Turbine heli clutch

Ideas
, , ,
1

Hi
This topic has been brought up in FS2020, and FS2024 discussions in 4 threads. It has been well described and detailed by several Dev’s.

It is still a problem that has yet to be addressed.

Currently it has become a developement blockage to us.

We have under developement, 2 turbine heli’s that hit a brick wall on this issue.
IRL, both heli’s are able to be started with the rotor stopped. This is due the nature of maritime ship operations, and associated safety and stability issues onboard with a rotor turning.

To do this, we need some way to eliminate torque from the engines going to the rotor.
The Cabri G2 has this already. Its called a clutch. This feature, although not explicitly stated (that I can find) seems to apply to piston heli’s only.

Please make this feature functional on turbine heli’s.

thanks
Stu

2

No need to stop at helis. There are propeller aircraft like ATR that have “hotel mode” where engine is running without turning prop. In Asobo ATR this is faked with animation/sound/guages; in the flight model the prop is still turning.

1 Like
3

I would like to add some words to this because I think it is very important to precisely differentiate between a piston engine clutch and the mechanism that is used in the vast majority of turbine engine helicopters.

Piston engine helicopters have a manual clutch that needs to be engaged by the pilot. When disengaged, the engine can be started without spinning the rotor, aside from that the rotor can be disconnected from the engine at any point the pilot wishes to do so. Reengaging the clutch in flight however, is something that is usually not possible, because it needs to be done in a specific low rpm range of the engine. On top of the manual clutch there is a freewheeling unit that allows the rotor to disconnect from the engine to ensure that the rotor rpm is not pulled down by the engine during autorotation.

In contrast to that the majority of turbine engine helicopters include only the freewheeling unit. The freewheeling unit is a one-way clutch that automatically connects the engine to the rotor system when the engine is providing enough power, but instantly disconnects it if the engine drive shaft rpm falls below the rotor rpm in terms of percent.
This mechanical separation allows the rotor blades to continue spinning freely due to inertia (autorotation), ensuring the helicopter can be landed safely without the drag of a dead engine slowing them down.
However this is not the only situation in which freewheeling happens, it also does during startup and shutdown. The first engine that is started will drive the rotor and the second engine will be freewheeling until it has catched up to the same rpm. During shutdown the engines usually spool down faster than the rotor which puts them into freewheeling mode as well.

The engine start without spinning rotor you are describing @StuLes20 is usually done via the rotorbrake since turbine engine helicopters do not feature a manual clutch. Before start the rotorbrake will be fully applied, the engine will then be started and the brake is usually released somewhere around idle.
Please let me know in case the helicopters you are working on are somhow unique, but honestly I have never heard of a manual clutch in twin engine helicopters.

1 Like
4

I don’t think that this is correct.
The second engine during start is not free wheeling. It is a free turbine, and hence not connected to the gearbox mechanically. You can see this on engines like the pt6 during start, a mechanic can hold the prop and stop it rotating. Characterisitic of free turbine, not freewheeling.

mechanically in a different position.
Freewheeling unit is between geabox and rotor.
Clutch is between engine and gearbox.

Turbine heli’s can have clutches. eg Lama, alouette II and alouette III all have clutches and seperate freewheeling units.

Depends upon the heli operating proceedure.
The first that we are working on, both engines are started to idle with the rotor brake on. Checks are done. Brake released, eng’s spun up, then takeoff minimising time on deck with rotor turning.
the Second heli. First eng started with rotorbrake on, then run up to flight rpm. Checks done. Number two eng started, and run up to flight rpm. Rotor brake released, takeoff. Once again min time on deck with turning rotors.

I am not saying that these heli’s have a clutch.
I am after a method that we can use, eg a clutch to emulate this disconnection of torque between engine and rotor.
This is not a sprag/freewheeling unit. This would be the reverse of a clutch, and disconnect torque between rotor and gearbox.

3 Likes
5

I think both of you are talking about the same thing.

Both terms refer to the same component, just using different terminology. There is the Freewheeling Unit and the Sprag Clutch. The term Freewheeling Unit is more commonly used in Europe, whereas in North America the term Sprag Clutch is more frequently used.

Both are installed downstream of the input drive shaft coming from the engine into the gearbox, and they prevent torque from being transmitted back from the main rotor system to the engine if they aren’t engaged.

How do they engage/disengage?

Simple: whenever one engines N2 is below rotor RPM. Mechanically, a freewheeling unit disengages through wedged elements (such as sprags or rollers) that lock when torque is applied in the driving direction. When the rotor rotates faster than the input shaft, these elements automatically un-wedge and tilt or roll out of their locking position, allowing the two shafts to rotate independently (usually when the difference between the two shafts is > 2%).

Why is this needed in MSFS?

  • Autorotatio: whenever you need to autorotate, you want to keep as much energy as possible in the rotorsystem and not getting absorbed by the engines being driven by the rotor
  • OEI animation: if you set one engine to idle or off, the respective N2 should decrease (to idle N2 or 0)
  • Startup/shutdown: the first started engine drives the rotor (NR = N2) while the second N2 would be 0 before engaging the second engine. After starting the second engine, N2 would spin up freely driven by the N1 until it reaches rotor rpm and the clutch/freewheeling unit engages. In the current implementation if you only start one engine, the N2 of the second engine will also spool up even though this engine hasn’t been started.
1 Like
7

Hey @StuLes20

As explained by @exil35 the freewheeling unit/sprag clutch sits between the engine drive shaft and the transmission. This means that in a twin engine helicopter there are two freewheeling units and this is also the reason why engines are freewheeling not just in autorotation but in other situations like startup or shutdown as well.

In the video below you can see a nice demonstration of that. Starting somewhere around minute 6. He just talks about autorotation but the same principle applies to all situations in which an engine´s rpm would be below the rotors rpm.

The point I was trying to make in my first comment is that a manual clutch is something that usually does not exist in twin engine helicopters. Freewheeling units/sprag clutches cannot be manually intercepted.
Is there a specific reason why you are not using the rotorbrake for your startup? If the real aircraft does so, I have not understood yet why you would want a manual clutch instead. If you make the rotorbrake strong enough it should be able to stop the rotor from spinning.

1 Like
8

Guys
thanks for contributing, but its derailing the thread.

I am after a clutch solution not a sprag clutch solution.

Different things.

Stu

1 Like
9

Yes there is.

  1. Even if you make it strong enough , it doesn’t stop slow rotation. Also the engine is maxed out trying to turn the rotor. If we had a free turbine, it would work, but to my knowledge, we don’t.
  2. Then, you have lost the rotor brake function in slowing the rotor. If its set high for the start, then it stops the rotor after shutdown it about 1 sec.

I agree that its not a turbine heli parameter to have a clutch normally.
The issue is that we don’t have a free turbine in the sim. A clutch parameter would allow the free turbine characteristics to be emulated.
Just trying to find a solution in an unrealistic sim.

1 Like
10

Based on how you formulated your text we had the impression that this your understanding of the real aircraft and not that you are just asking for the manual clutch as workaround for your rotorbrake start. If I had known that earlier I would have helped you directly.

I have send you an explanation regarding how to approach the rotorbrake start without having a maxed out engine via discord. I have it working in my BO 105 project.

2 Likes
11

@StuLes20 After thinking about it more in depth there are actually some reasons why I would fully support a manual clutch and voted for it, despite it not being realistic. One of them would be that this would allow us devs to code the long awaited freewheeling unit ourselves. So in in that regard it would address several aspects at the same time.

@FlyingRaccoon It seems to me that this would be a great way to adress both the missing freewheeling unit and the specific startup case @Stules would like to implement.
We´ve been struggling a lot with that bug since MSFS20 because it causes so many problems in other areas. Having an optional manual clutch that we can trigger via custom code is all we would need to address the situation ourselves and it would considerably reduce the required work on Asobos side.
The only thing that is very important is that it should be one clutch per engine and those clutches need to be in the correct position, meaning in between the engine drive shafts and the transmission.

12

Hello

I can’t tell yet the way this will be resolved but I forward your requests to the team and I’ve asked that when this is addressed, we communicate ahead of the release to explain how we tackled the issue to make sure it fits your needs.

Regards,
Sylvain

1 Like
13

Thank you very much!

I think @b1e1n7e idea would be a great tool and i can already see how we could implement such a clutch to act as a freewheeling unit with some minor cutsom code. This would help all of us helicopter devs a lot!

14

Hey Sylvain,

the reason I think the manual clutch might be the way to go is that it would allow the Asobo team to address the situation quicker due to the reduced complexity and simpler implementation. We understand that there is probably a lot on the to do list for the team and that capacities are limited. This would be a good compromise and would ensure that the bug can be adressed within a reasonable time frame.
I know that I have told you this before but I would like to point out again how incredibly inefficient the situation has been for us during the last 2 years and still is. Working around the current bug, fixing the problems it causes and finding creative ways to fake indications etc. has so far required months of full time work.
If we can get a manual clutch we would happily accept that because it provides us the foundation to be creative with it depending on the individual project.
So thank you very much for putting it forward to the team.

15

Hi Sylvain
Thankyou for adding traction to our request. Apprciated.

Thanks @b1e1n7e and @exil35 for getting behind the concept.
I think that you now recognise the usefulness of a simple clutch.
It allows us the flexibility to code it how we want to achieve multiple issues that have hampered our developement of heli developement.

regards

Stu

16

Hello everyone,

The issue being discussed here is known to the developers and actually involves three fundamental problems that need to be addressed:

  • The transmission kinematics require improvements in the area of boundary operating conditions.
  • The turboshaft engine model needs to implement a true free turbine.
  • The free turbine must be capable of producing negative torque.

For realistic modeling of helicopter autorotation and engine start procedures on single- and multi-engine turbine helicopters without rotor spin-up, all three of these issues must be solved together.

At present, as you probably know, in the turboprop and turboshaft models N1 is always equal to N2. In other words, we effectively have a single-spool gas turbine engine, which does not reflect reality.

In addition, the torque produced by the (non-) “free” turbine is defined via a lookup tables as a function of corrected N1 and prop/rotor RPM, and it reasonably contains only positive values. However, disengagement of the Freewheeling Unit is only possible when the free turbine shaft decelerates faster than the rotor - that is, when the free turbine shaft experiences negative torque.

Furthermore, as far as I understand, the transmission system itself requires a review with respect to boundary conditions: for example, ensuring that friction (the rotor brake) do not allow the system to be driven out of a resting state.

All of this represents strategic improvements to the simulation. These tasks are already on our basket, and, indeed, require a significant amount of time.

That said, I believe we can implement a temporary workaround as you requested.
We will try to introduce clutch behavior for multi-engine turbine helicopters so that you can use it on your side. Inclusion in SU6 will depend on how quickly it can be implemented; otherwise, it would come in a later update.

Best regards,
Asobo/Andrey

5 Likes
17

Hello @An.Petrovich,

Thank you very much for the detailed response regarding the challenges that the team is facing with the helicopter engine and its interaction with the transmission and rotor. Knowing that the situation is indeed well understood makes us confident about future improvements.

And thank you very much for taking on the manual clutch as a temporary workaround. This will make our workflow much more efficient since several dev teams are currently working on twin engine helicopters. We appreciate the additional effort of the Asobo team.

1 Like
18

Hi Andrey
Thank you for explaining the situation.
Its been frustrating on our side, and to know that it will be addressed is reassuring.

Should it be implemented before the documentation is updated, we would appreciate a quick brief of the functions and parameters that are being used.
You did this with the ‘tensor’ implentation, and it was a valuable heads up.

Thanks again.
Regards
stu

1 Like
19

Hello everyone,

as discussed, a temporary workaround will be introduced in SU6 to support rotor autorotation in turboshaft-powered and multi-engine helicopters, as well as engine start-up with delayed main rotor spin-up. The workaround will allow you to explicitly connect or disconnect each engine from the rotor transmission through the new FREEWHEELING_UNIT_ENGAGED SimVar:

  • engines disconnected from the transmission no longer contribute shaft torque to the rotor system, and their RPM is integrated independently;
  • connected engines continue to use the existing clutch and transmission simulation.

This SimVar provides a writable Boolean flag and takes an engine index as a parameter.

The previous behavior remains unchanged when all freewheeling units (FWUs) are engaged, which is the default.

This workaround is constrained by the existing transmission model. FWU engagement and disengagement are discrete and instantaneous, with no slipping phase, RPM synchronization, or transient engagement torque. Reconnecting an engine may cause its immediate RPM change, and multi-engine configurations still rely on a shared engine RPM rather than independently simulated drivetrain dynamics. These transitions will therefore need to be managed on the aircraft side.

Given the limitations of the existing rotor transmission and turboshaft engine models, this workaround is probably as far as I can take the implementation I am currently working with without revisiting its underlying architecture. Further improvements would require a more fundamental redesign based on an explicit torque-balance model. I hope we will have a chance to address this more thoroughly when the development plan allows.

Best regards,
Andrey / Asobo

3 Likes
20

Hey Andrey,

Thank you very much for your effort to provide us with this solution in such a relatively short period of time. This will be a big step forward for all developers working on twin-engine helicopters.

The limitations you mentioned should not be too problematic once we adapt the code on our side to take them into consideration.

Maybe you can let us know if the change will already be part of one of the SU6 betas? In that case, I would offer to test it and could give you some feedback before SU6 is out for everyone.

21

Thankyou Andrey.
Appreciated.
Regards
Stu