Board index FlightGear Development Spaceflight

Space Shuttle - Bugfixes

Discussion about development and usage of spacecraft

Re: Space Shuttle - Bugfixes

Postby GinGin » Mon Feb 15, 2021 2:28 pm

That’s normal without the new L/D fdm.
My fault , I pushed a bit early the auto launch with lower vrel target that matches true data’s with the good l/d.

It works very nice , you will see :)
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Bugfixes

Postby Thorsten » Mon Feb 15, 2021 6:27 pm

I had a hunch after posting it might have been something like that... Well, that's why we call it a 'development' branch... no worries then.
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Space Shuttle - Bugfixes

Postby eatdirt » Tue Feb 16, 2021 9:36 pm

I've just tried a TAL to Banjul, and the MECO condition was dead wrong - despite activating low energy logic, I ended up some 400 miles before the runway in the ocean.


I played one tonight, from a trajectory to Hubble. Worked ok, but the AP made me land 2.2nm aside of Banjul runway, errors were looking fine in the filter (but I'll need to check this twice).

I have then tried a TAEM at Banjul, it works fine, but it looks like the errors between GPS and the rest are building quite fast, already out of the HAC the runway is quite off-side by 2 runway width. Anyone noticing something like that? Forcing GPS, put the runway exactly where it should.



@Wayne: little issue in the console :)

Code: Select all
 1.33 [WARN]:sound      XML sound: couldn't find file: 'Aircraft/SpaceShuttle/Sounds/crash.wav'
eatdirt
 
Posts: 1012
Joined: Wed Aug 15, 2018 3:06 pm

Re: Space Shuttle - Bugfixes

Postby eatdirt » Tue Feb 16, 2021 11:28 pm

but the AP made me land 2.2nm aside of Banjul runway, errors were looking fine in the filter (but I'll need to check this twice)


Tried another time, confirmed :-/
The errors are building very fast. TACAN, ADTA and DRAGH are all consistent between them, but not with GPS, that one being consistent with the real runway.

With quite a bit too much energy :D

Image

Look at GPS entry error there, how it grows fast (meaning all the others point elsewhere):

Image

2minutes later:

Image

Seconds later:
Image

1 minute more:

Image

HAC on sight, I resetting everything to GPS ITEM 44 EXEC

Image

At wheels stop, already 1.7 error between GPS and the rest:

Image
eatdirt
 
Posts: 1012
Joined: Wed Aug 15, 2018 3:06 pm

Re: Space Shuttle - Bugfixes

Postby GinGin » Wed Feb 17, 2021 12:09 am

I dont have the Kalmann numbers in mind, I have a nav workbook about those, but there is a huge weight difference between TACAN, Drag H and GPS, MLS.
I dont find abnormal that the error is quite consitent without a precision filter like an MLS or a Nav delta update from MCC to shrink the state vector towards more accuracy

I like the clouds on the first picture, they are a bit special no ?
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Bugfixes

Postby eatdirt » Wed Feb 17, 2021 1:01 am

I like the clouds on the first picture, they are a bit special no ?


Ah yes, that's a slightly tweaked Earthview with exact parallax mapping, so you see the relief even if you're close to the object. The superb idea to use their transparency for making them 3D is Thorsten's of course :)
eatdirt
 
Posts: 1012
Joined: Wed Aug 15, 2018 3:06 pm

Re: Space Shuttle - Bugfixes

Postby Thorsten » Wed Feb 17, 2021 8:00 am

I don't see any particular issues with the filtering. GPS and MLS are 'exact' within the simulation (and pretty much in reality as well, they can place you to cm-accuracy), TACAN accuracy is a few hundred meters, it nevergives precision range, so it'll get you to a runway but generally not on a runway, if you have no visual from 12.000 ft you need MLS, your GPS or you're screwed.

Air data altitude is special because in addition to the intrinsic uncertainty, it actually uses pressure altitude and will be wrong unless QNH is entered for the actual weather at the site. Drag altitude has an error range of a few hundred meters, so it's hardly constraining at all.

So it is conceptually possible that the inertial solution drifts to the wrong corner allowed by TACAN / Air data, in particular if they're by chance offset in the same wrong direction and in this case the correction when reality comes on is quite large without anything being wrong.

***

With regard to BFC lights on - inspecting the code, pretty much the only condition to trigger this currently is a PASS error. @eatdirt - we may thus assume you've actually flown with a faulty PASS, but errors must have been small enough to not kill you. The code doesn't seem to trigger the lights when PASS switches to OPS 2, and currently the only way to get rid of them is to take over with BFS and then switch back to PASS (the real procedure for the last step isn't implemented, it is quite lengthy...).

I/O resets hide the errors. I guess we could also make the light go out when the BFS GPC is moded to HALT (I frankly have no clue what happens in reality when you blatantly ignore the warning) - but the underlying situation is actually quite dangerous, so I'm not sure we should make it easy to get rid of this - in-sim, the same errors would have affected your entry.

***

Another small issue to fix - the indicator arrow of the body flap on the SPI page seems to be offset - it is well above the indication ladder on the upper end. I'm fairly sure this used to work, so it's probably a regression.

***

@Wayne - since I didn't have manual control when landing yesterday, I happened to test the sounds when you land without gear - they sound about right to me :mrgreen: Thanks.

***

Which leaves the loss of roll control in manual - I'll have to see whether it triggers again through some chance and then try to follow it through the property tree. It'd help if someone else can record circumstances should it occur so that maybe a pattern emerges. Admittedly that one disturbs me quite a lot. :(
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Space Shuttle - Bugfixes

Postby eatdirt » Wed Feb 17, 2021 11:25 am

Thanks Thorsten for the feedback, I'll try to make a step-by-step recipe to reproduce for various of these problems.
I am getting the BFC blinking each time I am flying during the transitions, so I must trigger it by some procedure.
For Banjul, I don't fly a TAL often, but I hardly believe drift errors can push you out by 2nm!! I'll test with various conditions, realistic etc... to dig into that.

Cheers.
eatdirt
 
Posts: 1012
Joined: Wed Aug 15, 2018 3:06 pm

Re: Space Shuttle - Bugfixes

Postby GinGin » Wed Feb 17, 2021 12:06 pm

that's a slightly tweaked Earthview with exact parallax mapping,


That looks interesting.
Could you post more screenshots about those with different lightings ? :)

the indicator arrow of the body flap on the SPI page seems to be offset


It is ok now with the last stages.nas with the modified Bodyflap delfection limits


It'd help if someone else can record circumstances should it occur so that maybe a pattern emerges


I will go through different aborts and nominal insertions to try to reproduce what I observed with the CSS roll stucked.
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Bugfixes

Postby GinGin » Wed Feb 17, 2021 1:59 pm

I have done several RTLS without any roll problems so far.
MLS runway 15 and GPS gave a very good SV, Very Accurate TAEM. I will give a try without MLS and GPS update

I pushed a small correction commit ( typo error in the pfd.nas) and I activated the BodyFlap logic for GRTLS before TAEM.
It helps a lot to have the elevons centered during Nz and alpha recovery

Lets go for the Hubble TALs now :)
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Bugfixes

Postby Thorsten » Wed Feb 17, 2021 2:08 pm

For Banjul, I don't fly a TAL often, but I hardly believe drift errors can push you out by 2nm!!


Where and when (and is this with BFC lights blinking or not? If you have a PASS software error on top of everything else, all bets are off.)

At 100 miles distance to the station, the 2.5 deg lateral error of TACAN correspond to a 4.3 nm lateral distance you can have as compared with GPS. If TACAN actually is offset in the same direction as the inertial navigation, you can be 8.6 nm away from where you think you are and still have completely acceptable ratios.
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Space Shuttle - Bugfixes

Postby Thorsten » Wed Feb 17, 2021 2:40 pm

I am getting the BFC blinking each time I am flying during the transitions, so I must trigger it by some procedure.


You're actually right, my code was too clever and I managed to confuse myself, sorry. So we suppress BFC warning upon 201 transition if no BFS is running anywhere.
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Space Shuttle - Bugfixes

Postby GinGin » Wed Feb 17, 2021 5:01 pm

With GPS active and Delta State Update, I managed some close to MLS accuracy in Banjul. I posted in Flight.
We might add an MLS frequency though (?) It was a prime TAL for equatorial launch, so NASA would have send there a portable MLS.
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Bugfixes

Postby Thorsten » Wed Feb 17, 2021 6:48 pm

We might add an MLS frequency though (?) It was a prime TAL for equatorial launch, so NASA would have send there a portable MLS.


I actually used some NASA list of landing sites to determine who gets MLS at what channel. But feel free...
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Space Shuttle - Bugfixes

Postby wlbragg » Wed Feb 17, 2021 6:54 pm

since I didn't have manual control when landing yesterday, I happened to test the sounds when you land without gear - they sound about right to me :mrgreen: Thanks.

Good, I also added some logic to change the pitch based on rolling friction factor of the material properties of the ground surface. So if you slide off the runway it will change pitch. Also Edwards salt flat VS pavement will sound different as well.
Kansas and Ohio/Midwest scenery development.
KEQA, 3AU, KRCP Airport Layout
Intel i7/GeForce RTX 2070/Max-Q
User avatar
wlbragg
 
Posts: 7587
Joined: Sun Aug 26, 2012 12:31 am
Location: Kansas (Tornado Alley), USA
Callsign: WC2020
Version: next
OS: Win10/Linux/RTX 2070

PreviousNext

Return to Spaceflight

Who is online

Users browsing this forum: No registered users and 1 guest