On Wed, 14 Oct 2009, Scott Hamilton wrote:
> I've noticed that a lot of JSBsim files got updated just the other
> day, and today was the first time I've had
> a chance to try it out.
>
> I'm working on the A380 which does use JSBsim, and during the engine
> start everything freezes just after
> the starter is turned false, the ignition is true and the cutoff is
> turned false. This is around N2 = 27%
>
> Fortunately I had compiled everything with debug, so a quick look in
> gdb reveals at stacktrace (on interrupt signal)
Good catch!
I've also (with high probability) run into this using ZLT-NT, but
prefered to blame it on the sound subsystem :) and didn't have time to
investigate it further. Strangely enough it doesn't happen with all
aircraft - I have flown other JSBSim aircraft after the update.
ZLT-NT has an electric motor among the engines, does the A380
also have something like that?
I crosspost this to JSBSim-devel.
Cheers,
Anders
> #0 0x000000000068aeb8 in JSBSim::FGTank::Drain (this=0xc4829a0,
> used=1.0180511166896056e-315) at FGTank.cpp:196
> #1 0x000000000066640f in JSBSim::FGEngine::ConsumeFuel (this=0xc47d4c0)
> at FGEngine.cpp:212
> #2 0x00000000006945aa in JSBSim::FGTurbine::Start (this=0xc47d4c0) at
> FGTurbine.cpp:290
> #3 0x0000000000699d61 in JSBSim::FGTurbine::Calculate (this=0xc47d4c0)
> at FGTurbine.cpp:155
> #4 0x00000000006069c4 in JSBSim::FGPropulsion::Run (this=0xbdf8550) at
> FGPropulsion.cpp:147
> #5 0x00000000005670bc in JSBSim::FGFDMExec::Run (this=0xc573220) at
> FGFDMExec.cpp:362
> #6 0x000000000055949d in FGJSBsim::update (this=0xc413fb0,
> dt=0.041666666666666664) at JSBSim.cxx:487
> #7 0x0000000000435ec8 in fgUpdateTimeDepCalcs () at main.cxx:159
> #8 0x00000000004371df in fgMainLoop () at main.cxx:449
> #9 0x000000000049e1ac in fgOSMainLoop () at fg_os_osgviewer.cxx:172
> #10 0x0000000000433e07 in fgMainInit (argc=9, argv=0x7fffffffda58) at
> main.cxx:900
> #11 0x00000000004332e1 in main (argc=9, argv=0x7fffffffda58) at
> bootstrap.cxx:228
>
>
> 196 double FGTank::Drain(double used)
> 197 {
> 198 double remaining = Contents - used;
> 199
> 200 if (remaining >= 0) { // Reduce contents by amount used.
> 201
> 202 Contents -= used;
> 203 PctFull = 100.0*Contents/Capacity;
> 204
>
> the value for 'used' is 0 (which doesn't seem right)
> the values for 'Contents' and 'remaining' are > 0 (they are correct
> values)
> 'Drain' is called for each tank
>
> It seems this loop in FGEngine.cpp is executed repeatedly with no way
> out;
>
> 183 while (FuelToBurn > 0.0) {
> 184
> 185 // Count how many fuel tanks with the current priority level
> have fuel.
> 186 // If none, then try next lower priority. Build the feed
> list.
> 187 while ((TanksWithFuel == 0.0) && (CurrentPriority <=
> Propulsion->GetNumTanks())) {
> 188 for (i=0; i<Propulsion->GetNumTanks(); i++)
> {
> 189 Tank =
> Propulsion->GetTank(i);
> 190 if (Tank->GetType() == FGTank::ttFUEL)
> {
> 191 if ((Tank->GetContents() > 0.0) && ((unsigned
> int)Tank->GetPriority() == CurrentPriority)) {
> 192 ++TanksWithFuel;
> 193 FeedList.push_back(i);
> 194 }
> 195 } else {
> 196 cerr << "No oxidizer tanks should be used for this
> engine type." << endl;
> 197 }
> 198 }
> 199 if (TanksWithFuel == 0.0) CurrentPriority++;
> 200 }
> 201
> 202 // No fuel found at any priority!
> 203 if (TanksWithFuel == 0.0) {
> 204 Starved = true;
> 205 return;
> 206 }
> 207
> 208 // Remove equal amount of fuel from each feed tank.
> 209 FuelNeeded = FuelToBurn/TanksWithFuel;
> 210 for (i=0; i<FeedList.size(); i++) {
> 211 Tank = Propulsion->GetTank(FeedList[i]);
> 212 Tank->Drain(FuelNeeded);
> 213 FuelToBurn -= FuelNeeded;
> 214 }
> 215
> 216 // check if we were not able to burn all the fuel we needed
> to at this priority level
> 217 if (FuelToBurn > 0.001) {
>
>
> The value of FuelToBurn is = 1.4821969375237396e-323
> The value of CurrentPriority = 1 and never changes.
> once I step past line 217, it goes back to line 183 and never seems to
> stop.
>
>
> I'm not a C++ coder, and not much of a gdb debugger, but I hope that
> helps enough if someone else is seeing similar problems.
> Scott.
>
>
--
---------------------------------------------------------------------------
Anders Gidenstam
WWW: http://www.gidenstam.org/FlightGear/
------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel
