Rune

Aha! Gotcha, you sonova... ...or, you know, I came really close. It's not my fault, I swear! That ~5º misaligned corridor is totally docked straight, according to the game. Rune. There are also two docking ports that are connected but without undock option, but you know, the ring itself is closed without misalignments, so I'm gonna call it "good enough to prove the concept".

And now the million √ question. Is my docking-fu enough to make it close? Rune. Who am I kidding, I have Docking Alignment Indicator and a SnapDock tolerance of 0.995, should be a breeze... if I built it right.

Yup, that's the one. You'll have to summon Azi yourself for that info, I'm afraid. Rune. Contrary to popular belief, I don't remember everything I've ever read.

Well, the obvious thing is KAS, but if you want to keep it stock, @Azimech did a thing with an ungodly amount of parts and playing with the rigid attachment tweakable that kind of worked like a winch, IIRC. I'm sure he'll be happy to refresh all our memories momentarily! What do you know, a few winglets, and it werkz! With core recovery, even. Given the core is something like 70k√, the boosters 4k√, and the whole thing 177k√ to put on the pad, that is kind of nice, works out to a launch cost of about 7k√ in fuel and 4k√ in SRBs, for the longest ~30mT payload you'll ever see. There is no actual second stage in there, BTW, the core makes orbit, I just got the payload nicely sectioned. Rune. The fins obviously had to go backwards, given the news of the day and the shape of that fairing.

Well I'll be darned, it fits. The only thing that bulges is the Mk3 center section! Now there's the problem of the ~30mT, 35m long payload on top of a rocket... not sure I'll gain much stability by launching tori this way. Rune. It'll sure cut on drag compared to previous versions, there's that...

You quoted the wrong guy

Glad to have inspired such a thing! But in my quest to to true SSTA, and cracking the nut of Tylo, I actually moved away from airbreathers. The jump to kerbin orbit is a bit trickier (not really, just a matter of an obscene mass ratio), but after that, having >3km/s of brute-force chemical power at high TWR, and after that another bunch on nukes, is kind of sweet. Now I'm actually thinking about making do without the wings entirely, it's even more mass-efficient, and this thing doesn't really need to precision-land. Not sold on the shape, though: A radiator would help, certainly (make sure it's bolted to the part you want to cool or the one next to it), but you shouldn't need it if you can manage your vertical speed on reentry. It takes a certain amount of wing per mass to be able to, and certainly the maneuverability to attain high angles of attack on reentry, but I've reentered with Mk1 cockpits on the front, and those have both a pitiful internal temperature rating (1,1000, meaning they need to reenter quickly or they'll soak too much temperature and explode from within), and pitiful skin temperature rating (1,500º, meaning you can't go steep and brake with high G). It wasn't the easiest thing in the world, certainly, but I did it. The how is a lot of up/down movement while you reenter, pitching down to near your skin temperature, then pitching up to brake, go up, and shed the internal heat. Oh, and you absolutely need a pilot to do this or plasma comms blackout will screw you. Rune. An engineering conundrum: winged designs have the highest temp ratings, yet they are the ones that need them less.

Yup, what @Thor Wotansen says is pretty much the perfect explanation for why Mk3 is plain better than Mk2. Rocket parts (and Mk1's) are even betterer, because as you just found out, they have significantly better mass ratios (though you usually get those by dividing wet mass by empty mass, so the higher the better, but your calculation gives the same insight, if you read it right). Wing parts with fuel ratings are also pretty cool, if you need their lift, but don't cut it as tankage for tankage's sake. Rune. A shame we still don't have a a 2.5m liquid fuel tank.

I'm trying to be as cheap as possible here. Throwing four separatrons in there means 300√, and the whole thing is automated... if the guy launching it can manage a circular orbit without switching off the main engine, and without getting the apoapsis above 75kms. Might take a rather good pilot/Mechjeb. I just did, but it takes a lot of fiddling with the throttle, at the very least. The alternative would be a Vernor-based system with a probe core, to use the residuals for the job, and while we are at it, provide rotation authority with the main engine off to allow coasting periods. Much more convenient, as you say, but... the lightest system I can think of is at least 6,000√, more like 10, and heavier. A monoprop-based system could be a nice compromise, tough, with the linear ports at 280√ a pop. 6xRCS, probe, fuel tank plus a couple batteries... (fires up KSP, intrigued) ...Yup, about ten parts and 3,560√. Then again, the thrust is so anemic deorbiting is going to take its sweet time, and it has the same amount of propellant mass as the separatron one... which can barely cope with lowering the orbit of the big core a few kms. Compromises, compromises... Rune. The good part is they are sleek and beefy enough that the 3,5km/s dV budget is quite excessive, and they have some margin to screw up ascent/bolt stuff on top.

Yeah, quick and dirty sounds like the thing for them. Once you have painstakingly designed you amazing-looking mothership, just chuck one of these underneath in about five seconds and be sure it'll make space. I'm debating whether to put deorbit systems now, or trust in the customer to keep orbits clean... Rune. How much weight can a solid-based deorbit system be? I'm betting just enough to upset the payload numbers.

I didn't say it was 5km/s to land on Moho. I literally said: Which means that leaving form a Kerbin orbit and getting into a Moho orbit should be 5km/s. But if you don't believe me, here is the transfer planner info on my next Kerbin-Moho window: As you can see, under 5km/s, capture included. And that is from a 100km LKO to a 100km LMO. If you are starting form Minmus, you can shave off some respectable dV from that, because you are starting in an almost-escape orbit just by taking off from Minmus. Then again, you have to know what those numbers are, especially the ejection inclination and ejection angle, and how to get them cheaply. Which is why I told you to check your navigating. Landing on Moho should add a bit less than a km/s to the budget, which is about the same that can be won by taking off of Minmus. With 6km/s, you should have a healthy margin. Rune. I notice we are no longer maintaining Mk2 is perfectly fine.

dV, absent TWR considerations, is driven by tankage ratio (the coefficient between wet and empty mass). Because Tsiolkovksy. A Mk2 Fuselage has a way worse tankage ratio than a Mk1 tank... and the wings that would give it several times the lift rating of a Mk2. Not to mention the drag issue, which means you need much more airbreathing engine weight for a given fuel load to go supersonic. You could keep at it, and maybe you'll get somewhere (I doubt it, considering the exponential nature of the rocket equation, your size will balloon with diminishing returns). But if you put the exact same amount of fuel in different tanks, you will need less engines, and you will get better range. Math guarantees that. Also, a Kerbin-Moho transfer should be 5km/s, capture included, not 6, so you should also check your navigating. You can actually shave off of that almost a km/s from launching form Minmus, of course. And dare I suggest you haven't seen all of what a Mk3 can do? Not that Mk3 has the best tankage ratio anyway. Note: this two SSTOs are both powered by a single RAPIER and draw air using a single precooler. The first one has a TWR of 0.67, and struggles to go supersonic, needing to do so at basically sea level, flying level or slightly downwards, and taking its sweet time. The second one a TWR of 0.58, and goes supersonic much quicker, while climbing, in about a minute... with three times the frontal surface. Note the resource numbers, too. And of course my Mk3 designs usually make do with TWRs under 0.5, the secret to their payload efficiency. Mk2 fuselages, I'm sad to repeat, are broken. Rune. Just trying to help!

Woah, you are using Mk2's for a SSTA! Cease and desist at once! No, seriously, Mk2 is the worst fuselage in-game for pretty much everything. It has four times the drag of a Mk1, more volume for the same payload, the worst tankage ratio in game, not enough lift to do anything with it, and basically it is flat out terrible. Switch to Mk3 (or for an even better tankage ratio, Mk1) for the liquid fuel, and plain old rocket fuel tanks for the LFO mix, and you will see a much smaller empty weight, thus much more mileage with the same takeoff weight, and all of it in a much more compact airframe, meaning it will be more maneuverable under RCS/reaction wheels. Rune. Right now I actually consider mk2 fuselages to be bugged.

Hey guys! A very interesting conversation in @Raptor9's thread got me thinking about rockets and well, one thing led to another... anyhow, now I've got this: From left to right, you have 25mT, 50mT, and 100mT-to-LKO boosters. All using 1.5 SRB staging, with very similarly nice thrust curves, and all under 26 parts. Pretty neat, right? The thing is that I'm still undecided how you guys would prefer them... simple subassemblies with minimal part count so you can slap them under your payload easily? Should they have probe cores, electrics and attitude control and stuff so that they can be used to lift inert payloads and still comply with the Clean Space Act™? Maybe even add complex reusability subsystems to the expensive cores so they become cheap-ish* reusable boosters? *:(Nothing is cheaper than a reusable airbreathing SSTO) Rune. Cast your vote now! Or even better, your opinion!

Errr I meant with the same fuel load on the upper stage! As it is on the pics, it's 85mT of payload. With an additional bigger tank like yours (a 40.5mT wet S3-7200), payload would be 45mT, and the dV would be the same as a Titan 4N. Meaning, the second rocket has the same performance as yours, which was kind of my point. Yeah, you would think using half the high-performance engines would save you more than that, but then KSP goes and makes tankage relatively expensive in comparison... and the twin boar costs less than a single Vector and includes tankage, that's also a thing. It would be a really good candidate for core stage recovery, though, all the cost is in the almost-orbital core. And, you know, that SRB plume looks gorgeous to me. BTW, speaking of nice-looking rockets, the family I was inspired to build would be complete with this one: If you just grab the core + Boosters on each, these three make a fine booster family indeed, with ~25mT (2xSkipper, 2xKickback), ~50mT (1xTwin-Boar, 4xKickback), and ~100mT to LKO (1xMammoth, 6xKickback), and a very similar thrust curve (Ok, the second was a bit sportier, but everything is solved if you take out two kickbacks). And then the upper stage would be the customer's business, and I can avoid publishing anything similar to your stuff. I'm still debating whether to include control and/or recovery stuff, as it is they make very nice 19-26 part subassemblies. Rune. It shows I was just halving core thrust on each iteration. Yay science!

Riiiight, the reuse, I must confess I totally forgot about that. And fuel lines, I have kind of totally forgot that those ever existed, lately. I still think a taller core coupled with SRB's like in the second picture I showed (where I was copying liftoff thrust and seeing where it went), would be cheaper, and you should see the awesome SRB spread when they stage, but now everything you did makes much more sense. I would correct you on one point, tough: on that big one in the second pic, there is pretty much the same performance as yours, if you add a S3-7200 to the upper stage instead of 40,5mT of the 85mT payload. Now, I haven't taken it to Duna and back to work it out, but the upper stage spends something like 200m/s or less circularizing in LKO, so I'd say pretty confidently that it should have just as much juice left as yours. So maybe at least now you know you could go back to SRBs, if you so desire? Plus, you have a good idea of what it'll take to lift the NITE on its own (the first booster has a payload that is almost the same weight as the s3-7200 tank my smaller upper stage lacks). Aesthetic choices, of course, are a very personal thing, but there are rocket out there with six or more boosters (Delta II, PSLV, off the top of my head). Rune. And I had lots of fun reverse-engineering yesterday. Hope you don't mind!

Gorgeous upper stage, but can I nitpick a bit? Just to further confuse the guys that know me as 'a SSTO guy'. Isn't that like, a huge overkill in liftoff thrust for an underpowered upper stage? I toyed around with the upper stage idea, and in the end, I came up with these two: What is under the fairing is pretty much a clone of your NITE upper stage, if I can clone going by a picture (there is a tank inside the 3.5-2.5m adaptor, right? a shame not to use the empty space), with just a different RCS/docking port arrangement (I'm a big fan of 'universal 1.25-2.5m ports'). The first one gets 45mT to LKO+1km/s (pretty much your standard trans-Duna injection), and the second one, a whooping 85mT to LKO+650m/s (45mT to LKO+2km/s with the 40.5mT heavier notNITE, i.e a carbon copy of yours). So yeah, and that is with less liftoff thrust in both cases, and cheaper, lower Isp boosters (better suited to 1.5 staging, IMO, especially when the core already gives roll control). How thrust-limited are you running that core that it needs to be boosted? Because I have to gravity-turn quite aggresively on these two to get a nice apoapsis at core burnout (I have actually found that if I give them a 1º inclination on the launchpad, and use the other ~3º of sag, I can fly them without keyboard just by setting SAS to prograde when velocity goes over a mere 50m/s). In any case, let this imitation be a form of flattery: I really like the look of the in-space stage, even if I think TWR on it is too low for the fuel load. Rune. I am tempted to start a newer booster line here... the General Lifters are really moldy.

Sure! Tell me if you run into any troubles, but really, knowing your dB budget and with an accurate reading of what your stages can actually do, rocketry is not that hard. Rune. Oh and remember comms! I always forget communications lately. ^^'

Well, that's pretty much what I built, minus the landing, assuming RL doesn't mean RSS (which would mean instead of Duna we are talking about Mars). That could also be phrased as: I failed at understanding what you wanted, i.e: did step 1 all wrong. But the step themselves are still fine! The same dV chart I quoted says landing on Duna all propulsively would be 1,3km/s. Considering the thin air will help you a lot, but you want a nice margin to be less than perfect, 1,3km/s will be overkill. That should be as easy as adding a 1.5Mr landing stage, meaning the payload to LKO is about 1.5 times the one on my rocket (since the in-space stage also needs to have 1,5 times the fuel to give the same dV to a x1,5 payload). So add a rocketcrane stage on top of your payload, and build a rocket 1.5 times bigger than one I built. I reckon just adding fuel to the in-space stage, the core, and two extra boosters would do it, but let KER be the final judge. In a quarter of an hour, tinkering the previous design made this, but I must warn you that I was also tricky here and used a drop tank to not use dedicated in-space engines and instead use the ones on the rocketcrane form LKO all the way to the landing at Duna: Saves some weight, and it means just uprating the core to its maximum fuel (TWR at booster cutoff is perilously close to one) and adding a couple boosters, which makes for a rather efficient booster that squeezes the most out of each engine. I like that in boosters! But remember, the easy route is just to put 2~2.5km/s stages one under another, for a similar dV budget (~6.5km/s, instead of the ~5 of going to LDO I gave you earlier). Rune. Good thing I told you how to do it instead of just giving you the file.

Give a man a fish, and you will feed him for one day. Teach him how to fish, tough... How to build a 50 mT rocket to Duna: 1- Figure out what you need: Some handy dV charts say that to get to Low Duna Orbit (LDO), we need about 1,6km/s from Low Kerbin Orbit. That is what our upper stage needs, to get the payload to Duna, so we'll start there and build the launcher to LKO under it. Seems we won't need that much actually, since the budget to LKO is twice that (3,5km/s). I'd say two-three stages will be ideal for the whole job, going by the rule of thumb of 'one stage per 2km/s'. 2- We grab our test payload on the VAB, and either get some handy tool like KER, or we work out our dV the goold old fashion way, with Tsiolkovsky's rocket equation and pen and paper (or the windows calculator). In that case, the good old russian says with an Isp of 350 (Poodle), we need a Mass ratio of about 1.6. We'll play it safe and build some margin, of course, because I don't like to put debris in orbit, and thus like to finish circularization on the departure stage. But let's build the dummy mass simulator first! 3-Around that, we build our in-space stage. TWR matters very little, so we just chuck fuel until we have the ratio we want, and add accessories like solar power, batteries, a probe and reaction wheels. This is our 'smart' stage, the rest will be just dumb rockets. Notice how I also did an interstate with a fairing, because I used a 2.5m engine. It looks very nice to do things like that. 3- And now we need to boost this to LKO, so let's build a rocket under it. The dV budget for LKO is 3.5km/s, so we need more than one stage, because 3,5 is right on the edge of what stock TWR and tankage ratios allow (it can be done, but it's tricky). I like the stage-and-a-half concept (a core with a lot of fuel and low TWR with boosters to get it off the pad), so let's go with that, but you can also build two stages. Still, for this the core is just putting one big engine and enough fuel to bring its TWR close to one: There, 1.15. Could be even less, because it will get help form the boosters, but the main thing is to get as much dV out of it as we can (almost 3km/s), most of what we need, leaving just the boosting job for the boosters. 4-Add the boosters, fix the staging so all engines are lit on liftoff, and look at what KER says: What do you know, it's almost bang on 3,5km/s and TWR 1.5 at liftoff! I'm getting good at eyeballing this, those are pretty much the numbers you want. Anyhow, if they aren't, you should fix things. To get more dV, you need more fuel on the core. To get more TWR at liftoff, you need moar boosters. TWR at second stage separation is less of a hassle, because by then the core will be half-empty and thus it will have much more than the 'about one' we allowed in the previous step. 5- Make the whole thing flyable. In order to make things fly good, they have to be sturdy, and they have to be stable. Sturdiness is mostly a function of struts, but since 1.2, we have the handy option of strutting things without them, and I love it for the clean lines. Here I strut the source of forces in the booster to the heaviest part on the stack, which should keep everything very sturdy. I do the same thing with the fairing holding the payload inside: Stability is mostly keeping the CoM before the center of aerodinamic forces, AKA keep it nose-heavy. The simplest thing is to make sure the tanks on the rear empty first, so you keep a heavy, dart-like nose as your rocket spends propellant. And now we can tweak that with tweakables! So the lower tanks drain first, which should keep our CoM moving upwards as we fly. And finally... Test it, see what breaks, fix it, repeat. It will mostly be staging, or stuff during the staging. Maybe you feel that a stage could have less fuel, and another more. Here I would say that booster separation could be a bit less close (add separatrons and stage them correctly), and that that booster is definitely more than enough powerful, so we could increase the upper stage weight, or move some fuel to it, or build another core with a weaker engine. Tinkering is encouraged at this point! Note how I actually cut the thrust on the circularization burn with a ~20km peri, and stage to end the burn on the upper stage. I had plenty of gas, but if I do that, I ensure that LKO remains uncluttered, so hey, let's be green and stuff. It also means the much smaller in-space stage can maneuver much easier. And there you go, a 50mT-to-Duna booster, explained. Rune. Some assembly required.

I was just thinking about the same thing the other day, only the other way around, with the bays vertical. Rune. At the very least, they'll be sturdy.

I've been very bugged with Mk2 spaceplane parts lately. It's not that they have horrible drag (they do, more than two Mk1 fuselages, especially the Mk1-Mk2 short adaptors), poor lift (they kind of do, for their actual size), or awful tankage ratios (again, one of the worst in-game). The thing that finally did me in was the mismatch in pitch/yaw torque values for the probe core, which makes the Espada a handful to maneuver out of atmosphere. I was really close to crashing the other day... landing on Minmus! That should definitely NOT happen! Well, Mk3 is a tad big, but big means more fuel: 4,3km/s on LKO with seven Kerbals on board, plenty of juice to go anywhere on kerbin's SOI on the same training flight, and also poke out of it for the full training-to-lvl3 experience. Twice the engines on the Espada, tough. Rune. Really simple to put together, BTW, I slapped it together in an hour or so.

Or we are misunderstanding each other further. I think I get that what you think I should have written instead is 'the thing for a maneuverable yet still stable plane...'. I sort of implied this was a recipe for stability, when it's the edge case that binds the region of stability. Rune. Don't worry, I get the thing. I may not be able to express it correctly in a foreign language, but I actually studied it in uni.

This. I think it is the source of most of the arguing in this thread. The 'blue dot in KSP' may be called CoL, but it has little to nothing to do with 'the resultant of adding all the aerodynamic force vectors acting on the ship at any AoA' (and there are some simplifications in that definition, I know, but let's not confuse the readers unfamiliar with vector calculus further). That thing would actually be called the Center of Pressure (CoP), at least by me, and it is the thing that is important for stability, and can by the way be influenced by control surfaces and/or other things that change the lift/drag characteristics. @Val calls it Aerodynamic Center (another name I've heard for it), and it is the thing KSP shows... when you fix it by installing your awesome Correct CoL mod. Thanks for that mod, BTW, very useful. The problem is that the 'blue dot in KSP', on a stock install, only takes into account the lift force created by parts with lift rating, and disregards body lift and drag like they didn't exist. We all know that in this thread at this point, I think, but we don't know that it is the thing the other guy is talking about! BTW, just to confuse things further and condense everything into a sentence: the thing to do for stability to get a maneuverable yet still stable plane is to put the CoL (the real one, not the 'blue dot of KSP') on top or slightly in front of the CoM, but put the CoD (Center of Drag) behind it. That would mean the CoP (which is short of an average of the two) is behind the CoM, so stable, but still as maneuverable as it can be. All of this for a given static AoA, of course, then you would have to evaluate stability in the rest of the AoA envelope, and then repeat everything but looking at it from a dynamic point of view... Rune. We need a CoD indicator, that's what we need!

Thanks for the compliments! It's easy to miss my thread these days, not so much activity as it once had (as this late reply shows ). Still, I do keep an eye on it, and even tough I don't post as much stuff as I used to, that's mostly because I am very happy with the vehicles I am using now. There is the occasional thing that I still haven't got around to update, of course... just yesterday, I launched the latest flotilla to Dres. And it did include a Lackluster to serve as general purpose lander, of course, because that rocket is just too cute and versatile not to use. I found yet another way to cram enough stuff into it to actually use it as a sort of SSTA, thanks to KIS! And I also took the opportunity to update the reusable boost stage for >2.5mT payloads, so it didn't blow up the ramp, and instead got auto-recovered as it used to (because it soft-lands on the pad before the Lackluster is outside physics range). All of that stuff fits into one of the small 2.5m containers and is pretty easy to set up. I was much happier with how I got the Orca to follow the flotilla, tough. The Lackluster is easy, with the huge dV in the tanks it's only a matter of having a LKO refueling depot. But the Orca, with its poor Isp engines and run-of-the-mill mass ratio... it was a tough job, I thought at first. But! Rearrange the cargo, bring some Drive Pods up and... presto! This particular configuration gives about 5km/s with enough fuel left over for at least one landing, and is surprisingly well balanced and easy to put together: The Drive Pods are still one of the most amazing ideas I have ever come up with, they turned the problem of 'how to move an Orca through a 4km/s transfer' into an opportunity to transport a refinery module and some additional cargo. Talk about killing two birds with one stone! And the rest of the expedition is also very neat, a single Magdalena class transport pulls the rest of the stuff: a cool-looking station, a Base-in-a-Box, the crew with their Dwagon escape capsule, and enough relays to set up a local planet-spanning (and scanning!) network: Oh, and you guys might have missed the stuff that I uploaded to KerbalX the other day. I didn't bother to write the post at the same time, and obviously I completely forgot about it and went on vacation and everything. Not much to talk about, really, just the stupidly-efficient four-RAPIER SSTO that I previewed on the SSTO thread a while back, and what I'm using as commsats these days. Still, you might wat to take a look at it, so here it is: Rune. So that is what I'm launching these days, plus doing lots of asteroid wrangling.