jsbsim-devel — JSBSim development issues

Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

[Bertrand C. <bco...@gm...>]

Le dim. 14 oct. 2018 à 12:16, Bertrand Coconnier <bco...@gm...> a
écrit :

> * The routine to compute the standard temperature in FGStandardAtmosphere
> was using equations above 360,892 ft that were not accounted for in the
> temperature computations in GetTemperature(). That was causing a drift in
> the temperature bias to keep the ambient temperature constant and equal to
> the value requested by FlightGear (-76.5°C/354°R). At some point, the
> temperature was even becoming negative in Kelvin and Rankine !!!
>

I just updated the test suite to detect that bug and pushed a fix to
GitHub. FYI the fix is the same than the one pushed to FlightGear.

Bertrand.


> Le mer. 10 oct. 2018 à 11:19, Alan Teeder <ajt...@v-...> a
> écrit :
>
>>
>>
>> *From:* Sean McLeod
>> *Sent:* Wednesday, October 10, 2018 9:19 AM
>> *To:* Development issues
>> *Subject:* Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017
>>
>>
>> Hi Alan
>>
>>
>>
>> So what was the fix to get it past 150,000ft?
>>
>>
>>
>> Cheers
>>
>>
>>
>>
>>
>> ------------------
>>
>> Sean
>>
>>
>>
>> Bertrand pushed a fix to flightgear. It does not look as if he pushed
>> them to JSBSim itself.
>>
>>
>>
>> Alan
>>
>>
>>
>> -----------------------------------------------------------
>> src/FDM/JSBSim/JSBSim.cxx | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/src/FDM/JSBSim/JSBSim.cxx b/src/FDM/JSBSim/JSBSim.cxx
>> index 106adb63f..004647c81 100644
>> --- a/src/FDM/JSBSim/JSBSim.cxx
>> +++ b/src/FDM/JSBSim/JSBSim.cxx
>> @@ -697,8 +697,8 @@ bool FGJSBsim::copy_to_JSBsim()
>>
>>      Atmosphere->SetTemperature(temperature->getDoubleValue(),
>> get_Altitude(), FGAtmosphere::eCelsius);
>>      Atmosphere->SetPressureSL(FGAtmosphere::eInchesHg,
>> pressureSL->getDoubleValue());
>> -
>> static_cast<FGStandardAtmosphere*>(Atmosphere)->SetDewPoint(FGAtmosphere::eCelsius,
>> -
>> dew_point->getDoubleValue());
>> +    //
>> static_cast<FGStandardAtmosphere*>(Atmosphere)->SetDewPoint(FGAtmosphere::eCelsius,
>> +    //
>> dew_point->getDoubleValue());
>>
>>
>> Winds->SetTurbType((FGWinds::tType)TURBULENCE_TYPE_NAMES[turbulence_model->getStringValue()]);
>>      switch( Winds->GetTurbType() ) {
>>
>> -----------------------------------------
>> .../models/atmosphere/FGStandardAtmosphere.cpp     | 32
>> +++-------------------
>> 1 file changed, 4 insertions(+), 28 deletions(-)
>>
>> diff --git a/src/FDM/JSBSim/models/atmosphere/FGStandardAtmosphere.cpp
>> b/src/FDM/JSBSim/models/atmosphere/FGStandardAtmosphere.cpp
>> index 3e455c7a6..688471215 100644
>> --- a/src/FDM/JSBSim/models/atmosphere/FGStandardAtmosphere.cpp
>> +++ b/src/FDM/JSBSim/models/atmosphere/FGStandardAtmosphere.cpp
>> @@ -254,36 +254,12 @@ double FGStandardAtmosphere::GetTemperature(double
>> altitude) const
>>
>> double FGStandardAtmosphere::GetStdTemperature(double altitude) const
>> {
>> -  double Lk9 = 0.00658368; // deg R per foot
>> -  double Tinf = 1800.0; // Same as 1000 Kelvin
>> -  double temp = Tinf;
>> -
>> -  if (altitude < 298556.4) {                // 91 km - station 8
>> -
>> -    double GeoPotAlt = GeopotentialAltitude(altitude);
>> -
>> -    if (GeoPotAlt >= 0.0)
>> -      temp = StdAtmosTemperatureTable.GetValue(GeoPotAlt);
>> -    else
>> -      temp = StdAtmosTemperatureTable.GetValue(0.0) +
>> GeoPotAlt*LapseRates[0];
>> -
>> -  } else if (altitude < 360892.4) {        // 110 km - station 9
>> -
>> -    temp = 473.7429 - 137.38176 * sqrt(1.0 - pow((altitude -
>> 298556.4)/65429.462, 2.0));
>> -
>> -  } else if (altitude < 393700.8) {        // 120 km - station 10
>> -
>> -    temp = 432 + Lk9 * (altitude - 360892.4);
>> -
>> -  } else if (altitude < 3280839.9) {        // 1000 km station 12
>> -
>> -    double lambda = 0.00001870364;
>> -    double eps = (altitude - 393700.8) * (20855531.5 + 393700.8) /
>> (20855531.5 + altitude);
>> -    temp = Tinf - (Tinf - 648.0) * exp(-lambda*eps);
>> +  double GeoPotAlt = GeopotentialAltitude(altitude);
>>
>> -  }
>> +  if (GeoPotAlt >= 0.0)
>> +    return StdAtmosTemperatureTable.GetValue(GeoPotAlt);
>>
>> -  return temp;
>> +  return StdAtmosTemperatureTable.GetValue(0.0) +
>> GeoPotAlt*LapseRates[0];
>> }
>>
>>
>> //%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>>
>>
>

Re: [Jsbsim-devel] Fw: [Flightgear-devel] RC issues

[Bertrand C. <bco...@gm...>]

Le mar. 16 oct. 2018 à 10:55, Alan Teeder <ajt...@v-...> a écrit :

> Bertrand, Sean.
>
> The shuttle team have put their hands up and found that the problem is
> with
> them , and not with the atmosphere model ;)
>
> Alan
>

Great !

Thanks.

Bertrand.


>
> -----Original Message-----
> From: Thorsten Renk
> Sent: Tuesday, October 16, 2018 7:25 AM
> To: FlightGear developers discussions
> Subject: Re: [Flightgear-devel] RC issues
>
>
> >> I've now hopefully addressed this issue, and am waiting for test
> >> data to confirm whether we're now good or not.
>
> Okay, after some more data and trajectory comparisons - I believe we're
> fine and the atmosphere model is working okay as is and we can go ahead
> with the release (see for a full report below [1]).
>
> * Thorsten
>
> [1] The issue turned out to be that under some (not well understood)
> circumstances the launch autopilot delivered the Shuttle to a 90 mile
> periapsis. This would usually be unproblematic, except that for a low
> inclination launch from KSC, the launch dynamics is such that the
> Shuttle descends after MECO and goes through a very shallow periapsis
> about a quarter of an orbit later. This would be unproblematic, except
> it reaches the equatorial bulge of Earth right at the periapsis, so
> effectively it feels an atmosphere at an altitude 20 km lower. And by
> another a quirk of fate, if you do nothing and hold inertial attitude
> after MECO, you meet this  relatively dense atmosphere at 90 degree AoA
> and maximize drag.
>
> The combination of these factors explains the observed decay from orbit
> and the subsequent uncontrolled atmosphere entry over India.
>
> Now, the unexplained part is that I had tested orbital insertion myself
> in some detail prior to tagging the milestone and did not observe any
> issue with the launch AP - so naturally I assumed when getting reports
> about decaying orbits without state vector data that the launch code
> would be working okay. I do not know the precise combination of factors
> which triggers the bug but hid it from me (and GinGin probably who had
> been flying quite a lot of tests) - in the end the credit goes to
> Jonathan who made the right observation and asked the right questions.
>
> Anyway, we now conducted one test with the changed launch code and one
> with OMS engines applied right after MECO to get into a higher orbit,
> and for both the trajectories check out against the solver.
>
> With the observed orbital decay explained and the trajectories in orbit
> tested, I believe we may now conclude that the atmosphere code after the
> last fix is not at fault. Sorry for any noise, but I believe given the
> initial lack of data, the initial suspicion and sub-sequent
> investigation were a reasonable cause of action.
>
> Thanks to everyone who contributed with the tests!
>
>

[Jsbsim-devel] Fw: [Flightgear-devel] RC issues

[Alan T. <ajt...@v-...>]

Bertrand, Sean.

The shuttle team have put their hands up and found that the problem is with 
them , and not with the atmosphere model ;)

Alan

-----Original Message----- 
From: Thorsten Renk
Sent: Tuesday, October 16, 2018 7:25 AM
To: FlightGear developers discussions
Subject: Re: [Flightgear-devel] RC issues


>> I've now hopefully addressed this issue, and am waiting for test
>> data to confirm whether we're now good or not.

Okay, after some more data and trajectory comparisons - I believe we're
fine and the atmosphere model is working okay as is and we can go ahead
with the release (see for a full report below [1]).

* Thorsten

[1] The issue turned out to be that under some (not well understood)
circumstances the launch autopilot delivered the Shuttle to a 90 mile
periapsis. This would usually be unproblematic, except that for a low
inclination launch from KSC, the launch dynamics is such that the
Shuttle descends after MECO and goes through a very shallow periapsis
about a quarter of an orbit later. This would be unproblematic, except
it reaches the equatorial bulge of Earth right at the periapsis, so
effectively it feels an atmosphere at an altitude 20 km lower. And by
another a quirk of fate, if you do nothing and hold inertial attitude
after MECO, you meet this  relatively dense atmosphere at 90 degree AoA
and maximize drag.

The combination of these factors explains the observed decay from orbit
and the subsequent uncontrolled atmosphere entry over India.

Now, the unexplained part is that I had tested orbital insertion myself
in some detail prior to tagging the milestone and did not observe any
issue with the launch AP - so naturally I assumed when getting reports
about decaying orbits without state vector data that the launch code
would be working okay. I do not know the precise combination of factors
which triggers the bug but hid it from me (and GinGin probably who had
been flying quite a lot of tests) - in the end the credit goes to
Jonathan who made the right observation and asked the right questions.

Anyway, we now conducted one test with the changed launch code and one
with OMS engines applied right after MECO to get into a higher orbit,
and for both the trajectories check out against the solver.

With the observed orbital decay explained and the trajectories in orbit
tested, I believe we may now conclude that the atmosphere code after the
last fix is not at fault. Sorry for any noise, but I believe given the
initial lack of data, the initial suspicion and sub-sequent
investigation were a reasonable cause of action.

Thanks to everyone who contributed with the tests!


_______________________________________________
Flightgear-devel mailing list
Fli...@li...
https://lists.sourceforge.net/lists/listinfo/flightgear-devel 

Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

[Bertrand C. <bco...@gm...>]

Hi,

I'm quite busy at the moment but I plan to open an issue on GitHub about
that issue.

Basically, there was 2 distinct issues :

* The humidity feature needs some additional care to handle very high
altitude flights because, above ~150,000 ft the saturated vapor pressure
can become higher than the ambient pressure and in that case the humidity
rate HR can no longer reach 100% or we get negative density.

* The routine to compute the standard temperature in FGStandardAtmosphere
was using equations above 360,892 ft that were not accounted for in the
temperature computations in GetTemperature(). That was causing a drift in
the temperature bias to keep the ambient temperature constant and equal to
the value requested by FlightGear (-76.5°C/354°R). At some point, the
temperature was even becoming negative in Kelvin and Rankine !!!

Bertrand.

Le mer. 10 oct. 2018 à 11:19, Alan Teeder <ajt...@v-...> a écrit :

>
>
> *From:* Sean McLeod
> *Sent:* Wednesday, October 10, 2018 9:19 AM
> *To:* Development issues
> *Subject:* Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017
>
>
> Hi Alan
>
>
>
> So what was the fix to get it past 150,000ft?
>
>
>
> Cheers
>
>
>
>
>
> ------------------
>
> Sean
>
>
>
> Bertrand pushed a fix to flightgear. It does not look as if he pushed them
> to JSBSim itself.
>
>
>
> Alan
>
>
>
> -----------------------------------------------------------
> src/FDM/JSBSim/JSBSim.cxx | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/src/FDM/JSBSim/JSBSim.cxx b/src/FDM/JSBSim/JSBSim.cxx
> index 106adb63f..004647c81 100644
> --- a/src/FDM/JSBSim/JSBSim.cxx
> +++ b/src/FDM/JSBSim/JSBSim.cxx
> @@ -697,8 +697,8 @@ bool FGJSBsim::copy_to_JSBsim()
>
>      Atmosphere->SetTemperature(temperature->getDoubleValue(),
> get_Altitude(), FGAtmosphere::eCelsius);
>      Atmosphere->SetPressureSL(FGAtmosphere::eInchesHg,
> pressureSL->getDoubleValue());
> -
> static_cast<FGStandardAtmosphere*>(Atmosphere)->SetDewPoint(FGAtmosphere::eCelsius,
> -
> dew_point->getDoubleValue());
> +    //
> static_cast<FGStandardAtmosphere*>(Atmosphere)->SetDewPoint(FGAtmosphere::eCelsius,
> +    //
> dew_point->getDoubleValue());
>
>
> Winds->SetTurbType((FGWinds::tType)TURBULENCE_TYPE_NAMES[turbulence_model->getStringValue()]);
>      switch( Winds->GetTurbType() ) {
>
> -----------------------------------------
> .../models/atmosphere/FGStandardAtmosphere.cpp     | 32
> +++-------------------
> 1 file changed, 4 insertions(+), 28 deletions(-)
>
> diff --git a/src/FDM/JSBSim/models/atmosphere/FGStandardAtmosphere.cpp
> b/src/FDM/JSBSim/models/atmosphere/FGStandardAtmosphere.cpp
> index 3e455c7a6..688471215 100644
> --- a/src/FDM/JSBSim/models/atmosphere/FGStandardAtmosphere.cpp
> +++ b/src/FDM/JSBSim/models/atmosphere/FGStandardAtmosphere.cpp
> @@ -254,36 +254,12 @@ double FGStandardAtmosphere::GetTemperature(double
> altitude) const
>
> double FGStandardAtmosphere::GetStdTemperature(double altitude) const
> {
> -  double Lk9 = 0.00658368; // deg R per foot
> -  double Tinf = 1800.0; // Same as 1000 Kelvin
> -  double temp = Tinf;
> -
> -  if (altitude < 298556.4) {                // 91 km - station 8
> -
> -    double GeoPotAlt = GeopotentialAltitude(altitude);
> -
> -    if (GeoPotAlt >= 0.0)
> -      temp = StdAtmosTemperatureTable.GetValue(GeoPotAlt);
> -    else
> -      temp = StdAtmosTemperatureTable.GetValue(0.0) +
> GeoPotAlt*LapseRates[0];
> -
> -  } else if (altitude < 360892.4) {        // 110 km - station 9
> -
> -    temp = 473.7429 - 137.38176 * sqrt(1.0 - pow((altitude -
> 298556.4)/65429.462, 2.0));
> -
> -  } else if (altitude < 393700.8) {        // 120 km - station 10
> -
> -    temp = 432 + Lk9 * (altitude - 360892.4);
> -
> -  } else if (altitude < 3280839.9) {        // 1000 km station 12
> -
> -    double lambda = 0.00001870364;
> -    double eps = (altitude - 393700.8) * (20855531.5 + 393700.8) /
> (20855531.5 + altitude);
> -    temp = Tinf - (Tinf - 648.0) * exp(-lambda*eps);
> +  double GeoPotAlt = GeopotentialAltitude(altitude);
>
> -  }
> +  if (GeoPotAlt >= 0.0)
> +    return StdAtmosTemperatureTable.GetValue(GeoPotAlt);
>
> -  return temp;
> +  return StdAtmosTemperatureTable.GetValue(0.0) + GeoPotAlt*LapseRates[0];
> }
>
>
> //%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
> .
> _______________________________________________
> Jsbsim-devel mailing list
> Jsb...@li...
> https://lists.sourceforge.net/lists/listinfo/jsbsim-devel
> _______________________________________________
> The JSBSim Flight Dynamics Model project
> http://www.JSBSim.org
> _______________________________________________
>
>

Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

[Alan T. <ajt...@v-...>]

From: Sean McLeod 
Sent: Wednesday, October 10, 2018 9:19 AM
To: Development issues 
Subject: Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

Hi Alan

 

So what was the fix to get it past 150,000ft?

 

Cheers





------------------

Sean



Bertrand pushed a fix to flightgear. It does not look as if he pushed them to JSBSim itself.



Alan



-----------------------------------------------------------

src/FDM/JSBSim/JSBSim.cxx | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/FDM/JSBSim/JSBSim.cxx b/src/FDM/JSBSim/JSBSim.cxx
index 106adb63f..004647c81 100644
--- a/src/FDM/JSBSim/JSBSim.cxx
+++ b/src/FDM/JSBSim/JSBSim.cxx
@@ -697,8 +697,8 @@ bool FGJSBsim::copy_to_JSBsim()

     Atmosphere->SetTemperature(temperature->getDoubleValue(), get_Altitude(), FGAtmosphere::eCelsius);
     Atmosphere->SetPressureSL(FGAtmosphere::eInchesHg, pressureSL->getDoubleValue());
-    static_cast<FGStandardAtmosphere*>(Atmosphere)->SetDewPoint(FGAtmosphere::eCelsius,
-                                                                dew_point->getDoubleValue());
+    // static_cast<FGStandardAtmosphere*>(Atmosphere)->SetDewPoint(FGAtmosphere::eCelsius,
+    //                                                             dew_point->getDoubleValue());

     Winds->SetTurbType((FGWinds::tType)TURBULENCE_TYPE_NAMES[turbulence_model->getStringValue()]);
     switch( Winds->GetTurbType() ) {

-----------------------------------------
.../models/atmosphere/FGStandardAtmosphere.cpp     | 32 +++-------------------
1 file changed, 4 insertions(+), 28 deletions(-)

diff --git a/src/FDM/JSBSim/models/atmosphere/FGStandardAtmosphere.cpp b/src/FDM/JSBSim/models/atmosphere/FGStandardAtmosphere.cpp
index 3e455c7a6..688471215 100644
--- a/src/FDM/JSBSim/models/atmosphere/FGStandardAtmosphere.cpp
+++ b/src/FDM/JSBSim/models/atmosphere/FGStandardAtmosphere.cpp
@@ -254,36 +254,12 @@ double FGStandardAtmosphere::GetTemperature(double altitude) const

double FGStandardAtmosphere::GetStdTemperature(double altitude) const
{
-  double Lk9 = 0.00658368; // deg R per foot
-  double Tinf = 1800.0; // Same as 1000 Kelvin
-  double temp = Tinf;
-
-  if (altitude < 298556.4) {                // 91 km - station 8
-
-    double GeoPotAlt = GeopotentialAltitude(altitude);
-
-    if (GeoPotAlt >= 0.0)
-      temp = StdAtmosTemperatureTable.GetValue(GeoPotAlt);
-    else
-      temp = StdAtmosTemperatureTable.GetValue(0.0) + GeoPotAlt*LapseRates[0];
-
-  } else if (altitude < 360892.4) {        // 110 km - station 9
-
-    temp = 473.7429 - 137.38176 * sqrt(1.0 - pow((altitude - 298556.4)/65429.462, 2.0));
-
-  } else if (altitude < 393700.8) {        // 120 km - station 10
-
-    temp = 432 + Lk9 * (altitude - 360892.4);
-
-  } else if (altitude < 3280839.9) {        // 1000 km station 12
-
-    double lambda = 0.00001870364;
-    double eps = (altitude - 393700.8) * (20855531.5 + 393700.8) / (20855531.5 + altitude);
-    temp = Tinf - (Tinf - 648.0) * exp(-lambda*eps);
+  double GeoPotAlt = GeopotentialAltitude(altitude);

-  }
+  if (GeoPotAlt >= 0.0)
+    return StdAtmosTemperatureTable.GetValue(GeoPotAlt);

-  return temp;
+  return StdAtmosTemperatureTable.GetValue(0.0) + GeoPotAlt*LapseRates[0];
}

//%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

. 

Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

[Sean M. <se...@se...>]

Hi Alan

So what was the fix to get it past 150,000ft?

Cheers

From: Alan Teeder <ajt...@v-...>
Sent: Wednesday, October 10, 2018 10:11 AM
To: Development issues <jsb...@li...>
Subject: Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

Thanks Sean

Yes, it seems that we are seeing the same problems.

The current position with the shuttle is that it now enters orbit successfully, which it did not do before. However the orbit decays rapidly and the vehicle starts tumbling after it passes over central Africa (launched from Florida).
The likely culprit is air density.  I want to check against a standard atmosphere at 400,000 ft altitude.

Alan

Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

[Alan T. <ajt...@v-...>]

Thanks Sean

Yes, it seems that we are seeing the same problems.

The current position with the shuttle is that it now enters orbit successfully, which it did not do before. However the orbit decays rapidly and the vehicle starts tumbling after it passes over central Africa (launched from Florida). 
The likely culprit is air density.  I want to check against a standard atmosphere at 400,000 ft altitude.

Alan

Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

[Sean M. <se...@se...>]

Hi Alan

The python tests appear to fail when run from the Visual Studio test GUI because the command line argument, the path to the JSBSim source isn’t set or is set incorrectly. I haven’t figured out how to fix it yet.

If I’m debugging a python test I create a Visual Studio python project and simply reference the python tests in the test directory and then in the debugging tab I explicitly set a script argument with the path to my jsbsim source directory and then I can step through and debug the python test.

Or I simply run the test on the command line as below.

I have also been able to get things to build, including the JSBSim python components using the regular stand-alone cmake, and in the process of getting that to work I ran into the same linker error you’re seeing.

If you want to see how to do it take a look at the appveyor.yml file in the root of the JSBSim source tree. The main ‘trick’ was to ensure the use of Ninja. You’ll see that ctest is also used to run all the python tests. This was done in order to use the cmake build and to have all the python tests run on every commit to the JSBSim source at github as part of the CI setup we have for both Linux and Windows.

I presume you’re looking into this in order to try and debug the X-15/Shuttle high altitude/high mach bug?

Cheers

From: Alan Teeder <ajt...@v-...>
Sent: Tuesday, October 9, 2018 11:37 PM
To: Development issues <jsb...@li...>
Subject: Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017



From: Sean McLeod
Sent: Tuesday, October 9, 2018 8:16 PM
To: Development issues
Subject: Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

Hi Alan

But your original email listed a linker error while building, this isn’t the equivalent in terms of all is OK except for the python based tests.

After building using the bundled CMake etc. as per the documentation link try this from the command line in terms of running some of the python based tests.

C:\Users\Sean\CMakeBuilds\3f00c6d9-d323-5a32-8a90-665138817fd4\build\x64-Release\tests>python .\TestDensityAltitude.py C:\source\jsbsim\

Obviously replacing C:\Users\Sean\CMakeBuilds\3f00c6d9-d323-5a32-8a90-665138817fd4 with your equivalent.

You should see lots of output scroll by and then:

ok

----------------------------------------------------------------------
Ran 1 test in 5.857s

OK

Cheers

From: Alan Teeder <ajt...@v-...<mailto:ajt...@v-...>>
Sent: Tuesday, October 9, 2018 9:04 PM
To: Development issues <jsb...@li...<mailto:jsb...@li...>>
Subject: Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

Sean

That helped, thanks. When I ran

C:\Users\Alan\CMakeBuilds\4d4a831a-6058-163b-b208-90f00c55faca\build\x64-Release\tests> .\TestPressureAltitude.py C:\
flightgear\jsbsim

The test failed  with a useful error message!
Traceback (most recent call last):
  File "C:\Users\Alan\CMakeBuilds\4d4a831a-6058-163b-b208-90f00c55faca\build\x64-Release\tests\TestPressureAltitude.py", line 21, in <module>
    from JSBSim_utils import JSBSimTestCase, CreateFDM, RunTest
  File "C:\Users\Alan\CMakeBuilds\4d4a831a-6058-163b-b208-90f00c55faca\build\x64-Release\tests\JSBSim_utils.py", line 23, in <module>
    import pandas as pd
ModuleNotFoundError: No module named 'pandas'

So I added “pandas” to Python with pip and then got the output you described. However running the Python based tests from within the Visual Studio GUI still failed.

Also my standalone CMake + JSBSim still fails when building PythonJSBSim. ( that is what I meant by saying it was equivalent).

Alan

Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

[Alan T. <ajt...@v-...>]

From: Sean McLeod 
Sent: Tuesday, October 9, 2018 8:16 PM
To: Development issues 
Subject: Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

Hi Alan

 

But your original email listed a linker error while building, this isn’t the equivalent in terms of all is OK except for the python based tests.

 

After building using the bundled CMake etc. as per the documentation link try this from the command line in terms of running some of the python based tests.

 

C:\Users\Sean\CMakeBuilds\3f00c6d9-d323-5a32-8a90-665138817fd4\build\x64-Release\tests>python .\TestDensityAltitude.py C:\source\jsbsim\

 

Obviously replacing C:\Users\Sean\CMakeBuilds\3f00c6d9-d323-5a32-8a90-665138817fd4 with your equivalent.

 

You should see lots of output scroll by and then:

 

ok

 

----------------------------------------------------------------------

Ran 1 test in 5.857s

 

OK

 

Cheers

 

From: Alan Teeder <ajt...@v-...> 
Sent: Tuesday, October 9, 2018 9:04 PM
To: Development issues <jsb...@li...>
Subject: Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

 

Sean



That helped, thanks. When I ran



C:\Users\Alan\CMakeBuilds\4d4a831a-6058-163b-b208-90f00c55faca\build\x64-Release\tests> .\TestPressureAltitude.py C:\
flightgear\jsbsim

The test failed  with a useful error message!
Traceback (most recent call last):
  File "C:\Users\Alan\CMakeBuilds\4d4a831a-6058-163b-b208-90f00c55faca\build\x64-Release\tests\TestPressureAltitude.py", line 21, in <module>
    from JSBSim_utils import JSBSimTestCase, CreateFDM, RunTest
  File "C:\Users\Alan\CMakeBuilds\4d4a831a-6058-163b-b208-90f00c55faca\build\x64-Release\tests\JSBSim_utils.py", line 23, in <module>
    import pandas as pd
ModuleNotFoundError: No module named 'pandas'

So I added “pandas” to Python with pip and then got the output you described. However running the Python based tests from within the Visual Studio GUI still failed.

Also my standalone CMake + JSBSim still fails when building PythonJSBSim. ( that is what I meant by saying it was equivalent).

Alan

Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

[Sean M. <se...@se...>]

Hi Alan

But your original email listed a linker error while building, this isn’t the equivalent in terms of all is OK except for the python based tests.

After building using the bundled CMake etc. as per the documentation link try this from the command line in terms of running some of the python based tests.

C:\Users\Sean\CMakeBuilds\3f00c6d9-d323-5a32-8a90-665138817fd4\build\x64-Release\tests>python .\TestDensityAltitude.py C:\source\jsbsim\

Obviously replacing C:\Users\Sean\CMakeBuilds\3f00c6d9-d323-5a32-8a90-665138817fd4 with your equivalent.

You should see lots of output scroll by and then:

ok

----------------------------------------------------------------------
Ran 1 test in 5.857s

OK

Cheers

From: Alan Teeder <ajt...@v-...>
Sent: Tuesday, October 9, 2018 9:04 PM
To: Development issues <jsb...@li...>
Subject: Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017



From: Sean McLeod
Sent: Tuesday, October 9, 2018 5:34 PM
To: Development issues
Subject: Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

Hi Alan

Are you performing the CMake build as per - https://jsbsim-team.github.io/jsbsim-reference-manual/mypages/quickstart-building-using-visualstudio/

Have you installed CMake yourself in addition to the version of CMake that is bundled with Visual Studio 2017?

Are you running any CMake commands yourself, e.g. something like:
cmake -G “Visual Studio 15 2017 Win64” ..
Cheers


Sean

I am not using the bundled CMake.«, and open the JSBSim.sln file that the standalone CMake generates using VS 2017.

I tried the bundled CMake system as per your link. There were plenty of compiler warnings, but no reported errors. However when I tried to run the tests all of the python based tests failed.

The result is that both build systems produce identical results – i.e. all is OK except for the python based tests.

Alan

Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

[Alan T. <ajt...@v-...>]

From: Sean McLeod 
Sent: Tuesday, October 9, 2018 5:34 PM
To: Development issues 
Subject: Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

Hi Alan

 

Are you performing the CMake build as per - https://jsbsim-team.github.io/jsbsim-reference-manual/mypages/quickstart-building-using-visualstudio/

 

Have you installed CMake yourself in addition to the version of CMake that is bundled with Visual Studio 2017?

 

Are you running any CMake commands yourself, e.g. something like:

cmake -G “Visual Studio 15 2017 Win64” ..

Cheers

 


Sean



I am not using the bundled CMake.«, and open the JSBSim.sln file that the standalone CMake generates using VS 2017.



I tried the bundled CMake system as per your link. There were plenty of compiler warnings, but no reported errors. However when I tried to run the tests all of the python based tests failed.



The result is that both build systems produce identical results – i.e. all is OK except for the python based tests.



Alan

Re: [Jsbsim-devel] PythonJSBSim link error with VS 2017

[Sean M. <se...@se...>]

Hi Alan

Are you performing the CMake build as per - https://jsbsim-team.github.io/jsbsim-reference-manual/mypages/quickstart-building-using-visualstudio/

Have you installed CMake yourself in addition to the version of CMake that is bundled with Visual Studio 2017?

Are you running any CMake commands yourself, e.g. something like:
cmake -G "Visual Studio 15 2017 Win64" ..
Cheers

From: Alan Teeder <ajt...@v-...>
Sent: Tuesday, October 9, 2018 5:25 PM
To: issues Development <Jsb...@li...>
Subject: [Jsbsim-devel] PythonJSBSim link error with VS 2017

I am getting the following error when tying to build the test suite on Windows with Cmake 12.3 ,  VS 2017  and Python 3.6

Any ideas ?

Alan


1>------ Build started: Project: PythonJSBSim, Configuration: Release x64 ------
1>Compiling Cython CXX source for jsbsim...
1>Building Custom Rule C:/FlightGear/jsbsim/python/CMakeLists.txt
1>CMake does not need to re-run because C:/FlightGear/jsbsim/build64/python/CMakeFiles/generate.stamp is up-to-date.
1>Building Python modules...
1>running build_ext
1>building 'jsbsim' extension
1>CUSTOMBUILD : error : command 'C:\\Program Files (x86)\\Microsoft<file://Microsoft> Visual Studio\\2017\\Community\\VC\\Tools\\MSVC\\14.15.26726\\bin\\HostX86\\x64\\link.exe' failed with exit status 1181
1>Done building project "PythonJSBSim.vcxproj" -- FAILED.

[Jsbsim-devel] PythonJSBSim link error with VS 2017

[Alan T. <ajt...@v-...>]

I am getting the following error when tying to build the test suite on Windows with Cmake 12.3 ,  VS 2017  and Python 3.6

Any ideas ?

Alan


1>------ Build started: Project: PythonJSBSim, Configuration: Release x64 ------
1>Compiling Cython CXX source for jsbsim...
1>Building Custom Rule C:/FlightGear/jsbsim/python/CMakeLists.txt
1>CMake does not need to re-run because C:/FlightGear/jsbsim/build64/python/CMakeFiles/generate.stamp is up-to-date.
1>Building Python modules...
1>running build_ext
1>building 'jsbsim' extension
1>CUSTOMBUILD : error : command 'C:\\Program Files (x86)\\Microsoft Visual Studio\\2017\\Community\\VC\\Tools\\MSVC\\14.15.26726\\bin\\HostX86\\x64\\link.exe' failed with exit status 1181
1>Done building project "PythonJSBSim.vcxproj" -- FAILED.

Re: [Jsbsim-devel] [Flightgear-devel] JSBSim issues in the RC

[Alan T. <ajt...@v-...>]

Sean

All of the reports are from Flightgear, so we can assume that they are not 
stand-alone.

I have a meeting today, so will come back  to you later.

Thanks for the check.

Alan

Re: [Jsbsim-devel] [Flightgear-devel] JSBSim issues in the RC

[Sean M. <se...@se...>]

Hi Alan

I'm very busy work wise at the moment, hence the reason I asked for a reproducible JSBSim only example that I could quickly run and then debug the issue. When I get some time I'll look at getting the JSBSim X-15 to fly up through 150,000ft at realistic Mach numbers to see if I can reproduce the issue.

The X-15 mention below regarding flying the X-15 and seeing the issue, was that JSBSim stand-alone or via FG?

But as a quick test based on the assumption that the issue may be related to the changes we made recently to VcalibratedFromMach() and comments about CAS ending up as NaN I tried the following quickly.

Basically check the CAS calculated for Mach numbers from 5 to 8.5 in 0.1 increments at 150,000ft (pressure - 2.8419), I also ran the test at 160,000ft (pressure - 1.9419) and didn't spot any NaN values for CAS.

  for (double mach = 5.0; mach < 8.5; mach += 0.1)
  {
    double cas = JSBSim::FGJSBBase::VcalibratedFromMach(mach, 2.8419);
    cout << mach << " - " << cas << endl;
  }

Mach - CAS (ft/s)
5 - 273.074
5.1 - 278.542
5.2 - 284.004
5.3 - 289.459
5.4 - 294.907
5.5 - 300.349
5.6 - 305.785
5.7 - 311.213
5.8 - 316.635
5.9 - 322.051
6 - 327.46
6.1 - 332.862
6.2 - 338.257
6.3 - 343.646
6.4 - 349.028
6.5 - 354.403
6.6 - 359.772
6.7 - 365.133
6.8 - 370.488
6.9 - 375.836
7 - 381.177
7.1 - 386.51
7.2 - 391.837
7.3 - 397.157
7.4 - 402.47
7.5 - 407.775
7.6 - 413.074
7.7 - 418.365
7.8 - 423.648
7.9 - 428.925
8 - 434.194
8.1 - 439.456
8.2 - 444.71
8.3 - 449.957
8.4 - 455.196
8.5 - 460.428

Cheers

-----Original Message-----
From: Alan Teeder <ajt...@v-...> 
Sent: Wednesday, October 3, 2018 12:43 PM
To: Development issues <jsb...@li...>; FlightGear developers discussions <fli...@li...>
Subject: Re: [Jsbsim-devel] [Flightgear-devel] JSBSim issues in the RC



-----Original Message-----
From: Sean McLeod
Sent: Wednesday, October 3, 2018 8:55 AM
To: Development issues ; FlightGear developers discussions
Cc: Anders Gidenstam
Subject: Re: [Jsbsim-devel] [Flightgear-devel] JSBSim issues in the RC

Hi

Is there a model (Shuttle, Vostok-1 etc.) that I can run in JSBSim stand-alone with an initialization file and/or script file that reproduces the error?

If so then it means the issue isn't related to an atmosphere interaction issue between JSBSim and FG, although I guess there could be an additional issue along those lines as well. I don't use FG so I wasn't initially aware that there is a separate atmosphere model in FG, I assumed FG simply used the JSBSim atmosphere model.

Cheers

---------------------

Sean

There are Shuttle and X15 models in the JSBSim distribution. Perhaps they will show this problem. If not the versions in Flightgear-FGAddon may be used.

However be aware that making Flightgear  JSASim aero.xml file run in JSBSim standalone can be trivial or it can be very complicated. It depends upon how many properties in the JSBSim aero.xml file are dependant upon the rest of the Flightgear model.

Alan



"Two more test cases - the X-15 actually reaches to 150.000+ ft, so I took it for a spin and... at 150.000 ft the usual properties go NaN - so this definitely looks atmosphere related and not at all Shuttle related." 



_______________________________________________
Jsbsim-devel mailing list
Jsb...@li...
https://lists.sourceforge.net/lists/listinfo/jsbsim-devel
_______________________________________________
The JSBSim Flight Dynamics Model project http://www.JSBSim.org _______________________________________________

Re: [Jsbsim-devel] [Flightgear-devel] JSBSim issues in the RC

[Alan T. <ajt...@v-...>]

-----Original Message----- 
From: Sean McLeod
Sent: Wednesday, October 3, 2018 8:55 AM
To: Development issues ; FlightGear developers discussions
Cc: Anders Gidenstam
Subject: Re: [Jsbsim-devel] [Flightgear-devel] JSBSim issues in the RC

Hi

Is there a model (Shuttle, Vostok-1 etc.) that I can run in JSBSim 
stand-alone with an initialization file and/or script file that reproduces 
the error?

If so then it means the issue isn't related to an atmosphere interaction 
issue between JSBSim and FG, although I guess there could be an additional 
issue along those lines as well. I don't use FG so I wasn't initially aware 
that there is a separate atmosphere model in FG, I assumed FG simply used 
the JSBSim atmosphere model.

Cheers

---------------------

Sean

There are Shuttle and X15 models in the JSBSim distribution. Perhaps they 
will show this problem. If not the versions in Flightgear-FGAddon may be 
used.

However be aware that making Flightgear  JSASim aero.xml file run in JSBSim 
standalone can be trivial or it can be very complicated. It depends upon how 
many properties in the JSBSim aero.xml file are dependant upon the rest of 
the Flightgear model.

Alan



"Two more test cases - the X-15 actually reaches to 150.000+ ft, so I
took it for a spin and... at 150.000 ft the usual properties go NaN - so
this definitely looks atmosphere related and not at all Shuttle related." 

Re: [Jsbsim-devel] [Flightgear-devel] JSBSim issues in the RC

[Sean M. <se...@se...>]

Hi

Is there a model (Shuttle, Vostok-1 etc.) that I can run in JSBSim stand-alone with an initialization file and/or script file that reproduces the error? 

If so then it means the issue isn't related to an atmosphere interaction issue between JSBSim and FG, although I guess there could be an additional issue along those lines as well. I don't use FG so I wasn't initially aware that there is a separate atmosphere model in FG, I assumed FG simply used the JSBSim atmosphere model.

Cheers

-----Original Message-----
From: Anders Gidenstam <and...@gi...> 
Sent: Wednesday, October 3, 2018 9:38 AM
To: FlightGear developers discussions <fli...@li...>; jsb...@li...
Subject: Re: [Jsbsim-devel] [Flightgear-devel] JSBSim issues in the RC

On Wed, 3 Oct 2018, Thorsten Renk wrote:

>
>> I committed the small change to Vostok-1 so that it at least loads on 
>> the launch pad. I noticed afterwards that it doesn't start ok for the 
>> orbit -set file, but maybe there is something special I need to pass 
>> on the command line for that one?
>
>
> I don't think there used to be... only if you want to go into a 
> particular orbit do you need the commandline.
>
> But it may just be that by starting at high altitude you start in a 
> region where the atmosphere bug strikes - which is why it crashes (?) 
> (cf. Richard's experiment with the F-15 at 145.000 ft vs. 150.000 ft)

The main question must be if it is just the vc / ve computation that has broken or if it (also?) is the JSBSim atmosphere model or if it is due to some unfortunate interaction between JSBSim's and FlightGear's atmosphere models.

Looking at the FlightGear git log we have:

2018-09-15: The new JSBSim vc / ve computation added.
2018-07-20: The JSBSim FGSwitch change added. (IMO unrelated to atmosphere
problems)
2018-07-01-
2018-07-07: The new JSBSim FGAtmosphere model added.

I think someone would have noticed earlier if the shuttle or F-15 high altitude flight had been broken since July?
So I think the vs / ve change and/or its interaction with the FG side must be our prime suspect.


Cheers,

Anders
--
---------------------------------------------------------------------------
Anders Gidenstam
WWW: http://github.com/andgi
      http://www.gidenstam.org/FlightGear/


_______________________________________________
Jsbsim-devel mailing list
Jsb...@li...
https://lists.sourceforge.net/lists/listinfo/jsbsim-devel
_______________________________________________
The JSBSim Flight Dynamics Model project
http://www.JSBSim.org
_______________________________________________

Re: [Jsbsim-devel] [Flightgear-devel] JSBSim issues in the RC

[Anders G. <and...@gi...>]

On Wed, 3 Oct 2018, Thorsten Renk wrote:

>
>> I committed the small change to Vostok-1 so that it at least loads on
>> the launch pad. I noticed afterwards that it doesn't start ok for the
>> orbit -set file, but maybe there is something special I need to pass
>> on the command line for that one?
>
>
> I don't think there used to be... only if you want to go into a particular 
> orbit do you need the commandline.
>
> But it may just be that by starting at high altitude you start in a region 
> where the atmosphere bug strikes - which is why it crashes (?) (cf. Richard's 
> experiment with the F-15 at 145.000 ft vs. 150.000 ft)

The main question must be if it is just the vc / ve computation that has 
broken or if it (also?) is the JSBSim atmosphere model or if it is due to 
some unfortunate interaction between JSBSim's and FlightGear's atmosphere 
models.

Looking at the FlightGear git log we have:

2018-09-15: The new JSBSim vc / ve computation added.
2018-07-20: The JSBSim FGSwitch change added. (IMO unrelated to atmosphere 
problems)
2018-07-01-
2018-07-07: The new JSBSim FGAtmosphere model added.

I think someone would have noticed earlier if the shuttle or F-15 high 
altitude flight had been broken since July?
So I think the vs / ve change and/or its interaction with the FG side must 
be our prime suspect.


Cheers,

Anders
-- 
---------------------------------------------------------------------------
Anders Gidenstam
WWW: http://github.com/andgi
      http://www.gidenstam.org/FlightGear/

Re: [Jsbsim-devel] Fw: [Jsbsim-users] Air -fuel ratio (Piston engine)

[Alan T. <ajt...@v-...>]

The forum thread has degenerated into yet another lengthy squabble about something totally unrelated. However my original response to the OP included this:-

“FGPiston.cpp does have code to calculate AFR, but this is commented out. It looks as if it was only used for debugging purposes when the module was written.”

and on that basis the bulk of the work has already been done.
Alan

From: Alan Teeder 
Sent: Sunday, September 9, 2018 9:57 PM
To: issues Development 
Subject: [Jsbsim-devel] Fw: [Jsbsim-users] Air -fuel ratio (Piston engine)

I sent this to jsbsim-users, but probably should have sent it here.

Alan

From: Alan Teeder 
Sent: Sunday, September 9, 2018 9:48 PM
To: JSBSim user questions 
Subject: [Jsbsim-users] Air -fuel ratio (Piston engine)

Is it possible to expose air-fuel ratio to the property tree?
A user on the FG forum (https://forum.flightgear.org/viewtopic.php?f=66&t=34762&sid=3ccba873d448bf34b9e84c643ab8a50b&p=336225#p336225) wants this so that he can trigger exhaust smoke with an over-rich mixture.
Thanks

Alan


--------------------------------------------------------------------------------



--------------------------------------------------------------------------------
_______________________________________________
Jsbsim-users mailing list
Jsb...@li...
https://lists.sourceforge.net/lists/listinfo/jsbsim-users



--------------------------------------------------------------------------------



--------------------------------------------------------------------------------
_______________________________________________
Jsbsim-devel mailing list
Jsb...@li...
https://lists.sourceforge.net/lists/listinfo/jsbsim-devel
_______________________________________________
The JSBSim Flight Dynamics Model project
http://www.JSBSim.org
_______________________________________________

[Jsbsim-devel] Fw: [Jsbsim-users] Air -fuel ratio (Piston engine)

[Alan T. <ajt...@v-...>]

I sent this to jsbsim-users, but probably should have sent it here.

Alan

From: Alan Teeder 
Sent: Sunday, September 9, 2018 9:48 PM
To: JSBSim user questions 
Subject: [Jsbsim-users] Air -fuel ratio (Piston engine)

Is it possible to expose air-fuel ratio to the property tree?
A user on the FG forum (https://forum.flightgear.org/viewtopic.php?f=66&t=34762&sid=3ccba873d448bf34b9e84c643ab8a50b&p=336225#p336225) wants this so that he can trigger exhaust smoke with an over-rich mixture.
Thanks

Alan


--------------------------------------------------------------------------------



--------------------------------------------------------------------------------
_______________________________________________
Jsbsim-users mailing list
Jsb...@li...
https://lists.sourceforge.net/lists/listinfo/jsbsim-users

[Jsbsim-devel] JSBSim Issues on Github

[Sean M. <se...@se...>]

Hi

You may have noticed I haven’t been spamming the JSBSim developers mailing list with my questions/issues over the last couple of months 😉

That’s mainly because most of the issues and even bugs that have been reported by various people recently have all been logged as Github issues. If you’re interested take a look at the open and closed issues that have been raised over the last couple of months.

https://github.com/JSBSim-Team/jsbsim/issues

I’ve just created a new issue that I normally would’ve emailed to the JSBSim developers mailing list. What’s quite useful is that the Github issues support markdown allowing for decent formatting including the embedding of images without the 100KB constraint we have on the mailing list and pretty code formatting for various languages for code snippets.

https://github.com/JSBSim-Team/jsbsim/issues/113

Cheers

Re: [Jsbsim-devel] Trim options in initialisation file

[Sean M. <se...@se...>]

Hi

I submitted a pull request for this which was merged recently.

https://github.com/JSBSim-Team/jsbsim/commit/a10c4385e71e7b0a59025a402996c02efadcaca6

In addition to adding support in initialization files for additional trim options over and above the ‘ground trim’ option the pull request also enables the use of the ‘pullup trim’ option from within script files via simulation/do_simple_trim.

For example the following snippet triggers a call to FGFDMExec::DoTrim(int mode) requesting 3 – pullup.

    <event name="Trim">
      <condition> simulation/sim-time-sec gt 0.1 </condition>
      <set name="simulation/do_simple_trim" value="3"/>

However there was no way to specify the targetNlf to use for the pullup trim via the script, it could only be set via the C++ API, although there is an initialization element <targetNlf>

Now with the pull request the initial condition targetNlf is used to initialize the targetNlf within the FGTrim class. The C++ API can still override targetNlf.

Which means that a script using simulation/do_simple_trim to request a pullup trim can specify the targetNlf to use.

On the initialization front there is also a fix to have <running> -1 </running> work correctly in terms of starting all engines.

- running (-1 for all engines, 0 ... n-1 for specific engines)

Cheers

From: Sean McLeod
Sent: Saturday, July 14, 2018 1:23 PM
To: Development issues <jsb...@li...>
Subject: RE: [Jsbsim-devel] Trim options in initialisation file

Hi

I guess one option which would keep backwards compatibility would be something along these lines:

<trim> 0 | 1 | Longitudinal | Full | Ground | Pullup | Custom | Turn </trim>

0 – no trim, 1 – ground trim etc.

Cheers

From: Sean McLeod <se...@se...<mailto:se...@se...>>
Sent: Friday, July 13, 2018 11:48 PM
To: Development issues <jsb...@li...<mailto:jsb...@li...>>
Subject: Re: [Jsbsim-devel] Trim options in initialisation file

Hi Dave

When I initially saw an example of <trim> 1 </trim> in an initialization file I assumed that I could enter any one the trim options, i.e.

integer equivalent to one of tLongitudinal (0), tFull (1), tGround (2), tPullup (3), tCustom (4), tTurn (5).

But the code in FGInitialCondition only returns a bool based on the trim value from the trim element.

  /** Does initialization file call for trim ?
      @return true if initialization file (version 1) called for trim. */
  bool NeedTrim(void) const { return needTrim == 0 ? false : true; }

Which is then used in JSBSim.cpp with DoTrim() defaulting to tGround.

  FDMExec->RunIC();

  if (FDMExec->GetIC()->NeedTrim()) {
    trimmer = new JSBSim::FGTrim( FDMExec );
    try {
      trimmer->DoTrim();
      delete trimmer;
    } catch (string& msg) {
      cerr << endl << msg << endl << endl;
      exit(1);
    }

Hence the reason that currently the only trim option available from the initialization file is tGround.

I’d like to be able to specify any of the trim options from the initialization file. One thing to consider though would be maintaining backwards compatibility for existing initialization files.

Cheers

From: David Culp <dp...@gm...<mailto:dp...@gm...>>
Sent: Friday, July 13, 2018 9:57 PM
To: Development issues <jsb...@li...<mailto:jsb...@li...>>
Subject: Re: [Jsbsim-devel] Trim options in initialisation file

As I recall, this was added by me to support QtJSBSim.  I'm away on vacation, so I can't look it up right now.

Dave

On Fri, Jul 13, 2018, 12:48 PM Sean McLeod <se...@se...<mailto:se...@se...>> wrote:
Hi

Is there a specific reason why the only trim options available in the initialization file are ‘ground trim’  and no trim?

- trim (0 for no trim, 1 for ground trim)

Cheers

Re: [Jsbsim-devel] Trim options in initialisation file

[Sean M. <se...@se...>]

Hi

I guess one option which would keep backwards compatibility would be something along these lines:

<trim> 0 | 1 | Longitudinal | Full | Ground | Pullup | Custom | Turn </trim>

0 – no trim, 1 – ground trim etc.

Cheers

From: Sean McLeod <se...@se...>
Sent: Friday, July 13, 2018 11:48 PM
To: Development issues <jsb...@li...>
Subject: Re: [Jsbsim-devel] Trim options in initialisation file

Hi Dave

When I initially saw an example of <trim> 1 </trim> in an initialization file I assumed that I could enter any one the trim options, i.e.

integer equivalent to one of tLongitudinal (0), tFull (1), tGround (2), tPullup (3), tCustom (4), tTurn (5).

But the code in FGInitialCondition only returns a bool based on the trim value from the trim element.

  /** Does initialization file call for trim ?
      @return true if initialization file (version 1) called for trim. */
  bool NeedTrim(void) const { return needTrim == 0 ? false : true; }

Which is then used in JSBSim.cpp with DoTrim() defaulting to tGround.

  FDMExec->RunIC();

  if (FDMExec->GetIC()->NeedTrim()) {
    trimmer = new JSBSim::FGTrim( FDMExec );
    try {
      trimmer->DoTrim();
      delete trimmer;
    } catch (string& msg) {
      cerr << endl << msg << endl << endl;
      exit(1);
    }

Hence the reason that currently the only trim option available from the initialization file is tGround.

I’d like to be able to specify any of the trim options from the initialization file. One thing to consider though would be maintaining backwards compatibility for existing initialization files.

Cheers

From: David Culp <dp...@gm...<mailto:dp...@gm...>>
Sent: Friday, July 13, 2018 9:57 PM
To: Development issues <jsb...@li...<mailto:jsb...@li...>>
Subject: Re: [Jsbsim-devel] Trim options in initialisation file

As I recall, this was added by me to support QtJSBSim.  I'm away on vacation, so I can't look it up right now.

Dave

On Fri, Jul 13, 2018, 12:48 PM Sean McLeod <se...@se...<mailto:se...@se...>> wrote:
Hi

Is there a specific reason why the only trim options available in the initialization file are ‘ground trim’  and no trim?

- trim (0 for no trim, 1 for ground trim)

Cheers
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot_______________________________________________
Jsbsim-devel mailing list
Jsb...@li...<mailto:Jsb...@li...>
https://lists.sourceforge.net/lists/listinfo/jsbsim-devel
_______________________________________________
The JSBSim Flight Dynamics Model project
http://www.JSBSim.org
_______________________________________________

Re: [Jsbsim-devel] Trim options in initialisation file

[Sean M. <se...@se...>]

Hi Dave

When I initially saw an example of <trim> 1 </trim> in an initialization file I assumed that I could enter any one the trim options, i.e.

integer equivalent to one of tLongitudinal (0), tFull (1), tGround (2), tPullup (3), tCustom (4), tTurn (5).

But the code in FGInitialCondition only returns a bool based on the trim value from the trim element.

  /** Does initialization file call for trim ?
      @return true if initialization file (version 1) called for trim. */
  bool NeedTrim(void) const { return needTrim == 0 ? false : true; }

Which is then used in JSBSim.cpp with DoTrim() defaulting to tGround.

  FDMExec->RunIC();

  if (FDMExec->GetIC()->NeedTrim()) {
    trimmer = new JSBSim::FGTrim( FDMExec );
    try {
      trimmer->DoTrim();
      delete trimmer;
    } catch (string& msg) {
      cerr << endl << msg << endl << endl;
      exit(1);
    }

Hence the reason that currently the only trim option available from the initialization file is tGround.

I’d like to be able to specify any of the trim options from the initialization file. One thing to consider though would be maintaining backwards compatibility for existing initialization files.

Cheers

From: David Culp <dp...@gm...>
Sent: Friday, July 13, 2018 9:57 PM
To: Development issues <jsb...@li...>
Subject: Re: [Jsbsim-devel] Trim options in initialisation file

As I recall, this was added by me to support QtJSBSim.  I'm away on vacation, so I can't look it up right now.

Dave

On Fri, Jul 13, 2018, 12:48 PM Sean McLeod <se...@se...<mailto:se...@se...>> wrote:
Hi

Is there a specific reason why the only trim options available in the initialization file are ‘ground trim’  and no trim?

- trim (0 for no trim, 1 for ground trim)

Cheers
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot_______________________________________________
Jsbsim-devel mailing list
Jsb...@li...<mailto:Jsb...@li...>
https://lists.sourceforge.net/lists/listinfo/jsbsim-devel
_______________________________________________
The JSBSim Flight Dynamics Model project
http://www.JSBSim.org
_______________________________________________

Re: [Jsbsim-devel] Trim options in initialisation file

[David C. <dp...@gm...>]

As I recall, this was added by me to support QtJSBSim.  I'm away on
vacation, so I can't look it up right now.

Dave

On Fri, Jul 13, 2018, 12:48 PM Sean McLeod <se...@se...> wrote:

> Hi
>
>
>
> Is there a specific reason why the only trim options available in the
> initialization file are ‘ground trim’  and no trim?
>
>
>
> - trim (0 for no trim, 1 for ground trim)
>
>
>
> Cheers
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> Jsbsim-devel mailing list
> Jsb...@li...
> https://lists.sourceforge.net/lists/listinfo/jsbsim-devel
> _______________________________________________
> The JSBSim Flight Dynamics Model project
> http://www.JSBSim.org
> _______________________________________________
>
>