sevenperforce

A theory which, I'm happy to say, has largely been deprecated. It was a terror indeed.

That's how I originally read the rules, which is why I jumped at making an SRB SSTO, but it might be a more interesting challenge to measure cost per kg of payload. Of course, then you have to exclude the cost of the payload, and you have to define what counts as payload, and you have to decide how to handle reusability. If you say recovered parts don't count toward your launch cost, then the only way to be competitive is to use a HTHL SSTO or other fully-reusable solution, and it ends up being a "lowest fuel cost per kilogram of payload" challenge, which is different still. If you say recovered parts DO count toward launch cost, then no one will bother with recoverability.

KABLOOIE!!!!!!!!!

Assuming the target cost/weight ratio is about what's on the launchpad... Here's my successful entry. SSTO & SRB only (Hard Mode x2), 0.164 kerbucks/kg on the launchpad. As you can see, all the thrust limiters are set very carefully to get a thrust and isp curve matching the ideal ascent profile all the way from launch to circularization. I've got a handful of Sepratrons for final orbital adjustment. From there, it was a lot of trial-and-error to get the right ascent. There is very very little control authority on this, so it was a challenge. Full album:

This is tremendous. I love it. Now to figure out a way to get a simple, reengageable bearing so that you can rotate your main engines for that VTOL and save on mass....

Woah, really? Thanks for the tip! I admit, I semi-cheat with this whenever I have a ship running low on RCS fuel. Set SAS to go to the heading I want, tap RCS on to give myself a kick in the right direction, turn RCS off to conserve it, and then hop in and out of timewarp the moment I'm pointed in the right direction. Warp freezes me at the correct heading with no further expenditure of RCS. If you are gaming on a PC with the standard keybindings, tapping period and comma (or period and slash) in quick succession hops you in and out of timewarp almost instantly. Timewarping is also good for releasing payloads or undocked craft or debris safely. Collision meshes are not activated during timewarp. You can eject a payload from a fairing without blowing the fairing if you decouple and immediately hit period, because the payload will simply drift right through the fairing without breaking it. I've also used this to successfully stage off parallel boosters (once in space) that otherwise would have collided.

On Eve, I've found propulsive landing and ISRU to be more reliable and more mass-efficient than chutes.

Clarification: is cost/weight ratio the total expense divided by the payload mass to orbit, or total expense divided by the mass on the launchpad? Because those are two completely different challenges.

All right, got it. 320x101 km. Slightly cheaper per kilogram than before. Now to make an album and get it uploaded.

I did a minimum-mass Kerbin Orbit Rendezvous Mun mission where my entire two-stage direct-ascent lander was also a rover:

Yes, but I'm not going to submit until I actually get 100x100 km.

Currently have a solid-fueled SSTO (score: 160.47 kerbucks per ton) that I've gotten to a 101km x 71km orbit, but I'm not sure if a more careful ascent profile would get me any closer or not.

This monstrosity will get 16 Kerbals and a pilot down to the surface of Eve (even to sea level, if you are a glutton for punishment) and back up into orbit again. Requires pretty steely nerves to fly. Of course, getting it TO Eve is left as an exercise for the reader. Flown properly, you end up back in Eve orbit with more than enough fuel to rendezvous with a tugship, though the tug will need a Klaw or somesuch. An entry burn at Kerbin is recommended; propulsive landing will be necessary but not terribly challenging). If anyone wants the craft file, let me know.

I didn't realize this, but the Mk2 crew cabin is actually more mass-efficient than the Hitchhiker Container, and is significantly less draggy. Of course, the Mk3 Passenger Module edges them both out significantly. That's if you don't want to use command seats.

Yes, the command seat approach works well. My smallest Eve ascent vehicle used three Vectors, a Dart, a Spark, and a command seat, and I had to use the jetpack to circularize.

Good thing you don't like spaceplanes, because they are useless on Eve. Getting 16 Kerbals off Eve is roughly the equivalent of getting the VAB off of Kerbin. It's NOT going to be easy. Your best bet is parallel staging, and LOTS of it. Take it slow. There are no gravity turns on Eve; expect to expend at least the first half of your stages burning straight up. You need a lot of thrust; try using a Rhino as your final stage. Do not try direct ascent, either. You need to dock in low Eve orbit afterward. If you have trouble with your ascent vehicle breaking up on the way down, try coming in empty (or burning retrograde all the way down) and using ISRU to refuel. It really helps. RCS does not work on Eve. You need reaction wheels and steerable fins.

On the subject of Buran -- does anyone wiser than I know what those rectangular structures on the outsides of the Zenit boosters were?

If I hop over to the Jool system I'm sure I can get quite a few flybys before either getting slung out of the system or getting lithobraked. No real way to design for it, though.

In my view, having a thrust vector that intersects the ground at all doesn't really count...otherwise you could build a plane with larger engine, TWR>1, and take off at an arbitrarily low forward speed. The rule ought to be "lowest speed in level flight", so you are staying aloft solely with aerodynamic lift and not with angled thrust.

Two points here. First, you don't really need the Ike encounter, not if you have heat shields. Aerocapture at Duna from a nice clean Hohmann transfer is pretty forgiving. And if you want to make the most of your dV, then you'd want to plan the Ike encounter during your Kerbin escape burn, which is easier said than done. Second, it's easier to get a gravity assist from the Mun than you think. The optimal transfer window from Kerbin to Duna has a Kerbol periapse at Kerbin and a Kerbol apoapse at Duna. So your ejection needs to be roughly in the same direction as Kerbin's direction of motion around Kerbol. Thus, you'll want to get close to your optimal transfer window (which is when Duna is 36.5 degrees ahead of Kerbin; use this calculator for precise help), then let the Mun's gravity fling you out along the correct vector. You can expect to get roughly 90 degrees of turn (with a nice solid gravity-assist kick during) out of the Mun. It would be less if you were burning to, say, Jool or Eeloo, but for Duna that's a good approximation. So simply wait until Kerbol, Kerbin, and the Mun are all lined up, do a standard Hohmann to the Mun, and simply keep burning, with your Munar periapse as low as you dare. Your initial trajectory is directly away from Kerbol, but the Mun will bend it into a nice prograde ejection vector along a Hohmann to Duna. LMK if you need a diagram.

Question (although I think I might be able to work it out for myself): is it actually more efficient to spend the dV to dive to a low periapsis to get more of an Oberth effect? Or does it depend on how much excess hyperbolic velocity you need? My gut is the latter. For example, if you're at Minmus and you only need to leave Kerbin SOI to go rendezvous with a nearby asteroid, it's trivial to show that you'd be better off simply burning straight from Minmus's orbit to escape. However, I'm suspecting that there's a point where a very large burn (like, down to Moho or out to Eeloo) would benefit from the Oberth effect significantly, and so the dv you "waste" dropping down to Kerbin is more than compensated for by the Oberth boost to your escape burn at periapse. No idea where that cutoff is, though. Duna is not too terribly hard to aerobrake at. You'd be better off planning a gravity assist at the Mun than worrying about a gravity brake at Ike.

Just wait until the Mun is going to be tangential to your desired Kerbin SOI ejection angle and do a low-altitude swingby to take you into the Duna SOI. Adjust at Munar periapse to get the Duna periapse you want.

E.g., make your burn, shut off your engines, and see how many SOIs you cross with the perfect multiple slingshot trajectory? Will come down to mods, probably. A bit Traveling Salesman.

Mun swingby, but that's about it.

I had never launched a true STS clone until a few weeks ago, but I had a lot of experience, so I got it on the first try: Want to try a Buran clone soon. One Vector has the launch thrust of 8.7 Thuds. Something to keep in mind.