Jump to content

Launching Escape Systems...


Recommended Posts

The challenge is to get to a stable LKO with only Launch Escape Systems and Sepratrons (ONLY FOR SEPARATING STAGES)1,2. No cheating (except infinite electricity), mods (except MechJeb for SMARTASS, or KER for DV readouts), or altering anything in the KSP folder or the game version itself (stay in 1.2.2). It really isn't as easy as it looks (or sounds), and remember, you need around 3400 m/s of DV to get to orbit ;). 

TO SUBMIT: A video of launch and orbiting, the craft file, and a backstory behind the craft (like Bradley Whistance and his Odyssey by Bill). Ill check back every day at around 4-6 p.m. EST.

1: If no one can do it (because it might be theoretically impossible, somehow), highest altitude wins.

2: There is no prize except bragging rights. (YouTubers pls shout me out i'm desperate;.;)

ENDS MAY 28th, 2017!

The craft does not need to be manned.

Edited by Aiden.J
Link to comment
Share on other sites

16 hours ago, Aiden.J said:

The challenge is to get to a stable LKO with only Launch Escape Systems and Sepratrons (ONLY FOR SEPARATING STAGES)1,2.

What do you mean by "ONLY FOR SEPARATING STAGES"? Do you mean the Separatrons need to be pointed retrograde and activated at decoupling, or what?

Link to comment
Share on other sites

2 hours ago, joshudson said:

This is ridiculous, but somebody just might do it anyway. I'd rather just do separations to orbit, but I can't capture video either way.

If you can't take video, click "F1" at launch and at various stages during launch, and in orbit for proof of orbit.

1 hour ago, sevenperforce said:

What do you mean by "ONLY FOR SEPARATING STAGES"? Do you mean the Separatrons need to be pointed retrograde and activated at decoupling, or what?

All that needs to happen is for sepratrons to separate two different parts of the craft, and not add any additional velocity to the main craft that will achieve orbit. They can be pointed in any direction, just as long as they separate the stages.

Link to comment
Share on other sites

Seeking clarity before I make an attempt:  

1) You're making a blanket ban on all mods except those specifically listed, and thus ruling out any entries with visual enhancers?  It's only a minor inconvenience to uninstall Stock Visual Enhancements and the like, but I'd prefer not making unnecessary writes to the SSD the game's on.

2) Are you asking for a launch with only a probe core, LESs, and Sepratrons (no decouplers, no structural parts, etc), or just that LESs provide all forward thrust for a vessel without wings?

If it's a "stock parts and physics, no autopilot, but otherwise go wild" sort of thing, I may give it a shot.

Link to comment
Share on other sites

If I did the math right, it's pretty thoroughly impractical to get to orbit. The LES simply has too much dry mass, and you need an absurd number of them.

The hard cap on wet/dry ratio is going to be 1.25, the ratio of LES full/empty mass. From there, you can start calculating how many stages are needed at various stage wet/dry ratios to achieve a target of 3000 m/sec... which is quite optimistic.

For a wet/dry ratio of 1.025, you need 69 stages to reach 3,000 m/sec. As you increase this ratio, the number of stages you need goes down, but the size of the N+1'th stage grows more and more; the theoretical optimum is to have a low wet/dry ratio, so you're effectively staging off empty LESs more often.

You can also calculate how much bigger the N+1'th stage is than the N'th stage. For a wet/dry ratio of 1.025, the N+1'th stage has to be 10/9 as large as the N'th stage (1.11111...).

So, the final number of LESs can be calculated as such:

n = sum(0 to # of stages - 1) of (stage ratio ^ i)

For that wet/dry of 1.025, this is sum(0 to 68) of (1.1111^i).

This number is 12,918.35.

Thus, given hopelessly optimistic assumptions (no staging equipment, no fins, no structural elements, vacuum Isp throughout, only 3000 m/sec to orbit), you need at least 12,919 launch escape systems to get to orbit. Drop the wet/dry ratio much lower than 1.025, and you'll get some serious TWR issues.

Link to comment
Share on other sites

6 minutes ago, Starman4308 said:

...math...

Just from experimental tinkering (with an aim of launching a couple of tons worth of necessary gear to get a Kerbal involved and safely back out of orbit), I'm looking at more than 70 tons just to get the first 350 m/s of delta-V (at an average TWR around 2, in two ascent stages so far.)  I could, theoretically, build it on out and make it to orbit, but it would wind up being even heavier than my brute-forced retrosolar rescue attempts, and even harder on a CPU that barely handled those.

Link to comment
Share on other sites

9 hours ago, Aetharan said:

Just from experimental tinkering (with an aim of launching a couple of tons worth of necessary gear to get a Kerbal involved and safely back out of orbit), I'm looking at more than 70 tons just to get the first 350 m/s of delta-V (at an average TWR around 2, in two ascent stages so far.)  I could, theoretically, build it on out and make it to orbit, but it would wind up being even heavier than my brute-forced retrosolar rescue attempts, and even harder on a CPU that barely handled those.

I'm also now 95% sure I screwed up the math by assuming stage N+1 was X times larger than stage N, but I should also be including N-1, N-2, N-3, etc.

So it's even worse. I'll see if I can mock some code up tomorrow.

EDIT: Finished what I think is valid code for this.

Spoiler



    public static void lesToOrbit(double dV, double twrTarget) { // Called with 3000 and 1.8
        //double g0 = 9.8063 // Guess who has g0 stored in his "random calculations" folder!
        double isp = 180.0;
        double dry = 0.9;
        double full = 1.125;
        double thrust = 750.0;
        double targetAcc = g0 * twrTarget;

        int numStages = 1;
        double cumdV = g0 * isp * FastMath.log(full/dry);
        int nLESs = 1;

        while (cumdV < dV) {
            double unrounded = (full * nLESs * targetAcc) / (thrust - (full * targetAcc));
            int newLES = (int) Math.ceil(unrounded);
            double stageFull = full * (newLES + nLESs);
            double stageEmpty = (full * nLESs) + (newLES * dry);
            double stageDV = g0 * isp * FastMath.log(stageFull / stageEmpty);

            ++numStages;
            nLESs += newLES;
            System.out.println(String.format("Stage %d with %d LES (total %d) adding %9.5g dV to %9.5g.", numStages, newLES, nLESs, stageDV, cumdV));
            cumdV += stageDV;
        }
    }


 

The result, requiring each stage to have a TWR of at least 1.8 and a delta-V goal of 3000 m/sec, is 216 stages with a total of 4696 launch escape systems. I think part of why the number actually improved was that you can stage more often than I assumed thanks to the fantastic TWR of the LES; the end result is a stage wet/dry ratio of just about 1.005.

To get to 3400 m/sec, you wind up needing 265 stages with 16313 launch escape systems.

In short, despite mucking up the math the first time, it's still hilariously impractical, as you're talking a lower bound of thousands of LESs to get to orbit.

Edited by Starman4308
Link to comment
Share on other sites

On 5/23/2017 at 9:57 PM, Aetharan said:

Seeking clarity before I make an attempt:  

1) You're making a blanket ban on all mods except those specifically listed, and thus ruling out any entries with visual enhancers?  It's only a minor inconvenience to uninstall Stock Visual Enhancements and the like, but I'd prefer not making unnecessary writes to the SSD the game's on.

2) Are you asking for a launch with only a probe core, LESs, and Sepratrons (no decouplers, no structural parts, etc), or just that LESs provide all forward thrust for a vessel without wings?

If it's a "stock parts and physics, no autopilot, but otherwise go wild" sort of thing, I may give it a shot.

These are all good questions, so

1: You can use visual enhancements: but only stock parts except KER and the mechjeb box.

2:(insert facepalm here) Obviously you can/NEED to use decouplers or separtors. Structural elements are a'ok. And, so i don't need more facepalms, you can use reaction wheel, control surfaces, electricity stuff, and anything else thats suits your fancy. I don't care if you have wings, just the only forward thrust is from LES's.

On 5/22/2017 at 4:18 PM, sevenperforce said:

My entry: 16,704 meters.

Probably could have done better, but my first attempt kept breaking up over and over, so I ended up sacrificing mass fraction and going super-modular.

 

Nice job, my best was only around 7,000 meters, and the rocket had a sh*t ton of parts. This is so far the best entry (and the only one) yet. But i am missing one thing: a ship file. (Cut to me falling asleep in puddles of tears). (its fine, ill accept it. If you want to though, go ahead)

23 hours ago, Starman4308 said:

I'm also now 95% sure I screwed up the math by assuming stage N+1 was X times larger than stage N, but I should also be including N-1, N-2, N-3, etc.

So it's even worse. I'll see if I can mock some code up tomorrow.

EDIT: Finished what I think is valid code for this.

  Hide contents

 



    public static void lesToOrbit(double dV, double twrTarget) { // Called with 3000 and 1.8
        //double g0 = 9.8063 // Guess who has g0 stored in his "random calculations" folder!
        double isp = 180.0;
        double dry = 0.9;
        double full = 1.125;
        double thrust = 750.0;
        double targetAcc = g0 * twrTarget;

        int numStages = 1;
        double cumdV = g0 * isp * FastMath.log(full/dry);
        int nLESs = 1;

        while (cumdV < dV) {
            double unrounded = (full * nLESs * targetAcc) / (thrust - (full * targetAcc));
            int newLES = (int) Math.ceil(unrounded);
            double stageFull = full * (newLES + nLESs);
            double stageEmpty = (full * nLESs) + (newLES * dry);
            double stageDV = g0 * isp * FastMath.log(stageFull / stageEmpty);

            ++numStages;
            nLESs += newLES;
            System.out.println(String.format("Stage %d with %d LES (total %d) adding %9.5g dV to %9.5g.", numStages, newLES, nLESs, stageDV, cumdV));
            cumdV += stageDV;
        }
    }

 

 

 

 

The result, requiring each stage to have a TWR of at least 1.8 and a delta-V goal of 3000 m/sec, is 216 stages with a total of 4696 launch escape systems. I think part of why the number actually improved was that you can stage more often than I assumed thanks to the fantastic TWR of the LES; the end result is a stage wet/dry ratio of just about 1.005.

To get to 3400 m/sec, you wind up needing 265 stages with 16313 launch escape systems.

In short, despite mucking up the math the first time, it's still hilariously impractical, as you're talking a lower bound of thousands of LESs to get to orbit.

@Starman4308, Props to you. The math and everything. I honestly thought that when i made this, it would only take a hundred or so, not 16313. Two thumbs up from me.

Edited by Aiden.J
Link to comment
Share on other sites

3 hours ago, silks said:

Kerbin launch might be possible with the use of Launch Stability Enhancers. Not sure how stable it is though, and if it's considered cheating. 

The reason why that helped so much for Scott Manley's RCS rocket is that RCS engines have a particularly horrible sea-level vs. vacuum Isp; you lose 58% of your thrust/Isp at sea level. By starting from a tall tower, he was able to start his engines much closer to vacuum Isp. The LES won't benefit nearly so much from that; with its Isp being 160/180 for sea-level/vacuum, I simply ignored sea-level Isp in my analysis above.

The LES's problem is caused by its atrocious full/empty mass ratio; while you can get a huge kick out of the first few idealized* stages (which have just one LES), by the time you hit steady-state, each new stage adds maybe 8-9 m/sec of delta-V, and you're getting exponential growth in stage mass.

*Minimum number of LES set to full thrust to achieve > 1.8 TWR. This lets you stage off empty LESs as quickly as possible, although compromises might wind up being made in practice. The analysis was primarily to set a lower bound on the number of LESs necessary to achieve a vacuum delta-V target.

Granted, if you can start above atmosphere, you can avoid drag losses, and minimize gravity losses with a high TWR, something at which the LES excels. With a more generous delta-V target of 2200 m/sec (approximately LKO velocity), you can get to that using either 472 LES in 58 stages (5.0 TWR goal to minimize gravity losses), or 415 LES in 119 stages (standard 1.8 TWR goal).

So, it's maybe possible with a launch-clamp-tower-of-doom to minimize your atmospheric drag losses and minimize your delta-V requirements.

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...