Board index FlightGear Development Aircraft

B-2 Spirit

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

B-2 Spirit

Postby glazmax » Tue Jan 30, 2007 12:32 pm

Hallo there!

Getting used to flying with flightgear over the past years, I recently discovered the pleasure of flying the B-52 in multiplayer. That put me in the "mood" for creating "the beast", a rather odd worldly looking aircraft for this sim. Persuing this thought I set up the FDM (YASim) and created a 3D model of the B-2 Spirit and "stole" some 3D panels from other aircraft.
Now time has come to share: Since it is a bit annoying not be able to see or be seen in multiplayer when the corresponding aircraft model is not installed, I would donate the plane to the project if there is interest.

Currently the FDM is flyable, but you have to correct for yaw permanently, the 3D model is ready with some sections needed to be remodeled more accurate, animations work and a spare 3D cockpit should be sufficient for the first flights.

I would be glad to receive some feedback on how the plane could be improved.

Have a nice day,
glazmax

The B-2 waiting for action
Image
Visiting the Nimitz
Image
User avatar
glazmax
 
Posts: 453
Joined: Sat Jan 27, 2007 11:38 am
Location: Austria
Callsign: jettoo/OE-JTO
Version: cvs

Postby Pensacola » Tue Jan 30, 2007 3:28 pm

Those are small shots but from what I can see it looks great!
Pensacola
 
Posts: 14
Joined: Tue Jan 02, 2007 6:18 pm

Postby glazmax » Tue Jan 30, 2007 7:48 pm

Thanks! :-)

Maybe this is more informative (aligning with the tanker cost me a pound of sweat)
Image

Image

And here she is:B-2.tar.gz

Enjoy, and prevent your rudder control from overstress :-)

glazmax
User avatar
glazmax
 
Posts: 453
Joined: Sat Jan 27, 2007 11:38 am
Location: Austria
Callsign: jettoo/OE-JTO
Version: cvs

Postby HeliFlyer » Wed Jan 31, 2007 9:58 pm

Hi Markus,

very nice bird judged from the screenshots as I cannot fly it with OSG-CVS and PLIB-CVS, both give me a Yasim error
YASim SOLUTION FAILURE:
Insufficient elevator to trim for approach
Did you test it with the latest FG CVS versions?

Beside this, the real developer's music is played on the FG developers mailing list (see FG homepage). You should point out to your project to get it into CVS after fixing the little problem.

Regards
Georg EDDW
HeliFlyer
Retired
 
Posts: 230
Joined: Thu Jan 04, 2007 12:56 am
Location: Bremen, Germany

Postby glazmax » Thu Feb 01, 2007 2:04 am

Thanks for the testing HeliFlyer!

During development of the FDM I experienced that I am walking on thin ice with this project, since it is a bit untypical for the FDM. Nervertheless that pushes me even stronger in willing to archive the apparently unreachable.
The tests were done with 0.9.10 as I wanted the plane for the user to be functionable but you are right when you point out that I have to test in cvs as well.
Seems that I have to join the mailing list, although this form of communication seems to be quite ancient to me. :twisted:

- Changed approach speed and elevator (maybe it works now with CVS, didn'n check yet - CVS about to be installed the next days)

- As I saw some people having problems to maintain straight flight, I misused the autopilots wing-leveler for sideslip/yaw correction:
fire up Ctrl w after takeoff, and the plane is much more stable and you can use the drag control (controlled by rudder) as well.

The new version can be downloaded at the link some postings above.

Thanks for the input,
glazmax
User avatar
glazmax
 
Posts: 453
Joined: Sat Jan 27, 2007 11:38 am
Location: Austria
Callsign: jettoo/OE-JTO
Version: cvs

Postby DFaber » Thu Feb 01, 2007 11:24 am

Hi glazmax,

Really nice model, full of surprises. I've seen some screenshots of your model before and wondered what FDM it uses. I was quite surprised to find out it is possible to build a Yasim FDM without a vstab!!

Just tested your Model with OSG CVS and it flies impressingly well, although I think the original uses some "rudder-emulation". Maybe a nasal script to control some spoilers will do. You could also try to split the wing into two parts and give the outer wing higher idrag.

Great work!
DFaber
 
Posts: 709
Joined: Fri Dec 01, 2006 8:51 pm
Location: Aachen, Germany
Version: GIT
OS: Linux

Postby HeliFlyer » Thu Feb 01, 2007 6:57 pm

Hi Markus,
after your changes the aircraft works fine under OSG-CVS and PLIB-CVS.
Detlef said already all regarding the rudder-stearing with other (than side rudder) computer controlled control surfaces of the original, one also has to implement for this model in this or that way.
I was really impressed that you have created a very nice glass cockpit. The default field of view is 55 deg, I would suggest 60 deg (or something around) to have a better instrument overview - especially once you have made the instrument buttons clickable and functionable :-). Just a hint.
Anyway, with or without improvements, this bird belongs into FG CVS!
Regards
Georg EDDW
HeliFlyer
Retired
 
Posts: 230
Joined: Thu Jan 04, 2007 12:56 am
Location: Bremen, Germany

Postby glazmax » Fri Feb 02, 2007 3:18 am

Thanks guys, that feedback really lightens me up! :-)

Regarding the drag control for yaw (in "normal" planes rudder) I think I am pretty close to the original sceme, since pilots stated out that the B-2 flies like a big plane (e.g. KC 135), so you have to have your rudder controls mapped to a device that effects yaw - in this case drag by spoiler (although its named flap in yasim, it's not more than a spoiler with a lot of drag). Although in the real B-2 all input is computed, clamped to a secure range and than the contol surfaces are operated accordingly to stay in that range. If you would own a 2 billion dollar plane you would also take no risks BUT I have no idea where the secure range of a B-2 is situated.
A bit of history: Since the development of the YB-35 (not to mention the Holten brothers ..) pilots rated the performance of a wing-only aircraft in a very different way, as it tended not to react as a known normal airplane. Instead pilots got fond of the idea (those who managed to figure out how it worked) or they disliked it because of that unexpected new strange behavior. This led to the conclusion that the behavior of the plane had to be clamped in order to allow more pilots to be able to fly it, which was very disliked by the pilots that managed to handle the plane because it meant to reduce the true capabilities of the aircraft.
That brings me back to the clamping, the B-2 (as far as I know) has 3 known different combat modes (takeoff, flight, approach) each with their own secure range.

The B-2 I modelled has so far no restrictions, and you can feel like a test pilot trying out what's possible (I hardly ever crashed :roll: ), you can take the autopilot as a help but only within a certain range of maneuvers, if you push it too hard, you have to correct by yourself. My suggestion is to disable the autopilot in ground dodging flight because you get used to the behavior of the plane, it sharpens your senses and its much fun ;-).
So when I want to model the capabilities of the original/clamped B-2 every input has to be compared to an allowed behavior of the aircraft. At the moment the only clamping possible is done by autopilot but you guys are right when you state out that maybe flight modes should be selectable,that map output to 1 or more control surfaces depending on the force of the input. <-- This sounds like deep science to me .. I have to think about

I like the idea of Detlef to split the surfaces because of the fact that the outer wing would have more taper resulting in more idrag which should give the plane more stability.

The default field of the cockpit is also not satisfying for me because in terrain dodging flight you need about 55deg but for the rest (i real life maybe 99%) it would be of advantage to see more of the controls, I'll check out.
Functionable buttons will follow, but I rely on the inventors on some of the controls and I don't want to invent the weel for a second time ..

At last I have written much , maybe too much but thanks to everyone who got that far. Thanks for your input Detlef and Georg since it is very important for me in focusing my thoughts and thinking of improvements.

Have a nice day,
glazmax
User avatar
glazmax
 
Posts: 453
Joined: Sat Jan 27, 2007 11:38 am
Location: Austria
Callsign: jettoo/OE-JTO
Version: cvs

Postby glazmax » Sun Feb 04, 2007 10:46 pm

New version avalaible: B-2.tar.gz

edit: bug-fix done (rudder-trim is now reset when selecting other AP mode)

Improved auto yaw/sideslip:
1.) bug sideslip damper/rudder input possible - engaged at startup suitable for approach/takeoff and flight(nav1-hold misuse Ctrl n)
2.) heading sideslip damper/no rudder input possible - suitable for flight (wing-leveler misuse Ctrl w)

Have fun,
glazmax

The new and the old flying together:
Image
User avatar
glazmax
 
Posts: 453
Joined: Sat Jan 27, 2007 11:38 am
Location: Austria
Callsign: jettoo/OE-JTO
Version: cvs

Postby glazmax » Mon Feb 26, 2007 6:22 pm

New release version 0011:

We have flymodes now :-), and many more!

The flymode buttons are clickable as some instrument buttons are, although sometimes they are loaded wrong and you have to click above the buttons( or restart flightgear, works most of the times).
If someone has some information regarding switches and buttons in the B-2 middle panel I would be pleased to receive it.

Download here:B-2.tar.gz
or at the links in the postings above.

Enjoy,
glazmax
User avatar
glazmax
 
Posts: 453
Joined: Sat Jan 27, 2007 11:38 am
Location: Austria
Callsign: jettoo/OE-JTO
Version: cvs

Postby grtux » Tue Feb 27, 2007 8:52 am

Hello, glazmax

Thanks for that strange Aircraft (the most expensive),
your model is wonderful,
that last version is nicely flyable.

I took pleasure to fly it.

Regards
g.robin
LFMO
User avatar
grtux
 
Posts: 431
Joined: Thu Dec 14, 2006 5:19 pm
Location: Provence France

Postby glazmax » Tue Feb 27, 2007 9:36 pm

Thanks for the "flowers" grtux, the pleasure is mine :D !

Now I will have to dig in osg for new functions to come.

Fly on,
glazmax
User avatar
glazmax
 
Posts: 453
Joined: Sat Jan 27, 2007 11:38 am
Location: Austria
Callsign: jettoo/OE-JTO
Version: cvs

Re: B-2 Spirit

Postby SP-NTX » Sun Jan 31, 2021 7:51 pm

Is this plane still in dev?
Freedom can be frightening if you've never felt it
User avatar
SP-NTX
 
Posts: 212
Joined: Wed Mar 18, 2020 12:14 pm
Location: Western Poland
Callsign: SP-NTX
Version: 2020.4.0
OS: Windows 11

Re: B-2 Spirit

Postby Delta5142 » Sun Jan 31, 2021 8:23 pm

Well.. this project is insanely old, from 2007, so i'm not so sure
Delta5142
 
Posts: 340
Joined: Thu Jul 02, 2020 4:18 pm
Location: Northern California
Callsign: DAL5142
Version: next
OS: Mac OS X

Re: B-2 Spirit

Postby tdammers » Mon Feb 01, 2021 11:12 am

Probably not.

And even if it were, I wouldn't expect too much from it - the 3D model looks alright, but the way the real-world B-2 works, yasim is with 99.9% certainty the wrong choice. The B-2 has extremely quirky flight characteristics; like other stealth aircraft of the era, it is unstable in all 3 axes and requires constant automatic corrections from the FBW system. YASim simply isn't made to model such aerodynamics - essentially, what YASim does is solve the FDM for the given constraints (wing shape, fuselage, etc.) to create a stable situation in the demanded flight situations (usually cruise and final approach) - but the B-2 isn't stable at all, ever, so whatever solution YASim finds is going to be wrong.

The right way to build this aircraft in FG would be to find wind tunnel data, or at least guesstimate those from the airframe shape, and implement that in JSBSim; this will produce an aircraft that matches the real one, more or less, but will be pretty much entirely unflyable. You would then add a closed-loop FBW system that takes pilot input and interprets it as target roll, bank and yaw rates, and keeps the aircraft artificially stable. You'd probably also have to add envelope protections to that (limiting the aircraft to safe attitudes within the normal flight envelope, e.g. preventing the pilot from accidentally overbanking the aircraft); I believe with an airframe this quirky, pushing past the envelope limits would be even more devastating than in a more conventional design. E.g., what do you think would happen if you threw this aircraft into a spin...

Long story short, this is an extremely complex aircraft, *and* a highly classified one, so if you want to build it right, that'll be an extremely difficult task.
tdammers
 
Posts: 391
Joined: Wed Dec 13, 2017 11:35 am
Callsign: NL256
IRC name: nl256

Next

Return to Aircraft

Who is online

Users browsing this forum: No registered users and 16 guests