Aegolius13

I think Matt Lowne would launch a giant SSTO powered only by Thuds, put the entire station inside a cargo bay, and fly it to Eeloo and back for good measure. But your way is good too.

A more reasonable engine for rough economic times, the Poodle Kestrel engine doubles as a BBQ when at low power.

I'm fine with some simple parts of this sort. We know from the extendable landing legs that the Kerbals have some experience making things that move. Just wish they could adapt the technology a little more broadly. Of course, this is the same species that invented like 6 different kinds of rocket engines before the rung ladder...

I believe the Bobcat beats in it terms of pure thrust. My take is that prior to the 1.6 rebalance, the Skiff was a bit of an awkward tweener engine (not optimized for sea-level or vacuum) that was redeemed mostly by its ridiculous TWR. After the rebalance, it's lost that advantage, and is not really spectacular at anything -- it's a jack of all trades, master of none. It's kinda decent as a core stage when combined with higher-thrust radial boosters, since it has serviceable sea-level ISP and pretty good vacuum ISP. I've also had some luck surrounding it with a ring of Reliants or Kodiaks, which can be turned off or staged off, Atlas style, when no longer needed. I tend to think of it as a slightly bigger and better Swivel.

If you're not familiar with the underlying.. thing... https://en.wikipedia.org/wiki/Nyan_Cat

Interesting idea. My experience has been that if your ship is too pointy, you won't slow down enough by the time you hit the really thick atmosphere, and things will get crazy hot. But doing this, with an inflated heat shield at the BACK of the ship, could be a great way to maintain stability. Of course, there's also the issue of covering up all the stuff in front. And since an uninflated heat shield only covers a 2.5m cross -section, if you never inflate it, you might be better off with the regular 2.5 heat shield. I imagine it's draggier, plus it can probably dissipate more heat through ablation.

I'm gonna go out on a limb and say the SLS's transition from the core stage (4 RS-25 / SSMEs giving over 80,000 kN thrust) to the second stage (one RL10 giving ~100 kN of thrust) is not quite the ideal ratio.

Manley's on it, including a KSP version.

Your assessment is correct. Crossfeed is bidirectional, so outer engines will continue to draw from inner tanks until you take action. To get the automatic shutoff but still allow for crossfeed, you need to use fuel lines (or possibly a mod). See also:

Anyone know why it's taking so darn long to get perigee up to lunar level, especially when the Falcon 9 is putting it into a pretty elliptical orbit to start with? Is it just low enough TWR that it has to do a bunch of periapsis kicks?

Those must be from mods - they're not stock experiments.

I guess it depends in part on what other engines you're using. I.e., if you're using Rapiers, my personal opinion is that it's never going to be worth it to put a Swivel or Reliant onboard. Yeah, the vacuum ISP is a tiny bit higher, but it comes at a very significant mass and drag penalty. Whereas if you have a Rapier anyway for jet mode, its rocket function is effectively free. As a general rule, the further your mission is going, the more important vacuum delta-v will be. If you need so much vacuum delta-v that you need an upgrade over the Rapier in rocket mode, the best answer is probably a nuke. If you're using a lower-ceiling jet engine like the Whiplash or Panther, then yeah, rockets will have more of a use-case. But even then, keep in mind that you're likely not lighting up the rockets until you're well over 10km in altitude. At that point, you can essentially disregard the sea-level performance figures and just look at vacuum. Given that, the Reliant is a pretty bad fit, since it has poor vacuum ISP and no gimbal to help with attitude control. The Swivel is a little better, but still not ideal for this role. You might also take a look at more vacuum-oriented engines: the Terrier, Poodle and (if you have Making History) the Cheetah, Wolfhound or Skiff. As mentioned above, by the time you turn them on, these "vacuum" engines should already beat the sea-level engines on performance. The Spark can work nicely on small planes, and I guess the Skipper is a possibility as well, though I would think the Poodle would have adequate thrust most of the time if you're going 2.5m. The Vector could work if you need a ton of thrust on a 1.25m stack, but its vacuum ISP is not ideal. I'd suggest using fewer, but bigger engines over a lot of small engines. E.g., a Poodle mounted to the back of a 2.5m fuselage might work better than 4 Terriers on 1.25m nacelles. The jet engines might have to go on nacelles, but you don't really have an option with those since they're all 1.25m anyway.

If you can post a picture, that would help to diagnose the issue. But in general, space shuttles in this game are really, really hard. The main reason is the center of thrust does not go through the center of mass. I.e.. you've got a heavy external tank sitting out to the side, which makes the craft want to tip in that direction. To deal with this, as with the real-life space shuttle, the engines have to angle so that they point through the center of mass. But making things worse, the center of mass changes as the SRBs burn out and the external tank loses fuel and gets lighter. I'm not aware of any easy fix to this issue; it just takes a lot of very careful tweaking part placement and the like. Another problem is that the space shuttle analog parts are not balanced correctly, so it's very hard to make a 1-for-one copy. The Kickbacks are way too weak compared to the shuttle SRBS, and the Vector engines are way too powerful compared to the SSMEs. But Vectors are quite helpful due to their very wide gimbal range, so a common solution is to use less than 3. The issue about rolling to the side is a bit more of a mystery -- if all the parts are symmetrical that theoretically should not happen. It could just be a result of your control surfaces and engines trying to deal with the unbalanced torque, or you could have some actual asymmetry.

Of course, even if a toss-up on delta-v, there are other differences to keep in mind. The extra thrust for the Wolfhound usually counts in its favor. On the other hand, a Wolfhound stage would have more total mass, which would decrease the delta-v obtained from earlier stages (though the difference here is quite minor).

Re: the possible subsequent Eve flyby, you'll likely run into the same problem where Eve is not in the right phase angle when you hit its orbit. One option to deal with this is to: Burn from Duna such that your perhelion matches Eve's orbit (i.e., your orbit is tangent to its orbit). Set a maneuver node at perhelion, which will likely be in deep space if the phase angles don't match. Set up a retrograde burn to bring your aphelion in. But mess around with the size of the burn until, on your next pass (or two), Eve is in the right place. This is similar to a regular vessel rendezvous, and to the the "node" method to get to Moho. Of course, you lose some Oberth efficiency on the retrograde burn to bring aphelion down, but it's not that big of a deal, and you can use really low TWR engines. I suppose you could use the same approach to get back to Kerbin, but this will likely take more in-game time. (Since Eve revolves faster, it'll be in the right place sooner).

As @bewing said, the core is a HECS. Nice upgrade over the OKTO in terms of SAS choices, and pretty much renders the OKTO obsolete. Even in late game I like it in this role, since it's light but still has a reaction wheel, unlike some of the super-light probe cores (e.g., OKTO-2). The engine is an Ant - the smallest node-mounted conventional engine. The stats don't look too impressive, but it's well-suited to tiny probes like these due to extremely low mass. It's very easy to put a small Spark stage below it too. EDIT - never mind about the engine. The little engines are Sparks, just with an older model. The Ant would be good for space probes of this size, but these things were supposed to land on Jool's moons so I wanted a little more kick. The bigger engine below the Spark on the middle probe is a Terrier.

You can also do multiple copies side-by-side in a fairing, by using symmetry and a cubic octagonal strut (or similar part) to attach the problems. @Ratwerke_Actual's method is almost certainly going to be more effective if you just want two probes, since it keeps a nice narrow rocket shape. But if you need a bunch of ships at once (say for a Jool science mission or a whole relay network), mixing the two approaches can be helpful.

Agree with @5thHorseman. Delta-v is not a constant property of a given engine, like ISP or TWR. It's the output of a multi-variable equation, so one engine may not always generate better or worse delta-v than another. In the case, the Wolfhound has better ISP and more thrust than the Poodle, but higher mass. So even if the Wolfhound has better TWR than the Poodle (think this used to be true but maybe not after the 1.6 rebalance), that doesn't tell the whole story. On a small ship, the extra dry mass of the Wolfhound brings your delta-v down more than the extra ISP, brings it up. After all, this is why you wouldn't put a Rhino on a tiny space probe, even if it has pretty good ISP. Conversely, when you have a really big ship, the dry mass of the engine is a drop in the bucket, so the extra ISP of the Wolfhound will likely result in more delta-v.

Fuel cells. Personally I find it hard to overcome the stigma of using up a finite resource to generate electricity. But a lot of the time they make for an easier / cheaper / more reliable setup compared to solar panels or RTGs. The fuel cell arrays can also do some pretty good things when combined with ions.

You're correct - your antenna is not working because it has no line of sight back to Kerbin, but that can be fixed with relays. As far as which one to choose, this page has a nice table showing the effective range of different combos of antennae. https://wiki.kerbalspaceprogram.com/wiki/CommNet As a very general rule of thumb, it works well to use the 2gm* for relays on Mun/Minmus, the 15gm for relays at Moho / Eve / Duna, and the 100gm for all other bodies (and for master relays at Kerbin to deal with eclipses and whatnot). There's also an argument to be made for using bigger ones everywhere you can, as sometimes you can get unexpected but helpful transmission paths. You don't need a relay on the craft you're attempting to fly, just on the other antenna forming a bridge. That said, the only disadvantage of a relay antenna is weight/size/cost. They can do everything a regular antenna can do, and more. *You can probably get away with the 5mm one for the moons, but I've just never liked that one.

Here's an illustration of what I was thinking of. It's far from optimized, and doesn't have all parts (no electric supply, etc) but it can get to LKO with at least 4,500 m/s delta-v remaining, which should be plenty for a Mun mission. But it's an agonizingly slow trip. Likely could remove a little LF for better performance, and not lose much / any delta-v by the time you're at LKO.

That's a tough mission profile. You'll almost certainly need nukes. Might be worth a shot to see if something like 1 Whiplash + 2 nukes would get you to orbit. If not, you could consider a couple small rocket engines (Twitch, Cub, Spark on a stack of Oscar-B''s) to help the push to orbit. BigS wings will be very helpful due to the free (mass-wise) LF storage. I'll try and see if I can cobble something together that looks like it can do it.

You need all the normal utility stuff to keep the station running -- probe core, electric generation, batteries, antenna, attitude control (RCS and/or reaction wheels), etc. You probably want some crew modules, just in case. As mentioned above, it's common to store fuel. I'd add every science instrument you have. Even if you've already exhausted those experiments, you may periodically get a contract to perform science at the station's location. If you do, free money! On that note, it's good to add a science container. If nothing else it simplifies the act of collecting science from docked ships. The Mobile Processing Lab is a bit controversial. It can get you a lot of science, but it's fussy and confusing to use, and overpowered if you do get it working right. But in addition to the main function of generating science, it can level up your crew without retuning to Kerbin. I usually include it for this reason, and just for roleplaying I suppose. You'll definitely want docking ports. I'd add quite a few, add different sizes just in case, and try to arrange them so that docked ships pr expansion modules won't run into each other. I'd add some propulsion to the station. Occasionally you may need to do a bit of active rescuing, or get a contract to move the station to another orbit. I tend to radially mount engines, to keep the ends clear for docking ports. You probably don't need much thrust.

I don't have exact numbers at hand (and not sure if the wiki is current), but three RTGs does not sound like nearly enough juice for two mini-drills and three radiators. I've had some weird results as well with leaving drills to run outside of physics range. It can work, but doesn't always seem to. I think someone said at one point you have to leave them for a certain amount of in-game time before the mining registers, but not sure. I wouldn't worry too much about radiator numbers - the more important question is whether your drills are able to hold steady at their ideal temps. If you had just landed, the radiators might have also been rejecting extra heat from your engines or something like that. Loads could also vary depending on whether your ship is in the sun vs. shade. Or it could be a quirk with the mod you're using.

I nothing against a big vertical stabilizer, and I often put a big static triangular wing connector on the back bigger places. The reason I the Big S is overkill is that it has so much moveable area (can't remember the exact in-game term). If i recall, moveable area has around double the weight penalty of regular wing area. And I do think it has the potential for overcorrection. You can always turn control authority down, but then you're effectively paying the mass penalty without a corresponding benefit.