Testers wanted for my RUDDER FIX (Beta)

[nbz 11]

Hi,

I am the guy who suggested a fix for the rudder a while ago.

I have now hacked up two different gauges to adress The Rudder Issue and have brought them to a state where I would like some feedback to see if they work on other people's systems. (That means they work great for me.)

The first gauge is "just" the function of a rudder: you kick the right pedal, the plane yaws to the right.

The second gauge INCLUDES the manual function of the first gauge, but adds an auto-rudder function. Without such a thing, turns are not flown coordinated without manual rudder input.

The second gauge, the Auto-Rudder one, is what I'm really aiming for. The first one is a fallback in case the second one does not work or its functionality isn't wanted.

(Note: it may technically be called a gauge, but thankfully nothing will show up in that utterly fantastic virtual cockpit)

Requirements:

• Airbus X V1.22
• you use a controller AXIS for the rudder (no keyboard input)
• you do NOT deactivate Fly By Wire when flying the Airbus X - the gauges do nothing with FBW off
• you are willing to do either ALL of the following installation steps or NONE
  Known issues so far:
  
  • the Autorudder gauge does not like long and hard stutters. (This would probably be very hard to fix.)
  • the external rudder animation looks less than optimal. (This I can't fix at all. Just a side effect of how my gauge works.)

Disclaimer:

This is all meant in a friendly way, trying to be helpful rather than b*tching around. In the following instructions, a few lines of Aerosofts code get deleted, the reason is SIMPLY that these conflict with my gauge. Do not read anything into that, this is NOT a statement about Aerosoft, the Bus X, or whether I believe in life after death or aliens from outer space. Also, please note that I'm trying to adress a MICROSOFT bug here, not an Aerosoft bug.

These gauges are currently ONLY for the Aerosoft A321 and will likely need adjustment to work with the A320! I have ONLY developed this on, and for, the IAE A321, so please test this ONLY with the IAE A321. If it works, I can then start working on a version for the A320 too.

I am especially interested in the Auto-Rudder gauge. How does it cope with different framerates, fuel loads, use cases? Fly steep turns, try shutting down an engine in-flight and see if you can still land. I have not yet tried an auto-land myself.

If you want to try this, at your own risk, here are the exact installation steps:

1.)

In your SimObjects/Airplanes folder, open the folder "Aerosoft Airbus X A321 IAE".

2.)

Create a backup copy of: aircraft.cfg

Give that a good name, like "aircraft_beforeRudderMod.cfg" or such.

3.)

Open aircraft.cfg, search the line

rudder_trim_effectiveness = 1.0

this is in the section [FLIGHT_TUNING].

Change the value to: 4.0

I mean it!

4.)

Open the folder "Panel".

Open the folder "AB_Systems".

Make backup copies of the following three files:

AB_anims.xml

ASCLS.xml

RudderDisp.xml

Give them a good name, like...

AB_anims_beforeRudderMod.xml

ASCLS_beforeRudderMod.xml

RudderDisp_beforeRudderMod.xml

... or such.

Open the three original files in your favourite text editor. Preferably one that displays line numbers.

5.)

EDIT: AB_anims.xml

LINES: 1044 - 1067

Delete all lines of code below the comment

-- Rudder limiter --

and above the comment

-- Downtrim limiter --

6.)

EDIT: ASCLS.xml

LINE: 202

Remove this single line:

(L:ASC_Ruddertrim, number) (> K:RUDDER_TRIM_SET)

7.)

EDIT: RudderDisp.xml

LINES: 146 - 178

Delete that entire last "Element" block

8.)

Drop BOTH files attached to this post into the "AB_Systems" folder.

9.)

Go back up to the Panel folder.

Create a backup copy of panel.cfg

Give it a good name...

Open panel.cfg in an editor.

Below all the "gauge" entries in the [Vcockpit01] section, place.....

.....for the Rudder-only gauge (which is more likely to work):

gauge44=AB_Systems!NBZ_FBWRudderA321,0,0,1,1

.....for the Auto-Rudder gauge (which is better *when* it works):

gauge44=AB_Systems!NBZ_FBWAutoRudderA321,0,0,1,1

10.)

Happy flight testing!

NBZ_FBWRudderA321.xml

NBZ_FBWAutoRudderA321.xml

[Chris_327 42]

Thank you so much. This has really helped me to enjoy my Airbus X more!

I am using the "gauge44=AB_Systems!NBZ_FBWRudderA321,0,0,1,1" entry in my cfg as I find that this gives me more control over the amount of rudder input that I want.

The only thing that I would change personally would be the effectiveness of the rudder inputs which I assume is step 3 (rudder effectiveness) that we changed to 4.0? However, would this also change the effectiveness of the rudder when below 100ft where the default FBW is disabled by FSX? If so I imagine this would make a sudden change in input sensivity, i.e. suddenly swinging off the centreline.

Again, thanks for this, I can now perform the circle to land at LOWI, without praying that I can complete the turn before I crash into the mountains! I'll get some more testing done and let you know how it goes.

Chris

[KAPTEJNLN 237]

well nice mate think it is only req to edit down to point 5 or 6 for gauge44=AB_Systems!NBZ_FBWRudderA321,0,0,1,1 then it does what it should whit out moving the rudder. just to keep it to minimal chages. And i think i know why you have some problems whit the animation as it seems there is no effect over 100 ft there for it seems it gets your interface then try to get back to center. or is kiled by something els would try investegate and get back if i find some importan. but it is only a visual side effect.

[nbz 11]

The only thing that I would change personally would be the effectiveness of the rudder inputs which I assume is step 3 (rudder effectiveness) that we changed to 4.0? However, would this also change the effectiveness of the rudder when below 100ft where the default FBW is disabled by FSX? If so I imagine this would make a sudden change in input sensivity, i.e. suddenly swinging off the centreline.

Yes, for the "simple" gauge you are using, not the Auto-Rudder one, you can play around with rudder_trim_effectiveness as you see fit. A higher value will give you more authority. But the Auto-Rudder gauge has been balanced to work against that exact same value, so don't forget to set it back to 4.0 if you wish to use that one.

It is probably possible to tune this gauge so that it has a very similar effect to the "real" rudder below 100ft. But I did not fine-tune it to such a degree, so yes, there will unfortunately be a sudden change in rudder effectiveness when passing 100ft.

This is similar to the nose-drop bugfixes (the fixes for the "AP-off-nosedip" and the "100ft-nosedip") that were implemented by Aerosoft. These also work best if not much input is done on the stick at the moment they do their magic.

EDIT: oops, didn't really answer the question: no, editing that line should have no effect on how the rudder works below 100ft. But, my rudder works by flying entirely through FSX's rudder *trim*, and at that very 100-ft-moment, there will be some trim "left over" if you used the rudder at that moment. My gauge gradually, but still rapidly, returns the trim to neutral at that point, in order to avoid extreme trim settings. This is a bit messy, but I had to do *something*. This effect will of course also be stronger if you increase rudder_trim_effectiveness.

Another thing you may wish to tweak: both gauges have a deflection limiter - they will allow less rudder travel the faster you are flying. This is why you can kick the rudder hard at >300 knots without ripping the plane to shreds. Again, this can be changed ONLY inside the "simple" gauge, but should not be touched in the Auto-Rudder gauge (because the autorudder function uses the same limiter value - something I maybe should have done differently).

To tweak this, change that "42" you can find in the gauge code. A larger value will give more rudder travel at high speed.

I am absolutely delighted to hear of your improved experience in LOWI !!

Guess what - back in the days before the Airbus X, flying that turn into LOWI with the default MS-Bus was the reason why I just HAD to come up with a way to seize control over the Rudder with FBW on. But after having done the first version of the gauge, I thought it was not good enough to have to do all the inputs myself (like you would do on a Piper Cub) on such a sophisticated FBW-plane. In LOWI, you get to fly a high bank angle at low speeds, and there it shows that the Bus does not coordinate turns, the nose pointing "out of" the circular path you fly.

This is why I came up with that second "Auto-Rudder" version. You should be able to fly nicely into LOWI without ever touching the rudder, and be able to do a much narrower turn, with the nose pointing exactly where you are going.

If it works, of course ;-)

And i think i know why you have some problems whit the animation as it seems there is no effect over 100 ft there for it seems it gets your interface then try to get back to center. or is kiled by something els would try investegate and get back if i find some importan. but it is only a visual side effect.

I suspect that this is caused by either the standard Microsoft Yaw Damper trying to "fight" my gauge, or some code in the 3d-model that I can't see and couldn't change. Maybe that animation takes the rudder trim value into account. I do not particularly care that much, as I am usually looking at the external model while on the ground (which, on an AES-enabled Airport, looks damn near photographic... awesome!) and stay inside the plane while flying. On the ground, my gauge will (or rather: should) do nothing.

[KAPTEJNLN 237]

well tryeid to turn off under flight controls on the overhead panel SEC 1 then it works and mean it turns off FBW so a limitation again, as default plane works the same way.

[nbz 11]

well tryeid to turn off under flight controls on the overhead panel SEC 1 then it works and mean it turns off FBW so a limitation again, as default plane works the same way.

I'm not sure if I am reading you correctly, I don't understand...

... is the problem that my gauge is OFF when you turn off SEC1?

... is the problem that my gauge is still ON when you turn off SEC1?

... is the problem that my gauge somehow messes up the SEC or FlyByWire or something else?

What it is *supposed* to do, is to only be active (=providing an artificial rudder), if and only if

1.) you have FBW turned on

2.) you are more than 100 ft above ground

As far as I know, the *Microsoft* FBW stuff is all hooked to that SEC1 switch. Aerosoft may have hooked way more functions to the other switches but what is of concern here is to solve that the rudder axis is dead once Microsofts FBW kicks in. In all other cases the rudder is perfectly fine ... at least according to me :-)

However, thank you for the first bug found anyway - I just tested switching on and off SEC1 several times, and using the autopilot in both conditions, and found TWO bugs... pick yours:

1.) fiddling with the autopilot (MAYBE in combination with SEC1) can turn OFF my gauge. I guess this is what you mean. This MAY be due to Aerosoft fiddling with SEC1 themselves (and my gauge shutting off once that is no longer on), nothing wrong with that, I am *guessing* right now that I can fix that without touching any more original Bus X code.

2.) while the gauge returns the abused trim variable to zero when going below 100 ft, it FAILS to do the same when SEC1 is turned off, leaving you with a bogus rudder trim. This can then be manually trimmed in normal fashion, but should not happen in the first place.

[Chris_327 42]

I am absolutely delighted to hear of your improved experience in LOWI !!

Guess what - back in the days before the Airbus X, flying that turn into LOWI with the default MS-Bus was the reason why I just HAD to come up with a way to seize control over the Rudder with FBW on. But after having done the first version of the gauge, I thought it was not good enough to have to do all the inputs myself (like you would do on a Piper Cub) on such a sophisticated FBW-plane. In LOWI, you get to fly a high bank angle at low speeds, and there it shows that the Bus does not coordinate turns, the nose pointing "out of" the circular path you fly.

This is why I came up with that second "Auto-Rudder" version. You should be able to fly nicely into LOWI without ever touching the rudder, and be able to do a much narrower turn, with the nose pointing exactly where you are going.

Well, I shall definately give the Auto-Rudder another go then! Having said that, I do like the sound of editing the deflection for the 'simple' gauge. This sounds more like the actual airbus from what I gather from smartcockpit.com, which states that input can be automatic (like the scenario created by the 'autorudder' gauge) or mechanically via the rudder pedals, which are then electronically limited by speed of the aircraft.

I think that to my tastes, I prefer to have rudder control left to the pilot, as I like to be able to just yaw the aircraft without banking for small adjustments, for instance when counter-acting a crosswind on the localiser. As I say, I will try out the auto-rudder again, but also make the adjustments you suggested to the 'simple' gauge, and tweak it to my tastes. I will again let you know of any progress I have, but it might be a day or two, I've got a couple of long shifts at work coming up!

It's great to hear you also enjoy that LOWI turn, I'd say that a large majority of my flights within FSX depart or arrive at LOWI, just because of the procedures flown!

Chris

[KAPTEJNLN 237]

good news

Update and what i was talking about is that i think it seems to bee FBW you fight against as SEC1 turns it off FBW

[nbz 11]

This sounds more like the actual airbus from what I gather from smartcockpit.com, which states that input can be automatic (like the scenario created by the 'autorudder' gauge) or mechanically via the rudder pedals, which are then electronically limited by speed of the aircraft.

From what I gathered from PPRuNe, the way the Airbus deflection limiter works is actually a mess. Instead of translating a full rudder input into a small deflection at high speed, the rudder-input-to-deflection ratio stays the same, but an absolute limit is then set. This means that you may get only 10% rudder travel at some speed, but you get that when applying rudder to only 10% of pedal travel - the maximum rudder the system will deem good for that speed, 10%, is reached after only 10% of rudder travel, which makes it impossible to do fine inputs at high speed.

I was reading that, looking for info on how that stuff works, because I wanted to code such a thing. Now I'm not a pilot and may have misunderstood most of everything said there, but I guess not. The ruder pedals were described as "on/off switches" at high speeds, leaving not much room for erring on the wrong side. IF I am right, that limiter system is silly - I decided to do a proper one that spreads all of the allowed travel at a certain speed over the entire range of the controller axis.

I think that to my tastes, I prefer to have rudder control left to the pilot, as I like to be able to just yaw the aircraft without banking for small adjustments, for instance when counter-acting a crosswind on the localiser.

The AutoRudder gauge still leaves the pedal function there, but you are right, it's the *small* corrections that are harder to do with that gauge, because the rudder overrides the auto-rudder only progressively (one third of pedal travel or more completely shuts off the autorudder) but when only small inputs are done, the autorudder will "fight" the manual input. I may cook up something a little less rude here.

My tip for flying with the Autorudder gauge is to do *gentle* rudder inputs, slowly pressing the pedal down AND slowly letting it come back up.

I don't *know* this stuff but I *guess* that in a real Airbus, you only have your feet on those pedals to be able to react to an engine failure or such immediately, not to constantly manually fly turns, while both other axis are controlled by an army of computers.

I would prefer to fly approaches by crabbing, using only banking for directional control (if possible), as the AutoRudder is very responsive to banking and small turns can be made quickly without touching the rudder. Of course, a final rudder input has to be made for alignment with the runway, but this can usually wait until below 100 ft, where all this talk becomes irrelevant anyway as the normal rudder works splendidly there.

[Chris_327 42]

I had a quick go with the autorudder gauge yesterday to have a look at what you said about the gauge 'fighting' the rudder input, and I see what you mean, you can still use manual input which is great.

I also had a go with the autorudder gauge at LOWI with excellent results; if you have a look at the attachment (ok so I messed up the first turn; slewed out to the wrong position, the indicated one is the best), the gauge achieves a tight turn with no rudder input from myself: all done by the gauge.

Incase it's useful,

avg spd through the turn and approach ~ 150kts

weight - default when a/c loaded in fsx, fuel 30% in left and right tanks, 0% in centre

30' of bank maintained throughout turn

Also, since using the autorudder gauge, it's worked every time

Chris

[nbz 11]

I also had a go with the autorudder gauge at LOWI with excellent results; if you have a look at the attachment (ok so I messed up the first turn; slewed out to the wrong position, the indicated one is the best), the gauge achieves a tight turn with no rudder input from myself: all done by the gauge.

Thank you, this seems to indicate that it indeed works not only on my own system!

That is an amazing approach course that you flew on that screenshot. Almost looks like a schematic drawing on an approach chart!

Yay, we now have FBW on the rudder too!

Also, since using the autorudder gauge, it's worked every time

That's very nice to hear, thank you.

_____

I have not yet planned my weekend, but I hope to find the time to do some more work on that stuff. This will all be done inside the gauge, no more deleting of code. Most importantly to fix the bugs found so far:

1. Gauge turns off (under certain conditions I have yet to figure out) when the AP is used. Not acceptable.
2. Gauge leaves rudder trim messed up when turning FBW (SEC1) off. Not acceptable either.
3. One more I found: slight yaw oscillations when flying high with AP on. This is likely the result of an interaction between my gauge and the AP. May be easy to fix, maybe hard, don't know yet. But I know that this has to go away.

I also wish to figure out a better way to keep the AutoRudder from fighting the manual rudder inputs. I already have a promising idea for that, but have not tried it yet.

In case I can not do a better job in this regard, I'm considering to implement some sort of "approach mode". This would probably work just like the AutoRudder gauge most of the time, but will turn into the simple gauge (without the "Auto") when, for example, the gear is down, flaps set at more than 2, radio altitude below 400 ft, and bank angle less than 10° - or such.

I have also found a line of code that contains a superfluous division, that always results in zero! Now, the intended value that should be there actually IS zero, but why not just write "0" in the first place? It's rare in programming to do something really silly, which does NOT result in a bug!

Thank you KAPTEJNLN and Chris_327 for your tests.

I politely invite everyone who complained about the rudder on this forum to join in... :-)

[Chris_327 42]

1. Gauge turns off (under certain conditions I have yet to figure out) when the AP is used. Not acceptable.
2. Gauge leaves rudder trim messed up when turning FBW (SEC1) off. Not acceptable either.
3. One more I found: slight yaw oscillations when flying high with AP on. This is likely the result of an interaction between my gauge and the AP. May be easy to fix, maybe hard, don't know yet. But I know that this has to go away.

I've had a few more days playing around with the autorudder gauge. So far, I haven't been able to replicate the gauge turning off when AP is used. I have just been doing standard flights and disconnecting the AP once established on the localiser, aswell as obviously testing it in turns after disconnect, e.g. at LOWI and LFMN.

I've never played around with turning of FBW as for all it's bad points, I quite like the FBW, and wouldn't want to fly my 'Bus like a Boeing! I can't therefore comment on whether I get messed up trim when turning SEC 1 off and on.

Finally, I've never experienced these yaw oscillations when cruising at altitude. There is no visible movement, and I can't see any movement on the EICAS page either. I've tested both at normal sim rate, and at 4x speed, as I regularly use acceleration when I haven't got the time to sit back and enjoy the flight!

I hope this helps somewhat, I guess what we could do with is some other feedback but nevermind!

Chris

p.s. thanks for the kind words about my approach, tbh I think it's more luck than anything else, as you can't see the runway from the left seat and have to make an estimate of where you're going to end up. Not standard approach procedures but hey, I don't have a co-pilot so it'll have to do!