Free Flight, Free Speech

First we need to plan what we are going to do.
A) Do we want a "realistic" simulation with our "amateur" efforts?
B) Do we want to code a battlefield 3 style some-kind of a helicopter?
Mine is A and i respect your decisions. We need to push forward and go beyond our limits.

At the end, i achieved a stable takeoff. It was reaaalllllyyy hard to stay stable on hover. But, this model is almost perfect in forward flight except these cases.
1-This huey is under-damped on yaw axis, means yaw stability is under damped. Maybe the vertical fin on the tail is not coded or mis-coded.
2-Tail rotor produces more thrust than it has to. I cannot roll left, only 30 degrees. Also helicopter response to right cyclic input for rolling right is too fast. This situation confirms the over-thrust on tail rotor.
3- I guess stabilizer horizontal bar on the main rotor is not modelled. This reduces stability of the main rotor. You may consider to increase gains of the main rotor controller.
4- a. You said the turbine model is ok. Yes, the rpm stay still. But when you collective up very fast, rpm will drop dramatically at a moment.
b. RPM oscillates around %1 all the time.

With a weaker tail rotor and a little more stabilization , this model will rock.

bell_pilot wrote:
At the end, i achieved a stable takeoff. It was reaaalllllyyy hard to stay stable on hover. But, this model is almost perfect in forward flight except these cases.
.

I'd hoped that was the case

bell_pilot wrote:1-This huey is under-damped on yaw axis, means yaw stability is under damped. Maybe the vertical fin on the tail is not coded or mis-coded.

At low speeds the dynamic pressure is also low, so I wonder if the rotor downwash augments the dynamic pressure that the v-stab see's and thus creates a greater AoA during yaw rotations, thus damping the rotation.

bell_pilot wrote:2-Tail rotor produces more thrust than it has to. I cannot roll left, only 30 degrees. Also helicopter response to right cyclic input for rolling right is too fast. This situation confirms the over-thrust on tail rotor.

I'm not sure that's due to tail rotor thrust, as it's moment (thrust x distance from main rotor) has to equal and be negative to main rotor torque or you'll be constantly spinning

bell_pilot wrote:3- I guess stabilizer horizontal bar on the main rotor is not modelled. This reduces stability of the main rotor. You may consider to increase gains of the main rotor controller.

confused with the sentence, are you saying you want more or less roll/pitch stability ?
I assume here the stabiliser bar effects both axis.

bell_pilot wrote:4- a. You said the turbine model is ok. Yes, the rpm stay still. But when you collective up very fast, rpm will drop dramatically at a moment.
b. RPM oscillates around %1 all the time.

a) Is that not normal ? I know on this helicopter it has a mechanical link to help prevent governor lag
b) Hunting is a natural effect of the control system, which is internal to JSBsim... Could we live with the hunting at present and resolve some of the other issues, maybe return to it ?

bell_pilot wrote:With a weaker tail rotor and a little more stabilization , this model will rock.

More stabilisation... the answer to 3)

The tail rotor is the only concern I have here.
I'll increase the stabilisation of the axis, we'll see the results and then see what needs doing after that.

Thank you very much

Simon

I was thinking... it might be best with damping adjustment to make a property that you can adjust 'on the fly' until you get the right result and then we can hard-wire this value in.

And I notice Thorsten on the other forum is trying to prevent them from reviewing the fdm.

I'd say something inappropriate at this point, but I just hope that one day people wake up to the damage he does

bomber wrote:I was thinking... it might be best with damping adjustment to make a property that you can adjust 'on the fly' until you get the right result and then we can hard-wire this value in.

This is what we need. Superb idea if you can code it for me as i am not familiar yet with flight gear structure.

bomber wrote: confused with the sentence, are you saying you want more or less roll/pitch stability ?
I assume here the stabiliser bar effects both axis.

I am not sure yet. Maybe the only problem is tail rotor. Sorry for my writing mistakes. We will figure it out after tail rotor update. My suggestion is leave it as is for now.

bomber wrote: a) Is that not normal ? I know on this helicopter it has a mechanical link to help prevent governor lag

For normal inputs yes. But for rapid collective up, "droop compensator" can not hold rpm at %100. You may get "low rotor rpm" warning sometimes.

bomber wrote:b) Hunting is a natural effect of the control system, which is internal to JSBsim... Could we live with the hunting at present and resolve some of the other issues, maybe return to it ?

This is not the subject of this topic Hunting is reason of bad PID tunning. Struggling with tuning right now is not necessary right now. Agreed and passed away.

So as i understand, our next steps are:
bomber- Try to update tail rotor
bomber- prepare tunning with hand gump or similar thing.
bell_pilot-Tune stabillity coefficients.
IAHM-COL-Enjoy

bell_pilot wrote:With a weaker tail rotor and a little more stabilization , this model will rock.

My assessment of the FDM is quite contrary, but since you're a RL pilot and I'm not, I'd assume that my emphases are in the wrong areas.
But I still have the following questions:

1.) Is it realistic, when I'm on the ground with zero collective and I move the cyclic that then the helicopter tips over. My understanding is that with zero collective the blade pitch should be at or near zero and therefore there shouldn't be enough thrust to tip the helicopter over.

2.) ETL is not modelled. Maybe you haven't noticed it. But assuming you already have noticed it and since you haven't mentioned it in your review, does it mean you regard ETL as a not so important effect and therefore it must not be modelled?

I'd assume we're attempting to resolve the big issues first as these have the most impact. Once this is done this particular effect of ETL can be determined and added, although I added vortex ring state as I thought it'd be fun.

I assume yasim flight models and the other helicopter fdms have the ETL modelled to your satisfaction, maybe you could investigate their percentage of thrust reduction at 0 speed, 0 descent.

Regards

Simon

mue wrote:1.) Is it realistic, when I'm on the ground with zero collective and I move the cyclic that then the helicopter tips over. My understanding is that with zero collective the blade pitch should be at or near zero and therefore there shouldn't be enough thrust to tip the helicopter over.

I havent realized that while i have focused on hover and forward flight stability more. I guess this case you mentioned can be easily resolved by some little tunings if it still exist after the tail rotor or related stabilty updates.

mue wrote:2.) ETL is not modelled. Maybe you haven't noticed it. But assuming you already have noticed it and since you haven't mentioned it in your review, does it mean you regard ETL as a not so important effect and therefore it must not be modelled?

I just learned that ETL is needed to model individually. I assumed vortex ring state model was enough for representing ETL for JSBsim FDMs.

These cases you mentioned are noted for further uptades.
Thanks for your feedback. Keep up with the updates.

1) is a good question that I don't know the answer too in the same way as mue doesn't, we've never flown a helicopter or wiggled the cyclic when the collective is in idle.

And that brings me to the collective in idle position, pitch blade of zero and thrust of zero don't occur at the same point, especially with a 17 deg rotor blade twist.

I would also like to know if there's a jump, like going through a gate when leaving the idle position, or is it a linear progression, I can model either.

I have to put something up for others more knowledgeable to review, if that review is done in good faith with a view to improvement I have no problems with it. If however it's done aggressively, to put someone down, be it me or another I can't work with this type of person.

Thorsten is a classic example of someone that does this and as such we do not see eye to eye.

Simon

Looks like he went for trying to put someone down over on the other forum.

Well what did I expect