Centric nature of the cyclic in the Gazelle.

Okay this is one of those questions that I need to ask.

One of the most noticeable nuances between the Gazelle and other helicopters is the centric nature of the cyclic. It's one of those things that seems odd and almost as if it is a design feature that might say make using a joystick with centre detent so much easier as all the action is around that centre. It may well be that this is the way it is however I doubt it.

I fly the Gazelle almost exclusively with AP off, Trim off and sometimes also the Gyro off but I'm not real sure the latter makes a difference?? Also AP on or AP off does not seem to change this attribute just the way the Gazelle handles.

The "feel" seems about right if I ignore the the actual position of the cyclic in that input to the cyclic does what is expected however there is no general offset in the position of the cyclic to balance the aircraft from pick up to flight the cyclic is just centred and only slight blips or movement to correct as needed. Same holds true for set down and hover.

Also in forward flight there is no offset in the cyclic position once you balance the aircraft it seems to move forward (or what direction you want to fly) with the cyclic bought back to the centre.

I have no centre detent or spring action on my cyclic just some dampers that provide some resistance to movement and hold the cyclic where I leave it, (look Ma no hands :D) in effect I do not need to use trim. Also no curves so the cyclic is fairly sensitive but does have about 250 mm extension.

An oddity to this behaviour is pickup or set down on an uneven surface you need to bias the position of the cyclic away from centre to get the lift vector correct as I would expect.

Firstly, the gyro makes a big difference and ensures the AP is fully off.

Secondly.

Attitude, pitch /roll, 0/0, no SAS, no gyro, no trim, stick centered, rudder centered, flight straight and level, hands off, speed = 69 kts.

Increase pitch

" , -1 deg. speed = 82 kts.

" , -2 = 94kts.

" , -3 = 98 kts.

" , -4 = 106 kts.

" , -5 = 114 kts.

Direction will remain constant, even if you pull collective repeatedly, top to bottom. Rudder not used at any time. All conditions remaining the same.

Make your own mind up about all that.

..

Firstly, the gyro makes a big difference and ensures the AP is fully off.

 

Secondly.

 

Attitude, pitch /roll, 0/0, no SAS, no gyro, no trim, stick centered, rudder centered, flight straight and level, hands off, speed = 69 kts.

 

Increase pitch

 

" , -1 deg. speed = 82 kts.

" , -2 = 94kts.

" , -3 = 98 kts.

" , -4 = 106 kts.

" , -5 = 114 kts.

 

Direction will remain constant, even if you pull collective repeatedly, top to bottom. Rudder not used at any time. All conditions remaining the same.

 

Make your own mind up about all that.

 

..

Are you really talking about 2.1 latest release (and patched FM)????

Did exactly what you described (all OFF, no rudder, all centered and checked) with totally different results:

1. as you pick up speed (above 69km/h) direction changes to the right (or straightens out if you like) (because airflow over tail comes into effect - expected)

2. when you lower collective helicopter rolls to right, nose down then changes direction to the right

3. when you raise collective opposite to #2

4. never-ever hands-off possible with all channels OFF below 69 km/h, after picking speed above that margin, hands off for a short while possible (stabilization of tail).

Condition: standard atmosphere, no winds, 50% fuel, no weapons.

Similar result as you got DocWilly with 2.1 latest. Didn't get a chance to fire up 1.5.7 ver but the old version might well behave that way, I'll have to check but sounds about right.

Although it did accelerate a lot with very little power input even if I had to put input in.

What I do find now is that I am finding the Gazelle much more difficult to keep level in it's current form with and without AP etc.

Tested it 10 minutes ago and have nearly the same results as docwilly. You are sure you patched your game Holbeach?

Are you really talking about 2.1 latest release (and patched FM)????

 

 

4. never-ever hands-off possible with all channels OFF below 69 km/h, after picking speed above that margin, hands off for a short while possible (stabilization of tail).

 

Actually you can if you get her nicely balanced no further changes in input and it will follow for a goodly while but will eventually succumb to "other" influences but that's from a few seconds from stable hover to several seconds at low speed flight. It is a slippery beast I get that.

This is not so much what I'm asking about as really Huey and Mi8 do this very well also. What I am asking here is why is the Gazelle 's cyclic centered for pretty much all it's flight stages.

Pick up from flat ground cyclic electrically centered.

Forward (any direction really) low speed flight just into ETL cyclic electrically centered (20 to 40Kph)

Set down cyclic electrically centered

Of course you will need to put small inputs into cyclic but all it's action is over center unlike any other helicopter.

I get it that the cyclic can be centered at nominal forward cruse speed depending on CG etc that's okay. However the Gazelle still lives in this continuous state of being perfectly balanced.

That's what I am asking about. :):pilotfly:

@ Fragbum.

Would you mind specifying which version you are in, as you never mentioned it in your OP.

I am in 1.5.7 10175. release version, not 2.1 alpha.

..

@ Fragbum.

 

Would you mind specifying which version you are in, as you never mentioned it in your OP.

 

I am in 1.5.7 10175. release version, not 2.1 alpha.

 

..

He is flying the changed flight model in the latest 2.1 release as we others do...

1.5.7 will be updated later, maybe today...

He is flying the changed flight model in the latest 2.1 release as we others do...

1.5.7 will be updated later, maybe today...

I will be looking forward to trying it out.

..

@ Fragbum.

 

Would you mind specifying which version you are in, as you never mentioned it in your OP.

 

I am in 1.5.7 10175. release version, not 2.1 alpha.

 

..

Okay yes I should have added that it is the updated version, apart from the fact the Gazelle now has some new flight attributes and some improved characteristics and maybe some not yet implemented changes. I haven't broken one yet doesn't mean I haven't nearly broken one. :D

I think it's great that he Gazelle is moving forward

As for that exercise I had to apply inputs to hold level flight and also when accelerating with fairly low power. I'm not so sure about that still trying to follow induced flow.

One thing I find with the current "new"version is that level flight seems odd well at least obtaining it even after dialling in some saturation on pitch. As soon as I think I have the pitch attitude right the bias flips one way or another then it's bring pitch attitude back in.

I don't find any real evidence of this trait in roll except when getting her leaning over to nearly 90deg. :thumbup:

As for the centric cyclic question both versions <1.5.x 2.1.x> I have flown over about the last 12 months or so, have had that trait.

Cheers

Frag

1.5.7 will be updated later, maybe today...

The fix for the VCB mode switch is in today's 1.5.7.10872 update. I haven't checked to see if the flight model changes are in.

Alright some good input about other oddities and scheduled updates and such however none of that actually addresses my question regarding the centric nature of the Gazelles cyclic.

So is this a feature that makes using a joystick with it's centre detent easier at the expense of actually shifting the balance point and hence the position of the cyclic?

(If it is could we have an control input selector 1) for joystick and 2) for cyclic input.)

Or

Is it simply something that is not fully implemented yet?

Not bashing just curious and I really appreciate the changes to the Gazelle so far. :thumbup:

The most flagrant example of this is to do the following--very simple--test in 2.1:

With gyros and A/P channels off lift off into a hands off hover. Notice, as expected, the lack of roll. Settle back down and move the cyclic fully to the left and hit the force trim. Take your hand off the cyclic and lift off into a hover again. Same lack of roll, right? In fact, nothing changed. The cyclic had repositioned. The cntl-enter indicator repositioned. But the helicopter acted as if nothing happened. Put the skids down again and do the same test to the right with the same results. (Fore and aft cyclic behave more correctly.)

[EDIT] I believe that this is a known issue. I feel pretty confident that Polychop will get the Gazelle into shape in the next month or two.

The most flagrant example of this is to do the following--very simple--test in 2.1:

 

With gyros and A/P channels off lift off into a hands off hover. Notice, as expected, the lack of roll. Settle back down and move the cyclic fully to the left and hit the force trim. Take your hand off the cyclic and lift off into a hover again. Same lack of roll, right? In fact, nothing changed. The cyclic had repositioned. The cntl-enter indicator repositioned. But the helicopter acted as if nothing happened. Put the skids down again and do the same test to the right with the same results. (Fore and aft cyclic behave more correctly.)

Yes!

I normally fly Gazelle with AP off trim off and for good measure gyro off, it's just one of those questions I have. You would almost not notice the centric position with a cyclic type input control* unless you had the control indicator on, then you notice it. I have carefully checked the Huey and Mi8 and spoken to my instructor about it.

* All be it a home made one. :D

The most flagrant example of this is to do the following--very simple--test in 2.1:

 

With gyros and A/P channels off lift off into a hands off hover. Notice, as expected, the lack of roll. Settle back down and move the cyclic fully to the left and hit the force trim. Take your hand off the cyclic and lift off into a hover again. Same lack of roll, right? In fact, nothing changed. The cyclic had repositioned. The cntl-enter indicator repositioned. But the helicopter acted as if nothing happened. Put the skids down again and do the same test to the right with the same results. (Fore and aft cyclic behave more correctly.)

 

[EDIT] I believe that this is a known issue. I feel pretty confident that Polychop will get the Gazelle into shape in the next month or two.

haha nice catch, i tested that just now and its true...although my cyclic is completely deflected to the right or left and locked with the magnetic brake, it lifts off just fine, perfectly centred. If i deflect it directly by hand without locking the magnetic brake, it rolls over as expected.

Anyway, i still struggle with the notion of that perfect hands-off roll stability when lifting off. Its just that i don't know whether the real thing is for whatever reason really perfectly stable in the roll axis because some SAS or whatnot, or whether this is a part of the flight model thats not in its final state. We were talking about this a month back quite extensively, the latest modifications to the FM seem not to have changed the roll behaviour.

The devs are probably still waiting for input from the french guys....

I believe they mentioned that the trim solution implemented was a more game friendly one, this was around the release however so it seems mostly forgotten. I could also be misremembering.

 

As for how it works, it doesn't exactly have a magnetic brake trimmer, the stick position as a result of trimming is purely a visual implementation. What actually happens when you trim is you set the desired aircraft attitude, not stick position. You can test this by trimming at various attitudes with your joystick centered and you should see the controls indicator and cockpit stick move around to various positions that if you were to try holding the stick there yourself you would most likely crash rather quickly.

 

So to reiterate, the stick position shown after trimming has no bearing on the actual inputs given to the aircraft.

I think this is likely the crux of the centric nature of the Gazelle 's cyclic it's intended to make using a joystick with centre detent more "game friendly" rather then more realistic if using a cyclic. :(

For me the inputs during hovering and so forth "feels" about correct but the action is around the centre point of the cyclic (electrical centre) not the position that would balance the aircraft as you would move the cyclic back and to the right or left for pick up and hover in Huey or Mi8 this is especially evident when doing pick up (set down) and hovering in the Gazelle.

Of course in the Gazelle the cyclic needs to be off set from the centre when picking up from a non level position the same as for Huey and Mi8 so it seems like it's a definite choice in the software.

As far as pick up in the Huey or Mi8 or Gazelle if you pick up the correct way you can observe the outside world and adjust the cyclic to provide the correct thrust vector once set to balance the aircraft you can pick up up to a metre or so without cyclic input aka hands off and you generally wont get any roll if you have carefully balanced the aircraft with cyclic and torque input.

The difference is in the Huey and Mi8 the cyclic is not centred for aircraft balance point where as whilst you can use this method for the Gazelle you can also use the controls indicator to electrically centre the the cyclic and you will lift pretty much straight up. You can get into a stable hover in the Gazelle and carefully centre the cyclic and set down hands off as well.

In all aircraft you will need to start to use correction input to maintain hover.

The other aspect that this centric cyclic seems to hide is CoG, seems that the CoG doesn't shift with different load outs or fuel levels like it does in the Mi8 and Huey and like you need to account for in preflight for a real helicopter. Huey and Mi8 the cyclic position needs to be shifted to allow for different CoG the Gazelle seems to be happy centred.

Again I find the Gazelle great fun to fly and given the attention to the FM is receiving it's getting better.

I just wonder whether this will be addressed?

Even a check box in the controls setup Joystick or Cyclic input selection. ;)

The devs are probably still waiting for input from the french guys....

This was supposedly going to happen way back in June.

I understand being respectful and diplomatic, but we are looking at a year and a half of waiting for these updates and lo and behold they are yet another disappointment.

The flight model needs a fundamental overhaul or re-do, and that does not appear to be in the works.

@scarecrow: not a complete overhaul of a flightmodel is needed, just some adjustments. Do you know how long it takes to model a flightmodel?

criticis is nice, but they have to stay resonable in my personal opinion and talking about something that makes no sense is everything but respectful and diplomatic, to catch this point.

We will probably talk about this topic here on the weekend. Thanks for pointing it out

We will probably talk about this topic here on the weekend. Thanks for pointing it out

Indeed thank you. :thumbup:

Concerning CoG as Fragbum mentioned. When we were talking about this last month (https://forums.eagle.ru/showthread.php?t=186653&page=17), Jester986 explained that helicopters tend to roll in one direction (US or european style) mainly for two reasons (i hope i recall his explanation correctly).

1.) For catching a wheel or skid on the ground because of translating tendency when the tail rotor is pushing the helo to the left or right.

2.) Or for CoG being above or probably mostly below the rotational axis of the tail rotor. How strongly the helicopter rolls on pickup should thus change with different loadouts and fuel and its influence on CoG.

One aspect i myself am not completely sure about but might be important in addition to CoG, is the question of how violently you lift off (gently with barely enough collective or max collective). Hence, if you really quickly put in max collective and rip that helicopter off the ground, you needed to put in way more left or right pedal to counter that strong torque. This means stronger pitched tail rotor blades, more translating tendency, and thus a stronger roll around the CoG. Is that what should go on in real life?

If this is correct, you needed to deflect the cyclic even further when takin of fast/violently in addition to what is needed because of CoG alone.

@borchi_2b thanks for letting us know that you are aware and talking about this specific issue. Except for that missing rolling thing on lift off i am very happy with the Gazelle and i appreciate your continued effort to get this bird closer to the real thing with each patch.

Not a pilot.

Rolling tendency is the result of the horizontal thrust component of the main rotor due to the tilt of the rotor and the thrust of the tail rotor in the opposite direction. The resultant vector acts to cause the aircraft to roll. However the aim is to balance these forces with cyclic to mitigate the effect on pick up.

The effect can be reduced in design by bringing the height of the tail rotor closer to the height of the main rotor, off setting the position of the main rotor and correct loading of the aircraft.

However in the scenario of using more power to lift off it would take more correction to counter the forces on the aircraft.

''However in the scenario of using more power to lift off it would take more correction to counter the forces on the aircraft.''

and if the FM enhancements to come go as far as implementing this as well i shall be a very satisfied customer!

The most flagrant example of this is to do the following--very simple--test in 2.1:

 

With gyros and A/P channels off lift off into a hands off hover. Notice, as expected, the lack of roll. Settle back down and move the cyclic fully to the left and hit the force trim. Take your hand off the cyclic and lift off into a hover again. Same lack of roll, right? In fact, nothing changed. The cyclic had repositioned. The cntl-enter indicator repositioned. But the helicopter acted as if nothing happened. Put the skids down again and do the same test to the right with the same results. (Fore and aft cyclic behave more correctly.)

 

[EDIT] I believe that this is a known issue. I feel pretty confident that Polychop will get the Gazelle into shape in the next month or two.

My findings are different.

TESTED on 1.5.7 (latest release).

Conditions: gyro, all AP channels OFF, make only use of FORCE TRIM in phase 2.

1. hands off into hover, only correction by rudder input: helicopter rolls to the left ! (check this behaviour while SLOWLY raising collective, not ripping the helo off the ground - then I get a minor roll to left)

2. Cyclic fully to the left then FORCE TRIM: control indicator indicates cyclic "arrested" fully left, modelled cyclic moves to the left and stays right there!

Take off into hover.....rolling left.

3. same but deflected fully right: helicopter still rolls left on takeoff hover

Conclusion: Roll to the left is present and aggravated by pulling the collective (as expected) but I expected an influence of trimming the cyclic left or right and that clearly does not happen though control panel and modelled stick show the trim (L/R). The helo is not rolling over skids as it looses ground friction (due to height) before that occurs -not like IRL

Using HOTAS WARTHOG, before testing checked if everything was centered. And of course the physical position of the HOTAS stick does not reflect the position of the ingame stick in Gazelle´s cockpit but I am perfectly fine with it. It seems just to be a matter of "training" or muscle memory.

''However in the scenario of using more power to lift off it would take more correction to counter the forces on the aircraft.''

 

and if the FM enhancements to come go as far as implementing this as well i shall be a very satisfied customer!

I would expect part of this to be influenced by the addition of cyclic correction for translating tendency, in that some aft and right cyclic input to balance the aircraft with associated rolling tendency and hence additional cyclic input for the correction of that. And that's just pick up.