates the LPV OCS must certainly penetrate the LNAV/VNAV OCS unless I'm missing something.

ETA: Mark and I cross-posted and Mark just came up with an example of exactly this sort of case!

 

[azure]

An obstruction that penetrates the LPV OCS must certainly penetrate the LNAV/VNAV OCS unless I'm missing something.

Now I'm not so sure. This article from IFR magazine discusses the differences between the LNAV OCS and the LNAV/VNAV OCS and the different ways that the MDA and DA are determined. Apparently the LNAV/VNAV OCS has a flat part inside the DA. The LPV OCS is a sloping surface that is offset from the runway threshold I believe, but unfortunately the article doesn't show what it looks like in comparison with the LNAV/VNAV OCS.

As I understand it, in either case the DA is determined by projecting a horizontal line at the obstacle height outward until it intersects the OCS, and then putting the DA directly above that point. In order for the LPV DA to be positioned further out than the LNAV/VNAV DA, it seems the LPV OCS would have to lie below the LNAV/VNAV OCS and thus LPV would provide poorer obstacle clearance than LNAV/VNAV up to the obstacle. Maybe I'm thinking about this wrong.

 

[RussR]

@azure @midlifeflyer
There was another recent thread where this issue was addressed. Funny, it actually got very little interest...
Only 9 posts - https://www.pilotsofamerica.com/community/threads/reno-gps-approach-question.134105/#post-3139851

This situation is rare but does happen, and in general terms it all has to do with the location of the obstacle and the differing evaluation methods between LPV and LNAV/VNAV. The underlying surfaces are somewhat complex and are dramatically different between the two - one main reason as you suspect being the effect that temperature can have on Baro-VNAV systems, which are what the LNAV/VNAV criteria is designed for. Without seeing the source documents and actually doing the evaluation of the different surfaces it's hard to say. But I agree that it's hard to picture how it could happen, and even knowing the surfaces involved pretty well I can't say I can see how it could happen either. I'd have to look at the specific example and plot it.

However, there is a possibility that IS very easy for me to understand how it could happen. It is possible that the accuracy of the "controlling obstacle" is having an effect. Briefly, for these evaluations, the worst-offending obstacle is identified for each type of procedure - meaning the one that causes the most increase in DA. But this initial evaluation is based on the surveyed obstacle height alone. Then after this "controlling obstacle" is determined, the accuracy of the survey data is looked at to see if it conforms to the requirement for that segment. If it does not, then the DA is further increased to account for this inaccuracy. So it is possible that the LPV obstacle causes less initial DA adjustment than the LNAV/VNAV obstacle, BUT has worse accuracy, and once this is applied, the LPV DA gets a further increase. This is definitely a quirk of the criteria.

The obstacle accuracy scenario comes up reasonably often with things like circling area evaluation. For example, the Cat C circling obstacle is 1000 MSL but has poor accuracy that requires a 50 ft adjustment. So the MDA would be 1000 + 50 + 300 (for obstacle clearance) = 1360 (rounded to next 20 ft). The Cat D obstacle, however, is taller at 1020 MSL but is accurately surveyed and requires no further adjustment. So Cat D MDA would be 1020 + 300 = 1320, even though the Cat D circling area includes the Cat C area. A lower Circling MDA for a higher Category is actually NOT prohibited, although it looks weird and likely isn't published that way too often for just that reason.

 

[azure]

Thanks @RussR . Still seems somewhat mysterious. One thing I note in this case, though, is that the difference is only 12 feet and thus makes no practical difference in how I'd fly the approach. The difference in the case in the other thread is more significant, roughly 50 feet. So the question is a valid one, I think: with LPV-level integrity as annunciated by the instrument, is it safe to fly to the lower LNAV/VNAV DA? If not, how can it be safe if using a generally more error-prone baro-VNAV system? If it's simply a "quirk of the criteria" then it should have no bearing on the pilot's choice of minima - the better the integrity, the lower the DA you can safely fly to. But if it's due to the different shapes of the OCSes and those shapes really are based on safety considerations predicated on different sources of vertical guidance, then maybe there is a valid reason to consider the higher DA as limiting when using WAAS vertical guidance.

ETA: thinking more about your "criteria quirk" possibility, I wonder now if that wouldn't actually be a flaw in the evaluation procedure. If the LPV obstacle has poor enough accuracy to push the LPV DA higher, then shouldn't the "error bars" in the initial LPV obstacle height force re-evaluation of the controlling obstacle for the LNAV/VNAV approach? Wouldn't you expect that obstacle with its adjusted height to often or even usually become the new controlling obstacle for the LNAV/VNAV?

 

[midlifeflyer]

If it's simply a "quirk of the criteria" then it should have no bearing on the pilot's choice of minima - the better the integrity, the lower the DA you can safely fly to. But if it's due to the different shapes of the OCSes and those shapes really are based on safety considerations predicated on different sources of vertical guidance, then maybe there is a valid reason to consider the higher DA as limiting when using WAAS vertical guidance.

...and I think that is the ultimate question.

the material you reviewed and cited all indicate that LNAV/VNAV mins can be used by WAAS GPS; that WAAS actually flies them more accurately than BARO VNAV. Makes it difficult to think safety is an issue. But...

There may actually be two "quirks" at work here. One is the quirk in the analysis which can lead to LNAV/VNAV having lower minimums. The other is the quirk in GPS systems which prefer LPV over LNAV/VNAV. So you get the interesting combination of a "better" and usable approach not being annunciated.

It's an interesting academic exercise, but definitely not one I would bring up on a checkride!

 

[azure]

...and I think that is the ultimate question.

the material you reviewed and cited all indicate that LNAV/VNAV mins can be used by WAAS GPS; that WAAS actually flies them more accurately than BARO VNAV. Makes it difficult to think safety is an issue. But...

There may actually be two "quirks" at work here. One is the quirk in the analysis which can lead to LNAV/VNAV having lower minimums. The other is the quirk in GPS systems which prefer LPV over LNAV/VNAV. So you get the interesting combination of a "better" and usable approach not being annunciated.

It's an interesting academic exercise, but definitely not one I would bring up on a checkride!

In the case of the RNAV 16L X at KRNO (in the thread linked to by @RussR), I think it's more than an academic exercise: if I was flying that approach and the ceiling was right in the 700 to 800 ft. range, I'd definitely want to know whether I can safely descend to the lower LNAV/VNAV DA using WAAS vertical guidance. If it's simply a "quirk" in the GPS, the answer would be yes, but I wonder if there is something else going on here. For example, is the baro-VNAV glide path angled differently, so that it arrives at the LNAV/VNAV DA at a safe distance from the obstacle, whatever it is? If that's the case, then you shouldn't attempt to descend that far using WAAS, and the LPV preference of our GPSes isn't a quirk at all, but in certain rare cases has a basis in safety considerations.

 

[RussR]

Thanks @RussRETA: thinking more about your "criteria quirk" possibility, I wonder now if that wouldn't actually be a flaw in the evaluation procedure. If the LPV obstacle has poor enough accuracy to push the LPV DA higher, then shouldn't the "error bars" in the initial LPV obstacle height force re-evaluation of the controlling obstacle for the LNAV/VNAV approach? Wouldn't you expect that obstacle with its adjusted height to often or even usually become the new controlling obstacle for the LNAV/VNAV?

You would think, but that's not how it's done in FAA TERPS. I believe the reasons are historical. By which I mean, if you're manually plotting each obstacle by hand, then it would possibly be very workload intensive to evaluate every single obstacle's height AND accuracy in order to determine the worst one. I mean, there could be hundreds of obstacles in a final segment (now, with LIDAR and other surveying techniques, it's not uncommon for there to be even more). So, back in the manual days, it was a simple expedient to be able to get the work done - find the tallest one (or worst one) FIRST, then apply any associated accuracy adjustments.

Now, of course, computers do most of the analysis, so the reason is no longer as compelling.

However, there is an exception. For RNAV (RNP) procedures, the accuracy of ALL obstacles is considered for determination of the controlling obstacle. This is also how it is done in USAF TERPS if I understand correctly. I do not know the "why" here, other than perhaps because RNAV (RNP) criteria was developed more recently than the others and were never really evaluated manually.