best gravity turn on rss

[king of nowhere]

I ask those with experience in real solar system how the gravity turn goes.

in the stock game, you generally want to turn very fast, starting already around 50 m/s, being down at a 45° angle already at 500 m/s. that because in stock you orbit at 2200 m/s, you want to be horizontal at that speed.

in real solar system, at 2200 m/s you still miss over 5 km/s to orbit. I had experiences where I did start angling down early, and I had to pull up later.

on the other hand, I am now trying to go straight up until around 15 km altitude, and while this enables a proper gravity turn afterwards, going straight up at 700 m/s doesn't feel right. it can't be optimized, can it?

and orbital launches in rss last way too long for me to want to go by trial and error.

any good tip on when I should start turning?

[fleventeen]

Hello,

Please confirm, do you also have RO installed, or some other form of balance tweak in addition to RSS?

Edit: Also if you can provide some characteristics of your rocket, it might help others help you here. You have much slimmer margins in RSS, and the ascent profile will change depending on your TWR, number of stages, aero characteristics, etc.

Edited April 15, 2022 by fleventeen

[jimmymcgoochie]

Have you tried using MechJeb's ascent guidance in Primer Vector Guidance mode? As far as I know it's only available in the dev version of MechJeb, which requires a bit of CKAN know-how to download (Settings > CKAN settings > under metadata repositories click 'New' then add MechJeb2-dev) but it's also the recommended way of launching when playing with RSS and RO; I don't see any reason why it wouldn't work just as well with another parts balancing mod such as SMURFF.

If you're trying to fly ascents by hand, remember that orbital velocity around Earth is about 7800m/s rather than the 2200m/s of Kerbin, and the atmosphere is twice as high (140km versus 70km). Flying straight up will quickly rack up gravity losses as you need to spend about 10m/s of delta-V per second just to counteract gravity, so you want to start turning as soon as possible- try ascending to 50m/s vertically, then pitching over 5 degrees and then holding surface prograde; with a higher TWR, pitch more aggressively to gain horizontal velocity as fast as possible and to avoid pushing your apoapsis too high, with a lower TWR pitch more gradually so you don't start coming back down before you've gained enough horizontal velocity and burn up in the atmosphere.

 

[king of nowhere]

2 hours ago, fleventeen said:

Hello,

Please confirm, do you also have RO installed, or some other form of balance tweak in addition to RSS?

Edit: Also if you can provide some characteristics of your rocket it might help people help you here. You have much slimmer margins in RSS, and the ascent profile will change depending on your TWR, number of stages, aero characteristics, etc.

nope, no RO or other tweaks.

after reading all the stuff RO does, i decided some of that was too annoying to keep track of. as for other tweaks, I decided to go rss to increase difficulty, balance tweaks would defeat the purpose.

The rocket is the one here

As it is, it keeps crashing the pc because it's too big and complex. I know that with one less stack of boosters the pc can handle it, but i missed orbit by some 200 m/s. I'm trying to see if I can optimize the ascent profile to recover those 200 m/s.

It's a big asparagus stack, it starts with twr 1.8 and it gradually goes down as more boosters are discarded. Last stage is twr around 1, with 2000 m/s. Everything before that is no less than 1.4

2 hours ago, jimmymcgoochie said:

Have you tried using MechJeb's ascent guidance in Primer Vector Guidance mode? As far as I know it's only available in the dev version of MechJeb, which requires a bit of CKAN know-how to download (Settings > CKAN settings > under metadata repositories click 'New' then add MechJeb2-dev) but it's also the recommended way of launching when playing with RSS and RO; I don't see any reason why it wouldn't work just as well with another parts balancing mod such as SMURFF.

If you're trying to fly ascents by hand, remember that orbital velocity around Earth is about 7800m/s rather than the 2200m/s of Kerbin, and the atmosphere is twice as high (140km versus 70km). Flying straight up will quickly rack up gravity losses as you need to spend about 10m/s of delta-V per second just to counteract gravity, so you want to start turning as soon as possible- try ascending to 50m/s vertically, then pitching over 5 degrees and then holding surface prograde; with a higher TWR, pitch more aggressively to gain horizontal velocity as fast as possible and to avoid pushing your apoapsis too high, with a lower TWR pitch more gradually so you don't start coming back down before you've gained enough horizontal velocity and burn up in the atmosphere.

 

I don't want to use mechjeb or anything. plus, seeing as how mods increase lag and my pc is already crashing, i don't want to add mods just on principle.

Normally, I would have gone for a more horizontal profile. Especially with a rocket so big, aerodinamic losses are less than gravity losses - despite the draggy payload. However, a flatter ascent profile means that I will level up early. This can easily be countered by pointing upward of prograde just enough. But that beast is already very draggy - even on a purely vertical ascent, maxQ is around 30% of weight; pointing above prograde while in the lower atmosphere would increase drag immensely. Which is why I was reluctant to do it.  Also, I've seen real rockets and they go straight up for several km before starting the gravity turn, so I figured it could be the best way.

I will have to try that, though. If nothing else, because I already spent 2 whole afternoons trying to launch that monster and getting the pc crash every single time. Once I even reached the second-to-last stage, wasn't even lagging so hard anymore, and then it still crashed.

[fleventeen]

Hmmm, this sounds like an interesting challenge, and I am  not aware of any other cases of people trying exactly your setup (it seems quite difficult). The stock parts are basically balanced for a 2.5x scale Kerbol system. The RSS configs however rescale to 10x, which really require some form of rebalancing in order to get to orbit reasonably. If you are against the complexity of RO, you might try SMURFF,  which is a more simplified rebalance (it doesn't touch fuel types and such, only fuel mass fractions and engine performance IIRC), but brings the stock parts up to match real world performance in a 10x system.

Edit: If you want an interesting challenge using stock parts with no rebalancing, you can look at JNSQ, which changes the system to 2.5x by default, and doesn't require any rebalancing to reach real world challenge.

Edited April 15, 2022 by fleventeen

[king of nowhere]

I tried to turn earlier. this is how it went

in rss atmosphere ends at 140 km, so I missed orbit again. by maybe 100 m/s. it's 200 m/s better than the previous attempt, where I stopped at 7600 m/s. but most of that, I gained by emptying the water tanks - 65 tons of water that I will resupply with another mission. So, by using a steeper turn, I gained maybe 100 m/s.

However, this gives me faith that it can be done. I can't add another booster without crashing the pc, but I can get to orbit, just barely, with what I have. Also because I had some engine malfunctions all on the same side that caused instability and forced to make some corrections.

So now I'm launching again. I further discarded some monopropellant, and some uranium, and I used longer fuel tanks for the first stage. All together, it should buy me the deltaV that's missing here.

EDIT: Yay! it worked! In orbit with 100 m/s

Edited April 15, 2022 by king of nowhere

[fleventeen]

6 hours ago, king of nowhere said:

EDIT: Yay! it worked! In orbit with 100 m/s

I can't imagine how long that must have taken at reduced physics speed, but congrats.