Board index FlightGear Development Spaceflight

Space Shuttle - Entry/TAEM rework

Discussion about development and usage of spacecraft

Space Shuttle - Entry/TAEM rework

Postby GinGin » Wed Feb 17, 2021 12:02 am

I pushed in the dev branch the latest work on Entry and TAEM displays, with some updates concerning the guidance also.


Entry
Image



TAEM
Image

I spent some time to make those displays fully usable and not just placeholder with lines here and there.
Hence, the following Workbook is really a must read to understand how to use well the New Entry Trajectory cues and philosophy behind those archaic green things ( Drag lines, Boundaries meaning,etc) and the Energy over weight management ladder on the right of the TAEM display.
There are also in the book a lot of situationnal exercises that are applicable in the sim ( High, Low energy, Straight in approach, Minimal Entry point etc)

Entry,TAEM, and approach workbook
https://gandalfddi.z19.web.core.windows.net/Shuttle/USA005512%20-%20Entry,%20TAEM%20and%20Approach%2021002%20Basic%2020060123.pdf



For those who want to go even deeper, there are the Handbook, that will explain all the steps of the Entry checklist with a lot of details.
A kind of deeper workbook with a zoom on specific flight data files procedures.

Entry Handbook
https://gandalfddi.z19.web.core.windows.net/Shuttle/JSC-11542%20-%20Flight%20Procedures%20Handbook%20Rev%20E%20200504.pdf

Approach,Landing,Rollout Handbook
https://gandalfddi.z19.web.core.windows.net/Shuttle/JSC-23266%20-%20Approach,%20Landing%20and%20Rollout%20Flight%20Procedures%20Handbook%20-%20Rev%20B%20200505.pdf


Some entry parameters for testing:


Entry

Nominal KSC entry (4500Nm of entry range) with a nice energy management at KSC for TAEM ( either a 360 ° HAC or a Straight in ) . Good to test a wide range of energy situations.

Code: Select all
--aircraft=SpaceShuttle-entry
--lat=12.96
--lon=-158.87
--heading=51.5



High Energy KSC (4000 Nm REI )
It might require to manually bank to 90 ° in CSS up to guidance is active above q bar = 8 ( value written in the entry display). Then back to Auto mode.
Pre Bank procedures are described in the books above.

Code: Select all
--aircraft=SpaceShuttle-entry
--lat=12.96
--lon=-151.87
--heading=51.5



Low energy KSC (5200 Nm REI)
Far less fuel efficient as a longer time is spent in the atmosphere. In real, it would have be on the limit for the "backface temperature" ie. the aluminium temperature behind the tiles that might get too hot in a long entry.
Entry early path and flight path angle was a trade off between the max tiles temperature and the max duration that could handled the parts behind the tiles ( explained in the book above)


Code: Select all
--aircraft=SpaceShuttle-entry
--lat=12.96
--lon=-174.87
--heading=51.5




TAEM

Nominal return to KSC runway 15 Overhead

Code: Select all
--aircraft=SpaceShuttle-TAEM
--lat=28.627
--lon=-80.2010
--heading=060.0



Low energy that might require a Straight In Downgrade ( with Fault message)

Code: Select all
--aircraft=SpaceShuttle-TAEM
--lat=28.627
--lon=-80.4510
--heading=060.0



Low energy also where you can try an Overhead Minimal Entry Point.
Quite impressive to be aligned just 4 Nm before the runway at 6000 feet,

Code: Select all
--aircraft=SpaceShuttle-TAEM
--lat=28.627
--lon=-80.5010
--heading=060.0


Dont forget to turn on the help/info messages :)

PS: To have an accurate distance to runway when flying to the HAC entry waypoint 1 after a long turn to be aligned with this point, it can be nice to re enter an item 3 or 4 EXE in the Spec 50 display to reload the function that will calculate the HAC entry point and distance.



Image

Image
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Entry/TAEM rework

Postby Thorsten » Thu Feb 18, 2021 3:56 pm

Code: Select all
--aircraft=SpaceShuttle-entry
--lat=12.96
--lon=-158.87
--heading=51.5


That parameter set killed me pretty efficiently... It seems rather far from the site and weird things were happening.

Anyway, I've now committed a new standard entry scenario parameter set selectable via the scenario dialog with 4200 miles to site, that seems to work rather well (at least I landed in the end and usually was close to the center line of every display) :mrgreen:

A few comments:

* the AP seems at time rather agressive with the roll management, leading to severe oscillations. Especially at ENTRY TRAJ 2 I saw 90 deg bank angle followed by 0 deg, followed by 70 deg basically executed as fast as the orbiter would roll. In essence, this is oscillating drag around the desired value with insufficient damping.

Given that the scenario inserts right into the center line of ENTRY TRAJ 1 which should be an easy case, the need for such hard maneuvers seems not pressing... I'm somewhat concerned that with the reduced roll performance of NO Y JET, even such an 'easy' standard scenario won't really be doable for the AP.

* I assume the tolerance of 20 deg RelAz before roll reversal is okay - I'm pretty sure I had a lower number taken from SCOM...

* the shrinking HAC at TAEM poses a similar issue - the AP simply isn't equipped to fly this properly, the logic would have to anticipate a cone rather than a cylinder and behave accordingly, but instead I get a violent reaction to the offset and oscillation around the desired value. I've taken over manually since I had serious concerns halfway through the HAC - I know this is how it used to be flown anyway, but the AP shouldn't fail violently to force people to take controls...

Otherwise the experience was pretty nice - lots of new things to discover and see - well done!

Low energy also where you can try an Overhead Minimal Entry Point.


Note that the intended way of regulating TAEM energy is to simply pass altitude - that actually should work fine (did when I tested last). :D
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Space Shuttle - Entry/TAEM rework

Postby GinGin » Thu Feb 18, 2021 4:18 pm

That parameter set killed me pretty efficiently


Weird, it is supposed to be quite nominal and worked well for me, I will have another run to see.

Edit: I ran an entry with those parameters and it was stable for me, no hard oscillations

the AP seems at time rather agressive with the roll management, leading to severe oscillations


Was it with the above scenario parameters ? It might be too much agressive when High on energy and above the nominal line. I will check that

I'm somewhat concerned that with the reduced roll performance of NO Y JET


Good point.
I will run some tests on it as I didnt try it for a long time. I might force the ap to follow just the v error fps with same factor than before when No Y Jet is engaged, and disregard the delta drag bias that might cause a too high roll demand with no RCS.

I assume the tolerance of 20 deg RelAz before roll reversal is okay


Above mach 4, roll reversals start at 17° and by the time it is complete, the RelAz might extend to 21 °ish max in Nasa books
Image

the AP simply isn't equipped to fly this properly


I am about to push a small commit that will take into account the Hac radius shrinking and radial error to have a better AP guidance until final.
That works far better normally.



Otherwise the experience was pretty nice - lots of new things to discover and see - well done!


Thanks :)
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Entry/TAEM rework

Postby GinGin » Thu Feb 18, 2021 6:31 pm

For the no Yaw Jet, I reverted to the v error fps like before only when activated

Code: Select all
<fcs_function name="systems/ap/entry/vspeed-target-fps-function">
      <function>
         <sum>
         <property>systems/entry_guidance/nominal-hdot-target</property>
         <property>systems/ap/entry/thermal-protection-bias</property>
         <product>      
            <property>systems/entry_guidance/v-error-fps</property>
            <value>-0.2</value>
            <property>systems/fcs/no-y-jet</property>
         </product>
         <product>      
            <property>systems/entry_guidance/v-error-fps</property>
            <property>systems/ap/entry/v-error-fps-factor</property>
            <difference>
               <value>1.0</value>
               <property>systems/fcs/no-y-jet</property>
            </difference>
         </product>
         <product>   
            <property>systems/entry_guidance/delta-drag-bias</property>
            <difference>
               <value>1.0</value>
               <property>systems/fcs/no-y-jet</property>
            </difference>
         </product>
         </sum>
      </function>
      </fcs_function>



Like described in procedure, I waited until qbar of 20 and the first bank before to engage it up to Mach 7.
It went very smooth, didnt remember it was that funny to see that sluggy Shuttle oscillating above and below the nominal path.
The ap you coded for No yaw jet is really impressive.
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Entry/TAEM rework

Postby Thorsten » Thu Feb 18, 2021 7:10 pm

That parameter set killed me pretty efficiently... It seems rather far from the site and weird things were happening.


Hm, typo perhaps? :?: Perhaps I flipped a sign?

Anyway - I'll try to re-parametrize the scenarios so that we have easily accessible benchmark tests.

It might be too much agressive when High on energy and above the nominal line. I will check that


I didn't specifically pay attention, but does it do alpha modulation? I found that a really useful tool to dampen oscillations because it's usually a really long time till drag responds to bank angle...

The ap you coded for No yaw jet is really impressive.


Thanks - I've shied away from it for a long time, but once I actually started coding it, it turned out to be reasonably straighforward - and it's something that nicely illustrates what JSBSim as framework can actually do.
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Space Shuttle - Entry/TAEM rework

Postby GinGin » Thu Feb 18, 2021 11:35 pm

Anyway - I'll try to re-parametrize the scenarios so that we have easily accessible benchmark tests.


In the stages.nas, I put the following parameters for entry mission at 400000 kfeet and 4400 Nm REI given by the following entry lat/lon
That gives normally a very nominal entry.

Code: Select all
var rotation_boost = 1579.0 * math.cos(latitude) * math.sin(heading);
   #setprop("/velocities/uBody-fps", 25100.0 - rotation_boost);
   #setprop("/velocities/wBody-fps", 200.0);

   #setprop("/velocities/uBody-fps", 25500.0 - rotation_boost);

   

   #Inertial speed for STS 1 entry type (400 kfeet// 4300 Nm, Gamma -1.18°// 530fps vertical speed// Vi 25753 ft/s)
   setprop("/velocities/uBody-fps", 25753.0 - rotation_boost);

   #Vertical speed
   setprop("/velocities/wBody-fps", 710.0); # Inertial Gamma -1.18°


Code: Select all
--aircraft=SpaceShuttle-entry
--lat=12.96
--lon=-158.87
--heading=51.5



but does it do alpha modulation?


I restrained it a bit ( with low bank angle only) to not interfere too much when I first implemented the new delta drag bias thing
I will bring it back to normal and see.


it's something that nicely illustrates what JSBSim as framework can actually do


Really interesting.
In some Entry DAP updates, they went for a Entry DAP wrap mode using some Gain from the no yaw jet logic into the nominal Entry DAP to save some yaw RCS fuel ( using far less yaw rcs)
They went from 30 % RCS consumption to 10%ish during the latest flights
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Entry/TAEM rework

Postby Thorsten » Fri Feb 19, 2021 4:03 pm

In some Entry DAP updates, they went for a Entry DAP wrap mode using some Gain from the no yaw jet logic into the nominal Entry DAP to save some yaw RCS fuel ( using far less yaw rcs)


I know - it sounded like a really clever scheme, but I haven't mustered the energy to implement it yet...

***

So, a few more obervations: I've now created a high-E scenario which brakes on a shorter path and heats the Shuttle quite a bit more - the AP can do it fine, but I've noticed it does try to bring you to the center trajectory early and quickly - basically the AP has put everything I've thrown at it into the center 2/3 through ENTRY TRAJ 1.

This might not be bad as such, but unfortunately the high-E scenario has managed to overcome the NO Y JET mode i my test, I had loss of control and breakup :(

I'm reading through the documentation you linked, and I'll do some more low E / long range tests (is the low E logic still in?), so at this point these are just observations, not critique.

I've also added a scenario that should lead to White Sands coming from high latitude - the visuals are quite compelling as one crosses Nevada...
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Space Shuttle - Entry/TAEM rework

Postby GinGin » Fri Feb 19, 2021 4:16 pm

it sounded like a really clever scheme, but I haven't mustered the energy to implement it ye


Yes, I read through the thing and I closed the book :mrgreen:
But so far we are well within the Original RCS consumption without the Wrap Entry DAP, even lower. So pretty nice.


the high-E scenario has managed to overcome the NO Y JET mode i my test


I will push tonigh the AP.xml that removes the delta drag bias from V speed target for No Y JET, could you re try it then ?
I will give it a run also.


(is the low E logic still in?


Yep, still in.
It is not required for long range entry normally.
The AP works well to bring us back by chasing the equilibrium glide.
I mainly used the low energy logic for very low energy TAL .

I've also added a scenario that should lead to White Sands coming from high latitude


Very nice, I like thos entries also. I added a new Edwards Runway, the 23 Lakebed where STS 1 landed :)
I will push it tonight also
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Entry/TAEM rework

Postby Thorsten » Sat Feb 20, 2021 8:41 am

Yep, still in.
It is not required for long range entry normally.


Well, I'd like to have a scenario in which activating it is required (after all, that's what scenarios are for) - so it would seem we need really looooong range and/or low initial velocity, yes?
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Space Shuttle - Entry/TAEM rework

Postby GinGin » Sat Feb 20, 2021 3:55 pm

Thorsten wrote in Sat Feb 20, 2021 8:41 am:so it would seem we need really looooong range and/or low initial velocity, yes?


I think more a very low intial velocity.

From the TAL chart
TAL entry interface is around 350 kfeet / 2800 Nm, and when I am in the dashed zone ( 1500 ft/s ish Vrel less than the nominal), I really need the Low energy logic first to come back on the path

Image


So I think for a normal entry, REI 400 kfeet 4200 Nm ish and 2000 ft/s below the nominal Vrel ( around 22500 ft/s or even lower) should put us on the far right No bank equilibrium and might require the use of Low Energy logic

Being far right like that with a low Initial velocity

Image



I found that interesting chart about RCS consumption and No Yaw jet boundaries ( I didnt know it was meant to be flown in CSS)
Found in the Entry handbook linked above

Image
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Entry/TAEM rework

Postby Thorsten » Sat Feb 20, 2021 4:49 pm

Okay, that doesn't seem right...

Image

That's where the AP on the latest snapshot flew me with the lakebed landing scenario. It actually looked all fine during entry and early TAEM, I have good NAV data (even GPS ratio is quite good...) - the AP just didn't seem to steer towards a tangent point. I would have normally taken over much earlier (you can see the John Young message in the upper part) but I was curious what it was trying to do, so the shot is taken just after the 'turn left into HAC message was issued. You can also see that I'm a bit high on energy - not sure why as the energy was nominal at TAEM start. The whole scenario to this point was flown completely under AP control.

In the event I managed a landing (the lakebed is kinda hard to miss - but something seems quirky in the last snapshot with HAC shrinking logic...
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Space Shuttle - Entry/TAEM rework

Postby GinGin » Sat Feb 20, 2021 5:21 pm

I am testing your scenarios, very nice addition that tool and the scenarios work great so far ( testing the high energy for now).
I really like the pictures to load the scenarios.

I have done a lot of TAEM in White Sands with the new logic and it should work properly.
Did you do an "update" of the guidance at the beginning of the TAEM?
I usually wait to Shuttle to have a low Delta Azimuth and refresh the Wp1 calculation and the turn degrees with item 3 (4) at 50 Nm.

Anyway, it does not seem right at all indeed, I will dig that. It is like if it was aiming to the Hac center instead of WP1, never saw that.
I will give some runs with the lakebed and see
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Entry/TAEM rework

Postby GinGin » Sat Feb 20, 2021 6:30 pm

I ran the Lakebed Scenario, very nice scenics, yes.

I think the initial conditions are a bit too much towards a high energy entry ( high vrel, short REI) // 400kfeet / 3800 Nm // 24800ft/s
I think that the same with 4400 nm ish of REI would be perfect

Chasing the thermal boundary, and when the thermal bias protection kicks in to impose a sudden decrease in vertical speed, it starts to oscillate indeed

Image


For the TAEM, I think it was the fact to have not reloaded the target after TEAM.
The WP1 was a bit offset in your screen.
I had that after an item 3 EXE at 60 Nm

Image


and the pull up logic when entering the 360° HAC

Image
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Entry/TAEM rework

Postby GinGin » Sat Feb 20, 2021 7:42 pm

I have added 10 more degrees to the longitude for the lakebed scenario:

Code: Select all
<state>
   <lat>48.08</lat>
   <lon>157.37</lon>
   <alt>400000.0</alt>
   <u-body>24759.0</u-body>
   <v-body>71.938</v-body>
   <w-body>821.51</w-body>
   <heading>70.13</heading>
   <pitch>0.683</pitch>
   <roll>0.0</roll>
</state>


That gives more nominal entry, catching smoothly the third nominal.

What happened to your TAEM gave me an idea.
I forced a TAEM recalculation at 45 Nm to pinpoint the WP1 if we forget to do it
Something simple like that:

Code: Select all
#Recalculation of TAEM guidance target forced at 45 Nm for fine tunning

   if ((dist_to_go < 45) and (dist_to_go > 44)) {compute_TAEM_guidance_targets();}



I will push that tonight, I had another full Lakebed run with the new longitude and the recalculation forced, it worked well.
Those 360 ° HAC with high entry speed are good for testing the resilience of the AP.

I had a last try, I let the AP until 13000 feet. A bit of hac overshoot, but it came back into the hac and I was almost on the path at the end.
Tell me how it went for you with the new stuff :)
GinGin
 
Posts: 1580
Joined: Wed Jul 05, 2017 11:41 am
Location: Paris
Callsign: Gingin

Re: Space Shuttle - Entry/TAEM rework

Postby Thorsten » Sun Feb 21, 2021 8:59 am

if it was aiming to the Hac center instead of WP1, never saw that.


My suspicion would have been that it assumed a zero radius, but if you found it anyway...

For the TAEM, I think it was the fact to have not reloaded the target after TEAM.


When does the logic now compute the target? It used to be computed at transition to TAEM phase - so can you remind me why we have to reload the target manually? (I suppose I'll get there in the flight manual eventually, but I'm still at de-orbit burn complications...)
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Next

Return to Spaceflight

Who is online

Users browsing this forum: No registered users and 2 guests