(maybe Peter can copy/delete the content from the other thread here).
I’ll investigate the matter later (I should do at least some work or I will have to SELL the DFC90 anyway :-))
One of the problems in these discussions is that some argue from a practical standpoint (Jason, me) while some want to go into the elctronic details, which (i admit it) doesn’t interest me much.
What’s definitely true is that you can (in my Cirrus for sure) fly ILS and GPS approaches all completely in NAV mode and that there is no difference in precision.
And that’s what counts for me.
I don’t have time to move dozens of posts; the user interface I have enables the move of one at a time.
If you want to argue for flying without detailed aircraft systems knowledge, feel free to start a thread. But it may fairly quickly lead a a discussion of AF447, etc 
If you want to argue for flying without detailed aircraft systems knowledge, feel free to start a thread. But it may fairly quickly lead a a discussion of AF447, etc
That’s just polemic, and it is not my point. I don’t have to know anything about “ARINC 429” to be a safe pilot. I have to know what will work and what will not work.
You need to know which data sources are used in which modes.
So for example if you lose lateral guidance at the FAF because your GPS stops outputting the data which the AP is using in that particular mode, you could get yourself killed because you will end up flying an ILS in a wings-level mode. Not even a heading-hold mode… Normally, in northern hemisphere, the wind turns left as you descend, so if you get visual at the DH you could end up way to the right of the runway.
The wider issue here is a huge problem in GA – as this and the preceeding thread demonstrate. The “WTF is it doing now” is a really common moment. There is an increasing gulf between the very basic PPL training, the almost as basic IR training (and IR checkride standards which focus on traditional aspects of IFR) and the systems which can be found in modern aircraft.
NAV mode can be used for ALL data sources. The data source is selected on the PFD (GPS1, GPS2, VLOC1, VLOC2) or with the CDI button of the GNS (GPS, VLOC) . The Autopilot takes whatever data is coming from either the GPS or VLOC receiver.
Are we talking about the same thing?
Flyer59 wrote:
Are we talking about the same thing?
I think so. Peter talks about what could happen if the pilot uses the GPSS mode with the expectation that it would work the same as the NAV mode.
In that specific Cirrus installation is there a connection between the GNS430Ws and the DFC90?
If there is none, then I can see why the mode does not matter for the DFC90 – given the PFD, being the only factual data source for the DFC90, does not alter its output.
But then loss of the PFD unit would also mean that all input for the DFC90 is gone.
Yes, if the AHRS of the PFD fails (in my airplane), then the autopilot will go offline. Not so in your airplane, the R9 has two AHRS units, so your DFC100 (R9 version of DFC90) will continue to operate if one AHRS fails.
The DFC90 will continue to operate of course if only the backlighting of the PFD fails.
That’s actually the one point where the S-TEC55X offered more redundancy, It does not get its attitude from the PFD/AHRS but from the (hidden mounted) Turn Coordinator. With the DFC90 the autopilot gets the attitude data form the AHRS AND the Turn Coordinator (for verification). If that TC fails the A/P stays on if that happens in flight – but will not successfully self test if that happens on the ground.
Also not good (but only valid for SW version R2): If ALT 1 fails the A/P will fly on BAT1. But once BAT1 fails, or is empty, the A/P will go offline. Due to a stupid programming error the A/P will not fly on ALT2/BAT2 … There’s a field fix for that, but i don’t have that yet. Jim Barker developed it and he modified a couple of US planes (in the CB panel) in a way that the autopilot will also stay one on ALT2/BAT2.
What I have learned in the meantime … (some more information for Cirrus pilots included)
In my configuration (Avidyne Entegra 8.01 + DFC90 A/P + 2 GNS430W) the following is valid:
- All attitude and steering data for the A/P come directly from the PFD
- If the A/P is in NAV mode it will work exactly the same for GPS data coming from the PFD in digital form as if it was on “GPSS”
- VLOC data will be provided to the A/P in digital form (with the S-TEC55X VLOC data would come from the PFD in analog form)
- Of course if the A/P is in GPSS mode it will only fly GPS data
Other important facts for Cirrus pilots with DFC90 and Avidyne Entegra
- with SW version 2 (“R2”) the A/P will stop working if BAT 1 fails after an ALT1 failure
- the autopilot will work on ALT2/BAT2 with the release 1 firmware (“R1”)
- a field fix is available to get back this behaviour, one pin has to be changed in the CB panel (Avidyne obviously doesn’t care to fix this bug)
- if the PFD fails there will be no A/P, although the A/P is also connected to a (blind mounted) Turn Coordinator but that TC can not be used for attitude.
It is very unlikely that the AHRS in the PFD fails. Backlight failures happen, but those do not affect the autopilot. Even if you don’t have the PFD screen you can still use GPS
The GPSS mode with e.g. the STEC55 gave you the fly-by or fly-over differentiation on GPS approaches. By pressing the NAV button twice you would fly by the fly-by waypoint instead over over them. With only the NAV pressed without the GPSS annunciation, you would fly over the fly-by waypoint. Now, I have also in the past flown the DFC90 but can’t remember there being a function for that specifically. Other than that, the DFC90 was a lot better, but once my pitot heater failed and the pitottube froze up. All of a sudden the envelope protection on the DFC90 kicked in and thought I had lost all airspeed and wanted to protect me (cruising along at FL180 at high speed). So the DFC90 pointed the aircraft down in order to “protect” us from a stall. So our heads were against the ceiling of the aircraft and soon we were approaching maximum speeds. No good for me. Dumped the DFC90.
