Jump to content

The launch profile challenge


Recommended Posts

There have been a few posts recently in gameplay questions and tutorials about the best launch profile, and it seems many people have very different ideas.  Like most people I have a standard profile I fly, and I think it works well, so I figured it's time to put my money where my mouth is and see how my launch compares to yours.

The Challenge

Launch the Kerbal X craft from the standard Sandbox game in to an 80km orbit using your normal launch profile, and see how much dV you have left.  Then see if you can do better by playing with the profile a bit. 

Mods

Flight control and information mods like kOS, Mechjeb and KER are fine, but flight dynamics or part performance mods like FAR aren't.  The idea is everyone is using the exact same ship, aerodynamics etc.

Ship alterations

If you're using kOS or MechJeb you're going to need to add a part of two to the ship, that means you'll have a slight weight disadvantage but you're probably flying smoother so tough.  Please don't remove any parts or fuel from the craft, but feel free to apply thrust limits to the engines if you so desire. 

The report

This is part research project as well as a challenge, so please can people report the following

  1. A line or two describing the launch profile
  2. The final altitude when circularised (hopefully about 80km)
  3. The remaining dV once you've circularised (bear in mind you'll have 2 stages still with fuel in them)

The good the bad and the ugly

It would be nice to see people's standard profile, as well as how much you can improve on it, so please post your first attempt as well your best attempt.

 

 

Results

IronMaiden  - 3477 m/s - 45 degrees by 8km

RizzoTheRat - 3456 m/s - 1.77 TWR, 10 degree initial pitch then follow prograde, throttle down to maintain 60s to Ap until altitude hits 50km

CBase - 3456m/s - Mechjeb Turn shape 44%

CBase - 3454 - MechJeb, Classic Ascent profile, Turn start velocity 30 m/s, Turn end altitude 40km, Turn shape 45%

Laie - 3444 m/s - pitch to hit 65 degrees at 200m/s, throttle to hold Ap at 45s, full throttle when 50 degrees above horizon

RizzoTheRat - 3441m/s - 1.77 TWR, 6.5 degree initial pitch then follow prograde, throttle down to maintain 60s to Ap until altitude hits 50km

IronMaiden - 3440 m/s - 45 degrees by 5km

AHHans - 3393 m/s - tilt 5-10 degrees at 30m/s, follow prograde and throttle to maintain TWR=2

5thHorseman -  3361 m/s - 30 degrees by 30km, throttle down to maintain 60s to Ap until Ap hits 60km

Noname_Hero - 3342 - First 2 Swivels disabled. 10 degrees at 100m/s, prograde to 45 degrees then lock angle.  Ap to 75 km then circularise to 80.

Noname_Hero - 3316 - All engines. 10 degrees at 100m/s, prograde to 45 degrees then lock angle.  Ap to 75 km then circularise to 80.

Okhin - 3167 - Full throttle, 15 degrees at 80m/s the follow prograde.

Delay - 3120m/s - 1.4TWR, 5 degrees at 60m/s, 1.8TWR after last booster and 1 when Ap hits 55km

Okhin - 2801 m/s - Launch at TWR=1.3, thottle TWR to 1.4-1.6 below 40km, lock angle when reaching 40 degrees until ap hits 40km then, then 10 degrees every 10km 

 

 

 

 

Le me know if anyone thinks the line I've put in for your entry doesn't reflect your launch profile.

 

 

My first attempt was closer to an ideal gravity turn for the given TWR ending up at 45° by the time I reached 5km, this caused an awful lot of drag and as a result I got to orbit with around 3440m/s remaining. If I could add a fairing I could probably save another couple hundred or so m/s.

My second attempt was a bit less aggressive, the drag was still bad but I ended up with 3477m/s remaining. You can see my profile in the album but basically start pitching over immediately after takeoff then lock prograde, ending up at 75° by 165m/s, 60° by 265m/s and 45° by 8km.

If I were to try to improve on this more I'd take the 1 time drag hit and pitch down off prograde somewhere around 20km to be more aggressive once the atmosphere really starts to thin out.

 

Edited by RizzoTheRat
Link to comment
Share on other sites

My attempt

I’m using kOS so I added a kOS module and because my scripts update over the comms network I added a surface mount antenna.

I like to launch at about 1.8 TWR, so I limited to the 6 boosters to 85% thrust, which gives me a TWR on the launchpad of 1.77 according to KER.

My standard profile pitched it to 6.5 degrees at 30m/s, then followed prograde the rest of the way.  Once the time to Ap got to 60 seconds I throttled back to maintain it at 60s until I cleared 50km, then throttled back up until the Ap hit 80km, and circularised at Ap.

This left me in an 80200m orbit with 1579 fuel and 3441m/s of dV remaining.

After playing with some variation on my usual profile I found that a 10 degree initial pitch was about the most efficient.  I generally design vessels with lower powered upper stages which possibly accounts for this difference. 

This resulted in 81000m orbit with 1596 remaining fuel, which is 3456m/s.

 

Impressed that my standard pitch angle comes out so close to the optimum pitch angle for the rest of my launch profile, but interested to see what other people come up with.

Edited by RizzoTheRat
Link to comment
Share on other sites

I tried to do my "30 at 30" launch which has about a dozen steps but basically I slowly bring the nose down in such a way that when my Ap is 30km, my angle is 30 degrees. I say "tried" because KerbalX is moar boostery than my normal ships (I like 1.3twr) and it was hard to keep it going smoothly down. I also throttled down when my time to Ap hit 1 minute, to keep it near that 1 minute time until about an Ap of 60 or so, at which point I just locked it to prograde and let it go at about 1/4 throttle until the Ap was 81 (oops). At that AP, I did a small burn to circularize at 80x81km.

My ship in that orbit had 2553+761 = 3314m/s left. Consider this my first attempt.

I did it a 2nd time with a much better launch profile and had 2553+807 = 3361 m/s left, which I'm pretty happy with all things considered, so I'll call that my best score. I'm not going to beat a kOS script so hey :D. The craft has 6662m/s of vacuum dV in the VAB so that means I did it in 3301 m/s.

(One question, why fuel AND dV? As the ship is the same for everybody, surely one (the easier one, dV) should be all that's needed)

Edited by 5thHorseman
Link to comment
Share on other sites

 

33 minutes ago, 5thHorseman said:

I say "tried" because KerbalX is moar boostery than my normal ships (I like 1.3twr) and it was hard to keep it going smoothly down.

I'll call that my best score. I'm not going to beat a kOS script so hey :D

(One question, why fuel AND dV? As the ship is the same for everybody, surely one (the easier one, dV) should be all that's needed)

I had a bit of a dilemma about the ship to use, as different ship designs need very different profiles.  If there's a decent takeup on this challenge I'm thinking I'll do a second one with a different ship.  For my usual approach the first stage is fine but it's way overpowered at altitude.

As for beating a kOS script, it gives me an advantage in flying smoothly and repeatable, but it's only as good as the profile I've defined so I'll actually be pretty happy if someone comes up with a better profile.

I'm so used to using KER that I forgot you get a dV readout in the vanilla game now :D, I've just taken it out.

Did you launch at the full 1.9ish TWR or back off to try and get your usual 1.3?

Edited by RizzoTheRat
Link to comment
Share on other sites

Just doing some manual launches. The first few tries were total crap: "Oh, I should have staged that part away ages ago."; "Errr, no. Going horizontal at 12km is not the way to go to space." and so on.

My launch strategy is: tilt over by ca. 5 - 10 degrees at 30 m/s, set to prograde at 100 m/s -- 200 m/s (depending on craft and how much I actually turned), burn full blast until AP of 80 km is reached, coast and circularize.

My first real contender felt way too flat (spent ages at 16 -- 36 km altitude, AP went down from 80 km to 65 km during coast...), but: 1511 fuel and 3384 m/s

The second one was flatter than my usual launches at the start but "felt good" for the second part of the ascent: 1573 fuel and 3440 m/s

The third one was more like my usual launches up to 10 km altitude but then felt fairly steep even for my usually steep launches (with a 1000 m/s circularization burn): 1435 fuel and 3312 m/s

And the forth one was about like the third up to 10 km, but after staging away the first two stages I throttled down (and up after the 3rd stage) to keep the TWR at around 2, requiring a 610 m/s circularization burn: 1522 fuel and 3393 m/s

4 minutes ago, RizzoTheRat said:

For my usual approach the first stage is fine but it's way overpowered at altitude.

Same here!

What I take away from my tries is: throttling down does save dV when the alternative is a too steep ascent. And: any somewhat decent gravity turn will be within 150 m/s or so dV of an optimal one. (But I guess I knew that already before.)

Link to comment
Share on other sites

So, I tend to usually overbuild and build wide rocket (with lots of drag). So I do have an ascent profile which tries to get out of the light blue part of the atmosphere ASA while remaining under Mach 1.

I launched with a TWR of 1.3 (about 2/3 of the throttle), up to the skies, until I reached a TWR of 1.4, and I try to do that when I reach a speed between 80 and 100 m.s (depending on the drag), and then I go 5 degrees east until the Prograde vector catches up, at which point I lock prograde.

I maintain TWR between 1.4 and 1.6, at least while I do significant have atmospheric pressure (so, at least until 40km high). When I hit 40 degrees of inclination, I stability lock until I reach the last phases of the ascent. When apoapsis reached 50 km, I'll pitch 10 degrees more every 10 kilometers of apoapsis (so, 30 degrees at 60km, 20 at 70, 10 at 80 and usually 0 at 90), until I reach 80km apoapsis. Then I stop the engines, and I coast to the maneuver node for inserting myself into orbit.

After the insertion burn, I still have 2801 m.s delta V remaining, and got an apoapsis at 113 km and a periapsis at 79.9. Yeah, I do have eccentricity but I reached orbit :D

I'll try a second launch later today. I usually have two variables on which I work : starting TWR and speed at which I start gravity turning. But yes, there's obviously room for improvement when I see how much dV you have left (I'm 600 short). I'm just used to pushing weird stfuu through the atmosphere, kerbalX is way more aerodynamic than what I usually fly :D

Link to comment
Share on other sites

Sounds interesting! I'll have a go at it!

This will be particularly interesting for me because I know the KerbalX has a ludicrously high TWR in comparison to the 1.3/1.2 I'm always aiming for.

Is Principia allowed? N-body gravitation has negligible influence this close to Kerbin.

Edited by Delay
Link to comment
Share on other sites

Okay, after a few quick attempts I managed to get 3120m/s with a rather complicated profile.

I started with a TWR of around 1.4. On the navball, that's one line after the 2/3 marker. At 60m/s I pitched 5 degrees over and made sure that my TWR remained the same pre-staging to post-staging (so that it's always rising and never "reset").
After the last booster sep I throttled down to a TWR of 1.8, and as soon as Ap was at 55km, down to 1.0. Wait until Ap hit 80km, cut the engine and eventually circularized.

Edited by Delay
Link to comment
Share on other sites

3167 dV. By just going full throttle on the stock Kerbal X (without touching the engine thrust) and kicking 15° when I reach 80m.s, and then locking prograde ans staid full throttle.

The prograde vector did had a nice curve, and I reached 45° after the 40km altitude, but there should be some gain in dV by throttling back a little (I did have some atmospheric plasma on the rocket at some point, ad I think going above 320m.s under 12km cost a lot of dV just pushing through the wall of sound).

And I noticed that the ship tends to go nose down on low TWR (under 1.4, but maybe a bit more) which requires attitude corrections when reaching the 45°, and I suspect it drains dV too. So I tend to think this craft is more efficient at a higher TWM than what I'm used too.

Link to comment
Share on other sites

Nothing near the best here, 3342 dV. About 10° around 100 m/s, switch to prograde, switch to stability around 45°, switch from Surface to Orbit and to prograde around 19km, burn until apoapsis rises to about 75km, circularize/go above 80km.

What I find kind of frustrating is that I did this with first stage Swivels turned off and second stage Swivels restricted to 75%. I did this because Swivels have lower Isp ASL than Mainsails do and Kerbal X has high enough TWR for this to not matter, especially given better profiles here limit thrust too. And it feels kind of frustrating to both haul unused engines up and then drop them to their doom.

Edit:

Forgot to remember the rules. The launch above was not my first attempt, that was a similar profile but all engines on max, I was left with 3316 dV.

Edited by noname_hero
Link to comment
Share on other sites

I hate the vessel, but like the challenge.

The Kerbal-X has a much higher TWR than what I'd build for myself, it's also surprisingly draggy. But it's a workable vessel available to everyone, so here we go.

My usual numbers won't work, the rocket is too alien. So I dialled in different numbers right away.

  1. pitch so that I go through 65°@200m/s
  2. hold time to AP at 45 seconds
  3. when the prograde vector drops to 5° above the horizon, throttle up to make apoapsis.

3444m/s left after circularization.

I took off at full TWR, but locked the control surfaces and turned the vessel by 90° (so pitch goes east).

Link to comment
Share on other sites

3 hours ago, Delay said:

Is Principia allowed? N-body gravitation has negligible influence this close to Kerbin.

Maybe you should launch at night when there's no moon :D  I can't believe it would make a noticeable difference at these altitudes, go for it.

 

 

 

@Laie is that 65 degrees from the horizon nor a 65 degree from vertical initial push?

Interesting that @noname_hero did better with 2 engines completely disabled than using them all, one thing that this has taught me is that the Kerbal X is pretty rubbish design :D Although I believe it dates back to the old aerodynamics model.

A big flaw in this challenge is that the launch profile is strongly linked to the ship design.  I have a plan for a follow-on challenge that will address this though...

Link to comment
Share on other sites

To be honest, the difference I've observed might be due to my limited piloting abilities. I might launch the same design ten times and I'll always end up with different dV and different orbit - I've launched the same rescue ship into LKO enough times to know this for a fact.

I'm not ruling out the possibility my idea to turn those first Swivels off helped, because the Mainsail has 14% better Isp at sea level, but given we're talking about a small improvement in a stage that only has little dV to begin with, I'd bet on me getting lucky and flying a better profile. My guess would be turning those Swivels off might give me say 5 m/s, but I'm more used to rockets with lower first-stage TWR, so what I did seemed better than limiting *all* the side boosters equally.

As for this challenge being flawed, I've got to say I'm happy you came up with it. I know vanity is no virtue, but you sure helped boost my ego. Sure, I'm nowhere near the best. But I'm still a newbie-ish player who follows the tao of over-engineering, so my results are close enough to the good players to make me happy with how I did. In addition to boosting my ego (and maybe I'm not alone in this), shared launch profile ideas for this one vanilla design here might help me (and other still learning players) improve launch profiles for other designs too. And giving the challenge a try can be so short and simple I can take part even with limited play time available.

Link to comment
Share on other sites

3454 m/s dV left

MechJeb, Classic Ascent profile, Turn start velocity 30 m/s, Turn end altitude 40km, Turn shape 45%

Autostage and Autowarp, nothing else

Due to high TWR I actually had to tweak around with Turn shape and altitude.

 

Update: 3456 m/s as well with Turn shape 44%, but could not find a way to shave off one more :rolleyes:

Edited by CBase
Link to comment
Share on other sites

This thing has way higher drag and TWR than I'd ever use which makes this more/less interesting depending on your approach :P

Simply getting rid of the ladder will do wonders for the Δv as you'll be punished less for taking the appropriate gravity turn for the TWR.

My first attempt was closer to an ideal gravity turn for the given TWR ending up at 45° by the time I reached 5km, this caused an awful lot of drag and as a result I got to orbit with around 3440m/s remaining. If I could add a fairing I could probably save another couple hundred or so m/s.

My second attempt was a bit less aggressive, the drag was still bad but I ended up with 3477m/s remaining. You can see my profile in the album but basically start pitching over immediately after takeoff then lock prograde, ending up at 75° by 165m/s, 60° by 265m/s and 45° by 8km.

If I were to try to improve on this more I'd take the 1 time drag hit and pitch down off prograde somewhere around 20km to be more aggressive once the atmosphere really starts to thin out.

Link to comment
Share on other sites

7 hours ago, IronMaiden said:

If I could add a fairing I could probably save another couple hundred or so m/s.

I've been having a bit of a play with modified versions of the KerbalX for my follow-on challenge, and interestingly putting the upper stage in a faring only saved about 65m/s using my standard launch profile in kOS.  Admittedly that added a little bit of weight, but I expected drag to be a bigger factor.

Link to comment
Share on other sites

@IronMaiden: Could it be that you have a script or addon that empties Monoprop from command pods ?

I did several tries with Gravity turns to match your efficiency, but never got better than my classic ascend attempt until I noticed you had more dV in second stage left than me.

Edited by CBase
typo
Link to comment
Share on other sites

1 hour ago, CBase said:

@IronMaiden: Could it be that you have a script or addon that empties Monoprop from command pods ?

I did several tries with Gravity turns to match your efficiency, but never got better than my classic ascend attempt until I noticed you had more dV in second stage left than me.

No I figured everyone would just burn it off on the pad since it's just worthless mass for this challenge :P

15 hours ago, RizzoTheRat said:

I've been having a bit of a play with modified versions of the KerbalX for my follow-on challenge, and interestingly putting the upper stage in a faring only saved about 65m/s using my standard launch profile in kOS.  Admittedly that added a little bit of weight, but I expected drag to be a bigger factor.

The drag is a bigger factor if you take a more aggressive gravity turn which the TWR of this craft definitely allows. Here's a fairing flight with 3655m/s remaining, I would've had about 3685m/s had I not accidentally hit Z and shot my apoapsis up to 116km :P so that's about a 230m/s difference the fairing makes.

Another change you could make would be to revert the poodle to the old texture which causes far less drag. You can get the older version through the advanced mode. So anyone trying this challenge in a version older than 1.6 is going to have another 130m/s or so, not accounting for any changes made to the aerodynamic model since then. Here's my old poodle results. After the poodle, the next biggest culprit is the ladder, get rid of that and this thing flies much better.

 

Link to comment
Share on other sites

After some more tries with MechJeb Gravity Turn Profile:

3479 m/s, Turn start 25 m/s, Turn start pitch 10.5°

PhDhzc7.png

 

30 minutes ago, IronMaiden said:

No I figured everyone would just burn it off on the pad since it's just worthless mass for this challenge

:blush: I didn't, as I considered it payload. With burned off Monoprop above launch has 3519 m/s left.

Link to comment
Share on other sites

On 1/9/2020 at 8:30 PM, RizzoTheRat said:

@Laie is that 65 degrees from the horizon nor a 65 degree from vertical initial push?

65deg on the navball, or 25 deg from vertical. The code is pretty trivial, really:

# declared parameters (could as well be args from the command line)
turn_end_speed  = 200
turn_end_pitch  = 65

# make a simple linear function of that
turn_multi = float(90 - turn_end_pitch) / turn_end_speed

# then, in the main loop:
desired_pitch = 90 - turn_multi * airspeed

The linear function is too simple, if you want to make it perfect you have to throw in a TWR-dependent fudge factor that makes it turn more slowly at the beginning and faster at the end. I did that for an Eve lifter, once.

But even without TWR compensation, the vessel does a pretty smooth turn and remains oriented closely to prograde all the while. Which I find much more pleasing to look at than a sudden 10° jerk. It also has the side benefit that both agile and sluggish rockets will come out of the turn at the desired angle.

16 hours ago, RizzoTheRat said:

I've been having a bit of a play with modified versions of the KerbalX

I've tried replacing the Swivels with Reliants, putting on a Skipper in lieu of the Mainsail, and removing one Rockomax tank from the main stack. I'm afraid that this would be hated for having too low a TWR once the boosters are gone. But, fun fact, the nominal dV is nearly identical to that of the original Kerbal X.

58 minutes ago, IronMaiden said:

No I figured everyone would just burn it off on the pad since it's just worthless mass for this challenge :P

What a great idea. May I also run the poodle at low throttle before launch, to get rid of the useless fuel in the upper stage?

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.

×
×
  • Create New...