Board index FlightGear Development Spaceflight

Space Shuttle - Bugfixes

Discussion about development and usage of spacecraft

Re: Space Shuttle - Bugfixes

Postby Thorsten » Thu Jan 09, 2020 7:21 am

Not sure about the interaction between stick and trim - worth a question. As for the 'failed', the Shuttle never trims, because the 'complete' FDM loaded at init has ET and boosters attached, and as you can imagine that setup is far too heavy to respond to aerosurfaces in any meaningful way.
Thorsten
 
Posts: 11720
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Bugfixes

Postby wkitty42 » Thu Jan 09, 2020 1:50 pm

eatdirt wrote in Wed Jan 08, 2020 9:41 pm:@wkitty, I've just managed to destroy the Shuttle at ignition by pulling my joystick back and holding it locked there BUT before starting fgfs on the command line. That must be an initialization issue!

it may be... others on the dev mailing have been able to reproduce my symptoms with the reboot scenario... i'm still not sure where the problem is and haven't yet had a chance to reboot this machine to test further... i'm just glade that others can reproduce the problem of the joystick init after reboot before starting FG...

eatdirt wrote in Wed Jan 08, 2020 9:41 pm:If I pull the joystick well after starting fgfs, but before pressiong SHITF+I, nothing happens, as it should. Could you test if you get the same symptoms, I am able to reproduce it at 100%.

yeah, that is fine because as soon as i move the stick (or click a button IIRC) it settles down and everything works as expected/desired...

eatdirt wrote in Wed Jan 08, 2020 9:41 pm:Nothing in the console, up to this (which may be the consequence of the explosion)

i think it is, as well... i can confirm seeing similar here...
"You get more air close to the ground," said Angalo. "I read that in a book. You get lots of air low down, and not much when you go up."
"Why not?" said Gurder.
"Dunno. It's frightened of heights, I guess."
User avatar
wkitty42
 
Posts: 6603
Joined: Fri Feb 20, 2015 3:46 pm
Location: central NC, USA
Callsign: wk42
Version: git next
OS: Kubuntu 14.04.5

Re: Space Shuttle - Bugfixes

Postby eatdirt » Thu Jan 09, 2020 2:33 pm

yeah, that is fine because as soon as i move the stick (or click a button IIRC) it settles down and everything works as expected/desired...


Yes, I understand that. But what I meant is that, even with a perfectly fine and trimmed joystick, could someone also test that we can actually destroy the Shuttle at ignition time by doing:

1) Take your joystick and pull it (maximal elevon deflection) and hold it there
2) Start flightgear (and Shuttle-launch scenario) while still holding the josytick!!! (crucial)
3) With your remaining free hand, switch on APU, boilers, high pressure hydraulics, and then press SHIFT+I

=> Shuttle explodes at ignition because it tries to fly horizontally, as ordered by the joystick.

I was just trying to find something reproducible all the times, and in my case it is!
eatdirt
 
Posts: 624
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Bugfixes

Postby enrogue » Thu Jan 09, 2020 2:57 pm

eatdirt wrote in Thu Jan 09, 2020 2:33 pm:Yes, I understand that. But what I meant is that, even with a perfectly fine and trimmed joystick, could someone also test that we can actually destroy the Shuttle at ignition time by doing:

1) Take your joystick and pull it (maximal elevon deflection) and hold it there
2) Start flightgear (and Shuttle-launch scenario) while still holding the josytick!!! (crucial)
3) With your remaining free hand, switch on APU, boilers, high pressure hydraulics, and then press SHIFT+I

=> Shuttle explodes at ignition because it tries to fly horizontally, as ordered by the joystick.


I can verify that this destroys the Shuttle - I have a CH Fighterstick that has trim wheels that apply to the actual controls rather than usual trim, so all I have to do is set the elevator trim wheel to maximum up & start the shuttle -> boom!
User avatar
enrogue
 
Posts: 255
Joined: Mon May 19, 2014 6:40 pm
Location: London (UK)
Callsign: enrogue
OS: Ubuntu, macOS

Re: Space Shuttle - Bugfixes

Postby Thorsten » Fri Jan 10, 2020 7:24 pm

Now it'd be really good if someone would also report that to the mailing list, because I don't have a joystick, so I can't test anything, I have no clue how things should works, so that's the one probem I really can't do anything about.
Thorsten
 
Posts: 11720
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Bugfixes

Postby eatdirt » Fri Jan 10, 2020 9:18 pm

I'll do, in fact I did but I am using gmane as a news reader and my email never arrived :(

I am chasing another nasty bastards now, with next, sometime when I do a trajectory calculation on OPS 202 or Lambert targetter, I get a segfault. And again, that comes after a long while, it is crazy to reproduce. I am trying to install debugging symbols to chase it with gdb, but it is in libSimGear:

Code: Select all
fgfs[21709]: segfault at 93a4d2 ip 00007f875b82f9ba sp 00007f8749a12ad8 error 7 in libSimGearCore.so.2019.2.0[7f875b775000+152000]
eatdirt
 
Posts: 624
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Bugfixes

Postby Thorsten » Sat Jan 11, 2020 7:40 am

Try Richard's theory and de-activate the threaded Nasal garbage collector.
Thorsten
 
Posts: 11720
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Bugfixes

Postby Thorsten » Sat Jan 11, 2020 12:47 pm

I can verify that this destroys the Shuttle - I have a CH Fighterstick that has trim wheels that apply to the actual controls rather than usual trim, so all I have to do is set the elevator trim wheel to maximum up & start the shuttle -> boom!


Actually, let's narrow down the pathway a bit.

Is the issue actually that the engine gimbals move - or is it something else? What do the SSME and SRB engine pitch and yaw angles read at liftoff?

Also, what happens if you do the same exercise with SpaceShuttle-mission and start in orbit? Do you start with an uncontrolled spin? Or is everything normal there?

(I'm still trying to understand how the signal worms its way into the rate controller...)
Thorsten
 
Posts: 11720
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Bugfixes

Postby eatdirt » Sat Jan 11, 2020 10:34 pm

Gotcha!!! AI? FlightPlan while I was trying to dock?? :shock:

Code: Select all
Thread 146 "fgfs" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7f2258b7b700 (LWP 18179)]
0x00007f23341e29ba in naObj (type=type@entry=2,
    o=0xa61961 <FGAIFlightPlan::createClimb(FGAIAircraft*, bool, FGAirport*, FGAirport*, double, double, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)+1>)
    at /usr/src/debug/simgear-2019.2.0-9.mga7.x86_64/simgear/nasal/misc.c:33
33          o->type = type;


Another funny one, started on orbit from a previous crash, switched back to basic weather to avoid the atmosphere blinking, and sometimes it is snowing in space :)
I'll post on the mailing list the crash, and do the tests you suggest for the joystick.
eatdirt
 
Posts: 624
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Bugfixes

Postby eatdirt » Sat Jan 11, 2020 11:57 pm

Actually, let's narrow down the pathway a bit.


- Ok, I've tried the suggestion of Lego on the mailing list: Starting with --prop:/controls/flight/elevator=1 --prop:/controls/flight/aileron=1 does nothing, everything works fine.

- Then, with the holding technique, I clearly see on the pad the SSME engines in the position ordered by the joystick. For instance, pushing the joystick up, I see on external view the SSME trimmed to go down even after APU on and full hydraulic pressure. Same for the reverse, pulling the joystick down and SSME engines are trimmed for making a loop at ignition.

- Starting in space, it seems to work ok. Pulling the joytick before starting fgfs and the Shuttle appear with RCS on and start spinning at the rate ordered, releasing the joystick it stops.

I have tried to look under fdm/jsbsim/systems/fcs and in vectoring, but nothing look suspicious here. Maybe you could point me to a variable I could monitor? (full path welcome, I am not really at ease inside the property browser!)

Cheers,
Chris.
eatdirt
 
Posts: 624
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Bugfixes

Postby eatdirt » Sun Jan 12, 2020 12:03 am

I've encountered a minor problem/feature trying to play with survival possibilities for a 3 engines out during stage 1...

1) During ascent with SRBs, switch off the 3 engines, with the keyboard, CTRL+q +w +e
2) Have your last view of Earth by the windows while the SRBs are burning, or try to fly the Shuttle in CSS, boosters attached, that's more fun!

--> The boosters don't separate by themselves anymore, and I cannot manually separate them neither once they are extinct.

PS Yes, the conclusion is that the survival rate is currently 0! :)

Edit: I've tried also while they are burning, same thing, cannot separate (I've paid attention to push the first switch to manual).
eatdirt
 
Posts: 624
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle - Bugfixes

Postby Thorsten » Sun Jan 12, 2020 9:02 am

PS Yes, the conclusion is that the survival rate is currently 0!


3EO during first stage is a black zone in reality :shock: - you will die if that happens.

Edit: I can verify that the separation does not seem to happen - need to investigate...

Edit2: 3EO during first stage is a contingency BLUE abort that so far is simply not implemented. It seems, it needs a conditional to defer the 3EO logic from kicking in when stage 1 is still running...)

Edit3: Should now be possible to separate with 3EO during stage 1 - but it won't give you much joy, you'll tunble out of control and die...
Last edited by Thorsten on Sun Jan 12, 2020 3:42 pm, edited 1 time in total.
Thorsten
 
Posts: 11720
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Bugfixes

Postby enrogue » Sun Jan 12, 2020 3:41 pm

Thorsten wrote in Sat Jan 11, 2020 12:47 pm:Actually, let's narrow down the pathway a bit.


From my observations, whatever is commanded by the joystick at the point the 'hydraulic pressure select' is enabled is locked in & you can't change it with the joystick after that

disabling nasal gc threading makes no difference

Hope that helps

Brendan
User avatar
enrogue
 
Posts: 255
Joined: Mon May 19, 2014 6:40 pm
Location: London (UK)
Callsign: enrogue
OS: Ubuntu, macOS

Re: Space Shuttle - Bugfixes

Postby Thorsten » Sun Jan 12, 2020 4:15 pm

From my observations, whatever is commanded by the joystick at the point the 'hydraulic pressure select' is enabled is locked in & you can't change it with the joystick after that


I've just tried again giving a full stick pull with the keyboard when activating hydraulic pressure select, but I still get a clean liftoff. Again it begs the question - what is different when using the stick? I've also tried changing pitch trim prior to activating the hydraulics with the mouse - clean liftoff.

Could people please check

/fdm/jsbsim/propulsion/engine[0,1,2,3,4]/pitch-angle-rad
/fdm/jsbsim/propulsion/engine[0,1,2,3,4]/pitch-angle-rad-cmd

for a correct vs. screwed-up liftoff?
Thorsten
 
Posts: 11720
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle - Bugfixes

Postby eatdirt » Sun Jan 12, 2020 5:36 pm

I get things equal to 0 before switching on APU, and when APU and hydraulics are ON, I get a value. First two lines are with joystick pulled down, next without.

engine0
pitch-angle-rad= -0.45
pitch-angle-rad-cmd=-0.45
++++++++++++++++++++++++
pitch-angle-rad= -0.33438...
pitch-angle-rad-cmd=-0.33438...

PS: notice that there are no further decimals after 0.45, while the not touching the joystick's values have a lot, same remark below.


engine1:
pitch-angle-rad= -0.42100...
pitch-angle-rad-cmd=-0.42100....
++++++++++++++++++++++++
pitch-angle-rad= -0.26597...
pitch-angle-rad-cmd=-0.26597...


engine2:
pitch-angle-rad= -0.45
pitch-angle-rad-cmd=-0.45
++++++++++++++++++++++++
pitch-angle-rad= -0.42962...
pitch-angle-rad-cmd=-0.42962...


engine3:
pitch-angle-rad= -0.14
pitch-angle-rad-cmd=-0.14
++++++++++++++++++++++++
pitch-angle-rad= -0.064...
pitch-angle-rad-cmd=-0.064...


engine4:
pitch-angle-rad= -0.055...
pitch-angle-rad-cmd=-0.055...
++++++++++++++++++++++++
pitch-angle-rad= +0.069...
pitch-angle-rad-cmd=+0.069
eatdirt
 
Posts: 624
Joined: Wed Aug 15, 2018 2:06 pm

PreviousNext

Return to Spaceflight

Who is online

Users browsing this forum: No registered users and 2 guests