Author Topic: Kerbal Space Program or "Rocket science is harder than it looks"  (Read 373902 times)

0 Members and 2 Guests are viewing this topic.

Offline watsisname

Re: Kerbal Space Program or "Rocket science is harder than it looks"
In the interests of fuel conservation, there's two principles of orbital mechanics that everyone should know:

1:  Plane change maneuvers are EXPENSIVE, but there are good ways and not so good ways of doing them.  The lower the orbital velocity, the less expensive it will be to change the inclination.  Therefore, do this as far from the object you are orbiting as possible.  Let's look at an extreme example:

Imagine you are in a 100km altitude, prograde, equatorial orbit around Kerbin, and you want (for whatever reason) to switch this to a retrograde orbit.  If you were to try to plane change by 180° at this altitude, (basically cancelling all of your orbital speed, and then regaining it in the opposite direction), that would cost you a delta-v of about 4500m/s.  That's several times the cost of going to the Mun!  There's a far cheaper option:  Instead, make a prograde burn and ellipticize your orbit significantly, perhaps with an apoapsis at the distance of the Mun.  Then do the plane change maneuver at apoapsis, then recircularlize the orbit when you've fallen back to periapsis.  Even though you'll be be going much farther out of your way, the fuel cost is much less -- in this case, only ~2050m/s (840 to Mun, 370 to reverse the orbit, then another 840 to recircularize).  In all that's less than half the cost of the single low-orbit plane change!

2:  Prograde/retrograde burns are more efficient at higher orbital velocities, because the work done by the rocket is greater if it travels through a greater distance.  (This is the Oberth Effect). 

Example:  Going from Munar orbit (I experimented with a 133km circular prograde equatorial munar orbit) to Eve.  You might think to try a single ejection burn to get you straight to Eve's orbit.  This has a delta-v cost somewhere around 1000m/s, depending on the exact geometry of your ejection burn and where the Mun is in its orbit.  But there's a better option, by taking advantage of the Oberth effect: 

First cancel out your orbital velocity around Kerbin by burning prograde when your ship is in front of the Mun's direciton of motion (use the map and check the Mun's orbital path).  So you'll give yourself just enough speed to escape from the Mun, then slowly fall back toward Kerbin with as low of a periapsis distance as possible, then do your ejection burn at periapsis and go to Eve.  (This maneuver works best if the Mun is located behind Kerbin's direction of motion -- that way your ejection from Kerbin is retrograde and brings you to the inner solar system.)  This maneuver sequence has a delta-v of 230m/s (Mun escape) + 230m/s (Kerbin escape) for a total of 460m/s.  Again, less than half the cost of the more straightforward direct trajectory!
In my world of sleepers, everything will be erased.
I'll be your religion, your only endless ideal.
Slowly we crawl in the dark.
Swallowed by the seductive night.

 

Offline General Battuta

  • Poe's Law In Action
  • 214
  • i wonder when my postcount will exceed my iq
Re: Kerbal Space Program or "Rocket science is harder than it looks"
I'd add to 2 that if you are planning a mission to another planet with a Kerbin orbit rendezvous between mission spacecraft, you should establish the rendezvous (and begin your joint insertion burn) from as low an orbit around Kerbin as feasible.
« Last Edit: December 15, 2012, 11:09:54 pm by General Battuta »

 

Offline Nuke

  • Ka-Boom!
  • 212
  • Mutants Worship Me
Re: Kerbal Space Program or "Rocket science is harder than it looks"
i just found out you can double-dock parts, this solves my structural concerns about my kerbalstar.
I can no longer sit back and allow communist infiltration, communist indoctrination, communist subversion, and the international communist conspiracy to sap and impurify all of our precious bodily fluids.

Nuke's Scripting SVN

 
Re: Kerbal Space Program or "Rocket science is harder than it looks"
On the second rescue attempt, I took 253 screenshots.  I pared that down quite a bit, so if it seems like there's some bits of the mission that are missing from the below post, that's probably because there is.

Roight.  Rescue Attempt Number Two, Electric Boogaloo:

This time, instead of going to the station for a pit stop, prior to transfering to Duna, the pit stop came to the interplanetary vehicle.  Since the station's reserves were low, and the interplanetary stage just needed a little topping up, after reaching its low, eighty-five kilometer parking orbit, I just sent the oiler straight to the rescue ship.



After topping up the rescue ship and dumping the rest of its reserves in the station's fuel depot, the oiler returned to Kerbin for the most perfect landing ever.



The blokes at the Kerbal Space Center saw this as a wonderful sign for things to come, and so they ordered the rescue mission to depart Kerbin orbit.



Upon reaching Duna, the capture process was much the same as before, dropping the Duna periapsis, until we got a decent Ike encounter, and then doing a proper capture around the moon.



Ike orbit established, the automated lander was turned loose.  I had a strategy for landing on a slope, as level ground seemed a rare commodity on Ike, but as it turned out, the lander found a patch of the stuff about sixty kilometers from the stranded oiler.  That was too far for an EVA hop (at least too far for a safe EVA hop), but the knackered rover from the previous operation was still near the oiler, so sixty kilometers didn't seem unreasonable.  Moreover, any leftover fuel in the lander was going to make a handy reserve for the return trip, so I didn't really want to go burning all of it looking for another landing site.



Speaking of the knackered rover, that thing had been sitting around for about a year.  Hopefully the vaccuum kept the rover from breaking down the way cars do, when left unused too long.  After one last look around the oiler, it was time to find out.



One-fingered salute to the old lander!



If you do intend to send a ground vehicle to Ike, think rock-crawler, instead of rover.  There's no getting around the canyons, besides going straight through them, and the slopes on each side get pretty immense.  I hit a thirty-degree incline at one point, and twenty degrees seemed to be the norm for the steepest portion of any given wall.  You'll also notice that the indicator for the interplanetary vehicle passes by....twice.



Eventually, the oiler crew found themselves in a familiar position:  out of fuel.



Keldon, having a name vaguely Klingon-ish, was given the duty of diving head-first toward the second lander, because surely, his skull would be the most capable of surviving the impact, right?



After Keldon's compatriots joined him, they all paused, briefly awestruck.  After a year-and-a-half, most of it spent on a desolate, gray rock, it was hard to believe that their return trip to Kerbin was at hand.  The Kerbonauts orbiting above regretted not sending one of their number down in the lander, so that they could bring the KSP cattle prod to bear on those morons, who'd apparently exposed their brains to vaccuum one too many times.



I kept a shot of the map view handy, just to show the expanse of terrain that the stranded crew ultimately traversed.



Crew finally aboard, the lander pulled up stakes and set itself to intercepting the interplanetary vessel above.  It started out orbiting in the wrong direction, initially, so I wound up burning off all the extra fuel I hoped to keep handy, but mistakes are what reserve fuel is there for!



Now, I wanted to have a bit of a pause in the commentary here, since there was at least one complaint about difficulty docking, using side- or rear-facing docking ports.  If you right-click on a docking port and choose "control from here", your navball and controls will all reorient themselves so that the selected docking port is treated as the front of the ship.  I use the technique here for an effortless docking maneuver, using the rear-facing port on the lander, instead of having to switch to the interplanetary ship to use its forward-facing docking port.



After shuffling some fuel around, the interplanetary craft dropped its remaining boosters.



Then they plotted out a nifty course that would take it to a fifty-five kilometer periapsis above Duna, before blasting out into interplanetary space at ludicrous speed.  (Okay, maybe they didn't quite go plaid, but it was a fun escape maneuver that took me closer to Duna than I'd previously been.)



After a couple of corrective burns, the crew of the rescue craft realized that they needed a new mission planner, namely anyone who pays heed to phase angles.



Eventually, the ship made it back to Kerbin's sphere of influence, where an aerobraking maneuver was plotted out.  The blokes at Kerbal Space Center promised that this aerobraking maneuver wouldn't be quite so ambitious as the one they sent Nelfurt and company on.



After everyone survived the aerobraking manevuer, I did a corrective burn to keep the interplanetary vehicle in orbit.  It's got a 2,000,000 meter apoapsis, but a seventy-five kilometer periapsis, so if I can get an oiler out to intercept it for refueling, it will only take a relatively short burn to get the ship back out of Kerbin's sphere of influence, where that NERVA can get it damn near anywhere in the solar system.



After ensuring that section's safety, it was time to transfer the crew to the lander and let the two sections have their final parting of ways.



On the next close approach to Kerbin, the lander circularized its orbit at seventy-five kilometers to bide its time to land.  Remember that the ship came in on a highly inclined orbit, and the reserve fuel to be used for a plane change was spent correcting for a wrong turn over Ike.



Eventually, the lander decided the time was right, retracting its solar panels and beginning its descent.  Because it was an unusual angle of approach, I got to see some parts of the southern sea, that are usually over the horizon.



Land ho--holy ****.



With a narrow base and high dry center of mass, this thing was not designed for a water landing.  Still, there was nothing for it at this point, but to pull the chutes.



Beautiful!  Not even one bit torn off!  What somebody failed to mention was that even empty fuel tanks are highly reactive to water.



Still, checking the tins, we see that of nine six Kerbonauts deployed on this mission and subsequent rescue, all six are alive and well!



Now, those Kerbonauts just need to wait for a version of the game that will implement a nautical shipyard, so that the Kerbal Coast Guard will have something with which to retrieve them.


  

Offline Herra Tohtori

  • The Academic
  • 211
  • Bad command or file name
Re: Kerbal Space Program or "Rocket science is harder than it looks"
Meanwhile, my research into supersonic atmospheric transportation meets completely expected challenges.



I need to buy more space tape.
There are three things that last forever: Abort, Retry, Fail - and the greatest of these is Fail.

 

Offline newman

  • 211
Re: Kerbal Space Program or "Rocket science is harder than it looks"
That Concorde look-alike seems to think it's a bird...
You know what the chain of command is? It's the chain I go get and beat you with 'til ya understand who's in ruttin' command here! - Jayne Cobb

 

Offline Bob-san

  • Wishes he was cool
  • 210
  • It's 5 minutes to midnight.
Re: Kerbal Space Program or "Rocket science is harder than it looks"
Fugly structural pylons?
NGTM-1R: Currently considering spending the rest of the day in bed cuddling.
GTSVA: With who...?
Nuke: chewbacca?
Bob-san: The Rancor.

 

Offline newman

  • 211
Re: Kerbal Space Program or "Rocket science is harder than it looks"
@BlueFlames: That was a cool mission, I really like the two-engined lander design :)

If I might make a suggestion, while I do love a ton of pictures accompanying ksp missions, too much means your browser's going to jump around like a Jack Russel terrier while it's loading pictures and you're trying to read something. While annoying on a desktop browser, it's an utter pain on a mobile device. This is still ok if you don't post a bunch of pictures that show more or less the exact same thing. I'd suggest censoring that a bit - for example, 5 pictures showing almost the same angle of a ship making a burn over Ike aren't really necessary; posting one or two (if sufficiently different angles and you like them both so much) would suffice. You also don't really need 5 pictures of parachuting back to Kerbin, three showing the landed thing etc. Like I said, it's all cool but doing a bit of a critical selection would improve the readability of posts here.
« Last Edit: December 16, 2012, 05:21:41 am by newman »
You know what the chain of command is? It's the chain I go get and beat you with 'til ya understand who's in ruttin' command here! - Jayne Cobb

 

Offline Nuke

  • Ka-Boom!
  • 212
  • Mutants Worship Me
Re: Kerbal Space Program or "Rocket science is harder than it looks"
i try to use really efficient compression (quality level 4 in photoshop). i used to half the resolution but the jpeg compression makes the image filesize really small (about 80k per shot), so the halving of resolution really isnt neccisary. so long as you dont post pngs at 1080p everything should be ok.
I can no longer sit back and allow communist infiltration, communist indoctrination, communist subversion, and the international communist conspiracy to sap and impurify all of our precious bodily fluids.

Nuke's Scripting SVN

 

Offline newman

  • 211
Re: Kerbal Space Program or "Rocket science is harder than it looks"
True, though the sheer number of shots can be a problem too. 50 pics in a post aren't going to play nice no matter how efficient compression you use, and there really is no need to post 5 pics of the exact same thing from very slightly different angles.
You know what the chain of command is? It's the chain I go get and beat you with 'til ya understand who's in ruttin' command here! - Jayne Cobb

 

Offline TwentyPercentCooler

  • Operates at 375 kelvin
  • 28
Re: Kerbal Space Program or "Rocket science is harder than it looks"
I sorta like the photo-journal bits.

I only just picked this game back up and I'm getting used to all the new stuff. I launched a few satellites into Kerbin orbit and the beginnings of a space station, and now I just launched a probe to Eve. Unfortunately, my launch vehicle needs a bit of tweaking - I ran out of fuel before I could do the transfer and now I'm attempting it with an ion thruster. It's working...just...very slowly. Maybe I should have just aborted and tried again, but damn it! I've watched too many missions fail already; the Kerbal voters are getting restless and mumbling about cutting my funding. I'll show them! 760 m/s of delta-V on 0.5 kN of thrust may take 40 minutes, but I'm going to do it.

 

Offline newman

  • 211
Re: Kerbal Space Program or "Rocket science is harder than it looks"
I like the photo journal bits too. I'm just saying they could be a tad more browser and mobile device friendly by not having 5 almost identical shots. Select one out of 5, apply the principle to the whole post, and suddenly you get a post that offers the same amount of info and fun with far less frustration on load.

Also, ion engines are ok for small probes, anything else use nuclear engines (NERVAs). More efficient than standard LFEs but more powerful than ion engines, they can move heavier loads on interplanetary transfers and the burn won't last till your grandkids graduation day.

Eve, with it's thick atmo, is a place of delta-v happy hour; you can use the atmosphere to slow yourself down into orbit. Just save before you do it, if you set your periapsis too low you'll crash. If you mean to land a probe on surface, pack chutes - they work wonders in that thick soup Eve has for atmo.
« Last Edit: December 17, 2012, 02:20:44 am by newman »
You know what the chain of command is? It's the chain I go get and beat you with 'til ya understand who's in ruttin' command here! - Jayne Cobb

 

Offline watsisname

Re: Kerbal Space Program or "Rocket science is harder than it looks"
Eve is indeed fantastic for aerobrake/capture.  Conversely, I tried that with Duna and did horribly.  25 kilometers was too ****ing high and I barely lost any speed.  Tried again at 15km and nearly crashed.  I guess 18-20 is the sweet spot?

Yesterday I sent a lander to Minmus to mine it's delicious wintergreen mint flavor crystals.  Had barely any fuel left for shipping them back to Kerbin, but lucked into this awesome trajectory.



Thank you Mun!  I'm sorry I told you to **** off earlier.  You're cool.
In my world of sleepers, everything will be erased.
I'll be your religion, your only endless ideal.
Slowly we crawl in the dark.
Swallowed by the seductive night.

 

Offline Nuke

  • Ka-Boom!
  • 212
  • Mutants Worship Me
Re: Kerbal Space Program or "Rocket science is harder than it looks"
anyway i spent most of the day moving things around in kerbin orbit. i started working on the kerbalstar's spaceframe, but had to stop to launch some tankage which will be permanent internal tanks (there will be a number of places to mount external drop tanks). there will be 8 large tanks in all. plan is to build a ship capable of visiting every planet in the solar system without refueling and return to kerbin.

i also assembled a make shift space dock so i have a place to park my tugs, thruster pods, and bits when im not using them, i also docked one of my failed tanker tanks to it so it can also serve as a refueling station. i decided not to use the longer girders, and only some of the smaller ones. instead i will launch the space frame in what i think will be 4 or 5 pieces, put them together, and attach it to the drive section as one piece. i wonder how many simultaneous dockings can occur. ive had 2 ports connect simultaneously, and i hope i can get 6 simultaneous connections between the space frame and the drive section.

this ship is at least 120 tonns (crude estimate) right now. and it doesnt even have much fuel yet (just a few droptanks that never got emptied all the way, enough to move it about somewhat, though that is becoming structurally troublesome as the ship gets bigger). il post some pics once i get the first 4 tanks installed. i wish they would come up wit a sas that didnt cause large ships to vibrate violently. this makes docking large tanks troublesome. perhaps i can transfer the bulk of the fuel into a temporary tank and use a smaller ship to move the fuel back into the main tank one load at a time.
I can no longer sit back and allow communist infiltration, communist indoctrination, communist subversion, and the international communist conspiracy to sap and impurify all of our precious bodily fluids.

Nuke's Scripting SVN

 

Offline StarSlayer

  • 211
  • Men Kaeshi Do
    • Steam
Re: Kerbal Space Program or "Rocket science is harder than it looks"
Meanwhile, my research into supersonic atmospheric transportation meets completely expected challenges.

[lolshot]http://img248.imageshack.us/img248/4180/noodleplane.gif[/lolshot]

I need to buy more space tape.

Nice Ornithopter, now get it to Duna.
“Think lightly of yourself and deeply of the world”

 

Offline NGTM-1R

  • I reject your reality and substitute my own
  • 213
  • Syndral Active. 0410.
Re: Kerbal Space Program or "Rocket science is harder than it looks"
What I actually find most amusing is the nosecone trying to come off, not the wingflap.
"Load sabot. Target Zaku, direct front!"

A Feddie Story

 

Offline Sushi

  • Art Critic
  • 211
Re: Kerbal Space Program or "Rocket science is harder than it looks"
1:  Plane change maneuvers are EXPENSIVE
2:  Prograde/retrograde burns are more efficient at higher orbital velocities

How do you determine/quantify when and to what extend it's worth going out of your way to do these? For example, for #1, if you only need to shift a few degrees, is it really worth elipticizing your orbit first? By how much should you elongate it to maximize efficiency? For #2, if you're already in orbit at 100km, is it worth dropping to 70 for your Kerbin Escape burn? Is there a point at which it's not worth it? And if so, what point is that?

I'm as interested in when these tips DON'T apply as when they do, and where the boundaries are.

 

Offline Mongoose

  • Rikki-Tikki-Tavi
  • Global Moderator
  • 212
  • This brain for rent.
    • Minecraft
    • Steam
    • Something
Re: Kerbal Space Program or "Rocket science is harder than it looks"
I like the photo journal bits too. I'm just saying they could be a tad more browser and mobile device friendly by not having 5 almost identical shots.
I sort of feel like people deserve what they get if they bother using a mobile browser. :p

 

Offline Nuke

  • Ka-Boom!
  • 212
  • Mutants Worship Me
Re: Kerbal Space Program or "Rocket science is harder than it looks"
Meanwhile, my research into supersonic atmospheric transportation meets completely expected challenges.

[lolshot]http://img248.imageshack.us/img248/4180/noodleplane.gif[/lolshot]

I need to buy more space tape.

Nice Ornithopter, now get it to Duna.

and dont get swallowed by a worm.
I can no longer sit back and allow communist infiltration, communist indoctrination, communist subversion, and the international communist conspiracy to sap and impurify all of our precious bodily fluids.

Nuke's Scripting SVN

 

Offline watsisname

Re: Kerbal Space Program or "Rocket science is harder than it looks"
1:  Plane change maneuvers are EXPENSIVE
2:  Prograde/retrograde burns are more efficient at higher orbital velocities

How do you determine/quantify when and to what extend it's worth going out of your way to do these? For example, for #1, if you only need to shift a few degrees, is it really worth elipticizing your orbit first? By how much should you elongate it to maximize efficiency? For #2, if you're already in orbit at 100km, is it worth dropping to 70 for your Kerbin Escape burn? Is there a point at which it's not worth it? And if so, what point is that?

I'm as interested in when these tips DON'T apply as when they do, and where the boundaries are.


MATH.

Actually, this doesn't seem like it would be too difficult to figure out, if we make some (very) simplifying assumptions.  Basically, we need to compare the delta-v cost of the plane change as a function of i (for which we will assume a circular orbit, because otherwise the formula is nasty, and compare that to the total delta-v of the other maneuver.

Noting that the plane-change cost goes to zero as orbital velocity goes to zero, which happens as apoapsis goes to infinity, and further noting that the marginal cost to increase eccentricity decreases as e-->1, then this means the ideal case is to go to an escape trajectory.  (We'll further simplify this by ignoring other gravity sources once you escape -- e.g. the sun).  The total cost then is just two times the difference of the escape velocity minus the current orbital velocity.  (Two because I'm assuming you want to re-circularize the orbit afterward).

Delta-v for the plane change is:


Set that equal to the cost of the alternate maneuver:


do the algebra and solve for the inclination angle change desired, we get:


Hmmm.  We've got escape speed (at current altitude) divided by orbital speed (again at current altitude) there.  That's just root(2/r) divided by root(1/r), which is just root(2).  Should be true for all altitudes.  Well then, so that's just a number -- put it back in the equation and we get delta-i is 48.9°.  Any maneuver requiring a larger inclination change than that (from a circular starting orbit) should instead use the other method.  There ya go!  I learned something today. :)
(Now I think I did all that correctly; if not then point out the mistakes and I'll give that person a gold star.)

Again this is the simplified case of a circular orbit -- if the orbit is not circular then results will be more varied, and you'd have to consider both the eccentricity of the orbit and the location of the nodes.

TL;DR:  Just do your plane changes at whichever node happens to be higher. :V

edit: as for part 2 (Oberth Effect), that's always true.  Always aim for doing your transfer/capture burns at the lowest possible altitudes.
« Last Edit: December 17, 2012, 06:23:32 pm by watsisname »
In my world of sleepers, everything will be erased.
I'll be your religion, your only endless ideal.
Slowly we crawl in the dark.
Swallowed by the seductive night.