Project SpaceX SuperHeavy

[BrianJ]

@sberinde I tried an IFT-4 flight using your launch scenario, and it went pretty well.
Launch targeted 150km orbit initially, using autopilot. Then I switched over to IMFD "Base Approach" at ~138km for the last bit of the burn, targeting LZ_IND. It put me in a 230km x -80km orbit. Booster made a good landing at LZ_TX. No correction burns for Starship during flight, had about 100km crossrange error at reentry, made a good landing at LZ_IND, with not too much banking/turns activity.

 

[sberinde]

Hi Brian, thanks for your explanation regarding booster return trajectory. Now I want to discuss about ship reentry profile based on IFT-4 flight. I see your last post about using IMFD, but let's try something else.

From [1] and [2] we have estimates about the orbit (-15km x 213km x 26.8deg) and the landing point (a bit closer, if true). In the attached scenario the ship is right after the MECO. I succeded to match the orbit and the altitude (150km). After fuel dump, if the autopilot is engaged at the entry interface, it will land on the spot (with lot of banking). But if I try to fly manually a "smooth trajectory" (with minimal banking) the ship will overshoot by a lot .

May I ask you about atmospheric flight parameters of the ship? Are they adjustable somehow or I miss something here? Sorry if the question is way off.

[1] Jonathan's Space Report No. 834 draft
[2] Starship Flight 4 DISCUSSION : Starbase TX : NET June 2024

 

[BrianJ]

Hi,
thanks for the IFT4 orbit scenario, that will be useful to me for checking the aerodynamics.

The aerodynamics are hard coded in the .dll module and can only be changed if you edit the C++ code and recompile the .dll module (code included in the add-on, if you wish to modify it). Of course, changing the aerodynamics may well break my reentry and skydive-backflip-landing sequences (more fun!).

I don't know much about aerodyamics! If I can make something that behaves approximately realistcally in Orbiter, I'm happy.

Also, because we don't have accurate or complete information about Starship, or the sophisticated guidance and planning software SpaceX uses, - I find it almost impossible to do things in Orbiter exactly the same as SpaceX.

Having said that, I think it will be useful to use your scenario to compare the reentry profiles.
I'll see if I can get a better match without breaking everything else.

Mustn't lose sight of the original purpose of this vessel - I must get around to trying a Mars reentry and landing autopilot.

Cheers,
Brian

 

[sberinde]

Hi,
Thanks for your answers. I'm not a fan of C++ so I will not touch the code. But I want to help in other ways.

As you said, many parameters should be guessed, like for example drag/lift coefficients. I'm not an expert either. If you can adjust these params by comparing reentry profiles with "real" ones that would be nice. If too much trouble then no problem, we will adjust our simulated flights and land on Mars anyway .

Speaking of simulated flights here is a nice challenge for everyone (attached scenario). I revisited IFT-4 scenario in order to fly the reentry manually up to skydive. You are in Starship right after SECO in an orbit -75km x 213km (not the real one, but close to). Dump the fuel (very little needed) and at entry interface activate fly-by-wire using an AOA angle of 66.0 degrees. You will fly a very smooth profile with almost no banking required and little AOA adjustment near the end. Below 20km in altitude activate reserve fuel and skydive. You will land on the spot .

Notice: if I activate skydive direcly (Ctrl+B) ship will not fly as intended. I had to activate auto reentry instead (U).

 

[barrygolden]

I shared this with Brian but might be time to work on the Plasma . the video was breath taking

 

[BrianJ]

I shared this with Brian but might be time to work on the Plasma . the video was breath taking

I'll try and make the reentry visuals a bit more fun, but its never going to look like the Hollywood CGI

Currently working on improved Boostback fuel calculation, booster landing.

I think I'm going amalgamate the tanker function into one .dll - because maintaining two sets of code has now become a pain.

Cheers,
Brian

 

[sberinde]

This is an annotated recording of an IFT4 scenario using manual reentry. It shows a smooth reentry with a single banking from an orbit very similar to the real one. For this to work, I moved the landing zone a bit to the south to have a larger crossrange distance. The AOA at entry interface is 55deg, close to what I observed from the live broadcast.

Altitude is monotonically decreasing, as in the real flight [1].

My conclusion after this simulation: a lower AOA angle at entry interface and a larger crossrange of landing zone allows a smooth reentry. This is the closest match I get so far for the real flight, except maybe the initial banking for which I have no data from the live broadcast.

[1] Entry Profile for Starship Flight 4

 

[YourHistorian]

Hello everyone!
I am experimenting with starship and deorbiting using BaseSyncMFD's DEO page, and I've been experimenting with different values but I can't find the sweet spot, every thing I've tried as just either kept the starship in orbit or the starship has shorted the landing site, perhaps any of you guys could help. I have kept the reentry angle at 0.3, reentry anticipation at 50 degrees, and reentry altitude at 80, these have gotten me closest to the landing site, if anyone knows a more consistent set for BaseSync it would be greatly appreciated,

 

[sberinde]

From a circular orbit use the parameters:

Altitude (Alt.) 120km Re-entry Angle (ReA) 1.5° Anterior Angle (Ant) 45°

(taken from Starship add-on manual)

Unfortunately BaseSyncMFD is not so useful for suborbital flights (such as recent IFT tests). An alternative is to use IMFD, but personally I don't like it either, because it tries to cancel out the initial crossrange error to the landing site. This will change the insertion orbit by an unnecessary amount.

 

[YourHistorian]

Alright thanks a lot

 

[BrianJ]

Hi,
@YourHistorian re-entry parameters from LEO as per sberinde's post above, should work with IMFD Base Approach or BaseSyncMFD, and set it up for a good reentry. Hope you get down OK!

@sberinde I know IMFD has some "quirks" when dealing with targets 1/2 orbit away (sometimes it wants you to go polar) but the Base Approach has worked well for me for the IFT4 suborbital flight. I don't think it was made with suborbital hops in mind, of course.
Thanks for noting the AoA during IFT4 reentry - from the SpaceX visuals, it starts around 55deg, goes to 70deg during peak heating, then reduces back to ~55deg.
I always kind of assumed AoA would be critical/fixed, but if there is, say, 20deg freedom to vary it, that could be a very nice way to control reentry range! Better than fuel-expensive/uncomfortable s-turns (although the s-turns method is quite robust and forgiving). I might have a play around with that idea.
I also note the long stay (3mins+) at 70km during IFT4 reentry - I wonder if my add-on needs more lift at high Mach numbers? I'll try some experiments.

I recorded the alt vs. time from your IFT4 playback scenario, and compared it to the plot made by (somebody on Reddit? I can't remember who made it! sorry) from the SpaceX coverage. Orbiter IFT4 is the green line.


The add-on comes down a bit quicker, but then it is banking at first. More experiments required

WIP I'm going to make the landing-gear optional, have them as a separate mesh, then I can change 'em more easily when the time comes. Then I'll upload an update.

Cheers,
Brian

 

[sberinde]

Details about Starship entry profile from Elon himself.

Interview w/ Elon Musk under world’s largest rocket, Starship, with post launch follow up!

 

[YourHistorian]

Just one more question, how do you attach the caught starship to the booster?

 

[BrianJ]

Just one more question, how do you attach the caught starship to the booster?

Not possible, I'm afraid.
I tried my best to make that work, but I just can't find a way of coding it without CTD's, bugs, etc.
Maybe I'll have another look at that sometime.
Cheers,
Brian

 

[YourHistorian]

Unfortunate, however I have found another bug with the starship, running with the patch linked above, the tanker starship does not hotstage, yet the booster does hotstage if that makes sense

 

[BrianJ]

Unfortunate, however I have found another bug with the starship, running with the patch linked above, the tanker starship does not hotstage, yet the booster does hotstage if that makes sense

I forgot to mention the patch is only for the Starship, not Tanker.
Full update coming soon.
Cheers,
Brian

 

[BrianJ]

Update 240701

Added integrated "Hotstage ring" to booster.
Tanker functions (payload fuel tank, fuel transfer options) integrated into Starship.
Tanker variant selectable via .cfg file line: TANKER = true
Landing gear now optional, selectable via .cfg file line: LANDINGGEAR = true
Tweaked ascent autopilot engine selection/throttle sequence for hotstaging.
Tweaked booster reentry and landing autopilot.
Booster boostback fuel reserve now calculated "on the fly", option to override estimate removed from control panel.
Tweaked "Mechazilla" target selection (ignores landed vessels).

Probably some other stuff too...
Cheers,
BrianJ

 

[BrianJ]

Update 240805

Added reentry heatshield glow.
Added Mars approach scenario and Mars EDL playback.

Probably the last update before I relocate the forward fins and transfer the Reentry autopilot to an MFD and put together something to handle hyperbolic EDL at Mars and Earth (if I can!).
Cheers,
BrianJ

 

[Gieno]

Hi @BrianJ awsome work appreciate your dedication. Many thanks for your effort. Will Starhip maybe in the future have a Auto De-Orbit function like Crew Dragon?

 

[BrianJ]

Hi @BrianJ awsome work appreciate your dedication. Many thanks for your effort. Will Starhip maybe in the future have a Auto De-Orbit function like Crew Dragon?

Hi Gieno,
thanks, yeah, I think I'm going to try and clone the CrewDragonMFD for Starship. It already has a lot of code I need for the LEO de-orbit function and calculating IMFD settings for hyperbolic reentry. Then I can have fun trying to figure out an autopilot for Mars EDL. And maybe Mars ascent (has anyone tried Mars ascent with Starship yet? Is it possible to get to Mars orbit? I kind of assumed it would be...but I haven't tried yet). Anyway, plenty to think about and play with.
Cheers,
Brian