Board index FlightGear Development Aircraft

Space Shuttle

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

Re: Space Shuttle

Postby Thorsten » Sun May 13, 2018 6:59 am

Like the right decount for Body axis velocity


Yeah, I went through the Insertion/De-orbit workbook and added a bunch of small things here and there...
Thorsten
 
Posts: 9547
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby GinGin » Tue May 15, 2018 2:22 pm

I was playing with green Abort to Bermuda. Window is very tight to not over g and lose control of everything .
First failure at Vi 7000 and second at Vi 10400
Orbiter was too close to Bermuda and AP loscontrol duringNz hold with severe wings oscillations.
I am really impressed by the accuracy of the different steps you coded ( like nullify Bêta angle and yawing the Orbiter into The velocity vector when EAS superior at 4. Like in the book )


When you will have time Thorsten, could you try a RTLS around 3 mn for the failure.
Standard 32 degrees orbit inclination .

I found autopilot is leading to Loss of Control a lot during Nz hold. I am maybe doing something, wrong. I could remember that it was working great for failure between srb sep and 4 mn
GinGin
 
Posts: 294
Joined: Wed Jul 05, 2017 10:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle

Postby Thorsten » Tue May 15, 2018 5:11 pm

Since I probably won't have too much time any time soon (it's firewood-making and gardening season) - what is the pattern of the loss of control? Is the Shuttle unable to acquire a good beta and jittery in the yaw axis from the beginning, is there a roll issue (I remember having observed a pretty weird roll integrator windup once), or is there an issue with the pitch axis?

I assume trim is okay (that accounts for most of my Nz failures... with inability to acquire zero beta a close second)?

Frankly, control during Nz hold is a mess... the effectivity of thrusters vs. control surfaces changes so rapidly.
Thorsten
 
Posts: 9547
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby legoboyvdlp » Wed May 16, 2018 6:30 pm

Using the development branch, no payload, target 165 miles, 32.50 inclination the roll heads up causes a sink rate of up to 100 m/s. It recovered well, and I achieved orbit. I presume the sink rate is normal?

Also, is the propellant dump also done automatically now?

Also, following http://wiki.flightgear.org/Flying_the_S ... n_Advanced something went a bit wrong
Image

but it still works apart from flashing, so ok :)

Please don't ask me how I managed this:
Image

My console is also filled with such messages

Code: Select all
OMS burn for -2163 seconds
OMS burn for -2163 seconds
OMS burn for -2163 seconds
OMS burn for -2163 seconds
OMS burn for -2163 seconds
OMS burn for -2163 seconds
OMS burn for -2164 seconds
OMS burn for -2164 seconds
OMS burn for -2164 seconds
OMS burn for -2164 seconds
OMS burn for -2164 seconds
OMS burn for -2164 seconds


And when I try to execute the payload bay attitude maneouver, it results in FlightGear freezing me in this position, when it unfreezes it spins round for a few seconds around the roll axis:
Image

It then unfroze after five minutes and gave me this:
Image

Gone are the days of simple spaceflight :lol:

Edit: and while correcting the vector after completing the maneouver, I went into an uncontrolled spin due to excessive pilot inputs :D. After I recovered from that, I found myself in an increasing sink!

Edit #2:
Image

Sorry guys. A flat spin just doesn't work in a Shuttle. It was nice knowing you :)

And please don't be concerned that the Shuttle has major errors. This is just the history of how badly things go when a novice takes up a Shuttle!
User avatar
legoboyvdlp
 
Posts: 5412
Joined: Sat Jul 26, 2014 1:28 am
Callsign: YV-LEGO
Version: 2018.3.1
OS: Windows 10 HP

Re: Space Shuttle

Postby Thorsten » Thu May 17, 2018 5:58 am

Using the development branch, no payload, target 165 miles, 32.50 inclination the roll heads up causes a sink rate of up to 100 m/s. It recovered well, and I achieved orbit. I presume the sink rate is normal?


It's a complicated maneuver...

You can't just roll around the body axis, there's significant pitch and yaw movement involved. The reason is that the engines are mounted at an angle of ~14 deg, so the thrust axis is different from the body axis. In inverted flight the thrust points 14 deg down, if you'd just roll around the body axis it'd point left or right by the time you've rolled 90 degrees and point upward (and push you down) when you are in normal attitude. So somehow during the roll you need to yaw to avoid getting too much out of the inclination and pitch by 28 deg to avoid thrust accelerating you downward while the engine gimbals somehow have to take care of pitch, yaw and roll and might actually saturate since they don't come with too large gimbal angles. All the while the propellant depletes and the CoG shifts and the gimbal angles need to take care of that as well.

So yeah, you may lose some altitude and drift a few miles off course...

Also, is the propellant dump also done automatically now?


Dependent on what's i-loaded in the mission file yes or no.

Also, following http://wiki.flightgear.org/Flying_the_S ... n_Advanced something went a bit wrong


This is pretty much the reason I have decided to limit the use of in-sim checklists and have centralized the documentation in 'the manual'.

First, the devel version continually adds procedures. For instance, OMS-burns now require you to enter trim angles and to be in MNVR EXEC rather than MNVR COAST or similar, otherwise the engines won't ignite. That's different to the last milestone, but not documented anywhere - the documentation will only come when the next milestone is published and it may (or may not) be added to the wiki. The devel version needs understanding of real-life procedures, it's not documented anywhere else.

Second, with so many keystrokes to perform, chances are you make a mistake. So working down a checklist is far too risky, if you screw up a step, you need to understand what happened and correct it - and that requires a good understanding of the involved errors.

For instance, you can't just keep flying with things like an IDP poll fail error - if a problem occurs, you need to diagnose it and correct it.

My console is also filled with such messages


I'm vaguely guessing you never actually finished the burn - after the engines cut out you're supposed to burn the residuals using the RCS translational mode and then end the burn by transiting to the next OPS.

And when I try to execute the payload bay attitude maneouver


And I'm frankly not sure what happens if the OMS attitude logic still runs because you never finished the burn while you activate a different attitude logic...

Gone are the days of simple spaceflight


Not really - but if you set your eyes on re-configuring the GPCs in realistic mode, then that's not simple. Otherwise following 'the manual' will do the trick.

Edit: and while correcting the vector after completing the maneouver, I went into an uncontrolled spin due to excessive pilot inputs . After I recovered from that, I found myself in an increasing sink!


Climb and sink have little meaning in an orbit - they just tell you whether you approach perigee or apogee. You can have 50 m/s of sink in an elliptical orbit, but it doesn't mean you will end in the atmosphere - as long as the perigee is high enough, you'll eventually go up again and reach 50 m/s climb.

It's pretty unlikely that using the RCS you managed to degrade the orbit to the degree that you are forced into the atmosphere - there's not enough propellant in the RCS tanks to do that, when you de-orbit with the RCS, you have to set up a crossfeed from the OMS tanks.

It's more of a mystery to me how you can get into a spin with an attitude-holding DAP available when all you need to really do is center the controls, that commands zero rates and the logic will automatically figure out how to do it. I recommend giving the chapter on the orbital DAP a good read. :D
Thorsten
 
Posts: 9547
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby legoboyvdlp » Thu May 17, 2018 6:39 am

Many thanks :D
User avatar
legoboyvdlp
 
Posts: 5412
Joined: Sat Jul 26, 2014 1:28 am
Callsign: YV-LEGO
Version: 2018.3.1
OS: Windows 10 HP

Re: Space Shuttle

Postby GinGin » Thu May 17, 2018 3:01 pm

@Thorsten : Well, I tried again and did the alpha recovery manually to force the shuttle to have zero Beta angle and 50 degrees of AOA, then I put back auto for the Nz phase and it worked well.
I am still wondering what would have been the survivability of RTLS in real, not to mention another engine loss during RTLS :)

But indeed , I found that the smallest mistake in Bêta angle or AOA before Nz hold could lead to LOC easily ( and it is well enlightened in books that Nz hold like you said is a highly critical phase where parameters have to be nearly perfect to avoid oscillations and LOC)


Ahah Nice for’the Gardening, whish I Had a house with garden in Paris :)



@Legoboyvdlp: did you use Advanced gpc or simplified gpc?
I think I did procedure for both of them in the tuto
Indeed , something went wrong, all buses linked to one gpc , the cross on crt 3 is due to the fact that gpc 2 is not linked to a display buses, so no chance to have what is in the mind of gpc 2 on a screen.

For the trim angles, I mentioned them in the tuto.
Basically for a 2 engines burn , it is always the same. You want to have the OMS to fire in parallel with the CoG
For the right one , trim Yaw by +5.2 degrees and left one by - 5.2 degrees
For the pitch, 0.4 degrees

Well done for the spin ahah, acrobat :)

I have a good picture for you, look what is written above the HSI xD

https://goo.gl/images/KjDaKc




I found a nice article about the real space shuttle sim, where do I book ? :))))


https://www.wired.com/2012/11/nasa-full-size-shuttle-trainer/
GinGin
 
Posts: 294
Joined: Wed Jul 05, 2017 10:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle

Postby Thorsten » Thu May 17, 2018 3:10 pm

But indeed , I found that the smallest mistake in Bêta angle or AOA before Nz hold could lead to LOC easily


So is the problem you're seeing driven by non-zero beta prior to engaging Aerojet (which does not deal with non-zero beta gracefully)?
Thorsten
 
Posts: 9547
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby legoboyvdlp » Thu May 17, 2018 7:32 pm

Thorsten,
This is probably I have not updated my FlightGear nightly build for a few weeks,
However, when I use the "replicate CRT" button in the menu, the canvas window that comes up is transparent, with only the text displayed.
The new keyboard works perfectly.

Thanks,
Lego

GinGin :D Thanks!

Quick question, I accidentally moved to GNC SYS SUM 2 on my CRT1 instead of CRT 3. How do I restore MNVR EXEC on CRT 1? I tried OPS 105 PRO but it didn't load it. Luckily I still have MNVR EXEC on CRT 2 :)
User avatar
legoboyvdlp
 
Posts: 5412
Joined: Sat Jul 26, 2014 1:28 am
Callsign: YV-LEGO
Version: 2018.3.1
OS: Windows 10 HP

Re: Space Shuttle

Postby GinGin » Fri May 18, 2018 5:59 am

@Thorsten : Hard to say. I remember one time, alpha angle was too low before nz hold then autopilot just dive and couldn’t maintain the G. Same for beta angle . I can’t say it was a constant in my every attempts , but it might have be.

But sometimes, everything is good, alpha ok, beta ok, speed ok. And during nz hold, beta angle starts to increase inducing yaw and aileron movements and oscillations stronger and stronger etc.
I will do some more tests :)

I also wanted to test some engine stuck procedure . Expect shutting down an APU when thrust is at 67 %, is there a way to put an engine at 67% of thrust while the others remain at full thrust?


@legoboyvdlp: every time you are in a specialist page ( Spec) or display page to view data ( system summary /display), you can press resume (rsm) on the keyboard to go back to the main OPS page ;)

You didn’t answer, where you using advanced gpc settings or simplified one?
GinGin
 
Posts: 294
Joined: Wed Jul 05, 2017 10:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle

Postby Thorsten » Fri May 18, 2018 8:04 am

When you will have time Thorsten, could you try a RTLS around 3 mn for the failure.
Standard 32 degrees orbit inclination .


Okay, I just ran two patterns with engine failure at 2:40 and RTLS button pressed at 3:00 MET.

I was successful with a failure of the left engine, I got delivered into a TAEM pattern a bit low on energy, but it looked entirely recoverable by changing to a straight-in.

I crashed with a center engine failure - this ran into something I have not completely understood yet and which rears its ugly head occasionally.

The issue is that without the center engine the thrust vector that can be achieved by fully gimbaling the remaining engines is insufficient to go through the CoG of a near-empty tank. The issue is considerably mitigated by the presence of a payload though. If that happens, the vehicle starts to pitch up uncontrollably - which in this case happened some 2 seconds prior to MECO and turned out not to be recoverable. It seems really down to random chance whether it occurs in an RTLS prior to MECO or not.

I'm not sure what is ultimately behind it though.

It may be that I have engine positions and / or gimbal angles wrong for the left and right engine - but I've triple checked the angles with an engineering schematics and the position must be drastically off to make an effect.

It may be that my model of teh CoG of the empty ET is wrong - but I've played some with it, and again it has to be drastically wrong (basically the full ET is very top-heavy because the heavy oxygen is on top - but the empty ET can't be equally top-heavy - while there's more structure on top, it's all the same material).

The fact of the matter is that I don't really know - perhaps in reality you'd never fly an empty Shuttle and the payload makes the difference? Perhaps if the Shuttle is empty, powered pitch-around and MECO need to be done just a tad earlier?

I also wanted to test some engine stuck procedure . Expect shutting down an APU when thrust is at 67 %, is there a way to put an engine at 67% of thrust while the others remain at full thrust?


I'm not sure what you mean - you can't throttle individual engines in the SHuttle, but if there's an electric or hydraulic lockup, the engine ought to remain at the current thrust level no matter what you or autothrottle commands and only responds to the cutoff button. At least that's how it should work and how it tested out last time I tried... (I have never tried stuck at 67%, but I have tried stuck at 100% and then cutting one engine prior to MECO manually to avoid over g)
Thorsten
 
Posts: 9547
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby legoboyvdlp » Fri May 18, 2018 8:19 am

GinGin wrote in Fri May 18, 2018 5:59 am:
You didn’t answer, where you using advanced gpc settings or simplified one?


I used advanced earlier when I messed up everything... this time I managed a good launch in simplified from launch to oms 2 burn :) The mission was STS-129.
User avatar
legoboyvdlp
 
Posts: 5412
Joined: Sat Jul 26, 2014 1:28 am
Callsign: YV-LEGO
Version: 2018.3.1
OS: Windows 10 HP

Re: Space Shuttle

Postby Thorsten » Fri May 18, 2018 9:43 am

This is probably I have not updated my FlightGear nightly build for a few weeks,
However, when I use the "replicate CRT" button in the menu, the canvas window that comes up is transparent, with only the text displayed.


Probably not, since I'm not running a current FG at the moment. Might be a regression / bug having become apparent in new versions of FG (?) - anyway, I don't see it in either of my installations.

You probably know canvas, right? The code is in SpaceShuttle/Nasal/canvas/canvas_dialogs.nas as cdlg_mdu_clone class - maybe you can give it a quick look whether you can change some setting that makes it work for you?
Thorsten
 
Posts: 9547
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby GinGin » Fri May 18, 2018 9:52 am

@Legoboyvdlp: Ahh cool :) Advanced stuff is like a simulation into the simulation ahah


@Thorsten: Thanks for the feedback and test. I am almost always shutting down the center engine also. I will try another one to see.
Ok, so CoG linked.
Interesting what you said about the weight of Shuttle . I can't remember a mission without payload indeed. I will try a couple of RTLS with a payload to see.
By knwoing you a bit now, I don't think there is big mistake in coding, maybe it's just like that, everything as to be well within limits and good timing to make an RTLS work.
It is well emphasized in all documentations how important CoG, Beta and Alpha just to mention those paramteres are critical for an RTLS specifically.



Concerning the stuck thing.

I am talking about contigency stuck procedures where you have one engine out, one engine ok, one engine stuck at 67 % of thrust.
Only way I found to simulate that is too reduce manually all 3 engines to 67 %, then shut down the associated APU to stuck on main valve and let an engine to 67 %, then back to Auto thrust and shut down an engine.

I was just wondering if there was an option specifically for that, but no worries, I could try with this method some nice things to come :)



One question, Could you indicate me good readings about all the lod stuff and how to increase visibility to have a nice view without earthview in orbit ( want to try that)
So far, I get that

Image

I played with detailed and rough settings ( which bottlenecked my 16 go ram xD) I guess I have to play with the bare one to not have limitations ? Also, I read something about Z far camera group to allow the sim to go beyond a fix visibility, but I don't find the article anymore.




Finally, Quick sum up of my last test for 2EO out Bermuda ( which worked very well, AP worked like a charm and by the book, well done )


Love that launch trail

Image


First engine failure around Vi= 7k, TAL abort to ZZA, Second engine at Vi around 10 k, droop engaged but useless ( single engine OPS 3 boundary is still far)
So let's go for the 2Eo out Green to Bermuda



Hard Nz , struggling to maintain 3.6 G, and Black out incoming
Thanks AP, we survived the Nz Hold. Even if we don't make it to Bermuda, Bailout is available.

Image


We should make it, I put Straight in and MEP to have the most chance to make it

Image


Even a bit high, that's nice to have spare energy, I moved again to a Right Overhead to deplete easily the extra energy

Image



What a view

Image



No MLS in Bermuda, time for a delta Update to have an accurate State Vector

Image


Result is pretty good, we shouldn't have surprise

Image


On track, final turn

Image


Short final, Synthetic runway was almost on the real one. Delta update is really a thing

Image



I was too fast, touch down to far and high on speed ( 240 kt beurk)
Results : I need retraining in approach simulator xD

Image


great fun on that one.
Might start to update the wiki with contigency abort once I will have finish the intact one
GinGin
 
Posts: 294
Joined: Wed Jul 05, 2017 10:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle

Postby Thorsten » Fri May 18, 2018 11:27 am

Nice job with the Bermuda near-landing! (The runway is a bit on the short side - you might change short to emergency landing site speedbrake option - that helps a lot...)

Before you die with a full payload during Nz holding - upon preparing the scenario Durk had in mind for FSWeekend (RTLS for the planned mission where the Shuttle would carry the Centaur stage into orbit) - it turns out that there's also a region where for a heavy payload the elevons don't have enough authority during the initial phase of the Nz pullout - which kills you later. So we concluded that the Centaur fuel has to be at least partially dumped during an abort.

I suppose the payload simulation is also too simplistic at the moment - I need to give an option to shift the CoG of the payload somewhat forward and backward. I'm just a bit reluctant to do that because I don't know where the safe limits are...

And you probably mean that article, right?
Thorsten
 
Posts: 9547
Joined: Mon Nov 02, 2009 8:33 am

PreviousNext

Return to Aircraft

Who is online

Users browsing this forum: enrogue, Google [Bot] and 7 guests