Ibra wrote:
I think everybody in GA assume TSO-C146 box == SBAS box, do you know if there are TSO-C146 GNSS boxes that are not SBAS?
TSO C145/146 are for an SBAS enabled receiver/navigator/position source.
Peter wrote:
If you flew EGKA 20 with a GNS-W/GTN/IFD do you really get an angular lateral deviation? And what happens if SBAS is lost halfway down the “glideslope”?
Losing SBAS is different than the integrity dropping below limits. It would mean the geostationary satellite, which should always be in view for an LPV or LNAV/VNAV to be annunciated, is no longer providing the SBAS signal. In the US, this happened for a single 4 hour period since 2007, so 4 hours out of 14 years of operation. On the other hand, there were many many times where the integrity dropped below the limits and an indication was provided such as amber LPV annunciation and it could result in a downgrade to LNAV at a point one minute before the FAF. After that I would expect vertical and or lateral to be flagged and do a missed approach.
The most likely cause of a downgrade is the VPL drops below limits, in which case the approach is downgraded to LNAV. A downgrade only occurs before the FAF. If there were a SBAS outage rather than just a poor integrity value provided by SBAS, the equipment would revert to RAIM with FDE and as long as there were sufficient satellites to provide for a RAIM calculation, it would be used for LNAV. If even RAIM failed, then you would not get an LNAV annunciation although once on the procedure, as long as there was a valid GPS lateral position, you can continue the approach once started. Vertical guidance would never be allowed in this case, but +V which is advisory could be allowed. The former is official vertical guidance for a vertically guided approach whereas the latter is just advisory for descending to an MDA and any descent path is permitted so wrong information should not cause you to crash. Same is true of a visual approach guidance. you are supposed to be visual anyway.
RobertL18C wrote:
Doesn’t an LPV have a more precise data set of waypoints creating an ILS type funnel (half deflection being approximately 0.3 degrees TAN)? The other types of vertical guidance are in effect corridors and not funnels, so at some point some distance from the touchdown point, a half scale deflection fly up would provide a path to impact?
The WAAS vertical guidance is based on a FSD of +/- 0.25 times the GP angle. It is fixed outside a certain distance from the threshold to be +/- 150 meters. Also inside a threshold, there is an option for the vertical FSD to remain fixed minimum rather than continue as angular. For LNAV/VNAV, the fixed FSD minimum may be set to +/- 45 meters and for an LPV it can be set to +/- 15 meters. This is what I see on the Garmin units. Between the maximum and minimum fixed FSD values, the vertical guidance is angular and equal to +/- GP angle/4.
Airborne_Again wrote:
So it seems that getting the angular deviation is not dependent on the SBAS signal as such, but is simply a function that a TSO-C146 box should have
I think everybody in GA assume TSO-C146 box == SBAS box, do you know if there are TSO-C146 GNSS boxes that are not SBAS?
Maybe the case for FMS units using Baro & INS for fault detection & error correction
But you are probably right the angular is box specific and not related to the SBAS signal, LP angular sensitivity should work for LNAV in Tanzania with an SBAS box even if SBAB signal is not available down there? altough, you can argue that an non-SBAS box should also allow to compute and mimic angular devitation, it just does not have the means to self-test for fault detection & error correction to ensure its stays within the specs, maybe except KLN94 (that was for free Peter 
)
Peter wrote:
I know the 1nm FS sensitivity at FAF gradually reduces to 0.3nm at the DH, which sort of amounts to that (and this is how even a KLN94 delivers really accurate lateral guidance) but this is not related to SBAS because every IFR GPS does this.
No, it’s not related. I got the info from the PPL/IR PBN Manual. It refers to TSO-C146 as the source of that information. TSO-C146 refers to RTCA/DO-229C which you can buy for about $500…
If you flew EGKA 20 with a GNS-W/GTN/IFD do you really get an angular lateral deviation?
I’ve tried it with Garmin’s GTN650Xi simulator. I set up the EGKA LNAV 20 approach and flew it slightly offset. The navigator announces LNAV+V and, yes, you do get angular lateral deviation.
And what happens if SBAS is lost halfway down the “glideslope”?
That’s a good question. The major error in the non-SBAS position is due to the varying ionospheric delay. This does not change in the short term. So arguably, if SBAS is lost, the receiver could keep applying the most recent correction throughout the approach – at least for lateral guidance. I’m sure RTCA/DO-229C will tell you all about it. 
I tried this as well with the GTN650Xi simulator. I turned off SBAS shortly before the threshold and the deviation pointer didn’t move. I then tried an offset approach again and I got angular deviation again. So it seems that getting the angular deviation is not dependent on the SBAS signal as such, but is simply a function that a TSO-C146 box should have.
PS. I didn’t check what happened to the glide path. For another time…
With SBAS GNSS it’s LP performance and angular deviation for lateral profile on CDI near the runway even for an LNAV, this makes things hectic on SBAS LNAV with AP doing down to ground on APP mode bellow M/DH, the approach however, is designed as 0.3 lateral with the OCH that goes with, to maintain backward compatibility, it is still flown by non-SBAS navigators…
In a nutshell, SBAS LNAV = CDI sensitivity is angular, IAP protected area is linear
On non-SBAS, LNAV AP just flies beautifully an LNAV in it’s middle continuously: past DH, past MAPt, over THR, over centerline all the way to the other runway end, same effort in 3000ft as at 300ft, assuming it has load of satellites
What if you are flying a non SBAS approach (say EGKA 20) do you really get an angular lateral deviation?
I know the 1nm FS sensitivity at FAF gradually reduces to 0.3nm at the DH, which sort of amounts to that (and this is how even a KLN94 delivers really accurate lateral guidance) but this is not related to SBAS because every IFR GPS does this.
If you flew EGKA 20 with a GNS-W/GTN/IFD do you really get an angular lateral deviation? And what happens if SBAS is lost halfway down the “glideslope”?
Loss of LPV in an eighty ton aircraft with several seconds of inertia is a big deal in low IFR I would have thought.
Sure, but they are all flying a ILS.
RobertL18C wrote:
Doesn’t an LPV have a more precise data set of waypoints creating an ILS type funnel (half deflection being approximately 0.3 degrees TAN)? The other types of vertical guidance are in effect corridors and not funnels, so at some point some distance from the touchdown point, a half scale deflection fly up would provide a path to impact?
With SBAS, not only LPV, but all RNP approaches use localiser type angular deviation.
Doesn’t an LPV have a more precise data set of waypoints creating an ILS type funnel (half deflection being approximately 0.3 degrees TAN)? The other types of vertical guidance are in effect corridors and not funnels, so at some point some distance from the touchdown point, a half scale deflection fly up would provide a path to impact?
Loss of LPV in an eighty ton aircraft with several seconds of inertia is a big deal in low IFR I would have thought.
Extremely interesting replies – many thanks!
It does put the supposedly imminent loss of LPV in the UK in a better light.
