[SU11] TurboProp Auto Feathering is not working properly

Dear Asobo, During the last couple of week I have been trying to set a
TruboProp to auto feather as per POH, however I discovered several issues with
the configuration of this system in MSFS. 1) The auto feather ** will always
fail to arm
** regardless of any settings in the engines.cfg unless you add a
custom circuit on your system.cfg with the name of CIRCUIT_AUTO_FEATHER.
However there is nothing in the documentation explaining this requirement
causing lot of headaches since pretty much nothing works. In addition, many
auto feathering systems in real work don’t work using an electrical circuit,
they operate by using oil pressure and mechanical means to feather and un-
feather the prop. 2) max_n1_for_autofeather_actuated is not working as
intended, the description of this variable says this is the maximum percentage
N1 below which the armed autofeather system starts to feather the
propeller. Therefore the expectation is, if this parameter is set to
max_n1_for_autofeather_actuated =50, when N1 is below 50% and if the
autofeather is armed, the system should feather the propeller. However this
doesn’t occur… the propeller always feather only when N1 reaches zero
percent. 3) min_rpm_for_feather, this is not working as intended either, the
description of this parameter says, this is the minimum RPM at which the prop
will feather. Therefore if this is set to min_rpm_for_feather=500, the
expectation is to see the feathering occurring when
max_n1_for_autofeather_actuated kicks in + the prop RPM are below the
configured value.The result of setting this is that instead the propeller to
never feather when the engine is OFF prior startup, because the variable that
indicates the auto feather should be armed doesn;t kick in until the props go
ABOVE the 500 rpm. Most Turbo Props are feathered prior engine startup. 4)
None of the events to manually feather a propeller work, the feathering
switches, etc. even if you configure setting the parameter
feathering_switches=1. 5) It seems the only way to feather the propeller is by
setting the propeller leaver into a negative state, this works… however
there are many airplanes that do not have a propeller pitch lever!, so this
limitation impose some challenges to configure things up. 6) The simvar PROP
FEATHERED:index never fires into true value, even if you feather the propeller
manually by putting the pitch levers into full negative. Here is a picture
showing the problem, despite the blades being feathered… this is never
confirmed via the sim var:


NOTE: these variables set as follow for all these tests,
prop_feathering_available =1 and prop_auto_feathering_available=1 I have
resorted to writing my own auto feathering simulation in order to bypass all
these issues, however it is clear the TurboProp feathering system
configuration needs to be looked at, it shouldn’t be this hard to configure
something so fundamental for TurboProp simulation. Many thanks for taking the
time to read this long post, if you guys need anything from me to help with
this please do let me know. Best Regards, Raul

Hello @Simbol 1) We will clarify our doc on this and add cross references so
this is not missed. I don’t think this is the only case where you will have to
configure the electric system to operate a specific feature and I don’t think
we want to start exposing new parameters to make electrical subsystems
optional as setting them to true at all times seems a reasonable workaround. 2
& 3) We still have a bug where the config reading will look for the wrong
parameter for this. It’s looking for
max_pct_torque_for_autofeather_actuated instead of
max_n1_for_autofeather_actuated. The fix seems to have lost its way. I
will review this and make sure this is fixed in the future. Can you try this
change and tell me which ones of the following issues are still relevant?
Regards, Sylvain

  1. The PROP_FEATHERED simvar isn’t plugged to anything so I can confirm it
    will never switch to true and can be considered deprecated. You should use the
    beta simvar to determine when it’s been feathered instead.

Hi @FlyingRaccoon Sylvain, Thanks for looking into this, it explains why
things were going high wired. I wrote a custom solution where I am feathering
things manually via custom simulation. I will disable it now and test using
max_pct_torque_for_autofeather_actuated and revert back. While using this
parameter what you should use for the min_rpm_for_feather? the desired result
is prop feather on ground engine off, un-feather when engine on… auto feather
when engine off below certain N2 value. Best Regards, Raul

That’s where the prop lever is used in our model. From what I could see, most
of our TP start with prop lever set to negative values (feather range). Since
your aircraft doesn’t have a prop lever, how is it managed? There’s a system
that automatically feathers the prop on engine shutdown? What problem do you
encounter when programmatically setting prop lever positions to get the
behavior you want?

Correct the only option is to set the prop lever to negative and since this
airplane doesn’t have one, well it is kind of problematic… since my
expectation was to disable the prop lever controls in order to avoid issues
with users that have hardware axis assigned to this. So now I have the
complexity / difficulty of prop lever bind assignments… which I had to create
a custom input event procedure to cancel them out… and in order to set the
prop lever to negative (since I am using WASM, so simulate other things) I
found out that we have an issue with K:Events and negative values which I also
reported here. link: (https://devsupport.flightsimulator.com/t/4843) this is
still not working… I am not able to set the prop levers into negative via
trigger events and was forced to use Simconnect inside WASM to set the A:VAR
variable values directly via GENERAL ENG PROPELLER LEVER POSITION. You can see
how complex it all got just to feather the props… we need easier ways, Iam up
and running via custom solutions, but I wanted to providing feedback on this
matter since I think the platform would benefit for everyone from having
better ways to handle this. Best, Raul