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

[Paaaad]

1 minute ago, RealKerbal3x said:

I know that, but surely it should perform a more efficient ascent. It didn't pitch beyond 45 degrees during the launch, leaving me to perform a burn of >1km/s after it turned off.

The mod messes with the settings after each launch to try to optimize the accent profile- sometimes it gets worse before it gets better, but it will get better. Assuming the craft has a halfway decent design in the first place naturally.

[RealKerbal3x]

Just now, Paaaad said:

The mod messes with the settings after each launch to try to optimize the accent profile- sometimes it gets worse before it gets better, but it will get better. Assuming the craft has a halfway decent design in the first place naturally.

OK, thanks for your help

[danielboro]

14 hours ago, RealKerbal3x said:

Sorry if this is an incredibly noob-ish question, but how do I actually get the mod to put my craft into a circular orbit? I tried it with the default settings and it just puts me on a suborbital trajectory with a apoapsis of 80 km. Am I doing something wrong or am I just a complete doofus?

by fine tuning the parameter it its possebel to get the  PE above the atmosphere before getting the AP to max

[AVaughan]

17 hours ago, RealKerbal3x said:

I know that, but surely it should perform a more efficient ascent. It didn't pitch beyond 45 degrees during the launch, leaving me to perform a burn of >1km/s after it turned off.

Pure speculation, but is this a craft with a high thrust to weight ratio at launch, and some long burning SRBs?  If so then I'd guess that the gravity turn's default guess isn't pitching over aggressively  enough, and by the time the SRBs burn out, the craft is already heading out of the atmosphere without having pitched over enough.  If so lower your launch thrust to weight ratio.  (I like around 1.3-1.4.  Often gravity turn only needs around 20 m/s to circularise my rockets).

[linuxgurugamer]

20 hours ago, RealKerbal3x said:

Sorry if this is an incredibly noob-ish question, but how do I actually get the mod to put my craft into a circular orbit? I tried it with the default settings and it just puts me on a suborbital trajectory with a apoapsis of 80 km. Am I doing something wrong or am I just a complete doofus

Gravity Turn does not do the circularization of the orbit.   For that, either you have to do it, or if you have MechJeb installed, Gravity Turn will create a maneuver node for Mechjeb to execute 

[ARTOBAN]

how to run a buran with this mod ?

[Spaced Out]

What do the Hold AP time start and finish do and how do they affect the launch, also if I don't want them on how do i turn them off?

[DStaal]

21 minutes ago, Spaced Out said:

What do the Hold AP time start and finish do and how do they affect the launch, also if I don't want them on how do i turn them off?

You don't turn them off.  They are how this mod works.

What this mod does is turn the rocket, and adjust the throttle to hold the time to apoapsis - the flights 'highest point' - steady until that highest point reaches the desired orbit.  The 'start' and 'finish' times are the amount of time at the start of the gravity turn and the end of the gravity turn, respectively.  Typically you want them the same - but if you have rockets with particularly high or low TWR it might be worth adjusting them, or if you have solid rocket boosters you'll either want to adjust them or let them be automatically adjusted.

All this mod does is start the turn, and adjust the throttle to keep the hold AP time steady to whatever is defined until the AP reaches the desired height.  That's the entire launch algorithm.

 

[Cruesoe]

Gravity Turn and Fairings:

I have noticed Gravity Turn will trigger the fairings when at the right pressure.

This seems to only work for Stock fairings, I can't get the same to work when using the mod Procedural Fairings.  Does anyone know how to make GT trigger the ejection of Procedural Fairings?

[Tyko]

I have a feature request if you're still working on this mod.

Here's the problem - When a craft is at it's target time to AP GT moderates the throttle to keep it at that time...Great! The problem comes in when you're at 50 seconds to AP and you stage. In the time it takes for your 2nd stage to fire up you're now below 50 seconds to AP so GT reacts by increasing thrust and, if you're at max thrust, GT starts angling your craft up to try to increase time to AP. This is wasteful and I often find I have to overbuild my second stage with higher TWR than necessary just so GT can "make up" that time to AP lost during staging. I find that my 2nd stages have to be over 1.5 TWR to avoid this issue and that's a lot more than they need if they're just holding 50 seconds to AP. I can zero out staging delays, but that can cause problems when the 2nd stage fires.

Suggested solution - if the lower stage increased thrust in the final seconds before staging it would raise your time to AP. Once you staged the time to AP would drop, but if done right, it wouldn't drop below 50 seconds and you're next stage wouldn't have to be overbuilt to make up that loss. I've tested this manually and it solves the problem if you increase time to AP by about 10-15 seconds above the target number.

Would it be possible for GT to detect when a stage is almost out of fuel and boost the thrust for the last 5 seconds or so? Maybe make that burn time configurable because it will probably vary depending on your craft.

Edited September 12, 2018 by Tyko

[danielboro]

7 hours ago, Tyko said:

I have a feature request if you're still working on this mod.

Here's the problem - When a craft is at it's target time to AP GT moderates the throttle to keep it at that time...Great! The problem comes in when you're at 50 seconds to AP and you stage. In the time it takes for your 2nd stage to fire up you're now below 50 seconds to AP so GT reacts by increasing thrust and, if you're at max thrust, GT starts angling your craft up to try to increase time to AP. This is wasteful and I often find I have to overbuild my second stage with higher TWR than necessary just so GT can "make up" that time to AP lost during staging. I find that my 2nd stages have to be over 1.5 TWR to avoid this issue and that's a lot more than they need if they're just holding 50 seconds to AP. I can zero out staging delays, but that can cause problems when the 2nd stage fires.

Suggested solution - if the lower stage increased thrust in the final seconds before staging it would raise your time to AP. Once you staged the time to AP would drop, but if done right, it wouldn't drop below 50 seconds and you're next stage wouldn't have to be overbuilt to make up that loss. I've tested this manually and it solves the problem if you increase time to AP by about 10-15 seconds above the target number.

Would it be possible for GT to detect when a stage is almost out of fuel and boost the thrust for the last 5 seconds or so? Maybe make that burn time configurable because it will probably vary depending on your craft.

i recommend using 60 as start and 30 as end

this will help prevent/redoes the angling problem

option B is use KOS and write your lunch script

[AVaughan]

10 hours ago, Tyko said:

I have a feature request if you're still working on this mod.

Here's the problem - When a craft is at it's target time to AP GT moderates the throttle to keep it at that time...Great! The problem comes in when you're at 50 seconds to AP and you stage. In the time it takes for your 2nd stage to fire up you're now below 50 seconds to AP so GT reacts by increasing thrust and, if you're at max thrust, GT starts angling your craft up to try to increase time to AP. This is wasteful and I often find I have to overbuild my second stage with higher TWR than necessary just so GT can "make up" that time to AP lost during staging. I find that my 2nd stages have to be over 1.5 TWR to avoid this issue and that's a lot more than they need if they're just holding 50 seconds to AP. I can zero out staging delays, but that can cause problems when the 2nd stage fires.

Suggested solution - if the lower stage increased thrust in the final seconds before staging it would raise your time to AP. Once you staged the time to AP would drop, but if done right, it wouldn't drop below 50 seconds and you're next stage wouldn't have to be overbuilt to make up that loss. I've tested this manually and it solves the problem if you increase time to AP by about 10-15 seconds above the target number.

Would it be possible for GT to detect when a stage is almost out of fuel and boost the thrust for the last 5 seconds or so? Maybe make that burn time configurable because it will probably vary depending on your craft.

I've had similar experiences.  Often I design a rocket that will go to space fine if the first stage burns at 100% throttle, but with gravity turn's default settings the first stage throttles down too much and then the second stage pitches up trying to maintain time to Ap.  I work around this by setting sensitivity to 0.90 at lift-off and then turning it down after the 2nd stage has burnt enough fuel that time to Ap stops shrinking.  That means the time to Ap when the 2nd sage ignites is around 75 seconds or so.  That gives a slightly underpowered 2nd stage (often with a TWR of around 1.0 at ignition) time to burn some fuel before the time to Ap drops low enough that Gravity Turn would start to pitch up.

I'm wondering whether splitting the target Ap time into a user specified range would work (eg 45 to 50 seconds).  If the time to Ap is below the minimum time, then pitch up, if above the max then throttle down, otherwise burn full throttle prograde.  Would probably want to be defined per stage or by altitude or somehow scripted because you probably want a nice tight range when Ap is approaching your target orbit.

Edit: You might be able to tie the tolerance range into the existing provision for fading out of a high time to AP at SRB burnout, by allowing us to set the liftoff min and max time to Ap, and the ending time to AP.  Then interpolate min and max time to AP (or time to Ap at SRB burnout if higher) to end time to Ap, the same as currently.  

Edited September 13, 2018 by AVaughan

[Dunbaratu]

2 hours ago, danielboro said:

option B is use KOS and write your lunch script

[homer]Mmmm.... Lunch script.....gggrrrggggllllllll.[/homer]

 

[danielboro]

59 minutes ago, Steven Mading said:

[homer]Mmmm.... Lunch script.....gggrrrggggllllllll.[/homer]

 

a KOS developer (im not worthy )

[Tyko]

9 hours ago, danielboro said:

i recommend using 60 as start and 30 as end

this will help prevent/redoes the angling problem

This has it's own problems though. If you have a declining time to AP set GT is constantly having to tweak the throttle up and down over and over rapidly. 

Maybe if you could set time to AP by stage so it wasn't varying moment by moment causing GT to have to adjust the thrust so much.

[JAFO]

On 4/5/2018 at 3:38 AM, Gordon Dry said:

personally I don't like the "strategy" of guessing a first launch and then reverting and doing another one - because of the immersion... and of the KSP memory leak that still occurs when reverting too much...

If it's any help, from the immersion perspective, I think of the initial launch and the reverts as the engineers running launch simulations on the KSC's Kray SuperKomputer.

[Weavetoucher]

12 minutes ago, JAFO said:

If it's any help, from the immersion perspective, I think of the initial launch and the reverts as the engineers running launch simulations on the KSC's Kray SuperKomputer.

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.

[Jacke]

2 hours ago, JAFO said:

...on the KSC's Kray SuperKomputer.

...recently upgraded to the new Kray-Kray DualSuperKomputer.  Because everything Kerbal eventually gets kray-kray.

Edited September 24, 2018 by Jacke

[Craze]

I use RSS, but The earth's atmosphere is much thicker than Kerbin's. Mod is constantly trying to set the height of 100 km. But the earth is still the atmosphere!
Frequently bad start button "Improve Guess" still affects the parameters. Increasing the slope so that the ship eventually falls or burns in the atmosphere.

[danielboro]

11 hours ago, Craze said:

I use RSS, but The earth's atmosphere is much thicker than Kerbin's. Mod is constantly trying to set the height of 100 km. But the earth is still the atmosphere!
Frequently bad start button "Improve Guess" still affects the parameters. Increasing the slope so that the ship eventually falls or burns in the atmosphere.

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

Edited October 9, 2018 by danielboro

[JAFO]

15 hours ago, Craze said:

Frequently bad start button "Improve Guess" still affects the parameters. Increasing the slope so that the ship eventually falls or burns in the atmosphere.

As @danielboro indicated, GT is designed to improve its guesses over half a dozen or more launches before settling on the optimal settings. My normal approach is to launch to crash or AP, then revert and launch again, rinse and repeat until GT no longer changes any settings when you click "Improve Guess". (It's important not to revert until you crash or get to AP, so that GT knows how things turned out.)

I tend to think of the series of launches and reverts as the Kerbal engineers running a series of computer simulations before doing a launch "for real".

Edited October 9, 2018 by JAFO

[Zah]

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.

[linuxgurugamer]

42 minutes ago, 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.

@AndyMt  If you want to do this, take a look at Bob's Panic Box to see  how to do it, or ask me for help

[JAFO]

2 hours ago, 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.

Very nice idea!

[Gordon Dry]

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 ...