Board index FlightGear Development Spaceflight

Space Shuttle

Discussion about development and usage of spacecraft

Re: Space Shuttle

Postby Thorsten » Tue Jul 07, 2015 2:38 pm

By the sound of all of this, I would say we need to do something like X-Plane does--give the AI object its own FDM, instead of relying on the ai system to figure all this out, which seems to be impossible.


We need no more and no less than what I wrote - a numerically stable solver for orbital dynamics. The accuracy of analytical orbits seems quite enough though.

(You might not know this, but I am a theoretical physicist with more than a decade experience in numerical solutions of physics problems - so I frankly think it's unlikely that this discussion will significantly expand my insights what needs to be done - so unless someone else actually wants to start before orbiting AI objects reach the top of my to-do list, we're not in need of ideas but coded algorithms).

For those who have actually read the operations manual, we can now do a new class of interesting failure scenarios - helium pressure loss in an RCS or OMS pod, dropping fuel pressure due to blowdown dependent on the fuel fraction at the helium leak, once this fails cross feeding either RCS to RCS, OMS to OMS or OMS to RCS - with today's version, the cross-feeding valves now actually work as they should. Not sure why this would be needed, but we can also do partial cross feeding since we have separate 12 and 345 manifold valves - so you can cross-feed the 12 manifolds of the left RCS from the right RCS and feed the 345 manifolds of the left RCS from its own tank.
Thorsten
 
Posts: 10097
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby wlbragg » Tue Jul 07, 2015 5:22 pm

Thorsten wrote in Tue Jul 07, 2015 9:55 am:I've just noticed that the payload bay doors get odd lighting - did you do anything to them.

Not that I know of, but maybe shuffling the blend between japreja and myself caused some problem.

Thorsten wrote in Tue Jul 07, 2015 9:55 am:And the speedbrake seen from the rear has odd texture mapping...

That one I noticed, it may just need to be textured.

I noticed something else and forgot to ask about it. On the port side rear of the shuttle, are you venting something, this is during the approach phase?

@japreja, I may need your help with these, texturing issues, is not one of my bailiwicks, but I will look at it and see if I can correct any of it..
Kansas(2-27-15)/Ohio/Midwest scenery development.
KEQA (2-27-15), 3AU, KRCP Airport Layout
User avatar
wlbragg
 
Posts: 4279
Joined: Sat Aug 25, 2012 11:31 pm
Location: Kansas (Tornado Alley), USA
Callsign: WC2020
Version: next
OS: Win10/Debain/nVGT640

Re: Space Shuttle

Postby Thorsten » Tue Jul 07, 2015 5:34 pm

I noticed something else and forgot to ask about it. On the port side rear of the shuttle, are you venting something, this is during the approach phase?


Yepp - that's the hi-load flash evaporator running. I'll add a provision to switch it off below the required altitude soonish. It's actually neat - as described in the manual, the vent is asymmetric and slightly propulsive - if you leave hi-load on in orbit, you'll drift off attitude, it generates a pound of force or so.

In reality, during approach we might run ammonia boilers if cold soak isn't sufficient - they'd be switched on latest after touchdown till the craft is secured.
Thorsten
 
Posts: 10097
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby wlbragg » Tue Jul 07, 2015 8:24 pm

that's the hi-load flash evaporator running

I figured it might be something venting but I couldn't find it without taking more time than I cared to. I really wanted to make sure it wasn't some errant effect I added.

@japreja, I looked at the textures and the uv map in blender and there are some problem that have crept into the file.

I noticed the last shuttle_02.blend that I got from you was showing all pink (I read that means missing textures but didn't know it at the time). But I just appended from you work anyway and that shouldn't have mattered.

Just now going back into the blend to have a look, on my system it just decided to loose some texture locations. I don't know if the textures were ever packed with the file but I think that would be wise (I don't really even know how to tell if they are but I did see how to do it).

I managed to get all the textures mapped back in. I couldn't even export the ac until I cleaned up (remapped) the missing textures.
But then when looking at the uv map I noticed the speedbrake part of the uv map is definitely messed up and I don't know how to fix it. UV mapping is on my list of things to learn and master but so far all I really know how to do is map a simple cube (following a tutorial) :)
I am able to get to the uv editor and see that it is messed up, but don't know the steps to correct it.

time passed.....


OK, I am able to edit the uv map enough to clean up the speedbrake mess, at least it is presentable. I haven't looked at the bay door issue yet but will soon.

But I don't want to push this until I know it is right. If you know anything at all about the textures, please take a look.

So I am working on it, any experts want to beat me to it, go for it!
Kansas(2-27-15)/Ohio/Midwest scenery development.
KEQA (2-27-15), 3AU, KRCP Airport Layout
User avatar
wlbragg
 
Posts: 4279
Joined: Sat Aug 25, 2012 11:31 pm
Location: Kansas (Tornado Alley), USA
Callsign: WC2020
Version: next
OS: Win10/Debain/nVGT640

Re: Space Shuttle

Postby wlbragg » Wed Jul 08, 2015 12:21 am

So, more time passed and I think I corrected the UV and material issues.

I would feel a whole lot better about it if one of our many pros would look at Models.Artwork/shuttle_02.blend and make sure it is done correctly. I feel like there could be some cleanup needed.

But visually it looks like we are at least back to where we were.and the speedbrake inside surfaces look a bunch better.
Kansas(2-27-15)/Ohio/Midwest scenery development.
KEQA (2-27-15), 3AU, KRCP Airport Layout
User avatar
wlbragg
 
Posts: 4279
Joined: Sat Aug 25, 2012 11:31 pm
Location: Kansas (Tornado Alley), USA
Callsign: WC2020
Version: next
OS: Win10/Debain/nVGT640

Re: Space Shuttle

Postby Thorsten » Wed Jul 08, 2015 5:43 am

Quick question: Why do we animate the speedbrake with

/controls/shuttle/speedbrake

rather than

/fdm/jsbsim/fcs/speedbrake-pos-norm

which simulates the actual behaviour with finite transition times dependent on hydraulic system availability?

Similar for the body flap

/fdm/jsbsim/fcs/bodyflap-pos-rad

could be used.
Thorsten
 
Posts: 10097
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby wlbragg » Wed Jul 08, 2015 6:39 am

Because I didn't know any better, please change them to whatever is most appropriate

I really haven't played with the procedures and the choices available for realistic or easy so I really didn't know what or how automated you would make both the body flap and the speed brake. Whether or not you would automatically set them at the appropriate stages or if that was to be a line item on the procedures, or both.

After re-reading your post I do recall seeing those fdm props, I don't know how I came to use the others, I think I found them through the keyboard?
Last edited by wlbragg on Wed Jul 08, 2015 7:38 am, edited 1 time in total.
Kansas(2-27-15)/Ohio/Midwest scenery development.
KEQA (2-27-15), 3AU, KRCP Airport Layout
User avatar
wlbragg
 
Posts: 4279
Joined: Sat Aug 25, 2012 11:31 pm
Location: Kansas (Tornado Alley), USA
Callsign: WC2020
Version: next
OS: Win10/Debain/nVGT640

Re: Space Shuttle

Postby Richard » Wed Jul 08, 2015 7:22 am

Thorsten wrote in Wed Jul 08, 2015 5:43 am:/controls/shuttle/speedbrake

rather than

/fdm/jsbsim/fcs/speedbrake-pos-norm

which simulates the actual behaviour with finite transition times dependent on hydraulic system availability?

Similar for the body flap

/fdm/jsbsim/fcs/bodyflap-pos-rad

could be used.


It shouldn't be the /controls version as that is the demand/switch position etc..

It should be /surface-positions/speedbrake-pos-norm as the /fdm/jsbsim aren't transmitted by default over MP.

There are some properties that will need to go via sim/multiplay/generic/float (or int) for accurate animations over MP
Richard
 
Posts: 646
Joined: Sun Nov 02, 2014 10:17 pm
Version: Git
OS: Win10

Re: Space Shuttle

Postby wlbragg » Wed Jul 08, 2015 7:41 am

Richard wrote in Wed Jul 08, 2015 7:22 am:the /fdm/jsbsim aren't transmitted by default over MP


Does that apply to gear also?
Kansas(2-27-15)/Ohio/Midwest scenery development.
KEQA (2-27-15), 3AU, KRCP Airport Layout
User avatar
wlbragg
 
Posts: 4279
Joined: Sat Aug 25, 2012 11:31 pm
Location: Kansas (Tornado Alley), USA
Callsign: WC2020
Version: next
OS: Win10/Debain/nVGT640

Re: Space Shuttle

Postby japreja » Wed Jul 08, 2015 7:52 am

wlbragg wrote in Tue Jul 07, 2015 5:22 pm:I may need your help with these, texturing issues, is not one of my bailiwicks, but I will look at it and see if I can correct any of it.

That make two of us, I've tried to toy with textures in the past and all my attempts are fails. We may need to get someone who can remap the textures because the mesh geometry has changed wit the cuts. The landing gears could just be done without textures like IAHM-COL did one of the panels recently.

wlbragg wrote in Tue Jul 07, 2015 8:24 pm:I looked at the textures and the uv map in blender and there are some problem that have crept into the file.

I noticed the last shuttle_02.blend that I got from you was showing all pink (I read that means missing textures but didn't know it at the time). But I just appended from you work anyway and that shouldn't have mattered.

This happens when you append an external model on occasion, Just go into the "UV Editor" from the dropdown at the top area of blender, select the pink object in the righthand window and enter edit mode. Press the A key until all vertices are selected. In the lefthand window (still in edit mode with all verts selected) find the appropriate texture from that windows image selection menu (a thumnail of an image) and make sure it is set to the proper Texture file. On my system this is "spstob_1.png.001" for the body and heatshield tiles. Also the landing gear doors will be on this texture along with the body flap. If you give me a few hours I will make a youtube video on what I mean if you cant figure it out. This might be because blender cannot find the texture which is located just under the Artwork directory.

If you can't get it I will fix it. But I will do so by packing the images into the blend file if it is not already done. (I see you already have it done except the gear doors, just follow the instructions above and it should work :wink:

Blenders "File/External Data" menu has an option to automaticaly pack images into the blend. The .ac file exported will extract them to a subfolder called "Textures" wherever the .ac file is exported.
japreja
 
Posts: 330
Joined: Thu May 07, 2015 11:05 pm
Location: MT, USA
OS: Windows 10 Pro 64bit

Re: Space Shuttle

Postby Thorsten » Wed Jul 08, 2015 8:20 am

It shouldn't be the /controls version as that is the demand/switch position etc..


We have an unusually complex FCS section for the Shuttle (large parts of it also exist as systems) since we need to map control input in (so far) 15 different ways into reactions of thrusters, gimbal actuators and airfoils. Also, we may need to merge control input with automatic settings from flight control (yaw and roll stability), we may not be able to execute control input because the component is damaged, or because there is no electrical or hydro power,... and they're usually post-filtered by a kinematic to account for transition times, sometimes we have in fact three different kinematic tags to account for priority rate limiting, out of which we select one in the end...

So there's always a whole chain of similarly called properties, accounting for what is commanded by the user, what is the sum of user and automatic command, what is possible given the DAP, what happens given component damage...

What is authorative for the actual motion of the airfoil in the conventions I have used is only what is used in the flight dynamics section, not what appears in the FCS section. If it influences the aerodynamics, then it's a real motion, if not, then not.

Since I don't do MP, what is transmitted over MP is a secondary concern for me (with the current MP protocol, you won't be able to rendezvous with an MP craft in orbit anyway) - if something should be transmitted over MP, the flight relevant positions should be copied to what MP needs.
Thorsten
 
Posts: 10097
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby japreja » Wed Jul 08, 2015 10:25 am

Thorsten wrote in Thu Jun 18, 2015 11:49 am:What's next on the missing items list? Ku-band antenna perhaps? In principle we could also start cutting holes (ET umbilical, star tracker doors, vents, gear bays,...

I forgot about all the other stuff you noted in this post a while back, The ET Umbilical and maybe star tracker doors can be done the same way as the gear doors. I think it will look better if they are inset a bit but when they are open there is no texture yet, if each connection component is modeled it wouldn't realy be needed. As for the vents, I have never looked for them so I don't know exactly where they would be...

The Ku-band antenna i'll need to lookup some referance photos, which I might already have, but actual dimentional data would be nice and I am able to work with it quicker. There are 2 Ku-band antennas from what I saw, one on the left and one on the right of the payload bay. I am certain I saw these modeled someplace in a different shuttle model, I think from NASA 3D models???

One thing I have been trying to model is the cement structure for the launchpad, I have a basic outline from High Resolution Ortho Imagery but Hight is the only problem I am unable to find. I believe the crawler is capable of 6 degrees for leveling, from what I understand from various moonport videos, I don't know if anything was modifed for the Shuttle Program so finding the inclination angle of the Launchpads is a bit tedious. I have the images overlayed with TIGER data in QGIS.
japreja
 
Posts: 330
Joined: Thu May 07, 2015 11:05 pm
Location: MT, USA
OS: Windows 10 Pro 64bit

Re: Space Shuttle

Postby wlbragg » Wed Jul 08, 2015 12:38 pm

@japreja, thanks for the tips.

I think I have most of the issues fixed in the blend file. I spent some time on it yesterday and it looks pretty well cleaned up and back in order.

I made all the gear doors double sided and applied the spstob_1.png.001 texture to them. I also cleaned up the UV map on the inside of the speed brakes.

After looking at some images of the RL shuttle, I think the gear doors need to have some thickness applied to them and then add a new texture for the edges and the inside face.

I think I am starting to get the hang of some of this simpler stuff so I will work on that part of it.
Kansas(2-27-15)/Ohio/Midwest scenery development.
KEQA (2-27-15), 3AU, KRCP Airport Layout
User avatar
wlbragg
 
Posts: 4279
Joined: Sat Aug 25, 2012 11:31 pm
Location: Kansas (Tornado Alley), USA
Callsign: WC2020
Version: next
OS: Win10/Debain/nVGT640

Re: Space Shuttle

Postby Richard » Wed Jul 08, 2015 1:24 pm

It's important for the 3d model animations to use the values in /surface-positions rather than /controls or /fdm - because that way an observer will be able to see the correct movements in multiplayer.

Equally other elements of animation that aren't already handled by the MP protocol directly should be in sim/multiplay/generic/float[idx] - and the animation xml using these variables and using the alias

e.g. for the F-15
Code: Select all
                 <float n="0">0</float> <!-- free     -->
                <float n="1">0</float> <!-- free     -->
                <float n="2">0</float> <!-- free     -->
                <float n="3">0</float> <!-- wing torn, 1 = left, 2 = right, 3 = both(!)-->
                <float n="4">0</float> <!-- left elevator  -->
                <float n="5">0</float> <!-- right elevator -->
                <float n="6">0</float> <!-- refuel probe/door -->
                <float n="8" alias="/engines/engine[0]/augmentation-burner" />
                <float n="9" alias="/engines/engine[1]/augmentation-burner" />
                <float n="10" alias="/fdm/jsbsim/propulsion/engine[0]/alt/nozzle-pos-norm" />
                <float n="11" alias="/fdm/jsbsim/propulsion/engine[1]/alt/nozzle-pos-norm" />
                <float n="12" alias="/surface-positions/l-ramp1-position-deg" />
                <float n="13" alias="/surface-positions/r-ramp1-position-deg" />
                <float n="14" alias="/fdm/jsbsim/systems/electrics/ac-essential-bus1" />
                <float n="15" alias="/fdm/jsbsim/systems/electrics/ac-left-main-bus" />
Richard
 
Posts: 646
Joined: Sun Nov 02, 2014 10:17 pm
Version: Git
OS: Win10

Re: Space Shuttle

Postby Thorsten » Wed Jul 08, 2015 5:10 pm

It's important for the 3d model animations to use the values in /surface-positions rather than /controls or /fdm - because that way an observer will be able to see the correct movements in multiplayer.


Well, but what gets written into there? I don't see how FG automatically 'knows' what the correct surface position is as I might in fact compute it anywhere in JSBSim, including a systems section - so we presumably need to copy it adhering to some name convention.

Also, the inboard and outboard elevon deflection is a combination of elevator and aileron channel and only computed inside JSBSim for the animation - the Shuttle doesn't actually have an elevator deflection although we formally use it for the FDM.

I would bother doing this only with things which an observer actually has a chance to see though. We have potentially 50+ different thruster firings for instance... I believe the MP protocol is finite, so I'd think carefully what we actually need.
Thorsten
 
Posts: 10097
Joined: Mon Nov 02, 2009 8:33 am

PreviousNext

Return to Spaceflight

Who is online

Users browsing this forum: No registered users and 1 guest