Why GPS sensitivity?

[hais]

Isn't GPS accuracy in a small geographical area the same? Curious why we don't have 350 feet full scale in terminal environment for WAAS equipment? 

[midlifeflyer]

6 hours ago, hais said:

Isn't GPS accuracy in a small geographical area the same? Curious why we don't have 350 feet full scale in terminal environment for WAAS equipment? 

I'm making this up, but I think it's the same reason that some autopilot NAV and APR modes are different in terms of their lateral tolerances. The tighter tolerance isn't necessary when in the enroute/terminal environment and, if engaged, results in a lot of unnecessary back and forth movement.

[Jerry 5TJ]

The full-scale width of the GPS-derived localizer is set to 0.3 nmi at its most sensitive when in approach mode. That’s 1800’. 
A conventional ILS localizer has angular sensitivity that increases steadily as the distance to the runway decreases.   At the runway threshold the localizer full-scale deflection width is 700’ in total, 350’ to each side of the centerline.  

[hais]

6 hours ago, midlifeflyer said:

I'm making this up, but I think it's the same reason that some autopilot NAV and APR modes are different in terms of their lateral tolerances. The tighter tolerance isn't necessary when in the enroute/terminal environment and, if engaged, results in a lot of unnecessary back and forth movement.

That makes sense, but, for an LPV approach, what is the operational advantage of shaping the sensitivity? What is the downside of not having the same sensitivity, say, from IAF?

[midlifeflyer]

12 minutes ago, hais said:

That makes sense, but, for an LPV approach, what is the operational advantage of shaping the sensitivity? What is the downside of not having the same sensitivity, say, from IAF?

Beats me. I can't think of any major operational downside or upside offhand.. It may be as simple as, "program the unit to give them what is required for the phase of flight and don't overdo it." 

I said I was making it up  

[Vance Harral]

45 minutes ago, hais said:

That makes sense, but, for an LPV approach, what is the operational advantage of shaping the sensitivity? What is the downside of not having the same sensitivity, say, from IAF?

The operational advantage is not to you, the individual pilot; but to the TERPS staff who design the approaches, other pilots already used to flying them, and the overall integration and track record of the system as a whole.

There is a long history and body of work associated with high-precision approaches having angular rather than linear sensitivity (ILS technology was first tested in 1929).  TERPS staff know how to design such approaches using standards that have achieved unparalleled operational efficiency and safety, including Cat III autoland in 0/0 conditions.  There is no good reason to change all this just because GPS is "naturally" linear rather than angular.  It would likely introduce confusion and new problems in exchange for no benefit.

I'm not saying it's wrong to be miffed by this if you're new to the game and and a GPS "native".  But as with getting ATC vectors as headings rather than track, and consuming weather information with teletype-era abbreviations... everyone already in the game knows how to use these things, and there is no particular operational advantage to changing them.

In summary, why shouldn't a GPS-based "precision" approach have angular sensitivity just like an ILS?  What would be the operational advantage of having it work differently?

[Ragsf15e]

I believe the older non precision gps approaches did not change sensitivity as they got closer to the map.  I do think it’s easier to fly one using angular sensitivity.

[Vance Harral]

1 hour ago, Ragsf15e said:

I believe the older non precision gps approaches did not change sensitivity as they got closer to the map.

Yes, GPS LNAV approaches have fixed sensitivity from the FAF to the MAP.  These approaches are "older", I guess (I'm old enough that every GPS approach seems "new" to me...); but they're still in use everywhere obstacles or other factors preclude having an LPV approach.

The A-N/NDB/VOR transmitters around which non-precision approaches were originally developed are angular, so course width narrows as you get closer to them.  However, the practical behavior of the CDI needle on these approaches depends on where the transmitter is relative to the airport.  When the transmitter is on the airport, the non-precision approach narrows significantly from the FAF to the MAP, just like an ILS or LPV.  However, in many (my guess is most) cases, the navaid on which a ground-based approach depends is not on the airport, but instead far enough away that there is no significant difference in needle sensitivity between the FAF and MAP.  The best analog for that in a GPS implementation is fixed sensitivity, and I'm guessing that's why LNAV approaches work that way.  Probably a lot of discussion about it during development, but I don't have any direct insight into how the decisions were made.

[Ragsf15e]

24 minutes ago, Vance Harral said:

Yes, GPS LNAV approaches have fixed sensitivity from the FAF to the MAP.  These approaches are "older", I guess (I'm old enough that every GPS approach seems "new" to me...); but they're still in use everywhere obstacles or other factors preclude having an LPV approach.

The A-N/NDB/VOR transmitters around which non-precision approaches were originally developed are angular, so course width narrows as you get closer to them.  However, the practical behavior of the CDI needle on these approaches depends on where the transmitter is relative to the airport.  When the transmitter is on the airport, the non-precision approach narrows significantly from the FAF to the MAP, just like an ILS or LPV.  However, in many (my guess is most) cases, the navaid on which a ground-based approach depends is not on the airport, but instead far enough away that there is no significant difference in needle sensitivity between the FAF and MAP.  The best analog for that in a GPS implementation is fixed sensitivity, and I'm guessing that's why LNAV approaches work that way.  Probably a lot of discussion about it during development, but I don't have any direct insight into how the decisions were made.

And then there’s also VOR approaches where the vor is the faf and it’s angular, but getting less sensitive as you get closer to the map.  Lots of different ways they can work.

[Ragsf15e]

27 minutes ago, Vance Harral said:

Yes, GPS LNAV approaches have fixed sensitivity from the FAF to the MAP.  These approaches are "older", I guess (I'm old enough that every GPS approach seems "new" to me...); but they're still in use everywhere obstacles or other factors preclude having an LPV approach.

The A-N/NDB/VOR transmitters around which non-precision approaches were originally developed are angular, so course width narrows as you get closer to them.  However, the practical behavior of the CDI needle on these approaches depends on where the transmitter is relative to the airport.  When the transmitter is on the airport, the non-precision approach narrows significantly from the FAF to the MAP, just like an ILS or LPV.  However, in many (my guess is most) cases, the navaid on which a ground-based approach depends is not on the airport, but instead far enough away that there is no significant difference in needle sensitivity between the FAF and MAP.  The best analog for that in a GPS implementation is fixed sensitivity, and I'm guessing that's why LNAV approaches work that way.  Probably a lot of discussion about it during development, but I don't have any direct insight into how the decisions were made.

Oh sheesh, one other oddity… since you can use gps substitution on VOR approaches (while monitoring the vor), you could have an approach designed using angular vor sensitivity, but flown using fixed gps lnav sensitivity.

[Vance Harral]

17 minutes ago, Ragsf15e said:

since you can use gps substitution on VOR approaches (while monitoring the vor)

Well, you learn something new every day.  I was about to post how you can't do that, but then I thought "Ragsf15e is pretty smart, he probably knows what he's talking about".  Sure enough, with GTN navigators and per the AIM, you can legally use GPS guidance on a conventional VOR approach.  See https://bruceair.wordpress.com/2018/09/06/use-of-gps-on-conventional-approaches-update/  I'm a little embarrassed that I'm 5 years late on this "news".

[hais]

3 hours ago, Vance Harral said:

In summary, why shouldn't a GPS-based "precision" approach have angular sensitivity just like an ILS?  What would be the operational advantage of having it work differently?

Would make the IFR operational knowledge part of the check ride easier to pass

Thanks for your detailed explanation. I can see why designers would want to rely on established practices. 

[Ragsf15e]

22 minutes ago, Vance Harral said:

Well, you learn something new every day.  I was about to post how you can't do that, but then I thought "Ragsf15e is pretty smart, he probably knows what he's talking about".  Sure enough, with GTN navigators and per the AIM, you can legally use GPS guidance on a conventional VOR approach.  See https://bruceair.wordpress.com/2018/09/06/use-of-gps-on-conventional-approaches-update/  I'm a little embarrassed that I'm 5 years late on this "news".

No worries, I learn things that I probably should’ve already known like that all the time.  Just to clarify though, you can do it with any kind of approach certified gps nav, not just gtn.  You do have to monitor the underlying vor though.

[PT20J]

3 hours ago, Ragsf15e said:

No worries, I learn things that I probably should’ve already known like that all the time.  Just to clarify though, you can do it with any kind of approach certified gps nav, not just gtn.  You do have to monitor the underlying vor though.

I’d be really smart if I hadn’t forgotten so much stuff I used to know. 

I haven’t shot a VOR non-overlay approach for years. Got an IPC yesterday (good to do every so often, and it keeps the insurance company happy). Instructor wanted to do a partial panel (no G3X), hand flown, VOR-A with a circle to land. Not hard with a GTN, a G5 and a flight director. I sort of recalled something about needing to monitor the navaid, but then I got to thinking that maybe the GTN would auto switch to VLOC like it does for an ILS. It doesn’t. I felt kind of stupid, but at least I won’t make that mistake again. 

Skip

[midlifeflyer]

14 hours ago, Vance Harral said:

Well, you learn something new every day.  I was about to post how you can't do that, but then I thought "Ragsf15e is pretty smart, he probably knows what he's talking about".  Sure enough, with GTN navigators and per the AIM, you can legally use GPS guidance on a conventional VOR approach.  See https://bruceair.wordpress.com/2018/09/06/use-of-gps-on-conventional-approaches-update/  I'm a little embarrassed that I'm 5 years late on this "news".

Yeah, but don't worry about it. Outside of a checkride or IPC, I can think of very few situations in which you will fly a VOR approach with GPS. I avoid them when I can during recurrent training because I think it's unrealistic 99% of the time,

[Deb]

On 4/15/2023 at 10:35 AM, Jerry 5TJ said:

The full-scale width of the GPS-derived localizer is set to 0.3 nmi at its most sensitive when in approach mode. That’s 1800’. 
A conventional ILS localizer has angular sensitivity that increases steadily as the distance to the runway decreases.   At the runway threshold the localizer full-scale deflection width is 700’ in total, 350’ to each side of the centerline.

It’s actually a bit more nuanced; here’s the definitive answer from John Collins:

The LPV CDI FSD (Full Scale Deflection) is 1 NM until reaching a point 2 NM prior to the FAF. The FSD then transitions to the value for the FAF to threshold. At the FAF, this is equivalent to +/- 2 degrees measured from a calculated point beyond the threshold such that when at the threshold, the FSD is +/- 350 feet. It continues at a fixed value of +/- 350 feet until reaching a reference point (about 1000 feet from the threshold) and after that, is changes to a fixed value of +/- 0.3 NM.

So as a practical matter, it is similar to an ILS with a runway length of about 10000 feet. It is angular for the final approach course (+/- 2 degrees) and not a fixed value. It is +/- 350 feet at the threshold and continues at this value for another 1000 feet before increasing back to 0.3 NM.

My reference for this is from RTCA DO229D, which is the technical specification for a TSO C145/146 WAAS GPS Navigator. You have to be a member of RTCA to get this document for free or must pay $370 for a copy.

Here is some description from the 10/12/2017 AIM 1-1-18, d 4 (page 1-1-34):

4. Both lateral and vertical scaling for the LNAV/VNAV and LPV approach procedures are different than the linear scaling of basic GPS. When the complete published procedure is flown, ±1 NM linear scaling is provided until two (2) NM prior to the FAF, where the sensitivity increases to be similar to the angular scaling of an ILS. There are two differences in the WAAS scaling and ILS: 1) on long final approach segments, the initial scaling will be ±0.3 NM to achieve equivalent performance to GPS (and better than ILS, which is less sensitive far from the runway); 2) close to the runway threshold, the scaling changes to linear instead of continuing to become more sensitive. The width of the final approach course is tailored so that the total width is usually 700 feet at the runway threshold. Since the origin point of the lateral splay for the angular portion of the final is not fixed due to antenna placement like localizer, the splay angle can remain fixed, making a consistent width of final for aircraft being vectored onto the final approach course on different length runways. When the complete published procedure is not flown, and instead the aircraft needs to capture the extended final approach course similar to ILS, the vector to final (VTF) mode is used. Under VTF, the scaling is linear at ±1 NM until the point where the ILS angular splay reaches a width of ±1 NM regardless of the distance from the FAWP.

[hais]

On 4/24/2023 at 2:08 PM, Deb said:

When the complete published procedure is not flown, and instead the aircraft needs to capture the extended final approach course similar to ILS, the vector to final (VTF) mode is used. Under VTF, the scaling is linear at ±1 NM until the point where the ILS angular splay reaches a width of ±1 NM regardless of the distance from the FAWP.

Does that mean the scale is different if I select vector to final? 

If that's the case, I hope no DPE get ideas

[midlifeflyer]

10 hours ago, hais said:

Does that mean the scale is different if I select vector to final? 

If that's the case, I hope no DPE get ideas

No. It still goes into APR mode with vectors to final. The mode it is using is displayed. 

[Hank]

On 4/16/2023 at 8:17 AM, midlifeflyer said:

Yeah, but don't worry about it. Outside of a checkride or IPC, I can think of very few situations in which you will fly a VOR approach with GPS. I avoid them when I can during recurrent training because I think it's unrealistic 99% of the time,

Because you never know what may happen. The first year that I was instrument rated, Tower cleared me for an ILS back-course approach. I think I had flown one during training . . . But when I asked for the GPS approach to the runway, they gave it to me.

Then again, you may want / need to divert, and your new destination may not have many approaches to choose from; depending on why you are diverting, you may not have many fields to choose from, or time to check what approaches are available at each. It's best to be as prepared as you can be.

[GeeBee]

Actually there are approaches with high order accuracy all the way in from the IAP. RNP approaches, but our equipment in GA and training make them unavailable. The best example here is the RNP into PSP. I've shot the RNP into Quito Ecuador, RNAV (RNP) 1X RWY 36 (RNP .15) and it will take your breath away.

Take a look at the plate

https://ec.ivao.aero/sources/charts/SEQM - QUITO.pdf

[midlifeflyer]

2 hours ago, Hank said:

Because you never know what may happen. The first year that I was instrument rated, Tower cleared me for an ILS back-course approach. I think I had flown one during training . . . But when I asked for the GPS approach to the runway, they gave it to me.

Then again, you may want / need to divert, and your new destination may not have many approaches to choose from; depending on why you are diverting, you may not have many fields to choose from, or time to check what approaches are available at each. It's best to be as prepared as you can be.

I'll explain. Actually, you made the point for me. If there is an RNAV alternative, ask for it. ATC will give it to you. I can't imagine choosing a VOR approach in the real world when there is a RNAV alternative. Nor, for that matter, a LOC-only approach when there is a WAAS alternative.  On your diversion example, can you point to any airport in the lower 48 which has a VOR or LOC approach but no RNAV approach? There might be a few but I have not come across them. 

I was very specific about "fly a VOR approach with GPS" for a reason. The one (and to me only) time I would fly a VOR or LOC approach in the real world is during a GPS failure. But in that case, it will not be "with GPS." So, yes, definitely fly VOR or LOC  approaches to practice for "what might happen," but turn off your GPS or at least don't load the approach and switch away from the map page. That's what I meant by flying a VOR approach with GPS being unrealistic.

 

[PT20J]

It's good to remember that the pilot gets to select the approach regardless of what ATC is advertising. The only time you might not get it is if you ask for an approach to the opposite end of the runway that is in use because it messes up traffic flow, but I have even been able to get that when I wanted to land downwind to use an approach with lower minimums. 

[ilovecornfields]

5 hours ago, GeeBee said:

Actually there are approaches with high order accuracy all the way in from the IAP. RNP approaches, but our equipment in GA and training make them unavailable. The best example here is the RNP into PSP. I've shot the RNP into Quito Ecuador, RNAV (RNP) 1X RWY 36 (RNP .15) and it will take your breath away.

Take a look at the plate

https://ec.ivao.aero/sources/charts/SEQM - QUITO.pdf

I’m pretty sure I could do those in my Ovation with a GTN 750. As long as I had a tow rope attached to a jet.

[GeeBee]

You might get a call from the FAA since you have no operations specifications to execute the approach, they might be a little upset about the tow too

[ilovecornfields]

4 hours ago, GeeBee said:

You might get a call from the FAA since you have no operations specifications to execute the approach, they might be a little upset about the tow too

But, technically, is there an Equadorian air regulation that specifically forbids me from “flying the approach” while being towed behind a jet? ;-)