Goddard-Problem: Maximal launch altitude Challenge

[Red Iron Crown]

Interesting that, while in theory, this vehicle has 1715.12 Vacuum Delta-V, thus far, it only reaches about the height of a perfectly aerodynamic cannon ball launched at ~1050 m/s.

Gravity losses for a near vertical launch are significant.

[Pds314]

Not to mention aero-losses. In FAR, the thing has a Mach 1 drag area of nearly 6 m^2, even though its cross-sectional area never exceeds 4.778 m^2

And at Mach 1.3, its drag coefficient peaks at about 1.67. It is quite possibly the worst supersonic vehicle ever built, or the best supersonic parachute.

Of course, this can be avoided by staying below the transonic drag hump until sufficient altitude is reached, but that increases gravity losses again.

Ironically, if at an AoA of 15 degrees, it will achieve slightly lower supersonic drag coefficients.

Edited July 18, 2015 by Pds314

[Nich]

WHAT????

What kind of voodoo magic is FAR using?

[mhoram]

@sal_vager: cool! thanks for stickying.

Also I will keep this challenge supported as long as squad doesn't change the physics.

[Gaarst]

Had my attempt at the challenge with a manual piloted rocket: achieved 61525m.

The only mods I used here are KER and the Stock Bug Fix Module for 1.0.4.

My ascent profile was:

- Full thrust from launch to ~560 m/s

- Then reduced thrust (35-40%) from then to ~10.5 km

- Finally, throttle up to 60-65% until fuel runs out

This having all reaction wheels on, 100% gimbal and thrust, and SAS set on prograde.

35-40%:

60-65%:

[ihtoit]

I would keep thrust low until I hit 25km, curving it from 130m/s ASL to ~700m/s@25km (not even breaking 340m/s until at least 14km). I'm away from my sandbox right now, but I'll run the sim later on and post the result.

[mhoram]

@ Gaarst: welcome to the top position on the manual launch category.

@ ihtoit: I would like to see your results on that ascent profile.

[Sando Mutt]

Ok, after working in this interesting challenge, here comes my entry:

Manual control, mods: KER+VOID for launch data, part commander for fine tuning the launch profile and handling the engines in flight.

The launch profile:

1. At prelaunch, SAS on, engines at 100% and full throttle
   
2. After liftoff, SAS on prograde
   
3. When the rocket reachs about 620m/s of vertical speed, reduce the engines to 26%
   
4. Reduce throttle bit a bit trying to maintain the vertical speed till about 80% throttle
   
5. At 10000m altitude, engines at 100% till exhaustion
   
6. After the drag is low enough that the shock layer disappears, SAS on stability assist
   

I'm convinced that this launch is far from optimal, and tweaking and adjusting the profile can add may be even 100 metres of altitude. I think that 62Km is even possible, but probably very very tricky.

[mhoram]

@Sando Mutt: great entry ... you made it to the top and surpassed the scripted ascents!

[Nich]

Manual entry

61,762 was my actual I am not going to argue 1m

Manual 100% to 600

1/3rd to 8000

2/3rds to empty

[mhoram]

@Nich: The distance to 62k is getting closer and closer. Thanks for your entry!

[Padishar]

Good job folks. Looks like I'm going to have to find some time to do a few more runs...

[Nao]

Manual entry

61,762 was my actual I am not going to argue 1m

http://i.imgur.com/40sRZxj.jpg

Manual 100% to 600

1/3rd to 8000

2/3rds to empty

Idk if i'm more impressed with results or with the ancient program border

I use classic shell for my W7's, so it looks exactly the same on my side haha

... Great job on the ascent!

[Nich]

Ha ha ya I play windowed mode and use alt+PrtScn then paste them in paint so I can save them as jpg's. Converting default screenshots is a pain to me I wish we could set the default save type lol

[adamgerd]

Altitude: 62,892m

Control: Manual Control

Mods:

Mech Jeb 2 for Engine Control in flight

KER for Flight Information

http://i.imgur.com/ZaQLbWf.png?1 http://i.imgur.com/sJHuFKa.png?1

(The First Image shows it a meter lower than it was, as seen by the second image, because i did the first screenshot too early or too late)

The Launch Profile:

1.At Launch, turn on SAS (Stability Assist)and Throttle at 100%

2. After Launch put SAS on Prograde

3.After Mach 1.2 at 600m/s turn your throttle to 26%

4. At 10000m throttle slowly up to 80%

5. When the fuel bar shows barely any fuel, Throttle up to 100% until no more fuel remains

6. Change SAS to stability Assist

But i think you can optimise this flight plan to squeeze out a hundred or so meters

Edited August 17, 2015 by adamgerd
Information added

[Cool791]

Here is my entry

Edited August 20, 2015 by Cool791
Picture link fix

[Padishar]

Altitude: 62,892m

Control: Manual Control

Mods:

Mech Jeb 2 for Engine Control in flight

KER for Flight Information

http://i.imgur.com/ZaQLbWf.png?1 http://i.imgur.com/sJHuFKa.png?1

(The First Image shows it a meter lower than it was, as seen by the second image, because i did the first screenshot too early or too late)

The Launch Profile:

1.At Launch, turn on SAS (Stability Assist)and Throttle at 100%

2. After Launch put SAS on Prograde

3.After Mach 1.2 at 600m/s turn your throttle to 26%

4. At 10000m throttle slowly up to 80%

5. When the fuel bar shows barely any fuel, Throttle up to 100% until no more fuel remains

6. Change SAS to stability Assist

But i think you can optimise this flight plan to squeeze out a hundred or so meters

I'm a bit confused by your description. First, Mach 1.2 is much less than 600m/s, more like 400m/s. 600m/s is more like mach 1.8. Or are you saying that 600m/s is past mach 1.2? Secondly, how slowly do you throttle up to 80% at 10000m? Third, how much is "barely any fuel"? How long does it burn at 100% for?

When I fly a profile like that I get nowhere near your altitude. Throttling down at 400m/s results in ~52450m. Throttling down at 600m/s is a lot better at ~60800m but still a long way short of yours. What are your video settings (vsync, frame limit) and physics delta time? Have you checked that your physics.cfg matches the stock KSP 1.0.4 version linked in the first post?

I also tried to duplicate Sando Mutt's (confusing description, especially point 4) and Nich's (I assume 600 means m/s) entries and couldn't get either of them over 61000. However, given the vague-ish descriptions, the effect that small changes can have on the results and how close we seemed to be getting to the limit, their results improving on the previous best by ~100m are far more believable than yours improving by over 1100m.

[FleshJeb]

The consistency is becoming a problem for me too.

The atmosphere changes from night to day now. It's hell on repeatable testing.

[adamgerd]

I'm a bit confused by your description. First, Mach 1.2 is much less than 600m/s, more like 400m/s. 600m/s is more like mach 1.8. Or are you saying that 600m/s is past mach 1.2? Secondly, how slowly do you throttle up to 80% at 10000m? Third, how much is "barely any fuel"? How long does it burn at 100% for?

When I fly a profile like that I get nowhere near your altitude. Throttling down at 400m/s results in ~52450m. Throttling down at 600m/s is a lot better at ~60800m but still a long way short of yours. What are your video settings (vsync, frame limit) and physics delta time? Have you checked that your physics.cfg matches the stock KSP 1.0.4 version linked in the first post?

I also tried to duplicate Sando Mutt's (confusing description, especially point 4) and Nich's (I assume 600 means m/s) entries and couldn't get either of them over 61000. However, given the vague-ish descriptions, the effect that small changes can have on the results and how close we seemed to be getting to the limit, their results improving on the previous best by ~100m are far more believable than yours improving by over 1100m.

1. I am saying 600m/s is more than mach 1 and it was just after 600m/s at approx 615m/s. (You have to remember that human instinct makes everything a bit over the number)

2. I use mechjeb to limit it at 80% and then limit it at 10 000m, so nearly instant.

3. It's at 100% starting when you get 36 units of liquid fuel until 0.

4.

Vsync: Every VBlank

Frame Limit: 120 FPS

Physics Delta Time: 0.04s

Persistent Debris: 250

5. I checked my physics.cfg file and even copied the stock one, before playing it.

Also at these efficiency you need it exactly on the split second timing.

I hope I answered all your questions detailed enough.

[kookoo_gr]

Here is my attempt, stock ksp with manual flying

[mhoram]

Wow ... a week of vacation and lots of new entries came in. You are great!

Here is my entry

Thanks for your entry - added you to the leaderboard! And enjoy your time on the forum!

Here is my attempt, stock ksp with manual flying

http://i.imgur.com/3RSanuT.png

http://i.imgur.com/VQVCMCQ.png

Good launch! Also added you to the leaderboard!

The atmosphere changes from night to day now. It's hell on repeatable testing.

From what I read about the atmosphere, I gathered that the differences are not that large. And you are free to chose your launchtime as you wish - the "revert to launch" functionality even keeps the time identical.

Altitude: 62,892m

The Launch Profile: [...]

[more detailled infos]

This is a tough one. While I want to believe that you managed to beat 62k, please give me a bit more time to test your ascent profile and rate your entry.

@Padishar: Thanks for your trial runs and explanations. I appreciate it.

[mhoram]

@adamgerd:

I tried to replicate your results with the following kOS-script as best as I understood your description.

SET THR TO LIST(
1, // 0
1, // 1
1, // 2
1, // 3
1, // 4
1, // 5
1, // 6
1, // 7
1, // 8
1, // 9
1, // 10
1, // 11
1, // 12
1, // 13
1, // 14
0.26, // 15
0.26, // 16
0.26, // 17
0.26, // 18
0.26, // 19
0.26, // 20
0.26, // 21
0.26, // 22
0.26, // 23
0.80, // 24
0.80, // 25
1, // 26
1, // 27
1, // 28
1, // 29
1, // 30
1, // 31
1, // 32
1, // 33
1, // 34
1, // 35
1, // 36
1, // 37
1, // 38
1, // 39
1, // 40
1, // 41
1, // 42
1, // 43
1, // 44
1, // 45
1, // 46
1, // 47
1, // 48
1, // 49
1, // 50
0).

LOCK THROTTLE TO 1. // 1.0 is the max, 0.0 is idle.
LOCK STEERING TO HEADING(90,88.125).
WAIT 1. // give throttle time to adjust.

PRINT "Launch.".
STAGE. // same as hitting the spacebar.

declare i to 0.
UNTIL STAGE:LIQUIDFUEL < 0.001 {
print i.
LOCK THROTTLE TO THR[i].

set i to i + 1.
wait 1.
}
PRINT "Fuel Depleted @ ".
PRINT i.

declare maxalt is 0.
until maxalt > SHIP:ALTITUDE {
lock steering to SHIP:SRFPROGRADE.
set maxalt to SHIP:ALTITUDE.
wait 0.1.
}
log maxalt to alt.txt.
log i to alt.txt.
PRINT maxalt.
print "Ending".

However I was unable - just like Padishar - to get close to 62k altitude.

I was able to get consistent results in the 61.3-61.4 km range - perhaps you can test this script and point me to where it diverges from your ascent profile.

I would be happy if we were able to find the reasons for the discrepancies between the results and offer any help I can provide, but until then I am sorry that I can not accept your entry for the leaderboard. :-/

[Sando Mutt]

Ok folks,

I have made another tests batch (adding perhaps about 60 launches), in manual and kOS mode. I have bested my previous attempts by a good margin, see the results:

Manual launch: 61904m.

I have been very careful to get an uniform performance, repeating this flight profile at least 10 times, and getting consistent results above 61800:

1. Launch with SAS on at 100% throttle
   
2. SAS to prograde
   
3. At 620m/s of vertical speed reduce throttle to 28%
   
4. Maintain speed at all moment below 620m/s, reducing throtte bit a bit as needed.
   
5. At 10000m of altitude, throttle to 68%
   
6. At 25000m of altitude, SAS on stability assist
   

kOS scripted: 62116

// set vars for launch profile
declare maxs to 620.0.
declare maxalt to 10000.0.
declare ft to 0.68.

// prepare vessel.
lock throttle to 1.0.
sas on.
wait 2.0.

// launch vessel
stage.
wait 1.0.
set sasmode to "PROGRADE".

wait until ship:verticalspeed>maxs.

declare currt to 0.28.
declare tref to 0.28.
// throttle value for still vertspeed

// set control loop for vertspeed
declare i to 0.
print ship:verticalspeed + " vs".

until ship:altitude>maxalt {
declare ap to (maxs - ship:verticalspeed) / (250 * (currt / tref) ).
set currt to currt+ap.

if currt<tref-0.15 { set currt to tref-0.15. }
if currt>tref { set currt to tref. }
lock throttle to currt.
set i to i + 1.
print ship:verticalspeed + " updating throttle to " + currt + " on step " + i.
wait 0.1.
}

lock throttle to ft.

wait until altitude>25000.0.
set sasmode to "STABILITYASSIST".

// stategy 1 for triple-g Goddard problem

This script is a kOS translation of the previous manual profile launch. This night I will be doing a few more tries, to test parameter space.

I can say also that when you launch matters a lot. I have got more than 500m of difference between best and worst result at different hours.

Also I have tried adamgerd profile. So far I have been able to arrive only to 61600m. so I think also that his result is dubious and there is something different between his installation and ours.

Edited August 25, 2015 by Sando Mutt

[adamgerd]

@adamgerd:

I tried to replicate your results with the following kOS-script as best as I understood your description.

SET THR TO LIST(
1, // 0
1, // 1
1, // 2
1, // 3
1, // 4
1, // 5
1, // 6
1, // 7
1, // 8
1, // 9
1, // 10
1, // 11
1, // 12
1, // 13
1, // 14
0.26, // 15
0.26, // 16
0.26, // 17
0.26, // 18
0.26, // 19
0.26, // 20
0.26, // 21
0.26, // 22
0.26, // 23
0.80, // 24
0.80, // 25
1, // 26
1, // 27
1, // 28
1, // 29
1, // 30
1, // 31
1, // 32
1, // 33
1, // 34
1, // 35
1, // 36
1, // 37
1, // 38
1, // 39
1, // 40
1, // 41
1, // 42
1, // 43
1, // 44
1, // 45
1, // 46
1, // 47
1, // 48
1, // 49
1, // 50
0).

LOCK THROTTLE TO 1. // 1.0 is the max, 0.0 is idle.
LOCK STEERING TO HEADING(90,88.125).
WAIT 1. // give throttle time to adjust.

PRINT "Launch.".
STAGE. // same as hitting the spacebar.

declare i to 0.
UNTIL STAGE:LIQUIDFUEL < 0.001 {
print i.
LOCK THROTTLE TO THR[i].

set i to i + 1.
wait 1.
}
PRINT "Fuel Depleted @ ".
PRINT i.

declare maxalt is 0.
until maxalt > SHIP:ALTITUDE {
lock steering to SHIP:SRFPROGRADE.
set maxalt to SHIP:ALTITUDE.
wait 0.1.
}
log maxalt to alt.txt.
log i to alt.txt.
PRINT maxalt.
print "Ending".

However I was unable - just like Padishar - to get close to 62k altitude.

I was able to get consistent results in the 61.3-61.4 km range - perhaps you can test this script and point me to where it diverges from your ascent profile.

I would be happy if we were able to find the reasons for the discrepancies between the results and offer any help I can provide, but until then I am sorry that I can not accept your entry for the leaderboard. :-/

I will check the KOS Script ASAP, and maybe modify it slightly to match my ascent profile and then add my KOS script on the forum

[Padishar]

This is my best so far with my own script. 61689m

Sando Mutt, I tried your script a number of times and I can't get above 61.5k. Can you try my script above and see what you get?