Open legoboyvdlp opened 9 years ago
Hm, it's happened to me several times already, but now I wanted to try to reproduce it and it did not happen.
I am pretty sure that the bug exists, but it does not happen every time. If I just quickly start the engines, configure for take-off (flaps, spoilers armed, autobrake max) and take off, it does not happen. So it breaks by something during the normal preparation. But I don't know what.
Well, I always set up properly, with FGMC. It seems to happen at V1, so likely, if you set the V1 property to 80 knots, it would break at 80 knots.
Testing now. Set FGMC with SVMI/SVMG, FL180, Flight No, Alternate, and V speeds 80 120 140.
Confirmed, it happens at V1.
Yes, it happens when V-speeds are set; for me it happens at 72 knots, not V1, though.
So I am starting at the controller and it makes absolutely no sense however. The controller checks that throttle of both engines to /autopilot/autobrake/config/idle-throttle
, weight is on wheels and indicated speed is over 72 knots. Property browser tells me the idle-throttle
is still 0.1 and throttles are both still 1, so the first condition shouldn't be satisfied. And I don't see any other reference to the /controls/flight/ground-spoilers
property anywhere else either. Still, it becomes true. Only if V-speeds are set. Strange.
Ok, so I pulled out the howitzergdb and traced the system. What happens is that the auto-thrust controller in Systems/FMGC/auto-throttle.xml
:IAS hold FMGC
tries to control the thrust always, but is later overriden by Systems/thrust_levers.xml
:Left
/Right Throttle Position
. However the ground spoilers system runs between them, so it sees the thrust levers retarded. But the property browser, and FDM, both see them fully advanced, because that's where the autopilot system eventually leaves them.
Now I'll have to think about how to fix that properly (it can probably be hacked quickly by rearranging the controllers, but it would be extremely fragile too).
Yay! Nice backtrack!
I suggest thinking the real fix.... as you say.
(OTOH, this is already beyond my coding comprehension, so I limit myself to the role of Cheerleader!)
On Mon, Nov 9, 2015 at 1:09 PM, Jan Hudec notifications@github.com wrote:
Ok, so I pulled out the howitzergdb and traced the system. What happens is that the auto-thrust controller in Systems/FMGC/auto-throttle.xml:IAS hold FMGC tries to control the thrust always, but is later overriden by Systems/thrust_levers.xml:Left/Right Throttle Position. However the ground spoilers system runs between them, so it sees the thrust levers retarded. But the property browser, and FDM, both see them fully advanced, because that's where the autopilot system eventually leaves them.
Now I'll have to think about how to fix that properly (it can probably be hacked quickly by rearranging the controllers, but it would be extremely fragile too).
— Reply to this email directly or view it on GitHub https://github.com/FGMEMBERS/A320neo/issues/9#issuecomment-155193943.
I uploaded a fix for this Issue on my A330-200 repo. Should be the same for the A320neo. Would be nice if someone could check it out and tell me if it works for him.
I tested it and it works fine for me: Spoilers do not engange on take-off (if V-speeds are set or not). Spoilers retract on touchdown, when armed.
That sounds incredible.
@ALL. I recommend the following code to test the OrangeTiger repo
git checkout -b Spoiler-fix
git pull git@github.com:OrangeTiger/A330-200.git
That creates a branch in your repo named Spoiler-fix, and switches to it, then it pulls OrangeTiger repo into that branch
you can return to master with
git checkout master
and return to Spoiler-fix with
git Spoiler-fix
It this fix works, all is needed is that OrangeTiger (or someone else) initiates a pull request from OrangeTiger repo
IH-COL
Okay, plz remind me to test...
It seem not fixed yet in master.
Everytime I takeoff, I get a master caution, and speedbrakes engage. Ground spoilers engage with no reduction of throttle.