1.0 Ascent Profile

[A_name]

So up until .90 the rule of thumb seemed to be: 1) ascend at terminal velocity until 8-10 kms; 2) begin "gravity turn", keep turning gradually until craft is pointing at horizon by 30-50 km; 3) cut engines at desired Ap an coast.

Now in 1.0 I have no idea what I'm doing. What is the ideal ascent profile?

[Anquietas314]

I don't know if it's ideal (or even close), but so far this works pretty well for me:

1) launch vertically for the first 1 km or so to get some initial airflow on your winglets (you want these!)

2) turn over to about 85 degrees, let it stabilize, then disable SAS.

3) do nothing except staging the rest of the way until AP >= 75km - with my rockets I haven't even needed to touch the controls to make this work

4) cut engines, coast to AP, circularize as normal

I have had some issues with overheating though, so I'd recommend dialing down the thrust compared to 0.90 designs

EDIT: some screenshots of a design this works rather well for:

VAB:

(slightly messed up) turn to 85 / disable SAS:

letting it ride the prograde vector all on its own (tweaking throttle to control speed, but no touching QWEASD:

later in the ascent:

Edited April 29, 2015 by armagheddonsgw

[EtherDragon]

I have been experimenting with this a lot actually, since release - here is what I learned:

For most rockets that we are used to, they are actually over powered. They have too much TWR and go too fast.

Note: This works best with a rocket that has some kind of tail-fin installed!

0. Pre-flight Checklist: Enable SAS (T-Key). Enable Full Throttle (Z-Key).

1. Launch vertically to 1000m or so.

2. At 1000m, pitch your rocket over to 10 degrees off of vertical.

3. Once stable at 10 degrees of vertical, disable SAS (T-Key).

3a. Your tail fins will keep your rocket pointed in the right direction.

NOTE: Adjust throttle as needed to avoid atmospheric stress. Throttle back if you see the following signs: Supersonic air streams (they are white) or super heated atmosphere (red).

4. Let your rocket naturally pitch over to 45 degrees (do not use any WASD control input!). Your goal is to let your rocket hit 45 degrees at about 15km.

4a. If you are too high, throttle back, let gravity do it's job and help your rocket turn.

5. Once you reach 25km, Full Throttle, hold nothing back unless you see atmospheric stress.

6. Push Apoapsis to your desired height (I usually aim for 72km).

7. Coast to apoapsis with your rocket pointed prograde for the least dragon.

8. Done! (Circularize as normal)

Here is a Video where I demonstrate this in both the Kerbal-X (Stock Sandbox Rocket) and a typical rocket that a Career Mode player could build with Tier-3 Nodes unlocked:

[Francesco]

With 1.0 it seems important to keep the angle of attack low while flying through the atmosphere.

Below 20 km, I try to keep the AoA below 10°, actually more like 5° - which means staying inside or on the edge of the prograde vector's circle.

I begin the turn at 5/6/7 km.

And yes, a lower TWR is okay now, compared to the past.

[Geschosskopf]

I did some experimenting. You can reach 100km LKO for less than 3000m/s. I'm still zeroing in on the optimum profile but it appears to be in the neirghborhood of what worked for FAR in the pre-1.0 days.

1. You want a turn that starts out slow and then speeds up as you get higher, so that you're flying more or less level by the time you reach 35km.

2. You want a fairly high initial TWR (between 1.5 and 2.0, but probably not any higher) to get through the thicker air faster during the more gradual part of your turn. It's OK to go to a lower TWR (1.3 - 1.5) for upper stages once you get some speed and altitude.

In case you didn't know, MechJeb can be used as a pilot assist on launch, not just an autopilot. In the ascent guidance section, the very top button turns on a target icon on the navball. Edit the ascent profile (the very bottom button) to something like starting the turn at 1km, using some curve shape > 50% (I like 65%), and ending the turn at 35km. Then launch without engaging the autopilot, and keep your nose on the target on the navball. Or just let MJ fly the thing at least a few times so you can see what it does.

[SRV Ron]

With 1.0 it seems important to keep the angle of attack low while flying through the atmosphere.

Below 20 km, I try to keep the AoA below 10°, actually more like 5° - which means staying inside or on the edge of the prograde vector's circle.

I begin the turn at 5/6/7 km.

And yes, a lower TWR is okay now, compared to the past.

Most designs will now flip in the lower atmosphere if you try to turn too quickly especially if you are not using a steerable engine. Do the turn gradually watching your speed until you reach the upper atmosphere and you will be fine.

[Ape_Descendant]

Scott Manley just put up a really good video about this which i saw yesterday:

[A_name]

Thank you all for your great replies. I have a few follow up questions...

I have had some issues with overheating though, so I'd recommend dialing down the thrust compared to 0.90 designs

How much do you dial down the thrust? At what altitude would it be safe to go full throttle? We are trying to achieve orbital velocities after all.

NOTE: Adjust throttle as needed to avoid atmospheric stress. Throttle back if you see the following signs: Supersonic air streams (they are white) or super heated atmosphere (red).

So we're supposed to avoid the white air streams and red reentry effects? This seems a bit off to me, as my first test rockets were reaching Mach 1 (and thus getting the white air streams) at about 1 or 2 km (new cool sonic boom sound effect BTW). A bit related to my previous question in this post, but are we supposed to keep shooting for similar ascent speeds as with the pre-1.0 suposphere? Real rockets exceed Mach 1 very soon after launch.

In case you didn't know, MechJeb can be used as a pilot assist on launch, not just an autopilot. In the ascent guidance section, the very top button turns on a target icon on the navball. Edit the ascent profile (the very bottom button) to something like starting the turn at 1km, using some curve shape > 50% (I like 65%), and ending the turn at 35km. Then launch without engaging the autopilot, and keep your nose on the target on the navball. Or just let MJ fly the thing at least a few times so you can see what it does.

Does MechJeb know how to 1.0? In other words, I would have thought its ascent guidance would be useless with the new atmo, is this not the case?

Thanks again!

[Northlight]

I like to restrickt the 1st stage engine(s) so that my TWR starts at ~1.2

That gives me a nice and steady ascent, and I haven't had any troubles with overheating on the way up. It also frees you from having to control the throttle to focus on steering.

To find how much the engine(s) need I use this formula:

Thrust = TWR · weight_of_rocket · g

'g' is the standard gravity: 9.80665

The wheight of the rocket is found in the new engineer-panel.

Example:

1.2 · 18 · 9.80665 = 211.8236

This tells you that you'll need 212 thrust to get TWR of 1.2 with an 18t rocket.

Then when you find the required Thrust for your 1st stage, you can restrict the engines to that power.

[Sirrobert]

You have to use a real gravity turn now.

Starting from 1km up, SLOWLY turn sideways and you'll get to orbit

[Anquietas314]

How much do you dial down the thrust? At what altitude would it be safe to go full throttle? We are trying to achieve orbital velocities after all.

Obviously that will depend on your rocket design , but as a rule of thumb staying just below the red flames seems to be about right - even then it's risky, since I've just encountered an issue where I was below the threshold for that but still somehow managed to overheat the capsule on my rocket, which has a probe body above the capsule. Long story short, the capsule is hotter than the probe's maximum temperature and shortly after leaving the atmosphere *poof* because thermal conduction .

I'm not sure if the mach effects are still just a drop-in replacement for the heating effects below a set altitude though, but I haven't had issues with "reasonable" launch profiles that break the sound barrier early on. Also remember orbital velocity in KSP is far below what it is in real life, so launch speeds don't necessarily map neatly. Staying below 3-500m/s until ~5-10km and below ~1.2km/s until the 25km mark seems to be working out fairly well, although my rockets have a little too much TWR for the new aero. This is little more than learning through trial and error though, so by all means take my advice with a grain of salt.

[EtherDragon]

So we're supposed to avoid the white air streams and red reentry effects? This seems a bit off to me, as my first test rockets were reaching Mach 1 (and thus getting the white air streams) at about 1 or 2 km (new cool sonic boom sound effect BTW). A bit related to my previous question in this post, but are we supposed to keep shooting for similar ascent speeds as with the pre-1.0 suposphere? Real rockets exceed Mach 1 very soon after launch.

Thanks again!

If you don't have tail-fins, I would say definitely avoid low altitude super-sonic flight. The reason is the tendency for the air resistance to overwhelm your controls and cause your rocket to flip. If you have tail-fins, go for a fast and energetic launch - but that also means having to use control-inputs to steer your rocket.

So far, I'm really liking the lower thrust / authentic gravity turn (rather than the control-input-heavy gravity-turn-like ascent profiles we loved in 0.90) ascent profiles.

[Kerbart]

(...)Does MechJeb know how to 1.0? In other words, I would have thought its ascent guidance would be useless with the new atmo, is this not the case?

Problems are caused by an angle of attack that is too high, causing the rocket to flip. Mechjeb can follow the ascent profile very accurate, keeping the AOA very small. Mechjeb throttles speed if you go too fast, based on data it's getting from KSP, so that might or might not be correct

What's left is the profile itself and the existing profile will still get you in orbit. Maybe not an optimal ascent (yet) though.

[Wolf Baginski]

I am running a from-scratch science-mode operation and the first fins/winglets available seem more like the flights on an arrow in effect. So it's hard to make a faster start to the gravity turn. And the battery is so vital to maintain attitude during reentry. SAS-mode isn't bad for power use, but you need more battery than you might think.

Starting the gravity turn low-down, lower than v0.90, is good.

Don't be frightened of going supersonic, but it may be inefficient. It depends on the drag you get and how high you are.

MaxQ for ApolloXVI was at Mach 1.7, 14.3 km, but Kerbin is smaller. That was still the first stage. The MechJeb standard for starting the gravity turn was somewhere between 7km and 8km

[THX1138]

Are terminal velocities the same as they were?

[A_name]

Thank you for your super helpful, in-depth responses. I love you all.

Are terminal velocities the same as they were?

Given there's a new aerodynamic model, yes.

Edit: I'll leave this unanswered for now because this seems to be a topic that's still up for discussion as people continue to experiment.