Orbiter-Forum  

Go Back   Orbiter-Forum > Orbiter Addons > OrbitHangar Addons & Comments
Register Blogs Orbinauts List Social Groups FAQ Projects Mark Forums Read

OrbitHangar Addons & Comments Addons uploaded at Orbithangar.com will automatically generate a new thread in this forum for comments. The thread link will also be included on the addons page.

Reply
 
Thread Tools
Old 04-21-2018, 10:32 PM   #16
BrianJ
thing
Default

Quote:
Originally Posted by Marijn View Post
 It looks to me that the actual 1st stage flips around much more quickly after separation than the simulation in Orbiter and it did deploy its gridfins rather early as well. The whole move seemed very aggresive.
Yes, I've noticed that before. I need stronger RCS, I guess. I also think SpaceX ignite the centre engine at low throttle for extra torque before actual boostback attitude is reached.

Also, SpaceX may use a different boostback trajectory to my add-on - I need to go back and look at previous launch video to check 1st stage attitude during boostback burn(where possible). For TESS, the stage looked to be pointed slightly down(lower vert.vel, higher horizontal vel), whereas my add-on points slightly up(higher vert.vel, lower horizontal vel).

Plenty of other factors will make the add-on behave different to real-life.
Bear in mind, I don't really know what I'm doing when it comes to coding and control algorithms, so don't expect too much!

Cheers,
BrianJ
BrianJ is offline   Reply With Quote
Old 04-22-2018, 01:22 PM   #17
Kyle
Armchair Astronaut
 
Kyle's Avatar
Default

The first stage of F9 started doing "fast flips" for boostback burns starting with CRS-9 in order to utilize less propellant to return back to the launch side/ASDS (a 30 second maneuver for a rocket going ~2 km/s will cover a lot of distance compared to a 5-10 second maneuver). I also think SpaceX does ignite the center engine during this maneuver, too. You can see it happening on the incredible Iridium-4 launch @ the 1:30 mark. I wish SpaceX was as open with their launch profiles as NASA is.


Last edited by Kyle; 04-22-2018 at 01:24 PM.
Kyle is offline   Reply With Quote
Old 04-22-2018, 08:52 PM   #18
Marijn
Orbinaut
 
Marijn's Avatar
Default

Quote:
Originally Posted by BrianJ View Post
 For TESS, the stage looked to be pointed slightly down
For sure it did. The 1st stage seemed to do a clean and simple backward 180 degrees flip without any banking. Starting the flip only moments after separation at about 30 degrees up, it finishes less than 20 seconds later 30 degrees below horizon level. I'am guessing this a bit, but if an aft-facing camera clearly points above the horizon, it's front-facing end must point below the horizon. The SpaceX video shows @6:57 how the aft-facing camera of the 1st stage almost kills rotation having the 2nd stage centered.

So that would mean some thrust of the boostback is pointed down, adding to the entry speed. Only little later they extend the gridfins, inducing drag as early as possible. I wonder what the maximum altitude was of the 1st stage this time. I would say that it was rather low, perhaps not much more than 100-110km.

I am not sure what to make of this. Perhaps they want the landing spot to be as close as possible to the harbor when having the luxury of excess fuel. Cost-wise, that would make sense to me.

Quote:
Originally Posted by BrianJ View Post
 Bear in mind, I don't really know what I'm doing when it comes to coding and control algorithms, so don't expect too much!
You're doing a brillant job! While nobody else is doing it better, you are the expert. The F9 add-on pitches up 50 degrees and then banks it's way around. To me, that seems more complicated than doing a half-loop.
Marijn is offline   Reply With Quote
Old 04-23-2018, 02:21 PM   #19
Messierhunter
Orbinaut
 
Messierhunter's Avatar
Default

Quote:
Originally Posted by Marijn View Post
 So that would mean some thrust of the boostback is pointed down, adding to the entry speed.
Correct me if I'm wrong, but this doesn't seem to be necessarily true. If the stage still has positive vertical velocity then thrusting along a vector that will reduce the vertical velocity will ultimately act to reduce the entry speed. A higher apogee equals more vertical velocity at atmospheric entry, reducing vertical velocity therefore reduces the apogee, so as long as the vertical velocity does not start going negative during the boostback burn it should act to minimize entry velocity.

Anecdotally, I use a similar strategy to minimize entry velocity for re-usable rocket I made in KSP. In fact if I don't act quickly to kill the vertical velocity after stage separation, the apogee will be too high and results in catastrophic aerodynamic loads at entry. The trajectory is a very aggressive loft so that I can fly the first stage back to landing and then switch focus to the second stage to insert it into orbit when it hits apogee without having to use any automation mods. It's not directly comparable to SpaceX of course, but I think the principle at work is the same.
Messierhunter is offline   Reply With Quote
Old 04-23-2018, 02:39 PM   #20
Marijn
Orbinaut
 
Marijn's Avatar
Default

Quote:
Originally Posted by Messierhunter View Post
 Correct me if I'm wrong, but this doesn't seem to be necessarily true.
I think you are correct. It probably does not add to the entry speed. But the nose of the rocket does point below the horizon, so the apoapsis of the 1st stage is lowered during the boostback I would say. Maybe it barely leaves the upper atmosphere. That would explain why they deployed the grid-fins so early.
Marijn is offline   Reply With Quote
Old 04-24-2018, 10:03 PM   #21
BrianJ
thing
Default

Quote:
Originally Posted by Marijn View Post
 It probably does not add to the entry speed. But the nose of the rocket does point below the horizon, so the apoapsis of the 1st stage is lowered during the boostback I would say.
What it certainly does do is to reduce time-until-impact, so you need a higher horizontal velocity component to reach your target.

As ever, we don't know what the primary constraint is on the 1st stage boostback trajectory and reentry conditions. My add-on uses a "minimum-energy" ballistic trajectory, but SpaceX might use a "minimum-required-impulse", or perhaps flight-angle at reentry is important, etc. Go figure!

Cheers,
BrianJ
BrianJ is offline   Reply With Quote
Old 05-13-2018, 09:15 PM   #22
jgrillo2002
Conservative Pioneer
 
jgrillo2002's Avatar
Default

OK this trajectory is not easy. How does one accomplish this? TransX? or Lagrange MFD
jgrillo2002 is offline   Reply With Quote
Old 05-14-2018, 10:55 AM   #23
BrianJ
thing
Default

Hi,
not a straight forward lunar transfer, for sure.
You can use IMFD Surface Launch or TransX to plan the launch for lunar intercept in about 26days, or just launch at historical time/azimuth.

IMFD Delta Velocity or TransX for the burn to push ApA to 275000km, keep the ApA on the line of nodes of Earth/Moon orbit intersection.

Then you have 3.5 phasing orbits before lunar flyby.
You need to make correction burns at each(or some) perigee, for lunar intercept.
My method:
Use IMFD Course - Target Intercept
Target moon at the intercept date. Set burn ignition time to the perigee before lunar intercept (3.5-4days before intercept) (this will move as you make correction burns and your orbit changes, so adjust each time).
Make prograde correction burn at each perigee to reduce dV required by IMFD Course-Target Intercept by 50% (watch the dV required as you make the burn).

At the final perigee before lunar flyby, dV required by IMFD Course-target Intercept should be minimal and you can use IMFD to make the final intercept burn.

Once I have a lunar intercept set up, I switch over to TransX to plan the flyby for 37deg inc. and nominal PeA (can't remember the figure now ~117000km) orbit.

At first Earth perigee after flyby you can adjust the ApA to get a nice stable orbit for 1-2 years (check using IMFD Map or LagrangeMFD plot).

Hope this helps :-)
BrianJ
BrianJ is offline   Reply With Quote
Old 05-16-2018, 01:41 AM   #24
jgrillo2002
Conservative Pioneer
 
jgrillo2002's Avatar
Default

That is still a little complicated. the problem is burning at the right angle at the right time. Not to mention the correct altitude to keep up with the moon. What altitudes did you use for the 2nd and third burns?
jgrillo2002 is offline   Reply With Quote
Old 05-16-2018, 11:39 AM   #25
BrianJ
thing
Default

Quote:
Originally Posted by jgrillo2002 View Post
 That is still a little complicated. the problem is burning at the right angle at the right time. Not to mention the correct altitude to keep up with the moon. What altitudes did you use for the 2nd and third burns?
Hi,
do you mean 2nd and 3rd "phasing orbit" burns? IIRC, during the 1st phasing orbit, perigee is pulled up(by Moon gravity) to ~1000km alt, it doesn't vary much from there, so 2nd and 3rd burns are at ~1000km.

As per previous post, my "trick" is to set up IMFD Course - Target Intercept for a TLI burn close to final perigee (~4days before Moon intercept).
At each phasing orbit perigee, make a prograde burn to reduce dV required for TLI burn to minimum.


Cheers,
Brian
BrianJ is offline   Reply With Quote
Old 07-19-2018, 06:29 AM   #26
dropsofjupiter34
Orbinaut
Default

Several sources say that the total mass of TESS is 362 kilograms rather than 365 kilograms, and I'm working on changing the dry mass to match that number.

However, I'm not an expert in Visual Studio. What's the easy way to figure this out?
dropsofjupiter34 is offline   Reply With Quote
Old 07-19-2018, 09:23 AM   #27
BrianJ
thing
Default

Quote:
Originally Posted by dropsofjupiter34 View Post
 Several sources say that the total mass of TESS is 362 kilograms rather than 365 kilograms, and I'm working on changing the dry mass to match that number.

However, I'm not an expert in Visual Studio. What's the easy way to figure this out?
If you're recompiling from the source code, just change the EMP_MASS value, it's up at the top of tess.cpp. You'll need to link OrbiterSoundSDK40.lib also.



Or an easy way to do it, I think, is to add a Mass parameter to the .cfg.
Config/ Vessels/ Tess/ tess.cfg
Code:
Mass = 317
I think that will override the .dll mass.
BrianJ is offline   Reply With Quote
Thanked by:
Old 07-19-2018, 06:38 PM   #28
dropsofjupiter34
Orbinaut
Default

Thanks, BrianJ. Adding the mass parameter saved me a bunch of time.
dropsofjupiter34 is offline   Reply With Quote
Old 08-18-2018, 10:12 PM   #29
Ajaja
Orbinaut
Default

An interesting video tutorial for GMAT (General Mission Analysis Tool) where the TESS trajectory calculation is described.
The trajectory is slightly different (it was 2014) but there is the same idea.


And GMAT 2018a already has an optimizator (Yukon) which can be used instead of proprietary VF13ad or Matlab to calculate such trajectories, and it works fine even with these TESS tutorials scripts.
An amaizing tool for mission planning
Ajaja is offline   Reply With Quote
Thanked by:
Old 08-19-2018, 11:46 AM   #30
Nicholas Kang
Spaceflight News Reporter
 
Nicholas Kang's Avatar

Default

Or you can use the exact same software that JPL uses.

MONTE: https://montepy.jpl.nasa.gov/



I used GMAT once but I remember certain source codes and optimization algorithms are not made public and require commercial license to use them.

On the other hand, though MONTE is not open-source, all APIs are fully available, including the optimization algorithms. The source code is written in C++ and wrapped with Python 2.

MONTE only runs on Red Hat-based Linux OSes (RHEL and CentOS) and is not released for individuals but for academic research and commercial usage only, i.e. you have to apply it through your college/university/company.

EDIT: Just recall that TESS is a Goddard mission, not JPL's. So Goddard's in-house GMAT should give a result closer to prediction. But MONTE is worth trying though.

Last edited by Nicholas Kang; 08-19-2018 at 12:00 PM. Reason: Added video
Nicholas Kang is offline   Reply With Quote
Reply

  Orbiter-Forum > Orbiter Addons > OrbitHangar Addons & Comments


Thread Tools

Posting Rules
BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
Forum Jump


All times are GMT. The time now is 10:32 PM.

Quick Links Need Help?


About Us | Rules & Guidelines | TOS Policy | Privacy Policy

Orbiter-Forum is hosted at Orbithangar.com
Powered by vBulletin® Version 3.8.6
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
Copyright 2007 - 2017, Orbiter-Forum.com. All rights reserved.