Flight control

[Fox One]

From real aircraft flight manual:

Les commandes de vol sont reglees pour avoir des deplacements par G sensiblement constants au-dessus de 300kt environ.

Le manche permet donc de piloter un facteur de charge.

 

It seems to me they are describing a classic G-command type of control there.

 

First experiment

At low altitude with full AB with neutral trim in horizontal flight with 600kts I pulled the stick until I see 4G, then kept the stick fixed in precisely the same position. During the Immelmann the G continuously decreased and over the top speed was a little above 300 and G was 2.1

 

That's some serious variation in stick deflection per G. If I put the stick in a position that commands 4G, shouldn't the flight control try to maintain 4G, just like it would do in an F-15 or F-16?

 

Second experiment

At low altitude in horizontal flight with about 400kts constant speed with neutral trim I roll the aircraft inverted, then let go of the stick. The flight control will create about -1.1G. As if the aircraft was expecting inverted flight to last for quite some time. This doesn't make any sense IMO. What's the point of such kind of flight control behaviour? Help pilot recreate Top Gun inverted flight scene? As long as the speed is above 300kts, shouldn't the flight control try to maintain 1G, no matter the flight conditions or aircraft position in space?

 

Am I misinterpreting something? Maybe someone knowledgeable in Mirage 2000 could provide some answers if current simulator behaviour is right or wrong.

[jojo]

First experiment

At low altitude with full AB with neutral trim in horizontal flight with 600kts I pulled the stick until I see 4G, then kept the stick fixed in precisely the same position. During the Immelmann the G continuously decreased and over the top speed was a little above 300 and G was 2.1

 

That's some serious variation in stick deflection per G. If I put the stick in a position that commands 4G, shouldn't the flight control try to maintain 4G, just like it would do in an F-15 or F-16?

 

You speak about Immelmann. So I guess you went into vertical climb ?

 

Above 300kt CAS, the stick gives you % of max available G.

But max available G decrease with altitude.

 

If you do the test in horizontal turn, with idle throttle to slow down, it works as expected.

 

Second experiment

At low altitude in horizontal flight with about 400kts constant speed with neutral trim I roll the aircraft inverted, then let go of the stick. The flight control will create about -1.1G.

Agreed, there's is something to correct.

[Rlaxoxo]

Doesn't FBW system keep the aircraft inverted?

[Fox One]

Above 300kt CAS, the stick gives you % of max available G.

 

You probably mean "max available G for the current altitude". Are you 100% sure about that? The manual says " deplacements par G sensiblement constants au-dessus de 300kt environ ", it doesn't say this applies only if altitude is constant. Personally I interpret this "constant" as "constant in the entire flight envelope as long as speed is above 300." If at sea level the "max available G" is 9 and I pull the stick half, this will give me 4.5G. But if I'm at 40Kft altitude and the "max available G" there is 4, pulling the stick half will give me 2G. But this kind of behaviour surely can't be described as deplacements par G sensiblement constants. In fact, they are very far from constant. This type of flight control logic will give you along the entire flight envelope of the aircraft very large variations in stick displacement per G, in fact as large as an aircraft with conventional pitch control, like for example a MiG-23. What kind of fly-by-wire "smartness" is this?

 

In the F-15 if you pull 4G at sea level and enter an Immelmann and keep the stick in precisely the same position, the aircraft will remain at 4G(+/-0.2G) as long as this is still possible, it doesn't matter how hight you climb.

 

[jojo]

The purpose of Mirage 2000 FBW is to protect pilot from departing from flight envelope, AoA limitation, G limitation and spin protection.

 

Do you think you can sum up FBW logic by on phrase ?

P45 :

2.1.2 – Architecture

L’ordre électrique qui attaque les servo-commandes de gouverne sur un axe donné est la résultante :

- de l’ordre pilote sur cet axe, éventuellement pondéré par l’ordre de profondeur

- de termes d’amortissement

 

Les termes d’amortissement sont issus de détections gyrométriques (vitesses angulaires de roulis, tangage, lacet) accélérométrique (facteur de charge, accélération transversale) et d’information d’incidence.

 

Tous ces ordres sont pondérés par des gains variables (en fonction de Δp et Ps) ce qui permet un adaptation aux conditions de vol (Mach, Vc, Zp)

Δp = dynamic pressure

Ps = Static pressure

Vc = CAS

Zp = pressure altitude.

 

But if you really want to make a point: POST A TRACK !!!!!!

Then we will see what is your maneuver, your stick position and what happens. Otherwise we are just doing a point less talk...

Edited September 13, 2016 by jojo

[jojo]

Ok, I did the test into a loop.

 

I think the 2 problems are tied.

 

If you display stick axis (right Ctrl + Enter), when you get into inverted flight, you have increase in pitch down trim. That's probably the cause of G decrease.

[SCU]

Δp = dynamic pressure

Ps = Static pressure

Vc = CAS

Zp = pressure altitude.

 

I'm sorry for being off topic, but do you know if there is an English version of that manual?

[jojo]

It's mandatory for export customer. But available in the open...probably not.

 

The available manual is an early one, for Mirage 2000C RDM. It's incomplete.

[SCU]

Too bad :cry:, gonna try and learn French now :P. I don't know the first thing about French so it's gonna take a while.. Or just wait till a translation or the English version to come out.

[Fox One]

jojo, you are overcomplicating things when there's really no need to. What inputs the real flight control system uses to do its "magic" or how exactly does it is not what I'm asking here.

 

In the attached track I'm flying the F-15. At low altitude and supersonic speed I pull the stick until I hit 3G, then for the rest of the loop I don't move the stick at all. Despite the great variation in speed and altitude during the loop, G remains between 2.8 and 3.2. That could be described as deplacements par G sensiblement constants.

 

Le manche permet donc de piloter un facteur de charge. - that is exactly what I am doing here! With the stick in a certain position in pitch I am asking the aircraft to give me a certain G, and the flight control system will do its best to try to give me that, as long as it is still possible.

 

I believe the real Mirage 2000 at speeds above 300 behaves in exactly the same way as I described above for the F-15. Actually I would be very surprised if it does not.

15test6.trk

[Frederf]

It sounds like the Mirage translates stick position into load factor directly at high speed and blends into AOA based below some threshold. That's more or less how the F-16 works (which also has pitch-based control for AAR/landing/standby).

 

If the Mirage FCS isn't G-based (at least at high speed/ low AOA) then what is it?

[jojo]

That's it.

- G load above 300kt

- transition to AoA control below 300kt

 

Obviously there is an issue when flying inverted. It will probably be resolved soon :smilewink:

Edited September 14, 2016 by jojo

[Rlaxoxo]

Not sure how I feel about this : /

I kinda like how it is right now, but if it's closer to realism I guess I'll just have to get used to it :S

[Frederf]

I see what you mean about pitch semi-instability at the horizon. With dive it's not an issue as more dive is less load factor which at commanded load factor cancels the dive. That's positive stability and trim is functional. You're not supposed to maintain a dive stick neutral, 1G trim (in the F-16). The answer is simply you have to trim it.

 

With nose high the situation is reversed as more pitch is less vertical load factor which simply commands more pitch. That's negative stability. The F-16 gets around that by doing something clever which I will have to look up. I imagine the Mirage solution is practically identical.

[IvanK]

CptSmiley said:

 

"In conclusion, I've reverted the FBW G-load command logic to be more like it originally was and got the thumbs up from testers. The only "negative" thing (that is also experienced with every of the FBW based sim aircraft I could find in public domain) is a very slow and gradual pitch up tendency when climbing or descending in stick neutral and no trim. This is obviously because G-load is not the same as 0 vertical body acceleration on the aircraft. "

 

In fact this exact behavior occurs in the FA18A in real life. Its not a big deal but it was noticeable.

[jaguara5]

Hi all, if you don't care to read below, these behaviors noted have been corrected. Now for more details:

 

Just a bit of history, there are really two different ways to command based on acceleration. The first and the way it appears to be done across all other aircraft (F-16 [bMS], F-18 [VRS], F-15 [DCS] are good known sim examples) is via G-load that is the apparent G on the aircraft, that is, the apparent vertical acceleration in world frame plus body acceleration on the aircraft. The other and more simpler one is command based on only body axis vertical acceleration.

 

In the first, the aircraft always tries to maintain 1 "G" regardless of orientation the second the aircraft always tries to maintain 0 vertical body axis acceleration regardless of orientation. Now what does this mean you might ask? Well in the "G" command version going level and inverted the aircraft will still try to maintain 1G, thus causing you to go toward the ground, in the other the aircraft will stay level will always point in the direction it is going as a change in pitch rate would result in vertical body acceleration therefore it will actually trim at -1G.

 

So early in the M-2000C code base we had it the first way (G-load command) and the known way it is used in other aircraft. However, a few people complained about minor pitch increase when climbing/descending so I changed it to do the second way some time in the March/April time frame. The problem with the second way is exactly what FoxOne discovered, when you aren't perfectly level the aircraft behaves in a non-standard way especially when off the horizon plane and control load is unpredictable with respect to aircraft orientation. On the other hand with the vertical body acceleration way and stick neutral the aircraft will always point where it's going regardless of any orientation, which in many cases is a good thing.

 

In any event, I went back and looked at a bunch of public FBW diagrams for like aircraft, talked with some people in the know, and based on them along with those FBW aircraft tested out in sims and their respective behaviors the G-load behavior described by FoxOne appears to be the correct one.

 

In conclusion, I've reverted the FBW G-load command logic to be more like it originally was and got the thumbs up from testers. The only "negative" thing (that is also experienced with every of the FBW based sim aircraft I could find in public domain) is a very slow and gradual pitch up tendency when climbing or descending in stick neutral and no trim. This is obviously because G-load is not the same as 0 vertical body acceleration on the aircraft.

 

Hope this helps and is a sufficient explanation :)

I will visit tomorrow an M2000 (EG and -5) base for an airshow.

Is there a specific question to ask that might be helpful for you ?

[jojo]

I will visit tomorrow an M2000 (EG and -5) base for an airshow.

Is there a specific question to ask that might be helpful for you ?

 

It's off topic but I do have one question for them:

Does Mirage 2000EG have Rad Alt auto-pilot mode ?

It's useful for Exocet anti-ship mission with sea-skimming flight profile.

[jojo]

In fact this exact behavior occurs in the FA18A in real life. Its not a big deal but it was noticeable.

 

Mirage III and Hornet pilot...how lucky you were :thumbup:

[IvanK]

Still lucky .... still Fly French things :)

[jojo]

Still lucky .... still Fly French things :)

 

What type, Falcon ?

[IvanK]

One of these :)

[jojo]

A380 for Qantas then ? :smartass:

[Oesau]

A380 for Qantas then ? :smartass:

 

That's right, he's a Qantas driver on the A380.

[Sarge55]

Good to know.

 

I'll be taking a Qantas flight from New Zealand then a few around Australia in November.

[jaguara5]

The visit to the M2000 airbase was awesome! Here are some infos (some are slightly of topic:( but still interesting :)).

1) Indeed, the M2000 inverted will go down without any input from the pilot. No -1g force applied from the FBW. to keep the vertical flight path.

2)

It's off topic but I do have one question for them:

Does Mirage 2000EG have Rad Alt auto-pilot mode ?

It's useful for Exocet anti-ship mission with sea-skimming flight profile.

There is no radar altitude autopilot mode, and , according to the pilots, it would be not useful for air to sea attacks in the Aegean environment

3) Finally, i learnt what the heck are the W (0:53) and M (1:03) hud symbols in the following video

W is a warning for a (not serious) malfunction in the radar electronics, it can be cleared by pressing a button. M stands for memory mode. If the radar is about to lose lock from a locked target, target data are reserved for few seconds in the radar memory.

4) Through the buttons around the VTB, the pilot can not only input a virtual target , but (and that is the main use) he can program on the ground all the firing parameters for the exocet missile (i will ask what these parameters are in a future visit). The exocet is displayed on the PCA as '' M(Missile)39''

5) Also on the PCA , ADF stands for Air Defence Flight (= intercept of civil planes = Police mode) and AGG for Air to Ground Gun

6) The pilot retards smoothly the throttle at around 50 feet altitude above the runway.

Edited September 18, 2016 by jaguara5