5 hours ago, mSparks said:

It makes a big difference to end users - but indirectly.

End users dont "experience" BET/CFD or tables. What they experience is the aircraft and customer support - and it is in the aircraft development where it makes the difference: How simple is it to develop a precise aircraft,  and how easy is it to resolve a customers issue or customization request when they identify/make one.

That is also a much broader problem set than just how does the physics work - as important if not more so is how mature is the sdk and documentation, development and test tools, even something like how fast the sim itself starts makes a big difference.

basically this stuff:

https://developer.x-plane.com/author/austin/

 

I agree. However, there are many highly active developers for MSFS, and if we were to use frequency of addon updates or releases as a measure, then I'm not sure we'd see much difference between the two.

On 7/21/2024 at 2:52 PM, jcomm said:

And yet XP doesn't yet model VRS

Fingers and toes crossed for 12.2! 

 

6 hours ago, UrgentSiesta said:

justify your ongoing derision

Because just as IRL pilots flap their gums in these forums when so many of the topics hit home, this topic is one that flaps mine. You'd feel it too if you've ever tried to use CFD to any real effect for a real product. There are other reasons too (e.g. un-teaching marketing hot air). I agree with your outside-looking-in observation that it may not matter to the "hordes", but the hordes aren't 100% of the market population either.

As for what they're using it for....my bet (yes, pun) is that they're running CFD for airfoil or structural design and then plugging those resulting forces/moment curves into the sim just like one would export from Airfoilmaker. Then using that to predict overall performance behaviors (cruise, etc). If it's a good enough simulation, maybe they're using it to tune the autopilot and fly-by-wire system. Just like any quad copter, Alia's got some pretty cool stabilization features (at least, based on what I see in XP). That's where the dynamic response of the sim would be coming into play.

 

16 hours ago, mSparks said:

would require several weeks compilation on a hundred million dollar super computer

That's still very much a thing, though GPUs are quickly changing that status quo. We "could" be a decade away from practical real time CFD for low Reynolds numbers but the problem that I stated earlier remains: Boeing won't be handing out their CAD geometries any time soon and that's kind of a big deal for these types of predictions.

 

3 hours ago, Bjoern said:

Or the long faces when not even an approximated and comparatively simple CFD model can make their thing work.

Right about then is when we start talking about turbulence models to get their eyes to fully gloss over so that we can make our escape from the room.

7 minutes ago, blingthinger said:

Fingers and toes crossed for 12.2! 

That is what I though when I heard Austin talking about further developers to the rotaries... I hear you ! 👍👍👍

Edited July 22, 20241 yr by jcomm

6 hours ago, Humpty said:

Don't worry , I am an expert at VRS. 

You need any advise let me know 😎

None needed, i have a few choppers rarely fly them, Know my short comings,

11 hours ago, jcomm said:

Helicopter Flying Handbook | Federal Aviation Administration (faa.gov) from where I took the following excerpts and you can easily get how wrong XP's helicopter FM is in this particular region of modelling rotary wing aircraft:

Reading books and experience are not the same thing, I know that from 35 years in teh broadcast industry i am no expert, and happy to admit it.

7 hours ago, blingthinger said:

Fingers and toes crossed for 12.2! 

its estimated it since 11.10

sim/flightmodel/engine/vortex_ring_state

Wash-multiplier - vortex ring state - which are the airflows through the rotor disc normal flow is 0.5 (50%) VRS flow is 1.0 (100%)

ratio • float[ 16, 10 ] • v11.10.0+

 

Edited July 23, 20241 yr by mSparks

1 hour ago, mSparks said:

its estimated it since 11.10

sim/flightmodel/engine/vortex_ring_state

Wash-multiplier - vortex ring state - which are the airflows through the rotor disc normal flow is 0.5 (50%) VRS flow is 1.0 (100%)

ratio • float[ 16, 10 ] • v11.10.0+

 

Thx @mSparks !

The entering is there, and does work since XP11 indeed... Problem is the exit and how you can actually exit VRS by simply pulling on the collective, as much as you want, and always be able to recover ...

1 hour ago, jcomm said:

Thx @mSparks !

The entering is there, and does work since XP11 indeed... Problem is the exit and how you can actually exit VRS by simply pulling on the collective, as much as you want, and always be able to recover ...

well, aiui this is what Austin was just discussing at fsexpo - properly capturing the flow over blades, flapping feathering etc - we lost retreating blade stall early on in XP12 - but overly agressive VRS coming into landing was thankfully fixed (made it almost impossible to achieve a land into a hover).

That's why I said "estimate" rather than simulate.

imho its better to not have either at all rather than have them "simulated" but badly.

Getting them working is surely going to be a chore.

I tried the default helicopters and both can be brought into VRS (the relevant dataref is helpful to monitor how deep the VRS is; be sure to use the X-Plane "data output" function, which shows 2 VRS datarefs, I presume one is for the main rotor and the other for the tail rotor).

The VRS datarefs can go well past beyond 1, I think up to 2 or so, which presumably indicates a very deep VRS. In this case, the vertical speed increases and the collective is less effective in decreasing the vertical rate of descent. In my flight tests, the altitude lost for recovery certainly increases, potentially significantly, compared to a normal flight condition.

Although, using full collective usually seem to stop the descent eventually. This might be the main alleged discrepancy compared to real life behaviour. Neverthless, I think XP12 is doing an excellent job anyway: we're talking about a very complex situation to model accurately (unstable turbulent flow!), and the fact that XP is designed to try to model it in a general way for every virtual helicopter, is already a big achievement.

As everything else, maybe there's room for further refinements, but I already appreciate how XP12 is modeling it.

For the record, during my flight tests, I encountered a VRS situation in the default S-76 where it could not be recovered using only collective (and I crashed). Apparently it is more prone to unrecoverable VRS than the default R22.

VRS has killed me more than once in the beginning and pulling the collective lever to the stop is not always the solution, so I do not think that recovery is too easy at all. Maybe it seems too easy because transmission damage from overtorque from max collective is not accounted for by X-Plane, so one can simply power one's way out of it.

 

19 hours ago, blingthinger said:

Right about then is when we start talking about turbulence models to get their eyes to fully gloss over so that we can make our escape from the room.

Don't forget to collect your money on the way out and drop a remark why they wanted to cheapskate themselves out of actual testing with computational dice throwing.

4 hours ago, Bjoern said:

VRS has killed me more than once in the beginning and pulling the collective lever to the stop is not always the solution, so I do not think that recovery is too easy at all. Maybe it seems too easy because transmission damage from overtorque from max collective is not accounted for by X-Plane, so one can simply power one's way ou...

It's not a question of overtorque but rather of stalling / stalled blades, that will further stall when collective is added, no matter how much...

10 hours ago, jcomm said:

It's not a question of overtorque but rather of stalling / stalled blades, that will further stall when collective is added, no matter how much...

What about the air flow ? 

What I do is use the cyclic forward push or rather shake the cyclic to get out of the dirty air  in XP depending on the altitude I am at the time of hitting vrs

 

1 hour ago, Humpty said:

What about the air flow ? 

What I do is use the cyclic forward push or rather shake the cyclic to get out of the dirty air  in XP depending on the altitude I am at the time of hitting vrs

 

And that's the correct technique. Pulling on the collective should only aggravate VRS, but in xplane it doesn't...

23 hours ago, jcomm said:

It's not a question of overtorque but rather of stalling / stalled blades, that will further stall when collective is added, no matter how much...

Quote https://en.wikipedia.org/wiki/Vortex_ring_state:

Quote

...the best way to recover from them is to laterally steer clear of them, in order to re-establish lift, and to break them up using maximum engine power, in order to establish turbulence.

...

It is possible to power out of vortex ring state, but this requires having about twice the power it takes to hover. Only one full-scale helicopter, the Sikorsky S-64 Skycrane, is documented as being able to do this, when unladen.

To me, this means that VRS can be exited when enough power is available. And power, to me, means increasing collective, which in turn means more torque.

44 minutes ago, Bjoern said:

Quote https://en.wikipedia.org/wiki/Vortex_ring_state:

To me, this means that VRS can be exited when enough power is available. And power, to me, means increasing collective, which in turn means more torque.

Honestly, between wikipedia and the official helicopter syllabus by FAA, I would trust FAA....

"Vortex Ring State Vortex ring state (formerly referenced as settling-with power) describes an aerodynamic condition in which a helicopter may be in a vertical descent with 20 percent up to maximum power applied, and little or no climb performance. The previously used term settling-with-power came from the fact that the helicopter keeps settling even though full engine power is applied"

"When recovering from a vortex ring state condition, the pilot tends first to try to stop the descent by increasing collective pitch. However, this only results in increasing the stalled area of the rotor, thereby increasing the rate of descent."

Edited July 24, 20241 yr by jcomm

From my flight test, the default S-76 at MTOW cannot overcome deep VRS even with full power. Might upload a video maybe.

One contributing issue might be that it is probably more difficult to keep a hover with the limited feedback of a PC home simulator and the viewing angle of a monitor. So if the virtual helicopter is unadvertently moving forward, backward or sideways, it helps to contrast VRS.

This is not to say that the modeling of VRS in XP12 is perfect, indeed the default R22 seems to be able to overcome it just increasing collective.

1 hour ago, Bjoern said:

Quote https://en.wikipedia.org/wiki/Vortex_ring_state:

To me, this means that VRS can be exited when enough power is available. And power, to me, means increasing collective, which in turn means more torque.

The reference note in the wikipedia article seems a reliable academic article. I would tend to agree with what it says, i.e. VRS can be exited with enough power (but most helicopters won't probably have that power).

I'm still of the opinion that the simulation of VRS in XP12 is more than decent, even though can be probably further improved in some cases.