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 » Fri Sep 14, 2018 5:42 am

These messages were before loading scenery. Then, it continued reporting *CRASH DETECED* until SRB separation, when JSBSIM went completely haywire.


That's what's confusing me as well.

The Shuttle mission is divided into phases, and each phase does its own computations and limit checks (it's folly to check for aerodynamics in outer space).

You had before an issue with launch guidance evaluating time to MECO before SRB separation - which is only does after SRBSep.

So for some undetermined reason, the phase determination is off - it seems your install is trying to run all phases at the same time or running through phases rapidly.

Since the transition to phases is by well-determined events, I don't understand how that can happen (for instance, the phase transition to post SRB-sep guidance requires SRBs to be actually separated and the separation property to be set, so I don't see how you can see SRBs attached in FG and have guidance believe they're not attached.

So there's something weird on a really fundamental level - and I have no clue what.
Thorsten
 
Posts: 9897
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby legoboyvdlp » Fri Sep 14, 2018 8:15 am

Hi,
I've completely reinstalled the Shuttle.

On startup, I got messages:
Code: Select all
Touchdown
Wheels Stop


Therefore, I think it might be registering a touchdown as you drop onto the pad.
I'm also still getting the
Code: Select all
flight:3:C:\Jenkins\workspace\Windows-nightly\flightgear\src\FDM\JSBSim\JSBSim.cxx:564:41457: *CRASH DETECTED* 172.65 seconds: LEFT_MAIN
flight:3:C:\Jenkins\workspace\Windows-nightly\flightgear\src\FDM\JSBSim\JSBSim.cxx:564:41458: *CRASH DETECTED* 172.65 seconds: RIGHT_MAIN
, so we'll see if launch works.

Furthermore, when spawning the loose items on the pad, they essentially spawn at the front and go into the payload bay where they start to float - looks like there's no gravity in the payload bay :) But I thought the loose objects would only go inside the cabin and would bounce off the walls...?

Sorry for all the trouble...
User avatar
legoboyvdlp
 
Posts: 5697
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 Sep 14, 2018 8:30 am

I guess they don't do to many launches from Kansas.


Fixed it anyway.

Therefore, I think it might be registering a touchdown as you drop onto the pad.


There are listeners to WoW and the gear is (in the FDM) extended when on the pad (JSBSim doesn't like to init aircraft on the ground with gear retracted) so I guess it's somehow possible that WoW registers, although the wheels really do not touch anything (and this happens on none of my computers).

However, the associated loop terminates itself after wheels stop, so it should be harmless in itself.
Thorsten
 
Posts: 9897
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby Thorsten » Fri Sep 14, 2018 9:40 am

Okay, just flown one regular entry and one entry with NO Y JET mode from Mach 22 down to Mach 10 with no issues - Shuttle working like a charm.
Thorsten
 
Posts: 9897
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby GinGin » Fri Sep 14, 2018 6:57 pm

Awsome, one more thing to try the No Y Jet :)

I took another blast today with that Velcro and Finger Print on screens
Man, its awsome

Image


Image


One more test : 2 APU inop, Lock up of an engine at 100 %, No TVC for Engine 2 thus Shutdown and TAL to FMI
Worked well and I tried to land at Nimes near Istres, that was a fail xD


Image



Image
GinGin
 
Posts: 374
Joined: Wed Jul 05, 2017 10:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle

Postby legoboyvdlp » Fri Sep 14, 2018 7:26 pm

@GinGin,
you might want to clean your screen :D
User avatar
legoboyvdlp
 
Posts: 5697
Joined: Sat Jul 26, 2014 1:28 am
Callsign: YV-LEGO
Version: 2018.3.1
OS: Windows 10 HP

Re: Space Shuttle

Postby Thorsten » Sat Sep 15, 2018 7:38 am

Well, surely dirty screens do not cause multiple APU failures... :mrgreen: He probably was too busy flying.

(Actually, we can make the dirt level on the screens a parameter - it'd make sense)
Thorsten
 
Posts: 9897
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby Thorsten » Sun Sep 16, 2018 10:17 am

After a good hour of testing PDRS and orbital ops, I've now tagged the milestone - GinGin is right, this seems to be a rather stable version...
Thorsten
 
Posts: 9897
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby Thorsten » Tue Sep 18, 2018 11:58 am

It's been a while, but the next flight-tested milestone of the Shuttle is on the server in time for the next stable FG release (to appear soon). To accompany the milestone, the Shuttle Flight Manual has been updated and extended to reflect the new options and capabilities (the full version of the manual now has 237 pages). It aims to give a comprehensive introduction to Shuttle operations by not only detailing the procedures but also giving some engineering, spaceflight and physics background to understand better why the procedures are the way they are.

Image

The Basic Edition of the manual (describing how to get the Shuttle into space and back) is free for FG users (voluntary donations are gratefully accepted though), a license for the Deluxe Edition (describing off nominal procedures such as aborts and troubleshooting as well as interaction with the computer systems or details of orbital targeting) can be obtained for 15 Euro.

Download your version from here !
Thorsten
 
Posts: 9897
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby legoboyvdlp » Tue Sep 18, 2018 6:40 pm

Regarding my earlier issues, I had a successful launch today (got the AOA warning twice, but managed to stay in limits), however, I have a 400MB log by the time FlightGear crashes.
I had these symptoms:
Image
Vapor trail (not shown in screenshot for some reason) from there.

Image
Cockpit strange rendering for about 10 seconds, then back to normal

Image
Speedbrake extended itself?

Code: Select all
input is NaN.

console error: please see viewtopic.php?f=25&t=33237

also:
Code: Select all
getprop: property /velocities[0]/equivalent-kt[0] is NaN


The fact that you don't see any of this means it will be near-impossible for you to track it down :cry:
Maybe check the equivilent-kts calculation, as when I looked at /velocities[0]/equivilent-kts after seeing a console warning about it, was showing something like -nan?

I just pulled fgdata, got a new nightly, and pulled develop branch of the SpaceShuttle.

Edit:

see parameters.xml
Code: Select all
<tape>
      <name>KIAS</name>
      <condition>
         <and>
         <not>
            <equals>
            <property>/sim/current-view/view-number</property>
            <value>0</value>
            </equals>
         </not>
         <not>
            <equals>
            <property>/sim/current-view/view-number</property>
            <value>12</value>
            </equals>
         </not>
         </and>
      </condition>
      <x>-280</x>
      <y>-100</y>
      <width>28</width>
      <height>200</height>
      <option>right</option>
      <option>vertical</option>
      <major-divisions>10</major-divisions>
      <minor-divisions>5</minor-divisions>
      <display-span>50.0</display-span>
      <tick-bottom>true</tick-bottom>
      <tick-top>true</tick-top>
      <tick-right>false</tick-right>
      <tick-left>false</tick-left>
      <cap-bottom>false</cap-bottom>
      <cap-top>false</cap-top>
      <cap-right>false</cap-right>
      <cap-left>true</cap-left>
      <marker-offset>0.0</marker-offset>
      <enable-pointer>true</enable-pointer>
      <pointer-type>fixed</pointer-type>
      <tick-type>line</tick-type>
      <tick-length>variable</tick-length>
      <input>
         <property>/velocities/equivalent-kt</property>
         <min>0.0</min>
      </input>
   </tape>


and cockpit-orbital.xml:

Code: Select all
<!-- Gliding stuff - only shown in final stage HUD mode -->

   <label>
      <condition>
         <and>
         <equals>
            <property>/controls/shuttle/hud-mode</property>
            <value>3</value>
         </equals>
         <or>
         <equals>
            <property>/sim/current-view/view-number</property>
            <value>0</value>
         </equals>
         <equals>
            <property>/sim/current-view/view-number</property>
            <value>12</value>
         </equals>
         </or>
         </and>
      </condition>
      <width>2</width>
      <height>2</height>
      <x>460</x>
      <y>70</y>
      <format>Airspeed:       %3.0f KEAS</format>
      <halign>left</halign>
      <input>
         <property>velocities/equivalent-kt</property>
      </input>
     </label>


This property is seemingly created in:
https://sourceforge.net/p/flightgear/fl ... flight.cxx

Code: Select all
_tiedProperties.Tie("/velocities/equivalent-kt", this,
                      &FGInterface::get_V_equiv_kts); // read-only
User avatar
legoboyvdlp
 
Posts: 5697
Joined: Sat Jul 26, 2014 1:28 am
Callsign: YV-LEGO
Version: 2018.3.1
OS: Windows 10 HP

Re: Space Shuttle

Postby legoboyvdlp » Tue Sep 18, 2018 7:06 pm

Removing the references to that property in the HUD still results in this error:

Code: Select all
Nasal getprop: property /velocities[0]/equivalent-kt[0] is NaN


notice it now refers to nasal: before it just was getprop: .



1. I see this property is referenced in limits.nas, which my PC also seems to have trouble with. I'll have a look in there.
2. I'm sorry for spamming this topic perhaps, but if I find any errors at least I can fix them...

3. Also, if you don't mind me saying,

Is there anything non-standard about your FG installation or how you're using it? All the bug reports look like Nasal is going haywire (like after a bad reset), with a loop stuck running somewhere which ought to be started later, or a listener removed which should still be there - and no one except you seems to be able to see these issues. So I don't know at all what to make of this...


What I'm thinking is, perhaps on my system the timing of everything is off due to low framerates (20s - 30s). Seemingly you don't have any listeners to fdm-init - perhaps that means it works on your system, but on mine it doesn't. I'm not explaining this properly, but its to do with the discussions about listeners / loops on the mailing list?
Last edited by legoboyvdlp on Wed Sep 19, 2018 2:29 pm, edited 2 times in total.
User avatar
legoboyvdlp
 
Posts: 5697
Joined: Sat Jul 26, 2014 1:28 am
Callsign: YV-LEGO
Version: 2018.3.1
OS: Windows 10 HP

Re: Space Shuttle

Postby eatdirt » Wed Sep 19, 2018 2:21 pm

Hi there,
thanks for the new update, and the Velcro and dirt :)

I have started to play with DPS to maximum realism and I am following the Gingin's tutorial, for instance this:
http://wiki.flightgear.org/Flying_the_Shuttle_-_Launch_And_Post_Insertion_Advanced#Prelaunch_procedure

1) Is this expected that BFS is not following PASS up to OPS 105?
When we prepare for OMS-2 burn, we switch PASS to OPS 105. On the above tutorials, you can see on the pictures that BFS is not following, it remains locked in OPS 104. I confirm that this is still the case with the latest version of SpaceShuttle (see snaphots below)

2) I have an issue since a while with the timer for OMS burns. I think I am making the mistake sometime to load ITEM 23 before ITEM 22.... and I am getting negative values for the timer:

Here a snapshot, after loading ITEM23, look at the timer upper right:
Image

I would expect it to be fine, current mission time is there:
Image

This is catastrophic because I never manage to set it to zero afterwards and everything else fails. The Shuttle cannot be put in the correct attitude (namely it targets an attitude at a time which is crazy), the flashing EXEC never appears etc...
I would not expect the mistake of loading 23 before 22 to be so dramatic?

Thanks for any feedback!
Cheers,
Chris.
eatdirt
 
Posts: 96
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle

Postby Thorsten » Wed Sep 19, 2018 2:27 pm

1) Is this expected that BFS is not following PASS up to OPS 105?


Yes. BFS just has a this OPS to do OMS maneuvers. You do not actually need 105 in BFS.

2) I have an issue since a while with the timer for OMS burns. I think I am making the mistake sometime to load ITEM 23 before ITEM 22.... and I am getting negative values for the timer:


Never seen this (but then, I've never tried to set a burn timer before the burn is loaded).

Have you tried SPEC 2 ? It's just a CRT timer, it should respond to timer set and delete commands.

I would not expect the mistake of loading 23 before 22 to be so dramatic?


You're not running on the real avionics, you're running on my code - which was not developed by a dedicated team of specialists trying to think of every eventuality but by...me.
Thorsten
 
Posts: 9897
Joined: Mon Nov 02, 2009 8:33 am

Re: Space Shuttle

Postby eatdirt » Wed Sep 19, 2018 2:57 pm

I even know why I am making the mistake... You'll like it...
I am targeting special MT values to be able to do easier mental calculations for the TIG, which rushes me into forgetting the load... Yea, that's pathetic, this is what happens when you put lay man in space :)


You're not running on the real avionics, you're running on my code - which was not developed by a dedicated team of specialists trying to think of every eventuality but by...me.


LOL, yes! But it looks so real that we forget :)

I'll try SPEC 2, thanks!
eatdirt
 
Posts: 96
Joined: Wed Aug 15, 2018 2:06 pm

Re: Space Shuttle

Postby Thorsten » Wed Sep 19, 2018 5:02 pm

I'll see whether we can add a veto to the timer when it refers to the past to make it safer for you 8)
Thorsten
 
Posts: 9897
Joined: Mon Nov 02, 2009 8:33 am

PreviousNext

Return to Aircraft

Who is online

Users browsing this forum: No registered users and 10 guests