ATR72 autopilot - frequent disconnects

[GEKtheReaper 375]

14 hours ago, vitolo said:

Hans' ATR is nothing more and nothing less than a complete unflyable bugfest right now...

I also fly the plane on a daily basis and have zero issues with it except the BUG already known with the hydraulics in IDLE (already mentioned by @mryan75 ).

I also have the X-56 ( as the OP @Adamski_NZ) and had no AP disco in any of my flights.

I suggest to create new EMPTY profiles for all your inputs (KBD, Mouse, HOTAS) and assign basic buttons ONLY, to be able to test the ACFT. If it works then proceed to map other keys and buttons. If it fails then empty your community folder (leave GSX if you have it cause I also have it and it does not disconnect AP in the ATR).

Edited January 16 by GEKtheReaper

[mryan75 281]

30 minutes ago, dzmitry said:

I mean you're right, knowing the bugs you can somewhat go around them. I love the plane actually but it feels a bit a chore to fly it knowing that you can break it in so many ways. I really really really hope in an update this month.

Yeah, to me the airplane hand-flies wonderfully and with a really good update to address the known problems we will have a seriously cool airplane on our hands. I think it's a joy to fly, just the right mix of automation and pilot input.

[Adamski_NZ 967]

5 hours ago, GEKtheReaper said:

I also have the X-56 ( as the OP @Adamski_NZ) and had no AP disco in any of my flights.

I suggest to create new EMPTY profiles for all your inputs (KBD, Mouse, HOTAS) and assign basic buttons ONLY, to be able to test the ACFT. If it works then proceed to map other keys and buttons. If it fails then empty your community folder (leave GSX if you have it cause I also have it and it does not disconnect AP in the ATR).

Thanks for the tip. I already do something similar - as I use AAO. All my MSFS profiles are CLEAN/EMPTY and I then add AAO profiles on a per-aircraft basis. I used to do this with FSUIPC, but AAO is much easier.

So far so good ... it does appear to be a Simbrief or payload loading bug - rather than spurious control inputs.

[GEKtheReaper 375]

@Adamski_NZ I don't use AAO. Could that be the reason?

I do use the SIMBRIEF integration and load the ACFT through the EFB (clicking all 4 fields OFP / SET FPLN / SET FUEL / SET PAYLOAD). I somethimes skip SET FPLN and SET FUEL to do some manual work.

Maybe I could help you out comparing some of the values if you suspect a payload issue. If you share your FPLN and PAYLD, I could double check what a TO configured ATR gives me as e.g. trim setting etc.

[RALF9636 1,998]

I don't have an idea for a solution, but I can report that I also use AAO for the controls and always load the ATR via the EFB using the simbrief integration and never had this issue with the AP disconnecting.

[Adamski_NZ 967]

I'm part way through logging all the weights etc. before/after Simbrief import (via MCDU or EFB). As soon as I'm done I'll post them here. One thing that's apparent already is that inputting all the values (from Simbrief) manually into the PERF/WEIGHT page gives me a trim of 1.1 UP - whereas the EFB always gives me 1.4 UP.

Do people set the trim manually - or via the payload screen in the EFB?

I'll also share my test Simbrief flight plan when I've finished logging.

[RALF9636 1,998]

I tend to set the trim about 0.5 below the value given by the EFB to prevent the ATR from rotating on its own (which AFAIK is a known bug). But both with the trim value from the EFB or the lower one I never got the AP disconnect poroblem.

 

[GEKtheReaper 375]

@Adamski_NZ just loaded the ATR after the update and loaded it through SIMBRIEF with 68 PAX and a route that requiered 2069 BLOCK FUEL. I did not enter anything else except the PAX. The PERF CALC gave me a TRIM of 0.7 UP which was spot on.