Two pedal sets and one tiller ... what can be done for "soft" switching ?

[eudoniga]

Dear mr. Dowson,

lately I took the time to re-read a good part of FSUIPC documentation, including some parts of the SDK ...

I wish I had enough knowledge to benefit from this simple reading to understand how I could solve one problem, but it's not so.

I hope you won't regret to dedicate a few passages to clear out this subject, which deals with connecting to the server pc (the "flying" machine) three devices which control (in FS9) the same axis, that's the rudder.

I am not aware if FS9 has a dedicate axis assignment for the tiller, so that one part of the problem is already taken care of by FS.

Therefore I would like to know - apart from hardware solutions (thinking about a USB switch), what could be managed via software.

I know that there is one offset dedicated to flag the on ground/airborne status.

That's what I'd use to inhibit the tiller device or the rudder device, but I don't know how.

Then there's the double CH pedal set ... I do have a hardware USB switch, but would love to know if there's a "soft" way to serve only one set at a time, and that's the one where the flying pilot is manoeuvering with his feet ...

May you suggest some path ?

Thanks a lot for your patience and dedication.

Eugenio.

[Pete Dowson]

I am not aware if FS9 has a dedicate axis assignment for the tiller, so that one part of the problem is already taken care of by FS.

I don't believe FS9 has -- one was added in FSX I think.

Therefore I would like to know - apart from hardware solutions (thinking about a USB switch), what could be managed via software.

I know that there is one offset dedicated to flag the on ground/airborne status.

That's what I'd use to inhibit the tiller device or the rudder device, but I don't know how.

Then there's the double CH pedal set ... I do have a hardware USB switch, but would love to know if there's a "soft" way to serve only one set at a time, and that's the one where the flying pilot is manoeuvering with his feet ...

I'm rather puzzled about what you are asking, and why you are asking it here. However, if you want my advice and have a registered install of FSUIPC:

May you suggest some path ?

Assign the tiller in FSUIPC "direct to FSUIPC calibration".

Assign both rudder pedals to rudders in FSUIPC,also "direct to FSUIPC calibration".

Make sure none are assigned in FS.

Calibrate tiller and one of the rudder sets, folowing the steps in the User Guide, but also check that the calibration is good for both sets of rudders.

FSUIPC will do two things for you:

1. It will manage the change between tiller and rudder according to ground speed, as documented. By default by 60 knots GS the tiller has no effect, the rudders have 100% effect.

2. It will arbitrate between the two rudder inputs, obeying the one which is providing the greatest deflection -- which will of course be the one in use.

Pete

[eudoniga]

... :shock:

See what I meant ?

I must have surely missed the part of documentation which describes such FSUIPC capacities !!!

You're telling me what I was looking for is already there ... 100 % ... aren't you ?

:D

Now is when I'd be ready to pay for a second, third, fourth ... registration fee !!!

Thanks a lot ... will follow your indications and hope I won't bother you any longer ... but be prepared, anyhow ...

Cheers and best regards,

Eu

[eudoniga]

Okay, so easy and fast ... FANTASTIC !!!

Just tried with FS9 and two joysticks.

Saw how the "blending" action works below 60 kts.

I read about the MaxSteerSpeed parameter, but I am now more interested in the opposite concept: how do I set it so that at average taxi speed (let's take 15/20 kts) the rudder is much less effective, and practically almost null at lower speeds (0-10 kts) ?

Brgds,

Eu

[Pete Dowson]

I read about the MaxSteerSpeed parameter, but I am now more interested in the opposite concept: how do I set it so that at average taxi speed (let's take 15/20 kts) the rudder is much less effective, and practically almost null at lower speeds (0-10 kts) ?

You miss the point I think. When on the ground the blending works both ways. There's a linear blending all the way from 0 to the MaxSteerSpeed, which defaults to 60. The rudder has no effect at 0, 50% at 30 100% at 60. The tiller has 100% at 0, 50% at 30, 0% at 60.

Pete

[eudoniga]

No, sir, I guess I understood the point ...

But there's something which - I think - should be avoidable.

As a passenger watching the wing out of the window, I saw the airline pilots executing the flight controls check; some times before moving off, sometimes when the a/c is already taxing.

I was told this check is usually performed by the F/O, without disturb to the steering activity of the Captain, which holds the tiller.

With the actual FSUIPC settings, if the Captain is taxiing and steering at - say - 20 kts, and the F/O checks the rudder pedals ... well, they will noticeably zig-zag (I tried it with FS9), and it's not quite realistic.

:-D

I think this could be avoidable, somehow; that's why I asked.

;-)

Best regards,

Eu

[Pete Dowson]

I was told this check is usually performed by the F/O, without disturb to the steering activity of the Captain, which holds the tiller.

How do you see the rudder from inside the cabin? Are we talking here about a small GA aircraft with a rear window? Which of these have steering tillers?

With the actual FSUIPC settings, if the Captain is taxiing and steering at - say - 20 kts, and the F/O checks the rudder pedals ... well, they will noticeably zig-zag (I tried it with FS9), and it's not quite realistic.

:grin:

I think this could be avoidable, somehow; that's why I asked.

The only check they get from this is a graphical representation of the rudder control position on the screens in front of them. This can be obtained from the true rudder input value, supplied by FS, instead of the reduced input actually being supplied. Programs like Project Magenta use this for that display.

If what you are saying is that you are not happy with the rudder response, you can use the "slope" facility in the rudder calibration -- select one with a flatter response. If you really want the rudder axis to be reduced even further than the 66% already reduced at 20 knots (i.e. 1/3rd of the way to 60 knots), you could try setting the threshold speed a lot higher -- just make sure you still have enough rudder authority before rotation unless you want to be forced to carry on using the tiller till later on takeoff roll (when you should have your hands full already!)

Alternatively, if you want the proportions between tiller and rudder to be other than a linear crossover, I can only suggest that you try using a Lua plug-in. You can assign both tiller and the rudders direct to a Lua plug-in program which employs your own algorithm before sending them back to FS for actuation. If you want to proceed that way please first read up about Lua (see the FSUIPC documents subfolder in your FS Modules folder). When you know exactly what you want to do i can advise you on which FS offsets to write back to for either direct rudder control, or for calibrated rudder control as you wish.

Regards

Pete

[eudoniga]

Roger, wilco.

I am simply looking for a balance where simulating the rudder check during low speed taxi does not lead the a/c to veer off on one side and then the other.

For other simmers' benefit, I will report back here about what I shall find as a good working compromise.

Thank you very much, mr. Pete, you're the one, true, living SDK for us mortals !!!

:cool:

Eu.

[Pete Dowson]

I am simply looking for a balance where simulating the rudder check during low speed taxi does not lead the a/c to veer off on one side and then the other.

I'm pretty sure that rudder, aileron and elevator checks tend to be done when static. Gentle rudder changes might be performed when taxiing to cross-check the screen indications, but the feedback then being looked for is surely in fact some directional effect on the aircraft, else what's the point? Airliner pilots don't have wing or door mirrors like cars, they cannot see if the rudder actually moves.

Regards

Pete

[Pete Dowson]

You questions have actually reminded me of a request i received a few weeks ago, from someone who actually does pilot training for 737NG. He said that the rudder is always partially effective, even at very low speeds, but doesn't gain much effectiveness until around 30 knots or so. I didn't think of a solution to this till now, but I've devised this method:

1. Rudder effective at 10% of the pedal input up until 50% of the threshold speed (set at 60 knots default, so this would be 30 knots),

2. Above the 50% threshold mark (i.e. 30-60 knots), increases linearly from the 10% input to the full 100%.

I've made it optional so that it doesn't mess folks up who've got used to the earlier system, or who are flying aircraft which don't behave in this way,. To enable it, i've simple set the threshold speed parameter to a negative one. So, eg

MaxSteerSpeed=-60

would accomplish this change for the aircraft affected by this JoystickCalibration section.

This change was only going to be made available for FSX (i.e. in FSUIPC4), because i've really frozen FSUIPC3 now. But additionally I'm afraid I am not able to test this change at present -- my cockpit is undergoing some changes -- so if you'd like to check it out and effectively test it for me, let me know and i'll see if i can add it to FSUIPC3 after all (albeit reluctantly).

Regards

Pete

[eudoniga]

"Good, better, best ... never let it rest !", thanks mr. Dowson for generosity and dedication.

I am more than honored by your request, for testing purposes I've got these two scenarios available:

- FS9 still solidly running at the base of my 737NG home cockpit. The tiller is under construction, but I can connect one extra joystick for our purpose and steer with the R axis as the tiller.

- FSX installed on a desktop machine with one X-box style controller and a couple old joysticks (so that I can split rudder from tiller control).

- FSUIPC registered on both.

Just tell me where/how to download the modified version, and I will begin with FSX.

Then, if you are satisfied with the results, I'd be glad to see you overcome your reluctancy and make this little present for the good ole FS9 people like me ...

:oops:

Regards,

Eu

[Pete Dowson]

Just tell me where/how to download the modified version, and I will begin with FSX.

Okay, thanks.

FSUIPC4859e test2

Remember to find the MaxSteerSpeed=60 line in the relevant or default [JoystickCalibration] section and change the 60 to -60.

Regards

Pete

[Pete Dowson]

FSUIPC4859e test2

That link is now dead. Version 4.859e is now available via the Download Links subforum, where you will also now find 3.999y which has the same change included for FS9.

Regards

Pete

[eudoniga]

Okay, back from the testing activity with ver. 4.859e (which I could d/l before the link was dead).

It feels much better now, I tested it with the default B738.

By the way, I got confirmation from one B737NG pilot and also from the B737 flight manual that flight controls may also be tested during taxi.

Therefore I waited until I was taxing at 10-15 kts max (15 being the speed limit within apron perimeter) to see what was the behaviour like.

I tested the rudder at full extension, and I could see no veering of the a/c, which - IMHO - is good and real.

As for the "old" algorhytm, I tested it with the B737-700 I have at the home cockpit, but there was really much too rudder influence at low speed ...

I will now test version 3.999y there, but I feel it will work okay also with FS9.

Will give you final confirmation a.s.a.p.

Bye for now and again many many thanks !

Eugenio.

[eudoniga]

Second session, this time at the home cockpit with FS9 and the (old) PMDG 737-700.

The a/c seemed to feel somehow lighter, so the ruddel pedals proved to be more effective even at low speeds.

Taxing at 8 knots, I needed to be very fast in pressing the pedals full down left then full down right, if i didn't want to see the nose deviating from its course ... so here I'd rather find a way to lower the steering action produced by the rudder.

During the flight controls check, I watched the lower eicas screen, and noticed that - in order to inhibit the rudder from the steering action - its whole movement is limited. The same I saw from outside view. Therefore airline simmers will take advantage from this modification, in the sense that their taxiing won't resemble a drunk pilot manoeuvre; they shall be happy by checking the flawless movement of the yoke and pedals, but won't have to rely too much on what they see from the flt control display ...

Yours the final word, sir.

[Pete Dowson]

Second session, this time at the home cockpit with FS9 and the (old) PMDG 737-700.

The a/c seemed to feel somehow lighter, so the ruddel pedals proved to be more effective even at low speeds.

Taxing at 8 knots, I needed to be very fast in pressing the pedals full down left then full down right, if i didn't want to see the nose deviating from its course ... so here I'd rather find a way to lower the steering action produced by the rudder.

I've no idea why you get different results on FS9. The rudder effect is limited to 10% of its default action whilst below 30 knots, just as in FSX. It sounds rather like you haven't assigned both the tiller and the rudder pedals "direct to calibration" and calibrated them. And did you remember to change the MaxSteerSpeed parameter to -60 in the relevant calibration section?

Regards

Pete

[eudoniga]

Will double check and report back.

Best regards,

Eu

[eudoniga]

Okay, I did a few sessions, and made sure that

- MaxSteerSpeed was set to -60

- all joystick assignments were removed and their axis sent to FSUIPC

- all relevant axis were properly calibrated within FSUIPC

I also played a bit with the "Slope" function, values from 5 to 10, for the rudder.

In my very very humble opinion, I do still get too much rudder control at low speeds, in fact I could zig-zag with the rudder even taxiing below 10 kts.

Possibly, that 10 % alone under 30 kts is sufficient to react to full depression of the pedals on both sides, and make the a/c veer sensibly.

Maybe you want to wait for some feedback from the professionals, I might be doing something wrong, even if I don't know what it is ...

Thanks for keeping us FS9ers upgraded, though !!!

Brgds,

EU

p.s. I noticed - from session to session - that is was enough for all other devices, but the CH pedals, to clear any axis assignment within FS9 at the first time.

Those pedals, don't know why, show up again every time I run FS9, so I must always go first and remove their "self-assignment" to the rudder and the brakes.

What's going on ?

[Pete Dowson]

In my very very humble opinion, I do still get too much rudder control at low speeds, in fact I could zig-zag with the rudder even taxiing below 10 kts.

Possibly, that 10 % alone under 30 kts is sufficient to react to full depression of the pedals on both sides, and make the a/c veer sensibly.

Hmm. Strange. Could you please try version 3.999y2, now downloadable from the Download Links subforum? I've made one small change. I can't see anything wrong here, at least withthat version.

p.s. I noticed - from session to session - that is was enough for all other devices, but the CH pedals, to clear any axis assignment within FS9 at the first time.

Those pedals, don't know why, show up again every time I run FS9, so I must always go first and remove their "self-assignment" to the rudder and the brakes.

What's going on ?

Both FS9 and FSX (and all previous versions as far as I know) perform auto-assignments to devices they think are "newly connected". Maybe they fall asleep or give sme signal indicating such. I've always recommended either assigning all axes in FS or all axes and buttons in FSUIPC, and in the latter case disabling FS controllers completely to stop auto-reassignment.

Regards

Pete

[eudoniga]

Hmm. Strange. Could you please try version 3.999y2, now downloadable from the Download Links subforum? I've made one small change. I can't see anything wrong here, at least withthat version.

Roger, wilco.

I've always recommended either assigning all axes in FS or all axes and buttons in FSUIPC, and in the latter case disabling FS controllers completely to stop auto-reassignment.

In fact I'm ready to have all axes and buttons in FSUIPC ... but how do I operate to "disable FS controllers completely" ?

Regards,

Eu

[eudoniga]

Okay, tested version 3.999y2.

It did feel different, but - if I was allowed to chose - I'd go radically for 0 % of rudder control with speed < 10 kts: at least, this would truly be my "-60" option ...

As for the positive values, they could be coupled to the actual new blending logic, the "old" one being abandonable.

IMHO.

Brgds,

Eu

[Pete Dowson]

It did feel different, but - if I was allowed to chose - I'd go radically for 0 % of rudder control with speed < 10 kts: at least, this would truly be my "-60" option ...

Hmm. I'll consider that. I'll chat with my expert pilot guy first. With a linear climb to, what, 10% at 30knots? So still only 5% at 20 knots? I don't think that would be right for airliners on landing rollout. I'd have to reduce the change to 20 I think, giving:

0% at 0-10

0% rising to 10% from 10 to 20 knots

10 rising to 100% from 20 to 60 knots.

The last would give about 19% at 30% instead of the 10% you have now.

It's starting to sound as if you need a different gradient for each range of 10 knots befow 30 ...?

As for the positive values, they could be coupled to the actual new blending logic, the "old" one being abandonable.

Sorry, you've lost me there.

Regards

Pete

[eudoniga]

Well, I obviously have no direct experience to suggest you how fast the blend should grow properly and progressively from 10 kts on to simulate a real liner response ... I'm happy with the 0 % of rudder control below 10 kts, for the reasons I've already told you.

No problem with the "old way" that you want to maintain for positive values of the param ... I didn't lose you at all !

;-)

brgds (and thanks again), Eu

[Pete Dowson]

Well, I obviously have no direct experience to suggest you how fast the blend should grow properly and progressively from 10 kts on to simulate a real liner response ... I'm happy with the 0 % of rudder control below 10 kts, for the reasons I've already told you.

I've added more facilities so that folks can configure this even more flexibly.

By setting the MaxSteerSpeed=Q (with no numbers) I set 4 (Q=quad) zones for the rudder influence. By default these are

0% up to 10 knots

linear 0-10% from 10 to 20 knots

linear 10-30% from 20 to 30 knots

linear 30-100% from 30 to 60 knots

This is also equivalent to the parameter given as MaxSteerSpeed=Q10,20,30,60. Use that format to change the thresholds. If you specify any numbers you must give all 4 and they must be increasing.

The tiller effect is still linear 100% at 0 knots down to 0% at the max (60 knots by default).

The original MaxSteerSpeed=60 and MaxSteerSpeed=-60 are also still available.

I'm not releasing this generally yet. To try it download FSUIPC 3.999y3

Let me know. I've not made the same changes to FSUIPC4 yet.

[EDIT]

BTW, just got this from my fried who trains real pilots:

"For Airbus you check the rudder while taxiing but for this you have a disconnect button on the tiller to disconnect the nose wheel from rudder.

For Boeing you check the all controls before taxiing ones. Because Boeing doesn't have a disconnect button or switch for nose wheel wouldn't be able and also it is not allowed to do so."

Which aircraft are you flying?

Regards

Pete

[eudoniga]

WOW, this is super !!!

Will try it a.s.a.p. and report back.

Which aircraft are you flying?

Me ? A virtual B737-700, sir.

And on the 737 Flight Manual (Continental Airlines version), Rev 11/15/02 #41, sec. 3 - Page 91, I read:

"The Captain will: while holding the tiller securely, smoothly and slowly displace the rudder full travel in each direction, verifying full travel and freedom of movement. Ensure that the rudder pedals return to the neutral position and that no abnormal nosewheeling steering pressures are noted, unless required by a single engine taxi situation."

Also I got mention about testing the flight controls after a/c moves off from a former B737NG Ryanair pilot, now flying the triple seven with Emirates.

To a simple simmer, that was quite enough to imitate ... wasn't it ?

Brgds,

Eugenio