1.3 Rocket Ascent Profile and Gravity Turn

[wumpus]

1 hour ago, max_creative said:

Most of my rockets are about 3/4 as tall as the VAB. Does this still work? (I assume not)

Probably.  What is the average height and how does it compare to the cross sectional area of the base?  It all comes down to cross sectional area/mass.  If you have a ton of small side boosters, then the average height isn't much.  If you have a bunch of side boosters that are at least 1/2 as tall as the VAB, then you still have a tall rocket.

Assuming a tall rocket, your best [aero considertaions only] bet is to go horizontal immediately (I'm guessing under 45 degrees before 10km).  In reality, wrangling such a beast just isn't happening (I'm guessing you will hit 45 degrees no earlier than 30km).  Don't forget that while you can ignore aero losses with your tall rocket, you can't ignore aero heating.  Don't blow the thing up.

If you build something like Wackjob's "rocket cages", you might manage to tilt the thing sideways and launch at an angle (this will require a ton of launch clamps, and a miracle to miss them all during the launch).  My guess is this won't work at all and you will be stuck with aerodynamically inefficient launches 
(use a lot of AV-T1 winglets).  Launching at an angle works great with SRBs in early missions (and fairly hefty TWRs).

[max_creative]

I launch, when I get to the top of the atmosphere the boosters are gone and it's sorta easy to turn, so I turn 45 degrees. Once the apopsis hits 100 km I turn it horizontal and just burn for orbit. I know it's horribly inefficient but other wise the rocket tends to flip or explode if I turn sooner.

[max_creative]

Also I am a terrible pilot.

[JacktoseIntolerant]

42 minutes ago, max_creative said:

I know it's horribly inefficient but other wise the rocket tends to flip or explode if I turn sooner.

Gentler turn (don't let your nose dot get outside the green prograde circle), more fins farther back.

There's a virtuous cycle to exploit. If you turn more efficiently (your path is frankly horrible), you'll need a lot less fuel, you can build a smaller and more balanced rocket, you'll have an easier time controlling it, and fewer explosions.

[max_creative]

But I like explosions!

Just sometimes I want something in orbit.

[JacktoseIntolerant]

1 hour ago, max_creative said:

But I like explosions!

Oh, then make sure you're turning aggressively right when you stage.

[max_creative]

Just now, JacktoseIntolerant said:

Oh, then make sure you're turning aggressively right when you stage.

Ok. But what if I want to put a... 500 ton one piece space station in a 100 X 100 km orbit without explosions? (I know... No explosions?!? He must be crazy! )

[max_creative]

Also are my rockets ment to be on fire?

[max_creative]

And exploding?

 

Jeb: YES!!! ROCKETS GETTING BLOWN SKY HIGH ARE AWESOME!!!!!!!

[wumpus]

Best guess, turn slowly (try to keep the directional heading on the low edge of the prograde setting.  Keep gently turning the rocket over until right before it flips over and explodes, then stop.  Hopefully the rocket will have some sort of workable gravity turn before going supersonic (at which point you lose control and it tends to just go straight).  Going at about a 45 degree turn *through the atmosphere* is important.  By the time you get out of the atmosphere, you should be tilted nearly 90 degrees sideways (following prograde is more efficient, but it will likely put you "efficiently" into a higher orbit.  Eat the steering losses to get the orbit you want).

13 hours ago, max_creative said:

Also are my rockets ment to be on fire?

Sure.  As long as you aren't getting those heating indicators (little temperature bars on various parts) turning reddish on you, it is just a graphics effect.  If you somehow manage a low-angle "turn 45 degrees before 10km" that I suggested would make sense for such a rocket (and have TWR between 1.8 and 2.0 to take full advantage of it) you might have to worry about your rocket exploding due to heating.  Otherwise heating should only be a danger if your engines (and flames) are near fuel tanks.

[max_creative]

Umm... Ok... Mine are on fire and spontaneously combusting on the pad is that good? So far none of my rockets have been on fire going up (execpt when I put Jeb on a kickback SRB). Maybe because the payloads are like 500 tons... 

[JacktoseIntolerant]

For those interesting in optimizing a little more, reddit user only_to_downvote has some good science posted here.

[mrclucks]

Great guide!

While while there are a few pretty efficient ways to launch, all users should be aware that these are more of a rule of thumb and should not be afraid of expirementing a little to figure out ways of having their particular rockets to orbit.  

I for one use a starting twr of around 1.2 but more than 1.05, do my slight nudge around 50m/s so I get to 10 degrees at 3km and around 45 degrees at15 km. I also use mech jeb to keep prograde starting from my kick(the nudge that starts the gravity turn) so it's efficient and also because it moves much more smoothly than stock SAS. 

All of that being said, I do sometimes muck about and launch with twr of 2 or greater with a much steeper angle, but that's only because I have come to understand how typical rockets behave under different conditions in ksp

[Box of Stardust]

I try this, but it seems like lift forces just drag the rocket back up to vertical and then it spins out of control...

[precision77]

Are you guys sure you mean 10° at 50m/s, or you actually mean 80° (please check the navball picture: http://s1227.photobucket.com/user/zarakon2/media/ksp-navball.jpg.html). 10° is almost pointing towards the horizon which seem pretty radical to me (not a "slight nudge" as you describe it), rockets are no spaceplanes :).

Anyway, I've been trying to get a relatively small (still early in game) rocket to orbit but I've never managed to get to 75x75 orbit with less than 4000 deltaV. Even with mechjeb. I have good TWR (>1.3 at any stage) and 4 fins at the back. I followed the procedure in the first post but no luck. According to the dV maps I've seen 3500dV should be possible. What am I missing?

 

[Warzouz]

@precision77, yes it's 80° on the navball. It's 10° from your actual bearing.

[Noobton]

On 4/4/2016 at 8:18 AM, precision77 said:

Anyway, I've been trying to get a relatively small (still early in game) rocket to orbit but I've never managed to get to 75x75 orbit with less than 4000 deltaV. Even with mechjeb.

Are you checking the atmospheric deltav? there's usually a lot of dv "lost" on the first 20000m that used to mess my calculations

[precision77]

I mean  the vacuum dV from Kerbal engineer or MechJeb statistics. I am starting with 7100 and in orbit i have around 3000 m/s left. I got a little better at launching in the past 2 days but still a long shot from 3500.

Edited April 7, 2016 by precision77

[peewee69]

Many thanks! This was so helpful!

[Warzouz]

On 7/4/2016 at 4:29 PM, precision77 said:

I mean  the vacuum dV from Kerbal engineer or MechJeb statistics. I am starting with 7100 and in orbit i have around 3000 m/s left. I got a little better at launching in the past 2 days but still a long shot from 3500.

Did you have post a screenshot of your rocket ? What flight plan do you use ? What engines do you use ? 4000m/s is ultra strange.

• My "perfect" launch is 3150 m/s VAC
• Regular launch is 3200m/s VAC
• Missed launch is 3400m/s (or unstreamlined payload).
• I even have a experimental rocket that goes to LKO with 2850m/s VAC (but it has no payload and no real application : it has TWR = 6 on the launchpad)

I usually don't use SAS before 30 or 40 km, and I rarely hit any key (only light yaw or thrust adjustment).

Keep in mind that lighter rockets are harder to fly (1.25m). They usually have more drag, lesser ASL ISP engines (especially at low tech). That may explain your difficulties.

EDIT : Basic flight plan (for more detailed version, see OP)

• Don't activate SAS, take-off, full thrust
• At 50 m/s yaw 5 to 10° right (gently)
• Never go away from prograde circle
• Prograde should cross 45° at 10km (I like 8km with streamlined payload)
• At 25km activate SAS, and switch to prograde in orbit mode
• Switch to map view and cut engines when AP is at 75km or 80km
• You should be on a shallow 10 to 25° trajectory (orbit mode)

The most important part is under 5km.

Recommended TWR = 1.4 to 1.6. It your TWR is too high, you'll have to force the gravity turn and loose fuel. Further more, you'll be to straight and have a high cost orbital insertion. If you have a too low TWR, You'll loose a lot of fuel fighting gravity and lower atmosphere.

Edited April 11, 2016 by Warzouz

[precision77]

Thanks for the additional tips Warzous. I didn't think of the low efficiency of the engines - swivel for launch and at high altitude terrier. By now I got down to 3650 dV VAC so I am satisfied :). Are any of those crafts anywhere on the forum?

• Don't activate SAS, take-off, full thrust -> I take off with SAS until I do the turn at 50 m/s. Then it's off until the vessel is stable - with some fins up to 20km
• Never go away from prograde circle -> this I have a problem with . The circe falls towards the horizon because I lower my thrust so I don't get too fast 
• Prograde should cross 45° at 10km (I like 8km with streamlined payload) -> ok, something else to work on :). But won't the rocket lose to much due to drag losses since it will be going fast? What is your speed at this point?
• You should be on a shallow 10 to 25° trajectory (orbit mode) - I get there at less than 10°

 

On 4/11/2016 at 2:49 PM, Warzouz said:

• I even have a experimental rocket that goes to LKO with 2850m/s VAC (but it has no payload and no real application : it has TWR = 6 on the launchpad)

 

Edited April 14, 2016 by precision77

[Warzouz]

15 hours ago, precision77 said:

Thanks for the additional tips Warzous. I didn't think of the low efficiency of the engines - swivel for launch and at high altitude terrier. By now I got down to 3650 dV VAC so I am satisfied :). Are any of those crafts anywhere on the forum?

• Don't activate SAS, take-off, full thrust -> I take off with SAS until I do the turn at 50 m/s. Then it's off until the vessel is stable - with some fins up to 20km
• Never go away from prograde circle -> this I have a problem with . The circe falls towards the horizon because I lower my thrust so I don't get too fast 
• Prograde should cross 45° at 10km (I like 8km with streamlined payload) -> ok, something else to work on :). But won't the rocket lose to much due to drag losses since it will be going fast? What is your speed at this point?
• You should be on a shallow 10 to 25° trajectory (orbit mode) - I get there at less than 10°

Well, you can activate SAS it your rocket isn't perfectly balanced. I must say I hate unbalance. I use KER to tweak it so even without any gimbal, or torque, to rocket will go straight up.

If your prograde vector is going too low, that mean your TWR is too low. You loose a more dV to gravity, and if you have to burn sideway, you loose even more fuel to consinus loss (burning away from prograde) and you stay longer in lower atmosphere so you drag loss will be higher. Use KER to check you TWR for all you stage. 1.5 VAC on the launchpad is a good value. Finally, if you have to burn away from prograde, you expose yourselft to more drag that may flip your rocket and more wobble fighting.

I only go away from prograde when my rocket "slide" out of 90°, I need to correct that. But that usually happen around 25 to 30 km where drag is less of an issue.

As for speed at 45°, I don't know really, As a rule of thumb, when I have to "force" my rocket so it can be at 45° around 10km, it either I didn't do a big enough initial turn or I go too fast. With a 1.4 to 1.6 TWR on launchpad, your speed should be nice so every thing depends on your initial turn. I can't giove you a precise value, it depends on your rocket structure. Don't hesitate to revert to test different ascent.

Finally, if your angle is less than 10° before orbital insertion, that mean you did a very shallow ascent. That mean you've been longer it atmosphere. Test a bit straighter ascent, maybe you drag loss would be less.

I think MechJeb can display gravity loss, drag loss and  possibly cosinus loss during ascent.

In any cases, your Delta-V to LKO should be orbital speed (2300m/s) + drag loss + gravity loss + cosinus loss.

 

You can check my SSTO recoverable rocket launchers. It's quite well documented. But beware that was mainly designed for 1.0.4. In 1.0.5 you may loose your engines during reentry. But the ascent will go nicely. They are mostly fuelled up to 3400m/s, but that is needed to deorbit and do a power landing.

 

[grosser_Salat]

If you tend to flip your rocket, when steering in the atmosphere, I would like to ad one more tip: Activate Precision controles by hitting the ALT-button. Your controlebars should become blue instead of red.

I really like this guide. I allway tried to mimic the gravity by steering in a shallow arc, I really have to try it without.

But I also got a question: There are of course different ISP-values for atmosphere and vacuum. In which altitude do I have to switch between them on Kerbin (and how about Earth-RSS) to calculate the real amount of DeltaV I have / need? It feel like we are reaching vacuumvalues on 10km allready on Kerbin.

And a follow-up question: Is this relevant enough for the ascendpath? Should I go straight up in the beginning, if my rocket has 1.5 times better ISP in vacuum?

Edited April 15, 2016 by grosser_Salat
grammar

[H2O.]

I would like to point out that my Kassandra (link in the signature) perform gravity turn by herself in a stock install. Launch it once to give you a ballpark idea of what a good turn looks like.

[Warzouz]

18 minutes ago, grosser_Salat said:

...

And a follow-up question: Is this relevant enough for the ascendpath? Should I go straight up in the beginning, if my rocket has 1.5 times better ISP in vacuum?

On the launchpad, you should use engines which ASL ISP is neared VAC values. You should avoir Terrier, Rhino, Skipper, Poodle if you can, those are better fit for pure vacuum propulsion. For that matter, the Mammoth is a very good engine (but expensive) for take-off.

If you go straight up, you'll pay for gravity loss much longer. The only reason you should want to go straighter (not straight up)  is if you payload is incredibly unstreamlined. THEN, you adapt your launch stage (TWR, Delta-V...). Ideally, you would want to go horizontal as soon as possible, but this would require a very high TWR (to avoid cosinus loss), but in atmosphere your would get a lot of drag loss in addition to heat explosion. Ascent is a matter of balance.

In general, don't bother about your TWR after take off. Once you cross 45°, it can nearly have any (reasonable) value. A TWR > 1 is more comfortable, though.

I recommend training with test payloads until you get the ascent profile you're easy with (TWR, Staging, first nudge, altitude of 45°, angle of insertion). Then decline this setup to your various payload weight.

Keep your launchers designs as sub-assemblies and register them with proper payload capacity (for example : 20T launcher, 3500m/s).