Jump to content

[1.4][1.7.7] GravityTurn continued - Automated Efficient Launches


AndyMt
 Share

Recommended Posts

11 hours ago, Gordon Dry said:

The most disturbing fact to jump into the discussion "pro-or-contra MechJeb" is that GravityTurn plays with the engine's throttle ...

Try that in a RO/RSS/RealFuels install ... you know what I mean.

You just cannot "throttle around as you like" with a rocket engine ...

Somehow I didn't notice that @Craze was using RSS, or else I'd have informed them that GT isn't designed to work with RSS, and won't work with it.

So.. Having never used RSS myself, and considering all the requests we regularly get about using GT on RSS, what exactly IS the normal procedure in RSS for nailing a desired orbital altitude and circularising? (guessing here, but what I'm picturing is something along the lines of what real launches do.. which is to say, a combination of continuous-burn-to-orbit, and careful attitude/AoA control, along with a caca-ton of math to work out the correct launch profile in advance.)

Because it sounds to me like someone would have to develop a custom gravity-turn program specifically for RSS.

11 hours ago, Gordon Dry said:

You just cannot "throttle around as you like" with a rocket engine ...

Except that clearly, with some you can, and others you can't. For instance, the shuttles used to throttle down for passing Max-Q, then throttled up again, and SpaceX merlin engines do the same thing. Granted, the shuttles also had SRBs which were designed for a period of lower lower thrust burning around Max-Q, but still, the main orbiter engines were capable of being throttled.

In Apollo, the first and second stage flight profiles were designed so that the centre engines would shut off early, as a way of "throttling back" at Max-Q (stage one) and so that g-forces would not become excessively high for the crew (stage two).

Is RSS versatile enough to permit such things, or is it just a clumsy "no throttles allowed" blanket ban?

Edited by JAFO
Link to comment
Share on other sites

8 minutes ago, JAFO said:

what exactly IS the normal procedure in RSS for nailing a desired orbital altitude and circularising?

Something like this
https://github.com/lamont-granquist/MechJeb2/releases

8 minutes ago, JAFO said:

throttle down for passing Max-Q

Yes, but that is pre-defined. There usually is a low thrust setting, like kinda 80% on some engines, 20% on others.

8 minutes ago, JAFO said:

Is RSS versatile enough to permit such things, or is it just a clumsy "no throttles allowed" blanket ban?

Not RSS, but RO is very strict with the engine configs - also there is the ullage.

Edited by Gordon Dry
Link to comment
Share on other sites

4 minutes ago, Gordon Dry said:

:o Seriously? The go-to solution for all those reality-obsessed players is, "use MJ to get to orbit" and "overbuild your 1st/2nd stages to compensate for MJ's clunky inefficient version of a gravity turn"??? :0.0:

Ok.. that's seriously funny to picture.

 

9 minutes ago, Gordon Dry said:

Yes, but that is pre-defined. There usually is a low thrust setting, like kinda 80% on some engines, 20% on others.

So is that a one-shot thing too, or is it something that is repeatable enough that GT might be able to (albeit badly, perhaps) try and work with?

Link to comment
Share on other sites

1 minute ago, JAFO said:

Seriously? The go-to solution for all those reality-obsessed players is, "use MJ to get to orbit"

This is not the "normal" MechJeb ... this is the one with PEG guidance :cool:

2 minutes ago, JAFO said:

reality-obsessed

Wait, I have to add something ...

Do you think that even in the 1960s the rockets were steered by hand? So this is the reason why I don't have ANY problem in using MechJeb.

@JAFO that version even got a wiki:
https://github.com/lamont-granquist/MechJeb2/wiki

Link to comment
Share on other sites

8 minutes ago, Gordon Dry said:

This is not the "normal" MechJeb ... this is the one with PEG guidance :cool:

Ok.. first time I'd heard of that.. so I had to do a quick read of the wiki page for it..

Interesting.. looks complex enough with all those Pitch Program and other parameters, etc, that I think I'm finally starting to understand why so many (apparently) RSS noobs come here, hoping for something simpler.

 

17 minutes ago, Gordon Dry said:

Wait, I have to add something ...

Do you think that even in the 1960s the rockets were steered by hand? So this is the reason why I don't have ANY problem in using MechJeb.

Heck no.. I use MJ myself, mostly for GT circularisations and sometimes Munar landings (I suck at landing twitchy/tippy landers), and I justify it using exactly the same argument.

My comment was made before I realised this was a more realistically designed implementation of ascent guidance. I was imagining people who are sticklers for imitating real-life being forced to compensate for "stock MJ's" launch guidance inefficiencies.

30 minutes ago, Gordon Dry said:

Thanks, but I already found it..

Link to comment
Share on other sites

20 hours ago, Gordon Dry said:

The most disturbing fact to jump into the discussion "pro-or-contra MechJeb" is that GravityTurn plays with the engine's throttle ...

Try that in a RO/RSS/RealFuels install ... you know what I mean.

You just cannot "throttle around as you like" with a rocket engine ...

You can set a min throttle

Edit

Buy was i lyte :)

Edited by danielboro
Link to comment
Share on other sites

On 10/9/2018 at 12:15 PM, danielboro said:

use the lock button to prevent a value from changing.

improving a guess and getting a bad guess is normal
it will guess in the oposit direction the next time 

I do so, but then there is no sense in this function. Just setting my values and taking off.

Link to comment
Share on other sites

16 hours ago, Gordon Dry said:

Do you think that even in the 1960s the rockets were steered by hand? So this is the reason why I don't have ANY problem in using MechJeb.

Over the entire history of rocketry, computer guidance, whether on-board or on the ground by radio control, is the norm for rockets.  Next is inherent guidance by fins or spin stabilization without any other means.  Manual guidance is very very rare.

Edited by Jacke
Link to comment
Share on other sites

23 hours ago, JAFO said:

Heck no.. I use MJ myself, mostly for GT circularisations and sometimes Munar landings (I suck at landing twitchy/tippy landers), and I justify it using exactly the same argument.

I don't use RSS and found the way MJ interacts with GT weird. E.g. I build my rockets in such a way that GT puts me in a still suborbital trajectory at target AP, so the last ascent stage can just be ditched without leaving space junk and then do the circularisation manually using my "service module" (or similar) engine. Optimally that means Apo at LKO and Periapsis at nearly 0(technically higher would be better, but KSP doesn't reliably aerobrake on non-active vessels, if at all) with several minutes to Apo in the end. I found it very efficient(no more than 3400dv) and easy to plan my launches like this.

The first time I used MJ, GT handed off without break and used the leftover fuel in my ascent stage to start circularisation. There's probably a way to tell MJ not to do it this way but I haven't checked. Sure, I could design my rockets differently so that I stage before GT is done, but that's a lot more complex, I'd rather not. Now if I had an action group that fired when GT is done... :D

Edited by Zah
Link to comment
Share on other sites

On 10/9/2018 at 5:52 PM, Zah said:

I have a feature request: could GT be set to fire an action group(compatible with Action Groups Extended) when it's done?
I know it already hands off to MJ (which I don't want to use), but regardless being able to further automate a launch with an action group would be really useful.

That's an excellent idea! That way one could deploy solar panels, antennas etc...

Link to comment
Share on other sites

Regarding the comments about RSS and RO: I think I have to cleary state in the opening post that GT does NOT really work with RO in particular. You can get it to work with RSS scale, but you would have to fiddle with the values manually and not use the "improve guess" function. But with RO it does not work because of not restartable engines or not throttable engines etc.

So: GT is mainly intended to be used with stock scale. It might work with scaled up systems, but then you have to tweak the values manually. I often play in 2k or 3.2k and there I just reduce the angle value "a little" and increase the target height "a little, too".

Link to comment
Share on other sites

29 minutes ago, Frostiken said:

Why is this mod causing my rockets to insanely roll left and right, back and forth, at absurd oscillations?

We'd need to see your rockets to be sure, but to me it sounds like the control devices (engines and/or fins) are far from the command pod (which is actually pretty normal) and the rocket is particularly susceptible to flexing (which is common but undesired) and/or you happen to hit some resonance.

If your rocket flexes, the command pod is facing (say) 5 degrees to the left so the engines compensate by steering 5 degrees to the right. But the rocket is actually going the right way, so they cause you to go off course. Then your command pod waggles all the way over and now it looks like you're going 10 degrees to the left but in fact you're "only" going 5 degrees to the left. So the engines aim 10 degrees right, and you can guess the rest.

If it's not that, then like I said a picture would do wonders.

Link to comment
Share on other sites

4 hours ago, Frostiken said:

Why is this mod causing my rockets to insanely roll left and right, back and forth, at absurd oscillations?

4 hours ago, 5thHorseman said:

We'd need to see your rockets to be sure, [...] a picture would do wonders.

For that matter, a link to a copy of the .craft file could be helpful, too...

Link to comment
Share on other sites

On 10/11/2018 at 5:41 AM, JAFO said:

Somehow I didn't notice that @Craze was using RSS, or else I'd have informed them that GT isn't designed to work with RSS, and won't work with it.

So.. Having never used RSS myself, and considering all the requests we regularly get about using GT on RSS, what exactly IS the normal procedure in RSS for nailing a desired orbital altitude and circularising? (guessing here, but what I'm picturing is something along the lines of what real launches do.. which is to say, a combination of continuous-burn-to-orbit, and careful attitude/AoA control, along with a caca-ton of math to work out the correct launch profile in advance.)

Because it sounds to me like someone would have to develop a custom gravity-turn program specifically for RSS. 

Except that clearly, with some you can, and others you can't. For instance, the shuttles used to throttle down for passing Max-Q, then throttled up again, and SpaceX merlin engines do the same thing. Granted, the shuttles also had SRBs which were designed for a period of lower lower thrust burning around Max-Q, but still, the main orbiter engines were capable of being throttled.

In Apollo, the first and second stage flight profiles were designed so that the centre engines would shut off early, as a way of "throttling back" at Max-Q (stage one) and so that g-forces would not become excessively high for the crew (stage two).

Is RSS versatile enough to permit such things, or is it just a clumsy "no throttles allowed" blanket ban?

I usually set the initial angle from 1 to 5 degrees. Altitude is 150 - 400 kilometers. The pressure of the transition surface - orbit 20. The other parameters do not change.

And Yes, I know that GT is not intended for RSS. It it is more efficient and economical than mechjeb. Although there is much to improve.

Link to comment
Share on other sites

  • 2 weeks later...
On 9/24/2018 at 4:37 AM, Weavetoucher said:

On this note it would be good if you could run a few sims using KRASH and have it remember and port to your real launch, unless it does this already? That would help retain the immersion while "realistically" getting best computer estimate.

This is a good question. Does KRASH contribute to GT profile data for each iteration?

A bit of self-starting question-answering: Yup! KRASH totally works for building up GT simulation data. Role-Play revert all you like, people! ;)

I'm also echoing @tg626's question -- they always seem to jettison *slightly* too early, and I worry they're going to damage something/cause extra drag from their contents suddenly being exposed.

Edited by Beetlecat
Link to comment
Share on other sites

Hmm... I think the pressure unit is Pascals (KPa or Pa I definitely don't recall), but I don't remember if it's for dynamic or static/absolute pressure... I THINK it dynamic pressure. (Which I think makes more sense: if it was static/absolute, that would just be the equivalent of a fixed altitude in practice.)

Link to comment
Share on other sites

This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...