Question Glideslope / BaseSync reentry of STS/Shuttle Fleet Shuttle

PriestLizard

Member
Joined
Jul 10, 2009
Messages
95
Reaction score
3
Points
8
Location
Toronto
Hi all,

I'm using the Space Transportation System 2016 add-on, together with Glideslope 2.7 and BaseSync, and I'm having trouble understanding why my velocity after EI is not going down. This is the procedure:

  • Undock from MIR, at roughly 400x400 orbit.
  • Use Glideslope to select Cape Canaveral (tried other bases too) with appropriate runway, also make sure that the selected ship is "Space Shuttle" in the Glideslope config.
  • Sync BaseSync with Glideslope and find closest orbit to target.
  • Auto burn in Glideslope based on BaseSync's closest passage, EI angle roughly 0.7-0.8.
  • Auto burn happens, and coast to EI.

At around EI, I engage the autopilot in Glideslope to hold the reference AoA.

Starting EI, I control the bank angle to maintain a VSp roughly around reference value to ensure a descend rate that's according to Glideslope. I also use https://www.orbiterwiki.org/wiki/Intuitive_Atmospheric_Entry as a reference for rough bank angles but I suppose its to be expected that the actual bank angle needs to be different as the aerodynamics are not 100% accurate.

And this is where things really go astray.

Again - after EI, at altitudes around 120km to about 60km, I control the bank angles manually to maintain an appropriate descent rate along the glideslope, while the AUT autopilot maintains a 40deg AoA. But the problem is - I'm just not slowing down.

See attached image. I'm at an altitude of 70km, a bit over glideslope, and I'm not slowing down at all. Still flying at 7.5k TAS, and as per Glideslope reference, I should be down to 6k by now. This then of course messes up the rest of the reentry quite severely. I managed to land pretty close to runways a few times, but its usually a very awkward scramble as I'm totally overshooting the target.

I tried multiple things already:
  • Tried de-orbiting from different orbit altitudes (not sure if BaseSync/GS are assuming a MIR/ISS orbit altitude).
  • Tried different perigee of the orbit. The default burn from GS/BaseSync seems to lower the perigee to around 80km, but tried to lower it further to 40km as well. Same result.
  • Tried banking less, which would mess up the descent speed, and still doesn't slow me down.
  • Tried lowering propellant/mass during re-entry, no change.

I was thinking that maybe it is related to the Shuttle model in STS2016, but I tried de-orbiting with the default DG and I had similar results (even though I didn't go through that amount of extensive testing).

So - how can I slow down in the upper atmosphere at around 100km? Based on the reference data in Glideslope, I'm supposed to slow down right around the EI, and I can never make it happen.

Thanks in advance for any pointers,

Cheers
 

Attachments

  • reentry.png
    reentry.png
    208.8 KB · Views: 12
Last edited:

PriestLizard

Member
Joined
Jul 10, 2009
Messages
95
Reaction score
3
Points
8
Location
Toronto
Just to show another example: This time stock Atlantis, same procedure, same behavior.

At 100km altitude, Glideslope indicates I should decelerate by -0.67m/s2 - but in fact I'm still accelerating. AoA is 40deg, and I tried banking or not banking - same result.

Is Glideslope just generally very off with these reference values and should one not worry about them? But then again I am constantly overshooting my destination, so something is clearly wrong.

Any thoughts?
 

Attachments

  • atlantis.png
    atlantis.png
    160.5 KB · Views: 9

diogom

Well-known member
Joined
Aug 2, 2010
Messages
1,370
Reaction score
413
Points
98
It kinda sounds like the atmosphere in Orbiter is thinner on the upper ranges than the MFD expects. FWIW, I'm seeing the same with a Soyuz, that is, it's not losing speed fast enough and EI ends up happening 10 km too low. But, I'm also not 100% my aero configuration is correct. Still, could be something here.
 

PriestLizard

Member
Joined
Jul 10, 2009
Messages
95
Reaction score
3
Points
8
Location
Toronto
Yea, that makes a lot of sense. Even though Glideslope always used to be a very popular add-on AFAIK, so it would surprise me that the values are off by so much.

I did some digging in the config files, and turns out the Glideslope values aren't actually computed by any means but are just taken from data tables.
\Orbiter\Config\MDF\Glideslope has config files for the different ship types you can choose from.

Converting the first few rows into metric units, this is what you'd get:

Dist KmAlt kmTAS m/sVspd m/sAoA
8,282.14121.927,620.00-103.6340
4,878.1774.687,315.20-14.0240
3,996.6272.247,010.40-19.2040
3,335.4570.416,705.60-24.9940

...and that just doesn't seem to add up with what happens in Orbiter, unless I'm messing something else up fundamentally.

I checked Earth.cfg if perhaps some of the atmospheric values are off. If there was a major change in the latest Orbiter iterations around atmosphere physics, that would explain why Glidescope has such different expectations. The only value I could find is

AtmAttenuationAlt = 100e3; ; cutoff altitude for light attenuation

I will try setting this to 140e3 - perhaps that will mean that in the early stages of the reentry right after EI, there is a more significant drag effect occurring.

Apart from Glideslope, does anyone have a good experience with similar MFDs? Specifically Shuttle compatible ones?
I haven't looked into GPCMFD in detail yet, but AFAIK this one is based on data tables as well and relies entirely on manual control.
 

Snax

Space Cowboy
Joined
Jun 20, 2018
Messages
214
Reaction score
151
Points
58
Location
Brussels
Hello @PriestLizard, when I deorbit with a Shuttle-like vessel (Buran, Albatross, SSU/SSV, SF, STS2016), I use 1.23 degrees instead of the default 0.78 degrees. I don't remember from where I know that but that's what I use and it always works. I tried it with:
  • the default Shuttle (manual reentry, tho the RCS are very weak to maintain AoA, it works),
  • SSU/SSV (auto EI),
  • ShuttleFleet/BuranFleet (Orbiter 2010) (manual EI because AutoFCS won't work lol),
  • and STS2016 (manual EI too).
Except for SSU/SSV wich MM304 mode will do it automatically, I do the EI by hand just in case the autopilot would do anything but what it should :p
I hope it works for you too !

NotaBene: Your initial entry would be "lower" than the yellow line, I just wait with the wings leveled until my VVI reached 0 to start a left or right roll to maintain the VVI at 0 until the yellow line start to descend too.
PS: Sometimes if your velocity isn't reducing as expected, I sometimes raise the nose just a little bit more (from 40° AoA to 50° of AoA) to show more of the belly to create more drag (and by so forcing more on the S-Rolls to maintain the VVI where it want it to be).

Here's two pictures illustrating my results (I will show more to show the beginning of the EI):
1661622973866.png1661622986245.png
(don't mind the red/green boxes, I was showing to a friend where to click to turn on these displays)
 
Last edited:

PriestLizard

Member
Joined
Jul 10, 2009
Messages
95
Reaction score
3
Points
8
Location
Toronto
Interesting - so do you override the default EI angle in the BaseSync config?

Meanwhile, I have been experimenting with setting my own Glideslope data based on the actual behavior of the (STS2016) shuttle. I have been getting much better results, even though it still needs some tweaking. Ultimately it comes down to the EI being way too close when using the default Shuttle GS data.

This is the data I have been using so far, will continue to tweak and iterate:

UNITS US
BEGIN GLIDESLOPE
; Range Alt TAS VSpd AoA Comments
; NMi kft ft/s ft/s ° ; first please
; ------ ----- ---- ---- ----

7019 400.3 24934 -262 40
5842 246.1 24606 -230 40
4752 239.5 24606 -66 40
4320 229.7 24508 -66 40
3780 223.1 24147 -66 40
3240 210.0 23392 -66 40
2592 206.0 22507 -66 40
2160 200.1 21883 -66 40
1674 193.6 20997 -49 40
1080 183.7 19029 -98 40
810 173.9 17585 -115 40
675 160.8 16175 -131 40
540 144.4 13681 -131 40
405 131.2 11581 -66 40
270 124.7 9055 -197 39
162 111.5 4199 -197 38
141 108.0 4100 -250 36
106 105.0 4000 -250 34
75 89.0 3000 -260 30
61 82.0 2500 -260 27
49 76.0 2000 -280 23
37 65.0 1500 -320 19
28 51.0 1000 -250 16
0 0.0 195 -2 14
END GLIDESLOPE

When doing the default reentry at 0.7-0.8 degrees, I have a pretty smooth reentry "on curve" - but still overshooting slightly. So needs some iteration still - will provide an update once I'm happy.

Oh, and PS: How do you actually manage to change the MFD setting for the Shuttle VC? All I ever manage is to turn the display on - cannot seem to change what MFD is shown where.
 

Snax

Space Cowboy
Joined
Jun 20, 2018
Messages
214
Reaction score
151
Points
58
Location
Brussels
Yes, in GlideSlopeMFD I choose the Space Shuttle preset to load the ANG, ANT, ALT, then I link them with "GS" as Target in Basesync.
Once it's done, I unlink then by manually choose my target (Edwards in my screenshot) so I can change the ANG manually too in the MFD. That's all I do.
1661651195750.png

To change the other MFD displays, with the yellow box, I hold the left mouse button for 1 secon and the MFD menu selector will appear.
 

Snax

Space Cowboy
Joined
Jun 20, 2018
Messages
214
Reaction score
151
Points
58
Location
Brussels
Hi @PriestLizard, I did a few flight since, and I did one timelapse to show you my results of what I was suggesting.
Starting from 1 minute 10 seconds. I also realize that I don't really follow the yellow line (Alt) exactly, I always go 5km below to have more drag.
 

PriestLizard

Member
Joined
Jul 10, 2009
Messages
95
Reaction score
3
Points
8
Location
Toronto
Thanks for posting this Snax!

I tried it with your values:

EI 1.23deg
Ant 74.48deg
Alt 121.92km

Starting orbit: MIR orbit, so around 400x400.

I got totally different results compared to your video. What shuttle are you flying? And which Glideslope cfg? I'm using the STS2016 shuttle (installed via ORB), and I'm using the default Shuttle cfg for Glideslope in my example.

Some screenshots:
1662082284692.png

As you can see, I'm only just now starting to slow down at around 75km altitude, and my green line is already "too high" even though I tried to stay slightly below glidepath.

1662082452360.png

At around 65km you are already down to 6.5km/s, whereas I'm still shooting along at almost 1k more (7400m/s).

1662082582963.png

At 50km altitude, you are down to 3.6km/s, while I'm at 5.7km/s.

It actually ended up one of my best landings so far (came in a bit hot, but otherwise managed to land on the intended runway), but I was obviously overshooting by a wild margin, circling back over the Atlantic ocean, and then coming back in, circling one more time towards the target.

I don't really know what I'm doing wrong. I'm clear on the theory. If you are too fast, sink faster to slow down more. But if I descent too quickly, I have problems maintaining my AoA (here I'm already struggling to maintain 20deg - I had attempts where I was more aggressive where my AoA pulls down to 7deg even at max RCS thrusters).

Any pointers? :/

Oh, and BTW - what HUD are you using!? It's awesome to have all that data available on screen! As you can tell, I have to use the VC as its the only way to have more MFDs on screen. But I'd prefer just the normal cockpit with custom HUD like you have in your example).
 

Snax

Space Cowboy
Joined
Jun 20, 2018
Messages
214
Reaction score
151
Points
58
Location
Brussels
Thanks for posting this Snax!

I tried it with your values:

EI 1.23deg
Ant 74.48deg
Alt 121.92km

Starting orbit: MIR orbit, so around 400x400.

I got totally different results compared to your video. What shuttle are you flying? And which Glideslope cfg? I'm using the STS2016 shuttle (installed via ORB), and I'm using the default Shuttle cfg for Glideslope in my example.
In the video it was a Spaceplane by Kulch, but I also fly with the STS2016, SSV (and ShuttleFleet/BuranFleet on Orbiter 2010). I get similar results with these. For Glideslope I only choose the Space Shuttle preset, I don't touch anything else.

At 50km altitude, you are down to 3.6km/s, while I'm at 5.7km/s.

It actually ended up one of my best landings so far (came in a bit hot, but otherwise managed to land on the intended runway), but I was obviously overshooting by a wild margin, circling back over the Atlantic ocean, and then coming back in, circling one more time towards the target.

I don't really know what I'm doing wrong. I'm clear on the theory. If you are too fast, sink faster to slow down more. But if I descent too quickly, I have problems maintaining my AoA (here I'm already struggling to maintain 20deg - I had attempts where I was more aggressive where my AoA pulls down to 7deg even at max RCS thrusters).

Any pointers? :/
I think you're good on the theory, that's what I do too, by looking at your screenshots I don't understand why the Shuttle isn't losing velocity. I sometimes pull more AoA hoping it would help to show more of the belly (but not 100% working).

I'm not sure if deploying the speedbrake would do anything as the tail is in the blindspot during reentry :/
It took me a few tries to get it right, then all the time is was good, without having to do some unrealistic maneuvers.

Oh, and BTW - what HUD are you using!? It's awesome to have all that data available on screen! As you can tell, I have to use the VC as its the only way to have more MFDs on screen. But I'd prefer just the normal cockpit with custom HUD like you have in your example).
This HUD is coming with the Spaceplane I've linked before, it's hardcoded to work with this one only however.
 

PriestLizard

Member
Joined
Jul 10, 2009
Messages
95
Reaction score
3
Points
8
Location
Toronto
Thanks for the input Snax.

I suppose I could try increasing the anticipation angle so I'm farther away from the destination, and have more time to burn down velocity?
I don't think I meddled with any settings in Earth.cfg either to change the atmosphere?

The only thing I can think of is that maybe I bank too early. I usually start banking at EI to maintain a good descent rate - I will try to bank later to maybe burn some more speed before the bank maneuver. But not sure if it would make that big of a difference. The biggest discrepancy really is around 75km, deep into the reentry, where I'm losing WAY less speed than I'm supposed to, even though I'm lower than expected, yet with correct AoA. And at that point I have to bank in order to maintain a solid descent rate.
 

martins

Orbiter Founder
Orbiter Founder
Joined
Mar 31, 2008
Messages
2,448
Reaction score
462
Points
83
Website
orbit.medphys.ucl.ac.uk
I checked Earth.cfg if perhaps some of the atmospheric values are off. If there was a major change in the latest Orbiter iterations around atmosphere physics, that would explain why Glidescope has such different expectations. The only value I could find is

AtmAttenuationAlt = 100e3; ; cutoff altitude for light attenuation

I will try setting this to 140e3 - perhaps that will mean that in the early stages of the reentry right after EI, there is a more significant drag effect occurring.

Apart from Glideslope, does anyone have a good experience with similar MFDs? Specifically Shuttle compatible ones?
I haven't looked into GPCMFD in detail yet, but AFAIK this one is based on data tables as well and relies entirely on manual control.
The AtmAttenuationAlt config entry is used only for visualisation. It has no effect on the atmosphere physics used in Orbiter. The physical behaviour is encoded in a DLL, and in fact you have a choice of three different atmosphere models:
  • Orbiter 2006 legacy: do NOT use this. It is kept for backward compatibility only and is quite bad at altitudes > 120km (and not defined at all at altitudes > 200km)
  • Jacchia71 (modified)
  • NRLMSISE00 which is a derivation of MSIS90
You can select the model under the Extra tab in the Orbiter Launchpad. I haven't got Orbiter to hand at the moment, and I can't recall the exact entry, so you may have to search a bit. It is something like "Atmosphere configuration".
For a general discussion about the Jacchia and MSIS models, see for example https://link.springer.com/article/10.1007/BF03546372
For a discussion about the implementation in Orbiter and model modifications, see <Orbiter>/Doc/Technotes/earth_atm.pdf

Of the three models, NRLMSISE00 is the most complex (and slowest). Both Jacchia and MSIS take into account daily and seasonal cycles, so the parameters will depend (to a small extent) on time and date. Having said that, at altitudes < 120km I would expect the models (even the 2006 legacy model) to behave fairly similarly, in any case not to differ to the extent that explains your problems with a wrong choice of atmosphere model. Still, it wouldn't hurt to try the different models.

To debug what is happening on your descent trajectory, maybe the default "Flightdata" plugin can be useful. It plots various flight parameters, such as altitude, dynamic and ambient pressure, as well as lift and drag as a function of time (or writes them to file for post-mortem analysis). If you have any reference data to compare them to, maybe this will point to the problem.
 

PriestLizard

Member
Joined
Jul 10, 2009
Messages
95
Reaction score
3
Points
8
Location
Toronto
Looks like I'm using Jacchia71-Gill right now.

Just tried logging the data from the Flightdata plugin, good suggestion Martin, thanks! Overall reentry was like the ones posted above: Coming in way too hot, and overall burning too little velocity. That's the classic picture for me:

1662171232622.png

I feel like the "critical phase" where I don't lose enough velocity is between 70-50km. So focusing the Flightdata on that corridor (full data attached):

# ____TIME _______ALT _AIRSPEED ___TEMP _______STP _______DNP ____AOA ___SLIP _____LIFT _____DRAG _____L/D
710.45 65.9812 7450.92 230.6 0.0095 3.9767 42.9 4.4 519.36 127.33 4.079
808.36 62.9093 7289.69 239.0 0.0147 5.6909 42.5 0.0 750.91 183.80 4.086
871.96 60.9016 7125.66 244.5 0.0194 7.0202 32.4 -0.0 1142.49 281.97 4.052
965.83 55.6961 6838.27 258.9 0.0389 12.2259 24.1 -0.0 2086.34 519.02 4.020
1045.57 53.3245 6367.43 265.4 0.0527 14.0114 17.9 -2.4 2400.09 597.60 4.016
1108.15 50.8104 5988.43 270.6 0.0721 16.6437 21.4 0.1 2844.82 707.92 4.019
1155.58 46.7872 5600.54 269.1 0.1190 24.1478 19.2 0.1 4132.59 1028.64 4.018

So drag force only ever exceeds 1000kN below 50km. I'm not sure what the references should be. If someone gets a chance to check their reentry with Flightdata logging, that would be a great reference. Again, this is based on the STS2016 shuttle.

Reentry data:
EI 1.23deg
Ant 74.48deg
Alt 121.92km

Posting the top elements of my Earth.cfg for reference as well - maybe something is wrong here?

; === Configuration file for planet Earth ===
Name = Earth
Module = Earth
Module_Atm = EarthAtmJ71G ; default atmosphere model if no user selection
ErrorLimit = 1e-8
SamplingInterval = 79 ; interpolation sampling interval
; (interpolation error ~0.1m)
; === Physical Parameters ===
Mass = 5.973698968e+24
;Size = 6.378165e6 ; equatorial radius
Size = 6.37101e6 ; mean radius
JCoeff = 1082.6269e-6 -2.51e-6 -1.60e-6 -0.15e-6
; harmonic coefficients for shape description
AlbedoRGB = 0.7 0.85 1.0

; === Rotation and precession parameters ===
; ref: see www.orbiter-forum.com/showthread.php?t=8185
PrecessionLAN = 0 ; [rad]
PrecessionObliquity = 0 ; [rad]
PrecessionPeriod = -9413040.4 ; precession period (days) = 25771.5 years
LAN = 0 ; [rad]
LAN_MJD = 51544.5 ; [MJD]
SidRotOffset = 4.88948754 ; [rad]
SidRotPeriod = 86164.10132 ;
Obliquity = 0.4090928023 ; [rad]
;SidRotPeriod (days): 0.99726968
;SidRotPeriod (SI seconds - node to node): 86164.092
;Precession Period (years): -25771.5
;Obliquity (deg): 23.439291
;Ecliptic Obliquity (deg): 23.439291

; === Atmospheric Parameters ===
;AtmPressure0 = 101.4e3 ; pressure at zero altitude [Pa] (defined via module)
AtmDensity0 = 1.293 ; density at zero altitude [kg/m^3]
AtmGasConstant = 286.91 ; specific gas constant [J/(K kg)]
AtmGamma = 1.4 ; specific heat ratio c_p/c_v
;AtmAltLimit = 200e3 ; cutoff altitude [m]
AtmAttenuationAlt = 250e3; ; cutoff altitude for light attenuation
AtmHorizonAlt = 80e3 ; horizon rendering altitude [m]
AtmHazeExtent = 0.14 ; horizon haze extent
;AtmColor0 = 0.55 0.75 1.04
AtmColor0 = 0.42 0.72 1.02
AtmHazeColor = 0.55 0.75 1.0
AtmFogParam = 3.7e-5 2.0e-5 5e3 ; 4.0e-5 2.2e-5 5e3
AtmFogColor = 0.57 0.80 0.92
AtmTintColor = 0.04 0.05 0.20 ; additive tint to surfaces from high altitude
ShadowDepth = 0.65


Thanks everyone. Hope I can get to the bottom of this at some point ;)
 

Attachments

  • ReentryData.txt
    201.2 KB · Views: 3

ADSWNJ

Scientist
Addon Developer
Joined
Aug 5, 2011
Messages
1,667
Reaction score
3
Points
38
Necro post here, as I only dop in occasionally to see what's happening. I wrote the latest Glideslope MFD 2.x over 5 years ago, so I'm a bit rusty. But the theory is that you can save your own reentry to generate your own user-defined glideslope, and then promote it back into the Glideslope.cfg to pick up as a default for next time. Also - caution that 95% of my testing was in the XR-2 and XR-5, rether than Shuttle, so you may well be into new territory where you can make the default Shuttle glideslope better for everyone.

On a fantasy vessel like the XR-5, you can change the AoA by quite a lot without issues. More AoA for more drag, to slow down, etc. But for Shuttle if flying it fully realistic, what surprised me was that the AoA had to be kept at 40 deg +/- 3 deg for most of the reentry. So if you are high TAS, you would normally increase AoA to increase resistance, but in the Shuttle this is not allowed. Instead, you roll aggressively (e.g. 70-90, or even roll inverted a bit) to flip the lift vector to 0 or even to negative lift, so the vehicle can sink faster into heavier air. And then you reverse the roll to the other side to correct the drift off course.

TL;DR if too fast at a specific altitude in the Shuttle, with no AoA discretion, then your only way to correct it (without a retro-burn) is to roll to dump the lift vector, to dig into lower atmo, to get more air resistance. If all this is no good, then your original EI parameters need adjusting the next time to get lower sooner.
 
Top