Orbiter-Forum  

Go Back   Orbiter-Forum > Orbiter Addons > Addon Support & Bugs > Addon Developer Forums > Space Shuttle Ultra
Register Blogs Orbinauts List Social Groups FAQ Projects Mark Forums Read

Space Shuttle Ultra Support & development threads for Space Shuttle Ultra addon.

Reply
 
Thread Tools
Old 03-22-2019, 09:08 AM   #211
indy91
Addon Developer
Default

Quote:
Originally Posted by GLS View Post
 Plus, the pad being in the wrong place, means that even launching at the correct time will very likely not work perfectly.
Quickly ran the numbers for geodetic vs. geocentric latitude for LC-39, targeting 51.6 inclination. I might have simplified the calculation too much, but I am getting a LAN difference of 0.183. So definitely significant. You would need to adjust the launch time by 44 seconds to account for that.

NASSP of course has to account for that as well and we are using the historically flown targeting polynomials (inclination/LAN as a function of launch azimuth) for the lunar missions with the Saturn V. And the Iterative Guidance Mode really just does a little bit of yaw steering to correct for the wrong launchpad latitude, not very noticable, so it's not really a performance issue.

Quote:
Plus, for 28.45 orbits, launching on time almost certainly will not work perfectly as the pad isn't where it should, so the vehicle does a dog leg to correct the inclination, and doing this with no control over LAN... how knows exactly what happens.
I've tried launching STS-82 to the HST. The first attempt was at the actual launch time, which happened at the beginning of the launch window, so without yaw steering that gave me a very high relative inclination, 4 or so. Then I tried the middle of the launch window which resulted in about 1. Then I got tired of it, because I always had to adjust the HST state vector in the launch scenario etc. So I simply moved the Shuttle in the right orbit. I'm sure by manually tweaking the launch time I could have gotten better results, at least good enough so that the HST can be reached without cheating. Just as an example.

And I'll look into the time definition issue, maybe I can figure out how much of a problem it really is. It's also relevant for NASSP of course, but it hasn't caused any noticable problems so far.
indy91 is offline   Reply With Quote
Thanked by:
Old 03-22-2019, 09:18 AM   #212
Urwumpe
Not funny anymore
 
Urwumpe's Avatar

Default

Orbiter uses MJD, which means, leap seconds are not correctly included.

There is the \Delta T the difference between UT and TT, those could explain it (UT is based on earths real rotation, TT assumes a fixed day length like Orbiter). Its about 60 seconds between the origin of the Julian Date (1858) and today.
Urwumpe is online now   Reply With Quote
Old 03-22-2019, 09:27 AM   #213
GLS
Addon Developer
 
GLS's Avatar
Default

Quote:
Originally Posted by indy91 View Post
 Quickly ran the numbers for geodetic vs. geocentric latitude for LC-39, targeting 51.6 inclination. I might have simplified the calculation too much, but I am getting a LAN difference of 0.183. So definitely significant. You would need to adjust the launch time by 44 seconds to account for that.
If we had an ascent guidance that targeted an actual plane, that would be almost a no-issue, as 44s is pennies in a 10m launch window. I think working ascent is what I'll do first for SSU 6.0... just hope it doesn't take 1 year like the landing did...


Quote:
Originally Posted by indy91 View Post
 I've tried launching STS-82 to the HST. The first attempt was at the actual launch time, which happened at the beginning of the launch window, so without yaw steering that gave me a very high relative inclination, 4 or so. Then I tried the middle of the launch window which resulted in about 1. Then I got tired of it, because I always had to adjust the HST state vector in the launch scenario etc. So I simply moved the Shuttle in the right orbit. I'm sure by manually tweaking the launch time I could have gotten better results, at least good enough so that the HST can be reached without cheating. Just as an example.
The way SSU gets to a 28.45 orbit is by heading off to 90 and waiting for vehicle the latitude to drop below the target orbit inclination, then the "normal" orbital inclination logic kicks in... pretty much an open-loop solution, which is not very flexible or good in terms of performance, but it allows the vehicle to get to the correct orbit of a target.


Quote:
Originally Posted by indy91 View Post
 And I'll look into the time definition issue, maybe I can figure out how much of a problem it really is. It's also relevant for NASSP of course, but it hasn't caused any noticable problems so far.
I'm sure the difference is just a few seconds, but at 7km/s, the distances accumulate fast. Plus there could be 3 time systems at play here: Orbiter, TLEs and "wrist clock".
GLS is offline   Reply With Quote
Old 03-22-2019, 10:07 AM   #214
indy91
Addon Developer
Default

Quote:
Originally Posted by GLS View Post
 If we had an ascent guidance that targeted an actual plane, that would be almost a no-issue, as 44s is pennies in a 10m launch window. I think working ascent is what I'll do first for SSU 6.0... just hope it doesn't take 1 year like the landing did...
Many years ago I played around with the Shuttle ascent targeting, I even got the logic working for using throttling to intercept a target in orbit (for Design Reference Mission 3B). That code is not really in a very usable state, but if you are lacking documentation, I can at least point you to the useful documents.

And once SSU has that stuff implemented I can implement some launch window targeting in the FDO MFD.
indy91 is offline   Reply With Quote
Old 03-22-2019, 10:27 AM   #215
GLS
Addon Developer
 
GLS's Avatar
Default

Quote:
Originally Posted by indy91 View Post
 Many years ago I played around with the Shuttle ascent targeting, I even got the logic working for using throttling to intercept a target in orbit (for Design Reference Mission 3B). That code is not really in a very usable state, but if you are lacking documentation, I can at least point you to the useful documents.
DRM-3B, a.k.a., total insanity.
I've some pdfs with some general info on ascent guidance, and the coordinate systems, but I haven't sat down with them to know if they are enough or not. But, even if just the general scheme of things is presented, we can at least start to walk in the right direction.

Quote:
Originally Posted by indy91 View Post
 And once SSU has that stuff implemented I can implement some launch window targeting in the FDO MFD.
Yeah, that way you could estimate a nominal SV for post MECO (that SSU would have to meet), and the flight plan could be calculated on the ground pre-launch.
GLS is offline   Reply With Quote
Old 03-22-2019, 10:32 AM   #216
Wolf
Donator
 
Wolf's Avatar
Default

About adjusting for the errors and glitches above: with Non Spherical disabled it is just a matter of changing the launch time. Problem is how do we do that when Non Spherical is enabled. I wonder if there is a way to predict Nodal Regression at a future time (hence the magnitude of my desired RInc at MECO) if my chasing orbit keeps changing from launch to Rendezvous time.
Wolf is offline   Reply With Quote
Old 03-22-2019, 10:48 AM   #217
indy91
Addon Developer
Default

Quote:
Originally Posted by GLS View Post
 Yeah, that way you could estimate a nominal SV for post MECO (that SSU would have to meet), and the flight plan could be calculated on the ground pre-launch.
Yes. By the way, is the MPS dump after ET sep propulsive in SSU? That would be important to know for that. The rendezvous launch window targeting that was used generated a post-MPS dump state vector to then be used in all the orbital calculations.

Quote:
Originally Posted by Wolf View Post
 About adjusting for the errors and glitches above: with Non Spherical disabled it is just a matter of changing the launch time. Problem is how do we do that when Non Spherical is enabled. I wonder if there is a way to predict Nodal Regression at a future time (hence the magnitude of my desired RInc at MECO) if my chasing orbit keeps changing from launch to Rendezvous time.
It's difficult to do that in any manual way, because that difference in LAN that you want after the ascent is not constant, but will depend on the phase angle as well. Large phase angle needs a low orbit for the Shuttle to catch up and the lower orbit has a stronger effect on the differential nodal regression. Small phase angle will need a higher Shuttle orbit to avoid catching up too quickly and then the difference in nodal regression isn't very large. So far these numbers have (roughly) varied from 0.1 to 0.5 for me.

And inclination is also not constant, but varies over one orbit (with a mean inclination that always stays constant). So, I think, the insertion inclination you would want is the osculating inclination that the target has at the same point in orbit. Luckily that effect is smaller than than the LAN difference, but if you are simply using the mean inclination of the target then that might still cause NPC maneuvers that are 0-20 ft/s larger than they should be.

All of this can only really be solved by some prelaunch targeting that would give you the right launch time and inclination (and later the LAN).

Last edited by indy91; 03-22-2019 at 10:51 AM.
indy91 is offline   Reply With Quote
Old 03-22-2019, 10:59 AM   #218
GLS
Addon Developer
 
GLS's Avatar
Default

Quote:
Originally Posted by indy91 View Post
 Yes. By the way, is the MPS dump after ET sep propulsive in SSU? That would be important to know for that. The rendezvous launch window targeting that was used generated a post-MPS dump state vector to then be used in all the orbital calculations.
Yes it is. From memory I think it should be +/- 11fps... but then they started using it to pitch the vehicle up to put the ET in the overhead windows, so if that is made the dV is not all in one direction...
Plus, some missions had the +X RCS burn, so there's another variable to add.
GLS is offline   Reply With Quote
Old 03-22-2019, 03:44 PM   #219
Tim13
Orbinaut
Default

Quote:
Originally Posted by indy91 View Post
 
And once SSU has that stuff implemented I can implement some launch window targeting in the FDO MFD.
That statement almost made me wet myself.

Your MFD has made the SSU go from an A+ to an A+++

It's greatly appreciated!

Tim
Tim13 is offline   Reply With Quote
Old 03-22-2019, 04:52 PM   #220
GLS
Addon Developer
 
GLS's Avatar
Default

Quote:
Originally Posted by Tim13 View Post
 That statement almost made me wet myself.

Your MFD has made the SSU go from an A+ to an A+++

It's greatly appreciated!

Tim
Not yet.... but let's hope a rewrite of the ascent guidance/FCS doesn't take as much as the entry did...
GLS is offline   Reply With Quote
Old 04-02-2019, 09:42 AM   #221
indy91
Addon Developer
Default

One small update. I haven't had much time working on the MFD, but I am making some progress working on the Launch Window/Launch Targeting Processor. It consists of those two parts, a launch window calculation which calculates the liftoff time based on in-plane or phase angle and then a launch targeting part which calculates a more precise, zero yaw steering preferred liftoff time and also launch targeting parameters (inclination and descending node angle).

I have various, incomplete sources on this processor, which make the work interesting and challenging. There definitely is a direct heritage from Gemini (launching to intercept another Gemini spacecraft or an Agena) to the planned, but never flown AS-207/208 mission (two Saturn IBs launching a CSM and LM respectively, flying an Apollo 9 kind of mission) to Skylab and then to Shuttle. Other than the OMP the LWP/LTP will probably not end up working just like the real thing, as it was at the end of the Shuttle program. There are some numbers you would have to generate externally (the descending node angle for a specific inclination and time into launch window) and I don't fully understand how that would be used anyway. After all the documentation I have, other than the FDO Console Handbook, are all from 1980 or earlier. But it will still be very close to what the FDOs used on launch day to generate the exact launch window and targeting parameters.

While SSU doesn't do any second stage yaw steering yet it should probably be possible to intercept a specific orbit by using the right inclination and a biased launch time. The launch time is calculated with an analytical bias anyway, the Console Handbook uses the input value of 5m40s before the in-plane time. And for a set of specific inclination, launch weight and insertion parameters the necessary time bias should be fairly constant, it will just need a bunch of trial and error. My initial tests suggest that the 5m40s value needs to be changed by about 20 seconds. One LTP output is a post-MPS dump state vector, which can be used in the OMP to check if the orbital plane is right and any NPC maneuver in the rendezvous plan is near zero. That is especially relevant for the non-spherical gravity.

Parallel to this I'll implement this processor for NASSP as well and will try to fly a Skylab rendezvous profile with it. The simulated LVDC can do insertion into a specific orbital plane very well.
indy91 is offline   Reply With Quote
Thanked by:
Old 04-04-2019, 06:18 PM   #222
Donamy
Beta Tester


Default

Found this, in case you're interested.
Attached Thumbnails
Click image for larger version

Name:	RendvTraj.jpg
Views:	51
Size:	85.7 KB
ID:	16550   Click image for larger version

Name:	DM_1 approach.jpg
Views:	49
Size:	90.0 KB
ID:	16551  
Donamy is offline   Reply With Quote
Thanked by:
Old 04-08-2019, 10:10 AM   #223
indy91
Addon Developer
Default

Ok, the initial version of the Launch Window Processor is almost ready. Here is the workflow to use it, which may still change a bit, also based on feedback. I'm not very talented in making MFDs user friendly.

New button in the main menu:



There are several inputs screens for the LWP, although most of the values will not have to be changed every time you use the LWP:



-TGT is the target vehicle, whose state vector is used as an input to the LWP.
-LS can be used to input a specific launch pad (39A or 39B currently), so that the coordinates are automatically loaded. Default is LC-39A.
-LAT can be used to overwrite the launchpad latitude (if you want to launch to the HST you might have to bias the latitude southwards in order to get the LWP to calculate a solution).
-LNG to overwrite the launchpad longitude.
-YS is the maximum yaw steering the Shuttle can do for the ascent to orbit. Not really relevant for SSU right now and it should probably always stay 14
-PFA is the powered flight arc from launch to insertion. All the values on the right side of this screen are used in an analytical calculation of the Shuttle ascent. This is all the LWP can do, so updated values would need to be supplied for ascents that are very different from a usual one. But it's not super critical to have these values exactly right.
-PFT is the powered flight time from liftoff to insertion. Not MECO I guess, more the decay of all thrust or so.
-RAD is the insertion radius in feet. Might be a bit of an awkward unit to use for SSU, but that's what the FDO Handbook showed in a screenshot of the LWP inputs.
-VEL is the insertion velocity. This and FPA below will need to be adjusted for the specific mission, because it's very different from e.g. a mission to the ISS vs. to the HST.
-FPA is the flight path angle at insertion. All of the values on this page are usual values for a mission to the ISS, I think.

The next page has some rarely changed values. By the way, if you didn't know, pressing MNU will show descriptions of all buttons:



-DTE is the delta time from MECO to ET separation. Probably always 12 seconds.
-DVE is the DV vector of the ET separation maneuver. the DVX component should be either 0 ft/s (no +X translation maneuver or ET photo pitch maneuver) or 5 ft/s (to account for the +X translation maneuver). For DVZ I have seen 4 or 5 ft/s, not sure which value is closer to accurate. This will be the translation maneuver that is done automatically to get away from the ET. +5 ft/s if the attitude is heads down or -5 ft/s if the attitude is heads up.
-DTM is the delta time from MECO to the MPS dump; just like the ET sep maneuver this assumes an impulsive burn. The delta time is probably fairly constant.
-DVM is the DV of the MPS dump. The FDO Console Handbook has numbers that are averages from flight data:
heads down: +9.2, -2.6, -4.2
heads up: +9.7 +0.2 -1.0
-DELNO is the descending node bias that is used to account for differential nodal precession. This should always be 0 with non-spherical gravity disabled. There currently is no way yet to let the LWP calculate the value automatically, so it has to be done by trial and error. Calculate a LWP solution, then run the OMP with the state vector from LWP output and check the size of the NPC maneuver. A very rough estimate for this is phase angle at liftoff divided by 500. So for a phase angle of 300 it will be roughly 0.6 DELNO. This only applies to a FD3 rendezvous, for FD4 it will be more, for FD2 it will be less.
-OPT is the optimal, zero yaw steering delta time from the in-plane time. Launching at exactly the time when the target vehicles orbit is over the launchpad is not optimal because by the time you have left the atmosphere, a few minutes later, the orbit is already far away from the launchpad. This is an empirical derived number depending on time it takes to ascent to orbit etc. The FDO Handbook has -5:40, but I've tested it a couple times and for SSU the best value is -5:23. This will only be exactly for 51.6 inclination and a specific Shuttle weight and insertion conditions but it should be quite close in all cases.
-DTO is the delta time for the opening of the launch window. Launch window to the ISS is usually 10 minutes long, so DTO and DTC are -5 minutes and + 5 minutes. SSU can't do any yaw steering and this is only really used for the calculation of the phase angle right now. Just for information purposes.
-DTC is the delta time of the launch window closing.
-PHA and WRA are two rather unintuitive flags used by the LWP and currently it will only be used in the phase angle calculation. For example, this will decide if a phase angle is 30 or 390. Here is the numbers that are used for this, as per the FDO Handbook:




And here the output page:



The only button is CLC, to perform the calculation. This will populate the numbers on this page and also save a post MPS dump state vector for the OMP to use. Additionally it will overwrite the current launch time in the MFD. On the executive page you can see that the chaser vehicle is now showing "LWP Output":



The only numbers that are currently relevant for SSU are the optimum launch time and the inclination (IMECO) that are displayed on the LWP output page. I guess you will have to write down the numbers and then adjust the inclination in the mission file and the LaunchMJD in the scenario file. I could also add an export button so that all the insertion parameters and the launch MJD are saved in a file where you can copy them over to the mission and scenario files.

So this is how the LWP is going to work initially. There are many features which are currently not implemented or don't have a button yet: insertion at a specific phase angle instead of in-plane, calculating the descending node angle and with that an orbital plane vector for the Shuttle to target. And ascending/descending node launches, which would be relevant for a launch at VAFB. Any initial feedback on the user interface part?
indy91 is offline   Reply With Quote
Thanked by:
Old 04-16-2019, 03:11 PM   #224
indy91
Addon Developer
Default

Initial version of the Launch Window Processor has been released: https://github.com/indy91/Shuttle-FD...ag/0.2.0-alpha
indy91 is offline   Reply With Quote
Thanked by:
Reply

  Orbiter-Forum > Orbiter Addons > Addon Support & Bugs > Addon Developer Forums > Space Shuttle Ultra


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 03:33 PM.

Quick Links Need Help?


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

Orbiter-Forum is hosted at Orbithangar.com
Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2019, vBulletin Solutions Inc.
Copyright 2007 - 2017, Orbiter-Forum.com. All rights reserved.