-
Posts
333 -
Joined
-
Last visited
Content Type
Profiles
Forums
Developer Articles
KSP2 Release Notes
Everything posted by PLAD
-
Hello CitizenVeen, It is true that the text that comes with the download is pretty light on details. The Imgur album guides in the FF for RSS guide (in the very first post in this thread) answer those questions with a lot of details though. Here are the parts relevant to your questions. FF assumes you have a test spaceship in orbit around the start body (usually Earth) in an equatorial, circular orbit at the "boost altitude". I Use Hyperedit to put it there. You will never actually fly away in this ship, it is just there to show you the orbit you need to launch into. In the example below I want to launch from Earth and flyby Jupiter to get to Saturn. After some searching around I have decided I want the flight that is highlighted in Blue. What is called "start vZ" in the column on the left is called "start Equatorial Z velocity" in the detail panel on the right. It is the number you will enter in the Mechjeb (or precise node if you use that) field called "Normal +" that you see in the 2nd picture below. "Start equat. prograde velocity" is the number I will enter in the Mechjeb field called "Prograde" in the 2nd picture below. You are right that I do not suggest an angle to prograde. My approach is to have the user enter the start equatorial Z and prograde velocities into a manuever node placed on the test ship and then use "shift time" from Mechjeb to find the spot from which you need to execute the departure burn. I do this because even if you start with a suggested angle to prograde you will still have to adjust the start time from there until you find the exact spot. So I just start with the shift time at maybe 200 seconds and keep hitting "+" and "-" and watch the predicted path until it is close to hitting the target planet at the desired time, then cut the time step to about 10 seconds, then finally 1 second to zero in on the exact start time. "V infinity leaving start planet" is the speed your ship will have just before it leaves the SOI of the start planet. It is used as a check that your ship has the right energy at that point. The other "Vinf in" and "Vinf out" fields give the speed you should have just after you enter the SOI of a planet and just before you leave the SOI of a planet. They are also just safety checks that all is well. I could remove those fields and you could still fly the mission without knowing them, though if you do a lot of correction burns you might change your speed considerably and by comparing to the desired v infinities you might catch the problem early. In any case, here are the pictures. 1st picture: the most important numbers are the start equatorial Z and prograde velocities. 2nd picture: I have entered those two numbers in the correct fields in Mechjeb's manuever node editor, and then hit 'shift time' back and forth until I saw the flight Earth-Jupiter-Saturn. 3rd picture: This shows what the correct departure orbit looks like. The blue line is the circular, equatorial, 200x200km orbit of the test ship. The dotted yellow line shows the departure orbit that I will have to launch my ship into. I know it is inclined -32.4 degrees from the "start orbit inclination" that FF gave. My real ship is sitting on the launchpad at Canaveral, I must now wait until the Earth rotates Canaveral into the plane of the departure orbit. and then launch. Then I have to set up the maneuver node again for my real ship, but since my real ship will be moving on the right plane to start with (hopefully!) the maneuver will only have to be the "start Boost" entered into the "Prograde" field of Mechjeb. Radial always stays at '0' and normal will be '0' if your ship is exactly in the right plane. (Start boost is 6530m/s in this example.) It is hard to launch exactly into a line shown on the screen like that. I am trying different ways to indicate the best launch time but have some problems. The best I've come up with so far is to give the LAN of the desired start orbit, then sit at Canveral with Mechjeb's "Orbit info" page open, because Mechjeb gives the LAN of your ship even as it sits on the ground. The huge problem with that is that the indicated LAN at launch will be lower than the actual LAN once you are in orbit due to the variable nature of the ascent path. In any case this is what the info that FF gives out means. I've found the most important thing of all is the times of the launch and flyby periapsis, if you get those right everything else will follow. So for instance the example above says to leave Earth on Y24 D135 H0 and flyby Jupiter at Y26 D135 H0. If you get that right you should be able to make a small adjustment to your Jupiter (best is to do the correction 90 degrees before your Jupiter encounter) flyby to get flung to Saturn. In the example you might notice that my actual flight time E-J is 1 year 363 days 13 hours, about 35 hours shorter than FF said but that is good enough for this flight. If I was going on to several other planets it would be good to tighten that up a bit, experience will tell you what sort of flight time error is acceptable for certain paths. In no case should you need better than +/- 6 hours in RSS.
-
Here is how I make a folding centrifuge... This uses the centrifuge hub from DSEV, which is 2 or 3 parts. Then each of the arms have 6 parts including the Porkjet inflato hab and a hinge from Infernal Robotics. I don't think it could quite fit behind a 5-meter heatshield though. Maybe if you shorten the 'shoulders' and make the ship's body thinner.
-
How do you plan your launch windows in RSS?
PLAD replied to CitizenVeen's topic in KSP1 Gameplay Questions and Tutorials
For getting to the planets I use Flyby Finder for RSS (*blush*) as mentioned above. For getting to the Moon I also wrote a little spreadsheet in which you enter your launch site and the search start time, and it tells you the next two launch opportunities for a transfer to the Moon using a due-East launch azimuth. It's available at the bottom of the first post in the FF for RSS thread. -
How far did you get in Realisme Overhaul?
PLAD replied to CitizenVeen's topic in KSP1 Mods Discussions
CitizenVeen- Thanks! I've sent probes to all the planets, but only in a cheaty way- I test out Flyby Finder by Hyperediting unrealistic probes into orbit and then doing a series of flyby missions to make sure FF is set up right. I even did an Earth-Venus-Earth-Earth-Jupiter run that way. But right now I'm fixated on piloted missions and the technology just doesn't exist to get people past Mars/Venus. At some point I might cheat a better H2 boiloff rate and use Porkjet's nuclear mods to get to Jupiter. Already from my Moon missions I can see that Hydrogen is the key to the Solar System. I played one short career-mode game but am now only sandbox. I only use RO-approved engines. NathanKell- That was a superbly detailed Mars mission! -
I still check this thread regularly. Lately I've been using Realism Overhaul a lot and have not worked on the stock version though. I still have the short-term plan to improve the way start orbits are defined, and a long-term plan for double flybys. But it's the old problem of splitting time between the Real World, playing the game, and writing stuff. I have no plans do make a Linux version of FF since I don't have a Linux machine, and even if I had the OS it looks like there isn't a version of Delphi Pascal with full graphics support for Linux. I fear Pascal is a somewhat obsolete language so I'd have to learn another one to port FF over. Time, time, time.
-
How far did you get in Realisme Overhaul?
PLAD replied to CitizenVeen's topic in KSP1 Mods Discussions
I got as far as Mars, but lately I've been obsessed with piloted missions to the Moon, I'm trying to optimize my craft and develop vehicles for a permanent presence on the Moon with an eye to returning to Mars. The thing I like to see most from other people's missions is pictures with numbers, so here are some of my highlights... An early mission to Tsiolkovsky crater, this was before I used limited ignitions and ullage so is a bit simple. Here's my Mars mission, dang was this a lot of work (but mighty fun to finally succeed at). RO has allowed me to really understand why a piloted Mars trip is so much harder than a Moon mission. I used an opposition-class mission for a quick boots-and-flag run, but I swear I will be back in force someday! Here are some of the things I've flown to the Moon recently. Yesterday a flew my newest 2-person-to-the-Moon transport to Mare Moscoviense, I've got to put that at the end. It will be used to rotate crews and resupply the Moon hab. I would enjoy a thread that just shows piloted RO missions, right now I dig through old NASA and NASA-supplier papers for ideas, but I love seeing what others have done. But I don't know if many are made. Check the "MIssion reports" subsection of the "KSP Fan Works" section in this forum every few days. And of course the RO discussion thread. -
I like the lower H2 boiloff rates! Back in 0.90 I'd lose roughly 5% of my H2 in one orbit of Earth, now I lose the same 5% in the typical 4 days it takes to get to the Moon. This means that LH/LOX is now practical for Moon missions! Now I can use the nice selection of H2/O2 engines that RO offers for a lot more, especially the throttlable CECE variant of the RL-10, and the RL-60. So now I've redesigned my old Moon ships to see what LH2 can do. I tried big RL-10 landers but it is awkward with those huge LH tanks, in my largest the crew has to go down 10 meters of ladders to get from the cabin to the surface. (Reminiscent of the proposed Altair lander from the 2000's.) That seems too dangerous to me, as does the high cg at landing when the H2 tanks are empty. Then I found this Lockheed-Martin paper on concepts for landing LOX/LH on the Moon. The best idea to me is to use an efficient but unthrottleable cryogenic motor to decelerate out of Lunar orbit, then switch to small throttleable Aerozine/NTO for the last few hundred m/s, when the ship is much lighter. Ideally the same stage would do the TLI (3200m/s), the LOI (900m/s), and the descent (1900m/s), but the loss of about 220m/s of dV during the flight to the Moon puts the stage dV at 4300m/s before descent, and I found that past about 5200m/s there's usually no further savings. So I can either have the storable (low-isp) stage do the last 900m/s, or split the L2/O2 into 2 smaller stages of 3200m/s and 2600m/s with the landing motor doing about 300m/s. But the extra staging parts reduces the benefit of this. And if I'm landing near an asset I have to worry where the discarded H2 stage falls. It's all fun to test out. I keep an album of my experiments here. The thought of a 'lowest mass piloted Moon mission in RO' challenge is intriguing, but the huge variety of installs we have would make it tough to compare approaches. PS: They call it a "Dual Thrust Axis Lander", but I call it a "Space:1999 Eagle". PPS: I wrote a spreadsheet, "Moonfinder", that gives you the next launch windows to the Moon from any higher-latitude site (including Canaveral). Now you don't have to guess. It's at the end of the FF for RSS first post.
-
I travel to the Moon a lot in RSS/RO, so I made a little spreadsheet to find the best time to launch with a due East azimuth from sites that have a latitude larger than the Moon's orbtial inclination of 28.36 degrees. It can be used for lower-latitude launch sites with a caveat, see below. I've put the Dropbox link in Flyby Finder for RSS's (this thread) first post. A small album showing what it does is here. It only works for Earth's surface to Earth's Moon using RSS for KSP. The summary below is rather long, sorry about that. It is an exercise in spherical geometry, the advantage of the spreadsheet is you can see how I do it, step by step. Summary: Only change the cells in green! Input the desired search start time. Usually this would be the current time in your game. Then input your desired travel time from launch to Moon periapsis. The sheet will look for the 2 launch windows in the 24 hours after the search start time, that will arrive at the Moon at exactly (search start time)+ (minimum travel time)+1 days. Enter your launch latitude and longitude in degrees, note the format, for instance Canaveral is longitude -80 degrees or +280 degrees East, never degrees West. Southern latitudes are negative. I've written some real site locations down in the spreadsheet. The launch LAN fudge factor is trickiest. The sheet computes the LAN of the orbit you need to get into, and knows when your launch site will cross the plane with that LAN. But because you launch with the sideways motion of the Earth's surface your LAN will increase as you are ascending, and the amount of this increase depends on your launch profile, the acceleration of your ship, and your launch latitude. I've found values between 5 and 8 degrees work for most ships. You can figure it out by turning on the LAN indicator in Mechjeb and watching it change from launch to orbit. Every degree in LAN error causes about a 2-hour change in your arrival time at the Moon. The outputs are the two launch times in the 24 hours after the search start time at which you can launch due East and get to the Moon with a prograde-only TLI burn, with the flight time being between "minimum travel time" and "minimum travel time" +1 days. You have to figure out the TLI time and burn magnitude yourself but that is easy by just shifting the burn execution time around. From a 200x200km orbit and a 3 to 4 day flight time the TLI burn should be between 3120 and 3190 m/s. The nominal arrival time at the Moon is also given, along with the distance of the Moon from Earth at that time (smaller makes for quicker or lower-dv flights), the Moon's latitude at arrival time (useful for anticipating your re-entry latitude when returning to Earth) and the latitude on the Moon where the Sun will just be rising when you arrive, useful if you wish to land just after local sunrise like Apollo did. Note that sunrise is only accurate for the Moon's equator, the RSS Moon has seasons just like Earths (unlike the real Moon), if you are far enough North or South the Sun doesn't rise at all for extended periods. (This is a result of the KSP engine and RSS can't change it). I'm thinking about giving the 'season' of the Moon's North pole at arrival, but it will be the same as Earth's season so you can just look at the time of year (day 1 is January 1st) and guess it. Note that this program can work for launch sites close to the equator, but only of you will arrive at the Moon when it is closer to the equator than the launch site (which it is sure to be at least twice a month!). You can also launch straight into the Moon's orbital plane from near-equatorial sites but you will have to figure out those times on your own. Those will not be due-East launches, and this spreadsheet only finds due-East launch flights. This is all practice for defining the best launch time to get to flyby departures, which is a similar problem.
-
I've used flybys of Mun many times when I was trying to reduce the dV needed to get somewhere to an absolute minimum, normally in a challenge. The dV benefit varies depending on where you're trying to go. Summary follows, note I only give links to the albums because they are large, I try to give enough detail to allow others to copy what I do. Note that these were done in earlier versions of KSP, but no flight paths outside of Kerbin's atmosphere have changed for any of these. The most benefit I ever got from one flyby of Mun when traveling from LKO(75x75km orbit) to Eve is 87 m/s. It reduced what I need from 1068m/s to 981 (initial boost) plus 4 m/s of course corrections. The course corrections make an exact number difficult because you need more of them when using Mun but it is hard to get a repeatable number. Here is the album showing that flight. Note how low the Mun flyby was (180m above the surface!), which is why I think this is an absolute maximum gain for one flyby. http://imgur.com/a/cT18z On more typical flybys at 10-20km altitude the benefit is more like 80 to 70 m/s (still to Eve), and I've had to make extra corrections of up to 20m/s. If you are going straight to a further planet like Jool then the benefit is smaller because you are going faster and spend less time in Mun's SOI. You can make multiple passes of Mun to get a larger benefit. In the mission below I used 5 flybys of the Mun to reduce the dV needed to get from LKO to a Duna encounter to 878m/s. Normally this would take about 1050m/s, so about 170m/s savings total. http://imgur.com/a/cNG8P You can even use fybys of Mun to get you to a low Munar orbit from LKO for less. Normally going from a 75x75km Kerbin orbit to an 8x8 Mun orbit would take 856m/s+4m/s corrections+273m/s=1133m/s. Using the path in the album linked below I did it for 845+ 15 +209=1069m/s. It's an awful lot of work for a 64m/s savings though. You can follow the path in reverse to get home for the same savings, I do that later in the album. http://imgur.com/a/wxGvv You can get from low Minmus orbit to Kerbin for about 110m/s by flying by Mun. Doh... I think it's about 160m/s for a straight shot, but I only have an old note for that. So those are the savings, the problem with Mun flybys is the high precision required and the extra course corrections that leads to.
-
I was thinking about a problem like this recently. Though I wonder if it will really be necessary to set the periapsis and apoapsis values to the nearest meter per second. I see the contract gives it to that accuracy, but that's less than one part in 100 billion and I do not see how you can ever set those values that accurately. If you manage to get your periapsis to 108,006,024,920 meters then you would have to go 3322.77184373 m/s to have an apoapsis of 111,782,138,716 meters. If you go, for instance, 3322.77184372 m/s your apoapsis will be 111,782,138,715 meters. Maybe if you use one little RCS thruster on a thousand-ton ship? Anyway, here's how to ballpark the numbers. Pardon me while I do some mental calisthenics. Let's work backwards. Ra= apoapsis radius, the furthest distance from the sun of the desired orbit. Rp=periapsis radius, the closest distance from the sun of the desired orbit. a=the semimajor axis of the desired orbit= (Ra+Rp)/2, in this case 109,894,081,817 m. The further from the sun you are the slower you go, so we will do the do the big plane change and periapsis change at the apoapsis of the orbit. How fast do we need to go there? u=the gravitational constant of Kerbol, from the KSP wiki that is 1.1723328E+18 m^3/s^2. For all points in an elliptical orbit the speed, v, = sqrt(u*((2/r)-(1/a))). I'll call this "Formula 1". So the v of your target orbit at r=Ra is 3210.525247 m/s. Let's round to the nearest meter per second from here on, though you can see you might need to go to the nearest nanometer per second in the end... Now we need to find the transfer orbit from Kerbin to an apoapsis of 111782138716 m,= Ra. Kerbin's orbit is a circle of semimajor axis 13,599,840,256 meters, so we will use this as our transfer orbit Rp. The semimajor axis for this is (Ra+Rp)/2=62,690,989,488m. Using the Formula 1 from above we find that the v at Rp is 12398 m/s, and the v at Ra is 1508 m/s. So when we reach the target apoapsis from the sun we need to increase the speed from 1508m/s to 3211m/s and change the direction by 62.9 degrees. (Kerbin's orbital plane is inclined 0 degrees, and the Hohmann orbit we will transfer from Kerbin in will also be inclined 0 degrees, so the plane change will equal the inclination of the desired final orbit.) It is much more efficient to make both changes at once, so now we have a problem in trigonometry. Draw a triangle with on leg 1508 units long, and a second leg 3211 units long that is at an angle of 62.9 degrees to the first leg. We need to figure out the length and direction of the third leg, since that leg is the burn we need to do. From the law of cosines we find the mystery burn will be 2859m/s. Pythagoras can tell you what the prograde and normal components of that burn will be. Now the final problem is the departure from Low Kerbin Orbit. Kerbin's orbital speed around the sun is 9284.5m/s, so you need to leave Kerbin's SOI at 12398-9284=3114 m/s. If we can find the semimajor axis of the hyperbolic departure orbit from Kerbin we can find the velocity at any distance from Kerbin of that orbit. We'll need another formula: a=(r*u)/(2*u-r*v^2) in this case r is the radius of Kerbin's Sphere Of Influence, which from the wiki we find to be 84,159,286 meters. u is Kerbin's gravitational parameter, from the wiki it is 3.5316E+12. v is the 3114 m/s of speed we need to have at that SOI. The a therefore = -367,375 meters. I don't know if you have used these formulas before, if not don't worry about the negative value, that is how the semimajor axis for hyperbolas are given. In any case we can now use this value for a in Formula 1 and the value of 'r' will be whatever orbit you like to start from. I usually leave Kerbin from a 75x75km altitude orbit, but you must remember that Formula 1 needs the distance from the center of the planet, not the altitude above it's surface! Therefore I would use 675000 meters for 'r' in Formula 1. I then find my v in the 75x75km orbit needs to be 4481m/s, since the circular speed at that altitude is 2287m/s I will need a 2193 m/s burn to get to the plane-change burn point. In summary, starting from a 75x75km orbit around Kerbin, you will need to make two burns- a 2193m/s burn that gets you out to 111,782,138,716 meters above Kerbol, and then a burn at that point of 2859m/s that puts you right into the desired final orbit. Total = 5052m/s. In practice I'd allow for a few hundred m/s of correction burns, because there is no way you will be able to do those burns precisely enough! Note that that second burn must be made at the exact Longitude of Ascending Node of 63.7 degrees. How do you find this? Well, in our Hohmann transfer orbit from Kerbin to the plane change we will travel exactly 180 degrees around the sun. This means you must leave Kerbin when it is at 243.7 degrees. To figure out when this is we note from the wiki that Kerbin starts at a Mean Anomoly of 3.14 radians at 0 UT. 3.14 radians= 179.9 degrees. Kerbin's orbit is a perfect circle, so it always moves at the same speed. It makes one orbit of Kerbol (=360 degrees) in 9,203,545 seconds, so we can see it will move 243.7-179.9= 63.8 degrees in (63.8/360)*9203545= 1631072 seconds. So leave Kerbin at 1,631,072 seconds after the game starts, or every 9,203,545 seconds after that. Note that you could use a flyby of Jool to throw you into the right plane and out to the target apoapsis. This would save you about 100m/s on the leaving-Kerbin burn and about 2km/s on the 2nd burn. But that would be much more complicated to set up. You would need to flyby Jool when Jool is at 63.7 degrees, but determining what energy you need to arrive at it with is tricky. Considering that I think it's impossible to get into that one-part-in-100-billion orbit I wouldn't spend the time on it...
-
Physics help - gravity assists
PLAD replied to openatheclose's topic in KSP1 Gameplay Questions and Tutorials
I think this question can be cooked, because it contains a fatal flaw which Snark touched obliquely. If it says you will leave going back the way you came after the flyby, I read that as saying the planet will turn your path exactly 180 degrees. There is only 1 type of path that will turn you exactly 180 degrees, and that is a parabola. With a parabola your v infinity is always exactly zero. Therefore the ship will end with the exact same velocity it had before the flyby, which is zero relative to the planet! Now I see the contradictory word "hyperbolic" up there, which would mean the departure velocity vector would not be exactly 180 degrees from the entry vector, but if our entry speed was, say, 1 m/s then we would end going, say, 179.9 degrees from the start direction at almost 2m/s. Now, as Snark pints out, if we keep increasing the entry speed until it equals 4 times Earth's velocity around the Sun (where in the galaxy did it come from?) then the ship will have a v infinity relative to Earth of about 120000m/s and it will turn nowhere near 180 degrees. I think with a lowest possible flyby of 200km (avoid the air!) the ship would be turned about 0.5 degrees. In that case your final velocity would go from 4VE before the flyby to 4VE after, in pretty much the same direction, with a little extra either towards the Sun (if you flew by outside Earth) or away from it (if you flew by inside Earth's orbit). The test maker should have postulated an Earth-mass black hole as the object to be flown by. A flyby altitude of about 2 meters above a 1-cm Earth-mass black hole from a V-infinity of 120km/s would turn you 179 degrees. Oh wait, forgot the relativistic effects... After some rough spots I usually came to an understanding with my physics teachers. I remember a quiz where one asked what the path of a thrown baseball was (neglecting air friction), and I answered "a little piece of an ellipse". -
Herbal space program, that is brilliant. I had looked at the first encounter speeds for a Moho periapsis arrival versus an apoapsis arrival, found periapsis looked lower, and left it at that. Your realization that the periapsis decrease resulting from a flyby follows Moho's decreasing distance from the sun... that is freaking brilliant. That does make it sound like the apoapsis arrival is better. You might have to do more flybys of Moho to slow down enough, but if you are saving 20m/s or so per correction it could add up to the superior solution, especially when the lower departure-from-Kerbin speed is added in. I am sorry about your Kraken attack, I look forward to seeing your final results! Francois424 and Warzouz, I think Moho's sidereal rotation period is 52 Kdays, but since it's year is only about 102 Kdays long the solar day from a point on Moho's surface is longer than it's year, about 123 Kdays. (Kday=6 hours)
-
Landed my largest vehicle ever on the Moon in RO. 5331 tons at launch, 303 tons in LEO, 52 tons on the Moon. It's basically a Nova with 8x F1's at launch, but an upscaled Altair-type lander using the throttleable CECE variant of the RL-10 to land. I'm extra proud of this one because I couldn't use Mechjeb for the landing at all (it can't handle ullage or limited ignitions) so I brought it all the way down manually, a real challenge for me what with the low TWR and dV available. I am still stoked over this one. Wooo!
-
The other day I was building a large craft, the biggest ever RO first stage for me. After testing it with seven F-1's I added 2 SRB's to it and the framerate plummeted, then picked up noticeably as the SRB's burned out. For that one I changed a Smokescreen parameter and I never saw the problem again. I used Toolbar to change the Smokescreen config item 'maximumActiveParticles' from the default ( I think 5000?) to 100. I admit I didn't change it back to 5000 afterward to prove it was the problem, but it's still working great with 9 F-1's and 4 SRB's.
-
WeirdCulture, I cannot stop laughing at that "Kerbal" class ship. Oh my god. And the precise '362 t' rating. Now that is a Kerbal rocket!
-
The problem of your solar periapsis getting lower than Moho's after each flyby is a big part of this approach, since it limits how low your Moho encounter speed can get. I think you have to do Deep-Space Maneuvers to raise your solar periapsis back to Moho's orbit about 180 degrees after each Moho flyby. This is what defeated me though, I could not make the DSM's efficiently enough so the corrections cost more than the arrival speed savings. Metaphor did do them well. It is interesting to look at details of the Messenger probe's path, [URL="https://www.youtube.com/watch?v=otF2FjpCyZk"]here's a nice video[/URL]- you can see that they did a DSM a little over 180 degrees after each Mercury flyby. Looking closely I think the DSM is done halfway between 180 degrees from your pre-DSM periapsis and 180 degrees from your post-DSM periapsis. (It's also fun to see them 'stepping down' from Mercury resonance to resonance as the probe's SMA drops after each flyby.)
-
[quote name='herbal space program'][FONT=Times New Roman][SIZE=3][COLOR=#000000] [/COLOR][/SIZE][/FONT][FONT="Arial"][COLOR=#000000]Alas, it is slow going. If I do ever manage to beat Metaphor, it will not be by much, but I think there may be a slightly cheaper path this way. I’ve posted some pics of the first part below. What I found was that if I ejected from Eve’s descending node with Eve at like -25 deg, for around 100m/s more than the minimal Hohman transfer (~1140 m/s), I could set up an encounter at or very near the Eve-Moho ascending node. This in one flyby allows me to do nearly all of the plane change and to drop to a perfectly Moho-tangent orbit that is very close to 2:1 resonant. My biggest problem right now is I’m arriving at the wrong time. There’s also still 0.5 deg of orbital tilt because I’m not ejecting from exactly the right spot. I’m hoping that if I move my Eve intercept to the correct position, it will still cost about the same and perhaps put me in a better phase relationship with Moho. We’ll see…[/COLOR][/FONT] [/QUOTE] Thanks for the pictures, I see what you are doing. There are 2 points where Eve crosses Moho's orbital plane, 180 degrees apart. In my best missions I encountered Eve at the Moho-plane-crossing that is roughly 180 degrees from Moho's perihelion. I figured that it was better to encounter Moho as close to the sun as possible so that the Oberth effect would reduce the encounter speed. However, that meant that a single flyby of Eve was not enough to drop the orbit all the way down to the Moho perihelion. You are encountering Eve at the Moho-plane-crossing that is roughly opposite Moho's aphelion, so a lower-energy start from Kerbin can make it there after just one Eve flyby. I never ran all the numbers to prove which way is better, but my tests seemed to suggest that the meet-Moho-at-periapsis gives you a lower speed at your first Moho encounter. I could be wrong. My criteria for judging the first Moho flyby is the speed you are going at the moment you first enter Moho's SOI, in Metaphor's winning run that was 1584m/s (orbital). The higher this number the more times you will have to flyby Moho, correct your solar orbit, flyby Moho... Since every correction costs dV, the fewer you have to do, the better. Even at 1584m/s Metaphor didn't brake into orbit until his 9th Moho encounter. And yah, timing is a bear when dealing with Moho. After my 2nd Eve flyby I had to circle the sun 5.5 times before getting the first Moho flyby because I just couldn't time the flyby accurately enough to get the Moho encounter right away. Here's a link to a picture showing the shape of my Eve-Moho orbit, so you can see where the Moho encounter is relative to its perihelion: [url]http://i.imgur.com/SYZjn7g.png[/url] For your comparison, I figure that Eve first crosses Moho's plane on Y1 D14.22 (roughly opposite Moho peri) and Y1 D46.55 (Roughly opposite Moho ap) and every 65.486 days after those (Earth time!). Let's see, in Kerbal time I think that is Y1 D53.9 and Y1 D183.2 and every 261.94 days after those.
-
[quote name='herbal space program'][FONT=Times New Roman][SIZE=3][COLOR=#000000] [/COLOR][/SIZE][/FONT][FONT="Arial"][COLOR=#000000]After working away at it for like 5 hours last night, I’m pretty sure I can just edge out Metaphor’s record of 1700dV from LKO-LMO (if I don’t run out of patience), but it really is quite a pain in the end that should not point towards space. The whole key will be finding a path around Eve that is close enough to the Moho ascending node to minimize subsequent plane correction, but still intersects Moho’s orbit in a way that generates an encounter with minimal additional dV. Right now, I’m hitting Moho’s orbit at almost exactly 2:1 but 90 degrees out of phase with it. That unfortunately puts me in a spot where I need to burn for a good 250m/s to get an encounter even 5-6 orbits later. Doing that last night, I was actually able to hop down the next four resonant orbits: 11:6, 9:5, 7:4, and 5:3 for under 80m/s total (!). Going up to 12 orbits, I’ve now plotted 13 more resonant energy levels I could hit on the way down to Moho insertion. I’m pretty sure that with enough patience, I can find a similarly cheap path through those that wlll leave me with a trivial insertion burn, albeit 15 years after leaving Kerbin. As my current save stands, in 5:3 resonance with 1580dV expended, I think I could make LMO for around 1900 total, which is pretty darn good, but to beat Metaphor I’ll need to nail that first Eve encounter…[/COLOR][/FONT] [FONT=Times New Roman][SIZE=3][COLOR=#000000] [/COLOR][/SIZE][/FONT][/QUOTE] I find your post most intriguing. As one of the entrants that Metaphor beat so soundly in[URL="http://forum.kerbalspaceprogram.com/threads/74375-Lowest-Delta-v-to-Moho"] that challenge[/URL], I've often thought about how to beat his score. I determined when Eve crosses Moho's orbital plane, my plan was to find a Kerbin departure that arrives at Eve at one of those times, and then use the double-flyby path and then Messenger-style braking to get to Moho cheaply. Unfortunately I found that if you arrive at Eve with the lowest possible energy from Kerbin, you do not have enough energy to get flung to Moho's orbit without more burning. And the windows are so tight that you can count on needing many orbits of the sun between the second Eve flyby and the first Moho flyby, so I can't use my applications to figure them out. I sometimes consider a new 'lowest-dv-to..' type challenge for all the planets, but we'd have to agreed on a standard start orbit at Kerbin and final orbit at the target (or go surface-to surface like the old challenge). In any case I'd love to see your solution for a low-dV trip to Moho, especially if it beats Metaphor's!
-
Getting Tylo to capture me in Jool orbit
PLAD replied to More Boosters's topic in KSP1 Gameplay Questions and Tutorials
There was an [URL="http://forum.kerbalspaceprogram.com/threads/117415-SSTO-to-laythe-and-beyond"]SSTO to Laythe and beyond [/URL]challenge a while ago in which we used lots of flyby tricks to get from LKO to a Laythe encounter. Red Iron Crown came up with a beautiful Tylo capture maneuver, which I later copied. Here is the screenie showing the setup for a Tylo flyby to capture into Jool orbit. My ship comes from below on the dotted line, and since it is going faster than Tylo, comes up from behind and inside and passes around front to get thrown outward into the elliptical orbit that later encounters Laythe.: [IMG]http://i.imgur.com/QOIEhEh.png[/IMG] Note some details: -The Tylo encounter is made while moving almost tangentially to Tylo's orbit, and moving in the same direction around Jool. This way you reduce the encounter speed to the lowest possible value and enable Tylo to make the greatest change to your path. -I executed the 4.2m/s manuever almost 3 Kerbin years before the encounter. At this distance out you can easily adjust for when Tylo will be where you need it to be. For instance if you make a 1m/s retrograde thrust (so slow down 1m/s), the fact that it is still 30 million seconds to your Jool encounter means you will be about 30 million meters 'behind' when you get there and Tylo will have time to move tens of thousands of Km. You have to compensate for the change to your path that the retrograde adjustment makes by changing your radial and normal velocity a bit as well, but by fiddling with the three you can almost certainly get the Tylo encounter you want. -I needed to use Mechjeb's maneuver node editor to plan the burn accurately enough (to +/-0.01m/s). Precisenode works too. It is true that you can never make a burn that accurately, but what you are doing is finding that the right path is there and getting close to it. You'll almost certainly need another burn later (if you follow the album you'll see I made a 0.2m/s right after entering Jool's SOI) but it will be smaller that the early one. The sooner you get close to the right path the smaller the final adjustments will be! If you want to see the whole flights, RIC's entry in that thread is post #69, mine is post #85. -
How about Moho to Eeloo?
PLAD replied to More Boosters's topic in KSP1 Gameplay Questions and Tutorials
I love running the numbers, and we can run the numbers for this one. I notice Eeloo's longitude of ascending node is at 50 degrees and Moho's at 70 degrees, so their orbits are not too far from coplanar. But... Using Flyby Finder I see that, starting from a 20x20km orbit over Moho, and braking into a 20x20km orbit at Eeloo, the minimum start burn is about 4280m/s, and the minimum braking burn is about 1760m/s. Not with the same path though, the minimum total is about 6860m/s. For a direct Kerbin to Eeloo flight, starting at 75x75km over Kerbin, the minimum start burn is about 2020m/s and the minimum braking burn is about 1010m/s, and the minimum total is about 3230m/s. So going Kerbin-Eeloo is much cheaper than going Moho-Eeloo. I wondered about flybys, but the minimum to get to an Eve flyby from Moho is around 1070m/s, while the minimum from Kerbin to Eve is about 1030, so no sequence of flybys can be better from Moho than Kerbin, though it could be close. -
Interplanetary billiards is a good expression for what we're doing here (ha, I just remembered the Red Dwarf episode where it's used literally..). I have to agree with a point several here have made though, it's often a lot of work for only a small gain. The standard for getting from LKO to LMO is about 857+273=1130m/s in 8 hours, using every trick I could think of only reduced that to 1066m/s, and the trip took 322 hours! The way back is a better gain percentage-wise, from the standard 273m/s in 8 hours to 229m/s, but took 276 hours!. The only practical reason I would try it is to beat a challenge or try to get an underpowered ship there and back. For pretty much anything that stays in Kerbin's range of influence, flyby tricks and bi-elliptic paths won't save you more than about 100m/s. Perfecting your ascent from the KSC to orbit can make a bigger difference than that. Properly using the Oberth effect can save you more than that. That being said it can make a huge difference when going to other planets. Cutting a trip to Jool from 2000m/s to 1050m/s is a huge gain. It is still quite tricky and takes me a couple hours in real-world time to pull off, I suppose it depends on whether you enjoy the journey more than the destination, as it were. I personally love poking the laws of physics in the eye so I'll sit for hours planning a path and then nursing my ship through 0.2m/s course corrections to make an optimal flight. Have you seen the sneaky, sneaky trick the Juno mission is using to get to Jupiter? It can be simulated in KSP. You can cut nearly 500m/s off of a trip to Jool by using only a Kerbin flyby! A sharp KSP player posted a guide [URL="http://forum.kerbalspaceprogram.com/threads/122780-Juno-style-Kerbin-fly-by-to-Jool"]here[/URL]. In summary, if you want to learn flybys it will take some study and a lot of practice, but if you've got the time it can be a lot of fun. I've found that KSP is accurate enough to simulate the way it would work in real life quite nicely, indeed, most every path I find in stock KSP can be used similarly in the Real Solar System mod and the real world as well (allowing for the different planet arrangements). If you have some time you could check my Imgur album (linked to in post 20 of this thread), not only could you see how I started slow and gradually got better at it, I try to give enough details of what I'm doing to allow the reader to do the same thing. The guide for (blush) [URL="http://forum.kerbalspaceprogram.com/threads/80978-Flyby-Finder-for-KSP"]Flyby Finder[/URL] shows how to find, set up, and start a flight too. Oh yes, and you will absolutely need to use a node editor like Mechjeb's or [URL="http://forum.kerbalspaceprogram.com/threads/47863-1-0-2-Precise-Node-1-1-3-Precisely-edit-your-maneuver-nodes"]Precise Node[/URL].
-
When I really want to cut the dV required for a Mun mission I use something similar to that, it's basically a bi-elliptic transfer with a flyby thrown in. In the attached album I show my lowest-dv path ever from Kerbin to Mun and back. I did it in back in KSP version 0.23.5, when the Kerbal-X simply could not do a Mun-and-return mission, I was seeing how close I could come without refueling. (Then I refueled to get it home). You can see that starting from an 9x9km orbit around Mun a 207m/s burn takes you out to a 58,000km Kerbin apoapsis where a 20m/s retrograde burn give you a low flyby of Mun that flings you all the way to Kerbin's atmosphere with no further burns necessary. I did need a little correction burn so my total from LKO to Kerbin was 229m/s total. For comparison I had the rescue craft run a standard Mun-and-return so you can see a direct path cost 273m/s, as Slashy says above. The path you showed doesn't go out far enough, so the flyby of Mun doesn't lower your periapsis to Kerbin's atmosphere, I think you'll find adding just a couple more m/s will get you there, since just an extra 1m/s burn in LMO will add thousands of Km to your Kerbin apoapsis. There are very delicate adjustments of the burn sizes though, for those trying it a node editor will help a lot. And take care to aim for encountering Mun after Kerbin periapsis, not before, as I detail in the album, the path predictor gives a false solution otherwise! The album is rather large, sorry about that, I also detail a very low cost bielliptic way of getting from LKO to LMO using numerous Mun flybys to reduce both the LKO departure dV (from the standard 857 down to 845m/s) and the Mun orbit insertion burn (from the standard 273m/s to 208+13=221m/s). Plus the rescue mission thrown in for comparison. [IMGUR]wxGvv[/IMGUR] You will note that I could have reduced the return dV a bit more by using the same multi-Mun flyby path I took to get there, but I didn't have the patience to try that.
-
I'm studying T-K now, the real outer bodies have very high eccentricities and wouldn't be right for FF, so it doesn't look good.