The S-TEC55X works differently, and the above is valid only for the DFC90
- The S-TEC55X gets an analog VLOC signal
- The S-TEC55X needs the dedicated roll steering mode “GPSS” to anticipate the turns (you press NAV twice to get to GPSS)
Pressing NAV twice gave you the GPSS mode and therefore the A/P would flyby a waypoint.
The DFC90 does the “fly by” with all waypoints of the GPS flightplan, no matter if you are in GPSS or NAV mode.
The envelope protection will only lower the nose to protect you from a stall. Now, if you don’t have airspeed indication the A/P has to lower the nose even more, which is logical. Of course such a system for $ 10.000 cannot save you in ALL situations, it’s not an Airbus (and those fail too…), but it actually does protect you i almost all dangerous situations in IMC.
With an iced pitot tube hand flying would be better. In that case I’d fly by groundspeed for a while and compare with the wind vector. Autopilots can be very dangerous in these situations. I think this behaviour is not a reason to dump the DFC90
Reading this stuff shows how many permutations there are, and this is just one aircraft type!
One obviously doesn’t need to know the packet format of ARINC429 but one does need to know the system interconnections and the failure paths. An example is the Turkish B737 crash at Amsterdam where the pilots didn’t know that the failure of a particular RADALT (they had two) will have an effect on the engine power, causing the throttles to be retarded (IIRC).
There is no type rating in GA for systems of comparable complexity. You can just buy the plane, or if you had the insurance-mandated training for a fully loaded SR22, you can just sell that and buy a differently fully loaded SR22 and based on the firm I spoke to, they won’t question it.
…and that’s only one of many problems.
How much did your Flight school tell you about using the Autopilot when you did your IFR? Mine: zero. I hand flew for 35 hours, the A/P was always broken, and the Flight School had no interest repairing it.
The above is not known by at least 75 percent of all Cirrus pilots. I spoke with “Cirrus Instructors” who knew next to nothing about all this.
Peter wrote:
There is no type rating in GA for systems of comparable complexity. You can just buy the plane, or if you had the insurance-mandated training for a fully loaded SR22, you can just sell that and buy a differently fully loaded SR22 and based on the firm I spoke to, they won’t question it.
After our avionic upgrade we did a full evening of ground training as well as individual flight training with all pilots to familiarize them with the new equipment. I should think that it is in your own interest to do that, but of course it means you will have to have an FI or CRI who is trained on the particular airplane and knows the systems inside out.
Why do we always wait for someone to tell us we HAVE TO do something when we can do it ourselfs? Isn’t it in your very own interest to know your airplane? So if you change from one airplane to the other which has different systems then why assume anything and not do a good type intro?
AeroPlus wrote:
No good for me. Dumped the DFC90.
Well, now you know that the DFC90 will not operate as you expect it to in a case of unreliable airspeed. Neither will the Airbus AP btw, as the events around AF447 showed too clearly. A reason to dump the AP? Not where I am sitting, otherwise all Airbusses would have to have been grounded after AF447 (and the Qantas incidents) but were not, logically so, as a KNOWN issue is usually not a dangerous one. As a consequence of your experience it would mean you need to switch the AP off immediately if you loose your IAS indication.
Flyer59 wrote:
Of course such a system for $ 10.000 cannot save you in ALL situations, it’s not an Airbus (and those fail too…), but it actually does protect you i almost all dangerous situations in IMC.
As the AF447 and some other incidents show the Airbus protections are far from failsafe! Particularly pitot icing will produce very dangerous and unwanted effects, also with the Airbus logic. That is why protections are a two sided sword, they do safe some situations but actually worsen others. It is totally wrong to assume, as some do, that Airbus protections (or any others for that matter) are a licence to complacecy, just the opposite. You need to know your systems well enough to anticipate what happens if element “x” fails and what you still have and what you don’t. And moreover, any AP is no replacement for flying skills. AF447 has demonstrated that like no other accident, even though there are other examples such as the TK 737 in Amsterdam.
I would say the input AeroPlus gave re pitot icing and what the DFC90 does with it is a valuable input for anyone who has that AP and has not thought that scenario through, while it is actually quite a logical consequence. The more complex an AP gets, the more chances of failure modes it has. Which does not mean it’s a bad AP but vigilance must be greater.
It’s a common place that you hand fly the airplane in situations like that. The “envelope protection light” of the DFC90 is based on indicated airspeed only and the manual describes these festures well – and that you have to hand fly the plane in such situations.
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.
What is the Cirrus procedure for pitch trim runaway?
Pull the Circuit Breaker! But in the above case just AP off.
The Cirrus has no manual pitch control, if the electric system fails you have to manually overpower it on the stick. The forces are calculated to be bearable although I wouldn’t want to fly to the US from the Azores in a situation like that… although it would give plenty of time to come up with a McGyver sidestick holder 
I would rather have a manual backup … oh well
In the C510 you hold down a/p disconnect, pull breaker and keep speed as appropriate to reduce control forces. My training partner didn’t notice it quickly in the sim and was full aft trim before recovering. Try doing a OEI approach in that state.
