Board index FlightGear Development Aircraft

McDonnell Douglas MD-11

Questions and discussion about creating aircraft. Flight dynamics, 3d models, cockpits, systems, animation, textures.

Re: McDonnell Douglas MD-11

Postby Thorsten » Thu Jul 14, 2016 11:23 am

I'm guessing it is a fly-by-wire system, so the same comment holds - full pull on the yoke may not correspond to full surface deflection, the FCS can be made to adjust the gains dependent on situation.

I'm guessing you know the full deflection range, so it stands to reason that this is available under some conditions (otherwise the engineering would have been different).
Thorsten
 
Posts: 11331
Joined: Mon Nov 02, 2009 8:33 am

Re: McDonnell Douglas MD-11

Postby Octal450 » Thu Jul 14, 2016 12:25 pm

@Thorsten
Yeah, there is FBW, but it's much more minimal than in the Airbus. It *helps* you adjust attitude, but in the case of a failure (ex: LSAS off, or something like that) the plane is very heavy on the controls. It should be that way even without the FBW implemented. Ailerons are not FBW in the MD-11.

Best,
Josh
Waste of time. Goodbye forever.
Octal450
 
Posts: 4398
Joined: Tue Oct 06, 2015 12:51 pm

Re: McDonnell Douglas MD-11

Postby tikibar » Thu Jul 14, 2016 7:03 pm

Josh,
Don't go reinventing the wheel on the hyrdraulics. I have a good system in the 757/767 that can be easily adapted for three engines/systems. The hardest part is getting the panel all rigged up.
Boeing 747-8 (rename folder to 747-8i)
Boeing 757-200/300 (rename folder to 757-200)
Boeing 767-300/ER (rename folder to 767-300)
McDonnell Douglas MD-11 (rename folder to MD-11)
User avatar
tikibar
 
Posts: 515
Joined: Mon Mar 05, 2012 6:05 am
Location: Los Angeles
Callsign: CHT0009
OS: Ubuntu 14.04

Re: McDonnell Douglas MD-11

Postby Octal450 » Thu Jul 14, 2016 7:19 pm

@tikibar
OK, but I happen to think linking things is the easy part, making the systems is the hard part. ;)
note that the m.blahblahblah and me.blahblah and (getBoolvalue) and so on stuff in AFDS.nas and similar, I don't understand. But I'll give your hydraulics a go.

Josh
Waste of time. Goodbye forever.
Octal450
 
Posts: 4398
Joined: Tue Oct 06, 2015 12:51 pm

Re: McDonnell Douglas MD-11

Postby tikibar » Thu Jul 14, 2016 8:45 pm

:) Everybody's got their strengths. Why don't I get the system working, then you can plug it into the panel. Then we each do the parts we're good at.
Boeing 747-8 (rename folder to 747-8i)
Boeing 757-200/300 (rename folder to 757-200)
Boeing 767-300/ER (rename folder to 767-300)
McDonnell Douglas MD-11 (rename folder to MD-11)
User avatar
tikibar
 
Posts: 515
Joined: Mon Mar 05, 2012 6:05 am
Location: Los Angeles
Callsign: CHT0009
OS: Ubuntu 14.04

Re: McDonnell Douglas MD-11

Postby Octal450 » Thu Jul 14, 2016 9:56 pm

OK sounds good. Initial FLCH is installed, working decently, not I gotta setup the thrust modes.
Waste of time. Goodbye forever.
Octal450
 
Posts: 4398
Joined: Tue Oct 06, 2015 12:51 pm

Re: McDonnell Douglas MD-11

Postby ph-cor » Thu Jul 14, 2016 11:43 pm

To 411,
About your question in the Mumble popup chatbox this evening:

What I experenced two days ago, in a long flight with KL-666 from EHGG to LFSB, following heading instructions from LFSB ATC. The aircraft rolled suddenly away nose down. Disconnecting a/p and a/t did not help to get the MD-11 manual under control. A realistict crash was the result. I restarted fresh at LFGA, 31 miles north of Basel, and came in the same situation. A new Bermuda triangle?

Greetings,
ph-cor
ph-cor
 
Posts: 16
Joined: Tue Aug 19, 2014 4:46 pm

Re: McDonnell Douglas MD-11

Postby Octal450 » Fri Jul 15, 2016 1:17 am

@ph-cor

Cannot reproduce the issue, sorry.

Works fine on my end, try completely uninstalling.The latest is available here.

Josh
Waste of time. Goodbye forever.
Octal450
 
Posts: 4398
Joined: Tue Oct 06, 2015 12:51 pm

Re: McDonnell Douglas MD-11

Postby Thorsten » Fri Jul 15, 2016 5:47 am

Yeah, there is FBW, but it's much more minimal than in the Airbus. It *helps* you adjust attitude, but in the case of a failure (ex: LSAS off, or something like that) the plane is very heavy on the controls. It should be that way even without the FBW implemented. Ailerons are not FBW in the MD-11.


One of us is not getting something here - or I don't understand the filter discussions you're having at all.

You have your input properties - let's take elevator-cmd-norm for instance. That's a measure of how you move your virtual control device, with [-1:1] the full deflection range. Your actual device (say a joystick) can be configured to utilize part of that somehow, but that's no concern of the aircraft maintainer.

The real plane has a given airfoil deflection range, say -30 to + 30 degrees and that can be elevator-pos-norm [-1:1].

If we're not having fly by wire but pulling the stick physically pulls the airfoils, as long as the system is intact you have to use elevator-cmd-norm = elevator-pos-norm and pulling stick full means full deflection.

If you have fly by wire, the relation can be anything.

You can for instance only have the range elevator-pos-norm [-0.5:0.5] available via the stick (a gain of 0.5) such that full stick elevator-cmd-norm = 1 implies half airfoil deflection elevator-pos-norm = 0.5 (then you'll have to use relatively large stick movements to control the plane) and have the rest of the range available for auto-trim.

You can have a Mach-number dependent mapping such that for low airspeed the full range is available but as you pick up speed, sensitivity is lost - so you get approximately the same response of the plane at each airspeed even with the pitching moment caused by the elevator Mach-number dependent.

You can in fact dispense with the linear relation between elevator-cmd-norm and elevator-pos-norm and have the first as a bias in a PID controller such that stick motion never controls airfoils but pitching rate.

So there's always a

cmd-norm -> Map -> pos-norm

in fly-by-wire.

The point I'm trying to get across is that 'Map' (i.e. what the FCS does) can be different for the AP and the manual mode. Regardless of how the plane in manual 'feels' (heavy on the controls, like on rails,...) - that's NOT what the AP has to feel - the AP might well utilize a mode that might feel 'light' on the controls.

Also note that 'feel' is as much a function of the FCS as it is of the raw aerodynamics - 'feel' can be designed to be comfortable for the pilot. The Shuttle flown under 'stick controls airfoils' is very twitchy at hypersonic speeds, decent at supersonic and overly sensitive on subsonic speeds - under the Aerojet DAP a PID controller takes all of that away and you get a vehicle with a crisp and clean response which feels about the same in any regime.

It seems to me you're trying to design an AP that's essentially an AI pilot, i.e. has to cope with the control devices and mappings as exist for a human pilot - but that may not be at all how the system works underneath (there's no particular reason to design the logic that way).
Thorsten
 
Posts: 11331
Joined: Mon Nov 02, 2009 8:33 am

Re: McDonnell Douglas MD-11

Postby Octal450 » Fri Jul 15, 2016 1:01 pm

If this was the Airbus A320 or so, then that would apply. But the MD-11 FBW is only there to help the pilot on approach. The inputs he makes the yoke, go to a computer, but they go unchanged to the elevators. (mostly) MD-11's LSAS will sometimes adjust the trim, to help you, especially since on approach, the aircraft is a bit unstable.

I don't think it's right to fix FDM by using FBW type system. The fact still remains: The MD-11 is heavy on the controls, even without any computer changes.

So that is something to fix in the FDM, flight dynamics model, not using a pid-controller. Your FDM in the shuttle is correct, the real space shuttle probably does the same thing at supersonic speeds. So you built a computer to fix that -- as did NASA. (I think)

The AP was previously tuned for flying a heavy controls jet, so if we adjust the FDM, then nothing else needs to be changed = less time worrying about that, and more time spent on other systems, flightdeck, etc.

Best Regards,
Josh
Waste of time. Goodbye forever.
Octal450
 
Posts: 4398
Joined: Tue Oct 06, 2015 12:51 pm

Re: McDonnell Douglas MD-11

Postby Thorsten » Fri Jul 15, 2016 3:02 pm

I don't think it's right to fix FDM by using FBW type system.


Then I don't understand your conversations at all. If the FDM has an issue, you should not talk about adjusting gain filters but about pitching, yawing or rolling moments as functions of airfoil deflection.
Thorsten
 
Posts: 11331
Joined: Mon Nov 02, 2009 8:33 am

Re: McDonnell Douglas MD-11

Postby Octal450 » Fri Jul 15, 2016 3:15 pm

@Thorsten
In retrospect, it was a mistake to attempt to fix that with gain, I agree.

Best,
Josh
Waste of time. Goodbye forever.
Octal450
 
Posts: 4398
Joined: Tue Oct 06, 2015 12:51 pm

Re: McDonnell Douglas MD-11

Postby wkitty42 » Fri Jul 15, 2016 5:03 pm

it0uchpods wrote in Fri Jul 15, 2016 1:01 pm:The MD-11 is heavy on the controls

when you say this, are you saying that it is slow to respond and kinda sluggish? or are you saying something else?
"You get more air close to the ground," said Angalo. "I read that in a book. You get lots of air low down, and not much when you go up."
"Why not?" said Gurder.
"Dunno. It's frightened of heights, I guess."
User avatar
wkitty42
 
Posts: 5913
Joined: Fri Feb 20, 2015 3:46 pm
Location: central NC, USA
Callsign: wk42
Version: git next
OS: Kubuntu 14.04.5

Re: McDonnell Douglas MD-11

Postby Octal450 » Fri Jul 15, 2016 6:07 pm

@wkitty42
Yup. It takes more aileron to move than, per-say, a 767.
Waste of time. Goodbye forever.
Octal450
 
Posts: 4398
Joined: Tue Oct 06, 2015 12:51 pm

Re: McDonnell Douglas MD-11

Postby Octal450 » Sun Jul 17, 2016 11:29 pm

OK FDM is flying a bit nicer now, whatever you did tikibar, it's wonderful!

Here is what I have been working on.

advanced CONFIG. Will tell you what you need to do, to clear CONFIG on EAD, like SPOILERS, AUTOBRK, PARK BREAK, FLAP, LANDING GEAR, etc
EAD2 engine and Config displays, you can switch between them using the buttons below throttles.
Tape style engine displays on KLM and EVA Air liveries.
Fully custom MD-11 style FMS, FADEC. (Yes!)

I would like to install my IT-AUTOFLIGHT, but keep the current panel and dialog. (I built it extremely modular, any panel and dialog will work)

This allows for my FLCH system to work natively, and I can easily update it. Also, this will fix FMS/NAV to go direct when the next waypoint is changed, and then the MD-11 can get VNAV (PROF) when I finish that. It also has support for my FMS system (in private development), so when that is finished, it will listen nicely to my MD-11 style FMS. Also, it has working autoland (in testing, included with V2.11 or V2.12). Is this acceptable, or do you want to keep the current AP?

You can try my system out in several planes: my MD-88, MD-90, A346VIP, KC-137R, and quite a few more. But note that a few bugs with vertical mode using too much trim was fixed in V2.11 coming soon.

Regards,
Josh
Waste of time. Goodbye forever.
Octal450
 
Posts: 4398
Joined: Tue Oct 06, 2015 12:51 pm

PreviousNext

Return to Aircraft

Who is online

Users browsing this forum: asr and 2 guests