Jason Patterson

Ug. After an awful lot of work, I managed to get my vessel to Tylo in IVA and put it into a low orbit, ready for landing. I was trying to figure out where I wanted to land (it's hard to figure out where you're at when the world under you is spinning and you're headed around it backward.) After a quicksave/quickload, the brief bit of external view of the ship showed that my vessel simply fell apart - some sort of random game bug. I tried reloading an earlier version of the game and got the same thing, the ship is basically sliced in half. It's incredibly frustrating to have this happen after putting so much time into it. I was amazed at how well I had done with regard to my proposed delta-v budget. I had only exceeded my planned usage by about 500 m/s by that point, and I had built in 3-4km/s of excess. I don't have it in me to give this another attempt, sorry.

You really should practice aerobraking; it's a vital skill for interplanetary returns. That said, if you do it carefully, you should only have a few hundred m/s of excess velocity upon reaching Duna. It's hard to say where you're losing delta-v though, without monitoring your flight. 1. When you launch, how much delta-v are you using to get to LKO? If it's more than 4700 m/s, you need to work on your ascent. A skilled pilot can make it to orbit in a well designed ship for about 4400 m/s, but between 4500 and 4600 m/s is a much more realistic goal for most vessels. (~4600 m/s used) 2. Once you're in your parking orbit (which should be as low as possible), set up a maneuver node to plan your ejection burn to Duna. This will allow you to correct for any inclination differences and fix any timing mistakes in advance. It also gives you a good idea for how far in advance you'll need to start your burn to get there when you want to. 3. Make your burn in map view as much as possible. Sometimes you get to the target for less delta-v than expected, some times it takes a bit more. I usually delete the maneuver node 10-15 seconds before it finishes so that I can see what my orbit is actually doing without the node's interference. As you get close, throttle back and watch the close approach indicators. Stop burning when the indicator tells you you're starting to move farther from the target instead of closer (shocking, huh?) (~1100 m/s) 4. Timewarp until you're out of Kerbin's SOI, then set up a maneuver node to fix your encounter, getting your periapsis as low as possible (preferably under 250km at this point, but anything under 2Mm will do.) (~50 m/s) 5. Timewarp again until you're ~10 days out from Duna, then finally get your Duna encounter just right (use a maneuver node to plan it if you're uncertain in which direction to burn based on the map view and try to get your periapsis to ~70km if you're not aerobraking.) (~10 m/s) 6. Once you enter Duna's SOI, burn at the retrograde direction's heading, but with zero pitch, until your periapsis is where you want it. It should already be very close, but it will shift slightly upon changing SOI. (~10 m/s) 7. Timewarp until shortly before periapsis, then burn retrograde to capture. (~300 m/s) (Alternatively, set your periapsis to ~12km and enjoy the show as you coast through the atmosphere. Be sure to burn prograde at your apoapsis so that you don't make a disastrous second trip through the atmosphere.) Total delta-v - 6200 m/s, give or take

A couple of caveats - Your station will have to be roughly symmetrical. If not, it's going to be a nightmare putting rockets underneath it and actually getting it to orbit. Your station will have to be fairly rigid. If not, it's going to flop over and fall apart en route. Unfortunately this usually means struts, which tend to look ugly in the end. Your station will have to be sufficiently strong to support itself under acceleration. Once you've got a suitable station, a relatively simple rocket can get it into orbit. You might be surprised at how few parts it takes. Something like an LV-T45 attached to an X32 fuel tank makes a reasonable last stage. It will have far less delta-v than a similar stage using an LV-N, but the gain in TWR is nice if you have to use it to finish circularizing your orbit. Around that, use 4 stacks of X32 fuel tanks with mainsails on the bottom. Add as much fuel as you can before you push the thrust to weight ratio very far below 2. Use fuel lines to connect the stacks of mainsails and the central stage, asparagus style. Keep adding layers of X32/Mainsail stacks, using as much fuel as you can and keeping your TWR at around 2 for each stage. Connect it all to make an asparagus pattern and you're golden. You ought to be able to get sufficient delta-v to reach orbit with 3 layers of boosters without having to fly the thing perfectly.

If your TWR is too low you just can't make an accurate ejection burn (at least not without a whole lot of additional planning.) It also leads to additional gravity drag (since you're burning upward) and a loss of the Oberth effect (since your velocity will be decreasing as you climb during the burn, or at the very least it will be increasing very slowly.) Great Isp can definitely overcome the losses in the long run, but if you can spend a fraction of the delta-v and save yourself a couple of hours, why not do it?

Nope, it's not. If you look at the .craft files, you'll see that they are put together substantially differently from the vessel information you find in the persistence file. Would it be possible to write something that could do the conversion? Of course. Is it practical? Nope.

It's said on here that ascending at terminal velocity is the most efficient way possible. Going faster increases aerodynamic drag and going slower increases gravity drag with a balance of the two being most effective. I haven't done the math myself to demonstrate that the precise balance between the two is most efficient, but it's not unreasonable. In order to make this happen you would want your thrust to weight ratio to always be 2 in an atmosphere. Half of the thrust opposes gravity while the other half accelerates the ship up to terminal velocity. Practically speaking, anything from 1.8 to 2.2 does pretty darned well, and the TWR varies somewhat as the fuel in the boosters burns out. In space your TWR doesn't really matter, as long as it is high enough to get your ship moving in a timely fashion. Something like 0.2 works just fine for interplanetary burns on most missions. Isp is much much more important. When your ascent speed is limited by aerodynamic drag, you want a TWR of roughly 2. Once you are above an altitude where that is the case, you can have a TWR as high as you like (as long as it doesn't bring you up above terminal velocity, anyway.) On Kerbin you can pretty much open up the throttle above 15km or so, since it's not likely you'll be ascending at terminal velocity at that altitude on a normal rocket launch.

Getting to Duna's surface and back takes 7000-8000 m/s of delta-v, assuming aerobraking at both ends (which you should definitely practice doing, it's massively useful.) In addition to practicing aerobraking, one of the most important skills you can learn in KSP is how to build minimalist vehicles. It may not matter that you launch a 1000 ton monster to get to the Mun and back, but when you want to do more delta-v intensive missions, figuring out how to get all of the delta-v you can out of your ship becomes a priority, especially once the part count gets high and you can hear your computer's sad little cries for help. If you haven't already, downloading a building mod (like Kerbal Engineer Redux or Mechjeb) can really help with figuring out how to plan out your ship prior to launch. Once you've got a ship that ought to be able to do the job, it's time to actually get there. That involves good piloting skills, which can really only be improved with practice. Though watching videos (or watching an autopilot fly your ship) can help you figure out how to get around in space cleanly, it takes actually doing it several times to get it right. Here is a link to a fully functional Duna return vehicle. It has 74 parts and masses something like 65 tons (for the minimalists out there, yes, I fully realize that this can be done in a much much smaller vehicle, but this is reasonable.) I built it with ~8800m/s of delta-v, so there's plenty of wiggle room for messing up. The staging should work as it is, but if you decide to fire all three chutes at Duna instead of only the first two, you can repack them and deploy using action group 10. The ladder toggles with action group 4. If you're flying this well, you should be able to get into Low Kerbin Orbit (LKO) using the entire first two stages and about half of the fuel from the third. The remaining fuel in the third stage will get you to Duna and allow about 900 m/s of additional maneuvering if you miss or come in too steeply and need to use the rockets to brake rather than relying primarily on the parachutes. You should go to the fourth stage before landing though, or you'll wind up lithobraking. The fourth stage (just a command module with control stuff, parachutes and legs, an FL-T400 fuel tank, and an LV-909 engine) has enough power and delta-v to get you into Duna orbit and back to Kerbin with ~800 m/s to spare, if used wisely.

Specifically, of the information in your orbiting vehicle, you want to save the parts in bold. If you retain that and copy all of the remaining information from the vessel on the launchpad into into the broken vessel, it should work perfectly. VESSEL { pid = 943bb1ac193542bfa4ea1a62055a29ef name = Your Ship Name Here type = Ship sit = ORBITING landed = False landedAt = splashed = False met = 64355066.9567176 lct = 611978392.092731 root = 0 lat = -0.00320013195567924 lon = -121.098734135029 alt = 105664.194711881 hgt = -1 nrm = -0.9301952,0.002975476,-0.3670533 rot = 0.001699577,0.6806946,0.0009512879,0.7325647 CoM = -0.0002826358,-6.581337,0.001899925 stg = 15 prst = False eva = False ref = 2193686861 cPch = -0.094 cHdg = -1.882 cMod = 0 ORBIT { SMA = 805774.446879014 ECC = 0.000530133879218946 INC = 0.110768425626073 LPE = 73.1714880235999 LAN = 117.634944641204 MNA = 4.97400456735951 EPH = 676333459.149449 REF = 5 OBJ = 0 } There are other alternatives involving adding a single part to the vessel, but that's a real pain in the butt to do.

If you toggle SAS on the runway does it lock the flaps upward? Have you actually seen the flaps going up or does it nose up sharply as if that is what had happened? Does it only happen after you've used the rocket for a while, or does it happen immediately after liftoff too? (It looks like your CoM may shift behind your CoL as the fuel drains, since the plane is so lightweight.) Other thoughts: Have you got a joystick/controller pad plugged in that could be sending phantom signals to the plane? Have you accidentally set the trim upward on the flaps by pressing alt-up repeatedly (fat fingering it, presumably) while playing?

No, he certainly popularized it, but he was nowhere near the first to come up with the idea of a geostationary orbit (which is what a Clarke orbit is.) I don't remember who was the first, but I bet google knows... Arthur C Clarke wrote about worldwide communications satellites in this type of orbit in 1945 in Wireless World magazine. Prior to that, George O Smith wrote a series of scifi stories called the Venus Equilateral series that featured communications satellites in geostationary orbits (as well as satellites placed at the Lagrangian points.) Clarke apparently wrote an introduction to the collected stories noting the idea. Prior to that, in 1928 Herman Potocnik wrote about inhabited space stations in geostationary orbits. According to wikipedia this was also the first description of the architecture of a space station. Prior to that, Hermann Oberth (yes, that Oberth) wrote about satellites in geostationary orbits in 1922. Prior to that, Konstantin Tsiolkovsky (yep, that Tsiolkovsky) apparently wrote about the orbit in the late 1800's-early 1900's. I can't find the exact paper where this was supposed to have happened, but there are a bunch of sources that mention it. Clarke's contribution to the thing was to suggest that widespread communication could be done from these satellites. He didn't realize that such an orbit existed, or that it would be a good place to put a satellite, or a space station, or that it would be particularly useful for communications satellites.

No, it's not possible in game without constant monitoring. When you go to non-physics timewarp, move more than 2.2km away from a vehicle, or go to the space center or tracking station or quit the game, the vehicle is 'packed' and 'put on rails.' The relative positions and orientations of all of the parts of the vehicle are recorded and the properties of the center of mass of the ship are used to calculate the orbit that the ship is moving along. The upside of this process is that it allows the game to simulate dozens of vehicles simultaneously without having to actually simulate all of their individual parts. The downside is that the game ignores all physics other than straightforward orbital mechanics (and hitting the ground.) It will not retain the rotational properties of the vehicle, it sometimes turns it in unexpected directions, and it will not perform any aerodynamic calculations for a vehicle in a planet's atmosphere (though it will auto-destruct any vessel that gets too low, ~22km on Kerbin, as I recall.) If you really wanted your vessel to always point downward, you would have to introduce a rotation (an incredibly incredibly slow rotation) to it so that it spun once per orbit, in exactly the same way that the moon does as it orbits Earth. It's impractical to actually do this, since you'd be talking about moving 1 degree per minute.

I'm not 100% certain, but I don't believe that that function, delta-v(Mbo), has an inverse. I think that's what you're after, but it just doesn't look to be possible without resorting to infinite series.

Do you have a question? The first part of the equation is Tsiolkovsky's rocket equation. Is the second part your attempt to include TWR? I'm kind of confused about its purpose. If you're trying to calculate an effective delta-v by subtracting out part of the entire quantity to compensate for gravity drag, this won't work. You'd have to know how long the engines were burning upward and the characteristics of the gravity turn (assuming there is one) and all of that sort of thing. Additionally, as the rocket moves upward, its TWR changes because the force of gravity decreases. On several bodies you'd also have an atmosphere to deal with.

That would have been a neat trick, considering the fact that he wasn't the first to come up with the idea, or even the first scifi author to use it.

You can get UT visible via the debug console. Alt-F12 and a button press turn it on. Screenshot at the beginning of the lap (zero speed, wherever the starting line is) and screenshot at the end. If you want multiple laps timed, screenshot each one.

Good point. It does look like the front gear is turned to the left a bit. Can't load the plane to find out for sure without the mods though.

Ug, I keep topping out at 17.4Mm. I've got another idea, but it's going to have to wait, because my brain is starting to melt from 20 minute long sessions of watching an LV-1 burn at 4x speed.

It was never stated, and so far everyone has done this anyway, but is it safe to assume that we're only using stock building parts for this? Just waiting for the person to come along and tell us how they got an additional 4k m/s using FAR or what have you.

That's exactly what I thought when I first read it, but the challenge is something else entirely. It's not highest apoapsis with this much fuel, it's highest altitude at the moment that the fuel runs out.

I suspect that I know what it is, and I had the same idea and resisted. That said, I'm firing it up right now because I know you're in the contest, you sneaky bugger... ETA: There is definitely room for improvement here, but we'll have to see whether it is better to add staging or less fuel in the last stage (in exchange for greater initial velocity) or whatever on another day. Using an LV-1 gave me 52 minutes of burn time, and I was able to get to 4,778km before it ran out of fuel.

That would be equivalent to 10 FL-T400 tanks.

I was a bit confused about the rules at first, since so many of this type of challenge involves getting the highest apoapsis and the challenger doesn't realize that they've provided more than enough fuel for escape velocity. Then I got all literate and actually read what you wrote... Here is my entry, the screenshot is just before the engines ran out of fuel. 1,463km

Yeah, if the docking port is on the front of the ship it's pretty darned straightforward. This assumes a good view for the command module, of course, which excludes most of the vessels. It's also entirely possible to dock in IVA without any exterior view, as long as you know the orientation of the two vehicle's docking ports (north-south, for instance.) Line up the pink thingy with the yellow thingy while pointing in the right direction and you're golden.

My Tylo lander is now orbiting Jool. I had to guess quite a bit at the ejection velocity to get it there since it wasn't instantaneous, and I wound up guessing quite well. I knew that I was getting close to the right altitude to encounter Jool and so I tried finding it in the sky. I've heard that it's visible from Kerbin if you know where to look, and I definitely could find it from ~20 days out. I was prepared to make some ridiculous burns to force an encounter, but I bumped right into the thing. I managed to get captured, which took rather more delta-v than I had hoped since aerobraking isn't really an option. Right now I'm lowering my periapsis little by little until I wind up crossing paths with Tylo. I might also try a Laythe aerobrake if I get the opportunity. I really don't want to have to kill 3km/s of excess velocity just to get captured by Tylo. I've been filming the entire thing, including the original launch and 5 or 6 docking missions I sent in order to get it fully fueled and to add some supplemental tanks prior to leaving Kerbin, so when I'm done I'll edit and post what I've got. Thus far I have used map view only to set targets, and the normal camera only where IVA has no option to do the activity that I needed to get done (things like fuel transfers and undocking ports.) All of the actual movement of the ship has been done while in IVA mode.

Here are a couple of strange videos that I've hung onto for some time without using. Anyone else hang onto unexpected footage? Post your blooper reels! The Brave Little Booster Why You Shouldn't Stage Early Also, man I had forgotten how OP aerospikes used to be.