fdm.do_trim(1) and fdm['ic/gamma-deg']

[fernandafenelon]

Hello everyone,

I am trying to trim my aircraft with a sideslip angle of beta=14.7° which corresponds to a bank angle of phi=22°.
As inputs I use beta and gamma. I ask for gamma = 0°.
However after using fdm.do_trim(1), I end up with a gamma of -5°.

I expected to have a gamma of 0.0°.

Do you have any thoughts on this?

fdm['ic/beta-deg']=14.7
fdm['ic/gamma-deg'] = 0.0
fdm.do_trim(1)

fdm['flight-path/gamma-deg']
Out[5]: -5.466039161541537

fdm['aero/beta-deg']
Out[6]: 14.853146650331803

[seanmcleod]

How sure are you that the trim succeeds for this case?

For example take a look at my code at - #519 (comment)

In addition to checking for a trim exception you'll also see the trim results in the console like below. Particularly if you follow the thread of the comment link above you'll see that at some point at higher beta values it runs out of aileron power.

I haven't double-checked, but maybe fdm.do_trim(1) doesn't throw an exception but rather returns a result/error code?

  Full Trim

  Trim successful
  Trim Results: 
       Angle of Attack:   6.46  wdot:  9.05e-05 Tolerance: 1.000000e-03  Passed
              Throttle:   0.56  udot: -9.00e-05 Tolerance: 1.000000e-03  Passed
            Pitch Trim:  -0.41  qdot: -1.25e-10 Tolerance: 1.000000e-04  Passed
            Roll Angle:   0.00  vdot: -4.34e-16 Tolerance: 1.000000e-03  Passed
              Ailerons:   0.00  pdot:  9.18e-18 Tolerance: 1.000000e-04  Passed
                Rudder:   0.00  rdot: -4.53e-32 Tolerance: 1.000000e-04  Passed

  Full Trim

  Trim failed
  Trim Results:
       Angle of Attack:   5.56  wdot:  2.36e+01 Tolerance: 1.000000e-03  Failed
              Throttle:   0.48  udot:  1.42e+00 Tolerance: 1.000000e-03  Failed
            Pitch Trim:  -0.37  qdot:  1.24e-01 Tolerance: 1.000000e-04  Failed
            Roll Angle:  26.21  vdot: -1.42e+01 Tolerance: 1.000000e-03  Failed
              Ailerons:   0.00  pdot: -1.50e-01 Tolerance: 1.000000e-04  Failed
                Rudder:   0.00  rdot:  8.60e-02 Tolerance: 1.000000e-04  Failed

[fernandafenelon]

Well, it looks like the trimming was successful. When it is not successful I do not get any results here.
Also, this output you presented, I do not know how to print it here.

The funny thing is, if I ask for gamma=5.5°, then it returns a gamma=0°:

fdm['ic/beta-deg']=14.7
fdm['ic/gamma-deg'] = 5.5
fdm.do_trim(1)

fdm['flight-path/gamma-deg']
Out[3]: 0.026245560202056723

fdm['aero/beta-deg']
Out[4]: 14.85273617944729

[seanmcleod]

Take my code from the comment and cut and paste it into a file, e.g. steadyheadingsideslip.py

Then run it like this:

python steadyheadingsideslip.py

You'll definitely then see the console output with the trim results etc.

Also people generally post so little of their code that it's often a guessing game. So post a full example like I have in the comment I referenced and show us the console output.

For example I'm not sure if your code snippets above include a call to fdm.run_ic() # Initialize the aircraft with initial conditions like my example code does, because you don't or you've just omitted it from your question in the post.

[fernandafenelon]

Sorry for not providing the full code before. Here it is:

fdm = jsbsim.FGFDMExec('my_model_path')
aircraft = f'my_aircraft'
fdm.load_model(f'{aircraft}')

fdm['aero/alpha-max-rad'] = math.radians(20.0)
fdm['aero/alpha-min-rad'] = math.radians(-16.0)

fdm['ic/h-agl-ft'] = 10000
fdm['ic/h-sl-ft'] = 10000
fdm['ic/vt-kts'] = 233.8
fdm['ic/beta-deg'] = 14.7 
fdm['ic/gamma-deg'] = 5.5   
fdm['fcs/TrimWithHTorEle'] = 0.0 # Trim with HT (0 for HT; 1 for Elev)

fdm.do_trim(1)                   # Do Trim

gamma_final = fdm['flight-path/gamma-deg']
beta_final = fdm['aero/beta-deg']

print(f'Gamma_final={gamma_final} , Beta_final = {beta_final}')

Output:

 Full Trim

  Trim successful
  Trim Results:
       Angle of Attack:   1.61  wdot: -5.30e-05 Tolerance: 1e-03  Passed
              Throttle:   0.57  udot: -5.76e-06 Tolerance: 1e-03  Passed
            Pitch Trim:  -0.15  qdot:  6.88e-07 Tolerance: 1e-04  Passed
            Roll Angle:  22.56  vdot:  7.13e-05 Tolerance: 1e-03  Passed
              Ailerons:  -0.43  pdot:  5.49e-06 Tolerance: 1e-04  Passed
                Rudder:   0.99  rdot: -2.45e-09 Tolerance: 1e-04  Passed
Gamma_final=0.07857117367558498 , Beta_final = 14.783379940721945

If I change fdm['ic/gamma-deg'] = 5.5 to fdm['ic/gamma-deg'] = 0.0:

  Full Trim

  Trim successful
  Trim Results:
       Angle of Attack:   1.64  wdot: -2.83e-05 Tolerance: 1e-03  Passed
              Throttle:   0.34  udot: -2.02e-05 Tolerance: 1e-03  Passed
            Pitch Trim:  -0.14  qdot:  6.79e-07 Tolerance: 1e-04  Passed
            Roll Angle:  22.47  vdot:  7.73e-05 Tolerance: 1e-03  Passed
              Ailerons:  -0.44  pdot:  6.05e-06 Tolerance: 1e-04  Passed
                Rudder:   1.00  rdot: -5.62e-10 Tolerance: 1e-04  Passed
Gamma_final=-5.4639108956964595 , Beta_final = 14.853259764192138

No difference arose when I added "fdm.run_ic()" before the fdm.do_trim(1)

[seanmcleod]

Hmm, seeing a gamma final of -5.5 almost sounds like the sign issue that we fixed in this commit 95b4c86 a while ago. Although you're saying setting ic/gamma = 0 is ending up with a gamma of -5.5?

[fernandafenelon]

Yes, I am saying setting ic/gamma = 0 is ending up with a gamma of -5.5.
I don't think it is the same issue because back then, I was able to get the correct value output when I used fdm['flight-path/gamma-deg']. The problem was when I tried to print fdm['ic/gamma-deg'].

[seanmcleod]

So I modified my 737 example and set gamma = 5 and recorded and plotted the resulting gamma.

for beta in np.arange(0, 14, 1):
    fdm['ic/h-sl-ft'] = 1000
    fdm['ic/vc-kts'] = 200
    fdm['ic/gamma-deg'] = 5
    fdm['ic/beta-deg'] = beta
...
...
    # Trim
    try:
        fdm['simulation/do_simple_trim'] = 1
        betas.append(fdm['aero/beta-deg'])
        bankAngle.append(fdm['attitude/phi-deg'])
        ailerons.append(fdm['fcs/aileron-cmd-norm'])
        rudder.append(fdm['fcs/rudder-cmd-norm'])
        gammas.append(fdm['flight-path/gamma-deg'])

So there is something going on in terms of the trim with the combination of gamma and bank angle.

[seanmcleod]

And the result with gamma = 0.

[fernandafenelon]

That is what I am observing with my aircraft... It does not seem physical for me.
It is not a big issue, but it would be nice to understand...

[seanmcleod]

Take a look at the trim code.

The trim code will adjust the following in order to achieve [u-dot, v-dot, w-dot] = 0 and [p-dot, q-dot, r-dot] = 0, which is what you see when the successful trim results are reported. There is no guarantee that the initial gamma will be kept.

      case tFull:
        if (debug_lvl > 0)
          cout << "  Full Trim" << endl;
        TrimAxes.push_back(FGTrimAxis(fdmex,&fgic,tWdot,tAlpha));
        TrimAxes.push_back(FGTrimAxis(fdmex,&fgic,tUdot,tThrottle ));
        TrimAxes.push_back(FGTrimAxis(fdmex,&fgic,tQdot,tPitchTrim ));
        //TrimAxes.push_back(FGTrimAxis(fdmex,&fgic,tHmgt,tBeta ));
        TrimAxes.push_back(FGTrimAxis(fdmex,&fgic,tVdot,tPhi ));
        TrimAxes.push_back(FGTrimAxis(fdmex,&fgic,tPdot,tAileron ));
        TrimAxes.push_back(FGTrimAxis(fdmex,&fgic,tRdot,tRudder ));
        break;