herbal space program

So here’s a question for the group: if the jet engine indicator says it’s getting 100% of its propellant needs, then does that mean that intake air is not a limiting factor in thrust? Because I’ve been showing way lower than ~400 kN of thrust (maybe 200 or so) on my RAPIERs when I hit Mach 3.7 at 18km, but I’m getting at least somewhere in that ballpark if I hit it at 10km. Either way though, that indicator says 100% all the time, all the way up until I flame out at 29+km. What gives?

I guess that would explain why my 3Xmk1 plane seems to have less lift than it did before. I thought those fuselages had no lift before, but I guess they did!

I spent some time playing around with my SST014 last night (2 RAPIER, 1 nuke, ~46.5t wet, 3 shock cone intakes), and it seems like things have definitely changed in a fairly complex way. It seemed considerably easier for me to break the sound barrier at a lower altitude, making getting to the transonic realm a fair bit cheaper, but ten I also lost power at a considerably lower altitude. My lift seems like it’s less for some reason too, with the ship wanting to nose down significantly more than before, especially up high. Monitoring the engine’s performance, it seems like I need more air intakes up high now and not less to maintain the same thrust. I was only able to get close to the RAPIER nominal maximum thrust below 10km. In the past, I was accelerating steadily to around 1400m/s at 20km, then engaging the nuke and getting up to maybe 1500 in the 21-15km range, nosing up a bit, and dropping back down to maybe 1470 at a 29.5km switchover. I was able to make orbit with ~6km/sec dV fairly easily this way. Now I seem to lose all my thrust significantly lower down, and I think it’s more because of the intakes than the engines. When I slapped on a couple of inline intakes, performance definitely seemed to improve. I will try with even more tonight…

Well, I can’t say I’ve made a perfect ascent yet to LKO with this ship, but so far the best I’ve done to get to Minmus going sideways is 3108dV to LKO, around 60 m/s shy of what PLAD reported with a very similar stack, plus around 850 to get a Mun-Minmus transfer, for a total of 3955. The best I’ve done so far boosting straight up at the Mun is 3794 to the same encounter. Although there are clearly fewer gravity losses to LKO going sideways, all the sideways momentum required to get to Kerbin orbit is kind of superfluous if you can harvest enough lateral motion from your Mun encounter to kiss the SOI of Minmus at a fairly low rerlative velocity. I dunno, maybe if you have some time you could mess around with it yourself. I think it’s an interesting boundary condition to investigate. FWIW I also spent some time investigating various alternative configurations, using Mainsail, aerospike, etc. but nothing else matched the Mammoth-based stack that PLAD originally posted. I don’t think that performance is really beatable, so I think I’m done with this challenge. Congrats PLAD. AFAICT you nailed it to the wall!

Yes I was just figuring this out. Having everything based on vacuum dV kind skews the calculation. I was thinking that aerospikes might stand a chance of edging out the Mammoth, but their higher vacuum ISP will penalize them. One thing that might offset this is the fact that values close to vacuum ISP are reached pretty quickly. I monitored the ISP of the Mammoth on a stack very similar to PLAD’s, and found that it reached 310 at about 22 sec in, and engine cutoff was at 57 sec, so most of the burn actually happens at very close to vacuum ISP. I’m going to spend a little time investigating this. I’m also going to investigate the optimal atmospheric ascent speed a bit more. It seems like too fast too low might cause excessive drag losses that will more than offset the gravity losses of going a bit slower at first.

I think ISP is important both in atmosphere and vacuum because it indirectly affects the critical parameters of both TWR and drag. Here’s how I see the physics: Clearly the biggest source of deltaV loss in this exercise is getting off of Kerbin. Getting off of Kerbin entails possible losses from two sources: atmospheric drag and fighting gravity. The ideal solution to the first issue is to launch the most streamlined stack you possibly can as straight up as you can, minimizing the total amount of air you have to go through and how much energy you transfer to it. The solution to the second would be to instantaneously deliver the entire impulse required to get you where you’re going on the ground so that you can coast all the way to your destination. You could escape an airless Kerbin that way for around 3.4 km/sec, and whatever fraction of that it takes to reach the Mun SOI plus the drag losses is the absolute floor of how much dV you need to expend to get to Minmus. So what you need to do is accelerate as quickly as possible from the moment you take off, on a near-vertical trajectory that tops out just at the right spot on the inner edge of the Mun’s SOI. Your acceleration will depend on the TWR of the whole stack on the pad and on how much drag it has. The intrinsic TWR of the engine is one critical component of the first parameter, but its ISP is also critically important because it dictates how heavy the whole stack will be. The less fuel you have to boost, the faster you will accelerate. Less fuel also means less total drag, so ISP affects that as well. I don’t know what the correct formula to optimize this relationship is, but I Imagine it works out to something along the lines of maximizing TWR*LN(ISP)/drag coefficient. I think I might try applying that formula to all the available engines and seeing if any of them beat out the Mammoth. I also note the PLAD had an actual capsule on the ship, and you didn’t specify that there had to be a Kerbal aboard, so there could be some dV shaved off there by using a probe core. Anyway, I agree with PLAD that this is indeed a very interesting challenge.

I think OP engines will definitely help you if they have better TWR values than stock does for similar ISPs. As PLAD has shown us, the fundamental parameter of this challenge is how high TWR/drag is for your stack when it takes off. AFAICT, everything else is execution, although that part is by no means trivial. By my reckoning, this number will depend linearly on both the TWR and ISP of the engine in question, since the former dictates the acceleration on takeoff for a given mass and the latter how large that mass will have to be. Other than those, drag is the only real structural consideration.

I think there still must be some way to improve TWR/drag, but you totally nailed so many other aspects of this that I think you’re going to be very tough to beat in practice.

I did it mostly because I didn’t want to start over if one of my ships somehow got irretrievably corrupted, and also to get a sense of how much wiggle room there was in the E-K-K-J route. As to how to get to a 2:1 resonant Kerbin orbit from Jool, I don’t know if this is the best strategy, but here’s what I did: If Jool is at 12 o’ clock and the Kerbolar system is spinning counterclockwise, you want to have Kerbin at a little past 7 o’ clock. I ejected retrograde off of tylo into an orbit with a PE close to Kerbin’s, then boosted about another 200 m/s to encounter Kerbin at around 8 o’ clock, just under 2 years after ejecting from Jool. By tweaking this encounter, you can easily get to a 2:1 resonant orbit. Just keep twiddling the maneuver node from around your plane correction until until your Kerbin PE is like 137km coming from the outside. Note that this orbit still re-encounters Kerbin at excessively high speed, but from its Ap, you only need to burn around 250 more m/s to correct to a survivable tangential encounter. Hope that helps!

If you can assist off of Kerbin to a 2:1 resonant orbit, preferably with a higher PE than the one you have now, you should be able to make that orbit more tangential by boosting prograde and (I think) anti-radial at the Ap of that orbit. If you mess around with the maneuver node at Ap, you should be able to raise your PE to Kerbin’s orbit and push your Kerbin re-encounter towards your PE for a few hundred m/s. It took around 250 for me to go from a lethal angled encounter at ~8 o'clock to a very survivable tangential one near 6. Good luck!

So I finally got my fleet back to Kerbin safely: Please see my earlier posts on this thread from August 19th and September 8th for more details and pictures about the trip out. I had a fleet of four SSTO12 ships, with 3 RAPIER and 2 nuke engines, 55t wet, 21.2t dry. They made LK with around 4700 dV. They all went out via the PLAD e-k-k-j route, for around 1300 m/s all told to a Jool encounter. Using a tangential inbound encounter with Tylo, they dropped to a capturable Laythe encounter for very little dV, but they required significant fuel usage to land because of a high stall speed. After landing (first picture of this album), they all made orbit again quite easily, with no requirement for oxidizer burning. Outbound, I had to burn for around 690 m/s to get to the best retrograde Tylo assist I could devise. This was actually enough dV to put me very close to a Hohman transfer back to Kerbin orbit, but from my earlier experience of instantly burning up on contact with Laythe’s atmosphere and reading the accounts of others, I gathered that I would be coming in too fast that way. So I added another 200 m/s or so to my ejection velocity from Tylo and dropped down more steeply to an earlier Kerbin encounter. This was also way too fast, but it allowed me to set up a gravity assist to a lower 2:1 Kerbin resonant orbit. Even that orbit was not good enugh, because it intersected Kerbin’s at too much of an angle, so at apoapsis I had to do another ~250 m/s correction to get to a more tangential Kerbin re-encounter. From there it was pretty smooth sailing to get back home, although I’ll admit it took a few mulligans for me to get all the planes on the ground back at KSC together! MY best plane landed with about 600 m/s of nuke burn left, so I think I probably could have managed a Minmus landingif I had been willing to futz around enough. Anyway by my reckoning, that gets me on the board with 1050 points, for making it out and back and also for having under 100 parts… I learned a lot from playing with these ships, and my improved SSTO14 (also posted on this thread) is both much easier to fly and makes LKO with around 6km/s dV in nuke burn. I’ve actually got a version now that can do that with LT-2 struts mounted on the back, so depending on how full the fuel tank is, I should be able to land on airless bodies up to Mun size. I don’t think I can get to the insane ion powered point total this way, but I think I might be able to manage some bodies like Mun or Duna, that require more TWR. We’ll see…

As to the first issue, I was thinking that there's a pretty big difference in orbital inclination between the two bodies, so it would take a lot of dV to do the plane change, but if that's not the case, then I agree dropping PE to Tylo is not necessary. Certainly if it's only a couple hundred m/s, then by all means straight from one to the other would be the way to go. As to the second question, the most efficient way to land on a particular airless body in theory would be to cross into its SOI from behind at the minimum possible speed, on an approach trajectory that grazes the surface, then suicide burn at PE to bring horizontal velocity to zero. I believe that if you want to do this in stages, i.e. establish a highly eccentric orbit on the first pass and then lower its AP in stages from your low point, that would actually take roughly the same amount of energy as doing the whole suicide burn in one go. For Pol and Bop, as you pointed out, a lot of your energy budget will be spent matching orbital velocities, as their gravity is too feeble to change your speed very much.

I would say visit Laythe first, because you can get on the ground there for next to no dV from your Jool approach if you set up your Tylo-Laythe capture correctly. From Laythe orbit, it's then about 680 m/s to set up a Tylo encounter that could put you just about anywhere in the Jool system. If you use that to get to Pol or Bop, I think it would cost you fairly little dV to drop yourself from those bodies back to a Tylo encounter, first to go from one to the other and then to send yourself home. Also, I think the lighter your ship is when you're trying to land on these airless bodies, the better.

I hate to admit it after you said all that, but I had my dry wt. wrong in both my spreadsheet and my OP. Actual dry wt. is 13.4t, I had 13.1 in the sheet, so dV on LKO was really only 5852 . Anyway, I still think I can improve on that, so 6km/s is not out of the question. As to the low TWR, I agree it's a drag. It limits where I could land after Laythe, for example I don't think Duna would work with only 1 nuke. It also makes getting to orbit quite a bit more difficult, to the point where adding the extra nuke significantly (but not quite fully I think) offsets its own weight by letting you burn less LFO on the way up. I feel like with a 2:2 setup I could make it with almost no LFO burn at all, but of course the higher dry weight knocks off quite a bit of dV at the back end. Update -- after tweaking the ship a bit and making a few more attempts last night, I was able to get on LKO with 5917m/s, with 75 units of Ox left in the tank. The plane also lands quite well back on Kerbin, even mostly full. I could probably push this just above 6km/s if I replaced some the oxidizer with LF, but I want to keep that in case I need more power than the nuke can give. With that amount of dV on LKO, I think a Laythe-Duna-Kerbin return tour is feasible. I might just have to try it now...

Thanks! Lining them all up wasn't so hard, but not knocking off the center engine was so difficult that when one of them rolled to a stop in one piece a few hundred meters away, I decided to call it good enough. As to your design advice, I actually spent yesterday evening trying to use all the stuff I learned from you and others in this thread to improve upon that plane, and this is what I came up with (SSTO14): It has significant positive AOA in the wings (all wet) relative to the fuselage, and uses what seems to be the most efficient configuration, 2 Rapiers + 1 nuke. It thus has only 3 fuselages vs. 5 to further reduce drag and uses all winglets and canards to steer. I used the small hardpoint to center-attach the fuselages, mounted the rear wheels lower and further in, and added two pairs of struts front and back to increase the rigidity of the airframe. The resulting plane flies much better than the previous version. It weighs 41.4t on the runway, 13.4t empty, and as you can see from the pictures, it reached orbit on my first successful try with almost 5.9km/s deltaV left, a number I think I can still improve slightly upon. I believe this is close to the best one can do without resorting to cargo bay or part clipping abuse, but I haven't really scoured the forum for better performance. As to the swept wings, I just didn't notice they had worse numbers than the others. It looked to me like they were all the same, so I stopped paying attention. Anyway, it's been fun learning! I'm tempted to run this new plane through the challenge, but I only have so much free time...

So I finally managed to land all 4 of my ships on Laythe, all in (more or less) the same spot: I'll post a bunch more images when I'm all done. As it turns out, my plane was nearly impossible to land without knocking the center Rapier off, something I REALLY should have tested back on Kerbin . My stall speed was also alarmingly high, so it was quite difficult to set it down gently enough without any thrust. DeltaV from LKO to the ground on Laythe ranged from 1355 (based on nuke ISP) for my best ship to 1612 for my worst, in all cases significantly less than just getting to Jool's SOI on a straight Hohman transfer would have cost. I've gotten one of the four ships back to Laythe orbit with 1.8km/s left in the tank, which should allow me to hit some other destination. My best ship should have at least 2.1km/s left back in Laythe orbit. I'm tempted to try Duna, because the 2 nukes should be enough to lift me there, but we'll see. Transfer to a Tylo slingshot costs me only ~685 m/s by my estimates, so I should have a decent amount left to play with. Anyway, it's been quite a challenge, but I can't imagine any part of what remains will be as hard as landing those hastily designed planes on Laythe was.

With this approach I got captured at laythe after a gentle aerobrake at ~2.6 km/s. Once I knew where I needed to put the encounter (thank you - I was ready for a spoiler on this), It was pretty straightforward to get it dialed in. I haven't actually put any of my ships through yet, but it looks similar to your picture. Trying to fine tune my approach from just before the Tylo encounter, I'm finding that if I come at Laythe's orbit from just outside and as close to the other side of Jool as I can, I get really long projected Laythe encounters that look like I'm almost getting captured directly into a prograde orbit. I'm tempted to believe that I can safely aerobrake directly in these encounters rather than swinging around to encounter Laythe again, since they look a whole lot slower than what I had before. That's what I'll try the first time through tonight...

Hmm, have you actually done this in 1.0.3? At what velocity are you hitting Laythe's atmo? Because even from a high-AP Jool orbit, I'm exploding about 3 seconds after I hit Laythe's atmo, without even slowing down appreciably, regardless of my attitude. KABOOM! My whole ship is gone. Back in like 1.01, I seem to remember that by pitching a space plane up enough you could cheat the heat, but not any more I think. It seems like now you have to be coming in below 3km/s. i.e. from a fairly similar orbit to Laythe's, to stand a chance.

With this approach I got captured at laythe after a gentle aerobrake at ~2.6 km/s. Ahh. All my initial Tylo encounters had me coming in from the top with Tylo around 10:30 o' clock, then sending me close by Jool to a Laythe encounter on the other side. This always left me in highly eccentric orbits after both encounters, from which I couldn't seem to set up a circularizing slingshot of off any moon. Come to think of it, my second ship has a Laythe-Laythe trajectory that looks like it will work, and it hits Laythe right around 9 o' clock like your Tylo encounter. I should have realized that would work better for Tylo as well, but getting a Tylo encounter just where you want it is not as easy as it is for Laythe. I guess what I need to do is back up my initial set-up to my Jool midpoint through the magic of alt-F5 and try again...

Well I must say this challenge has given me more than I bargained for at every turn. I thought I had it made slingshotting retrograde off of Tylo into Laythe for my approach to Jool, but after many tries my velocity at Laythe has always been such that any attempt at aerobraking was instant death. It's cheap to get captured into Jool orbit, but all the orbits I seem to be able to get off of Tylo are so eccentric that I can't intersect Laythe at a reasonable velocity. I seem to be having better luck swinging around Laythe on my first go, but the jury is still out. I think my cheapest strategy is probably to get captured into an on-plane orbit with the highest possible AP, then raise PE to Laythe orbit from there and set up an encounter. If I hit Laythe's orbit at my PE, I seem to be able to set up encounters that put me in near-circular orbits around Laythe's altitude without having to put my toes in the atmo.. At any rate, like every other part of this mission this is not trivial... - - - Updated - - - It took some tweaking of other aspects of the design, but after messing around for maybe an hour with my wings tipped up a few degrees, I'm getting my SSTO12 on LKO with a good 400-500 m/s dV more than I ever could with everything in one plane. I can probably do better with a little more practice. It just helps at every turn.

Omg, I can't believe that never occurred to me. At last now I know why I can't get close to your tons/Rapier numbers! Back to the drawing board...

I never had any real trouble setting up any of the encounters before Jool either. What I found last night analyzing the projected Jool encounters of my fleet of 4 ships is that the timing of the second Kerbin encounter is critical. Since you have to be at a 2:1 Kerbin resonant apoapsis no matter what, your speed coming into Kerbin's SOI on the second encounter is pretty much fixed. That means that you have a very limited parameter space to maneuver in in terms of where you place your Kerbin PE and what your ejection angle is. Since you have to come from the inside, the closer you get to Kerbin the higher your AP will be, but also the earlier on Jool's orbital path. That means if your second Kerbin encounter is too early, you'll come in too far behind Jool with no way of correcting your course without burning anti-radially a whole lot. Presumably coming into Kerbin too late would put you out ahead of Jool, but that's actually quite a bit easier to fix since you'll be going so much slower than Jool when you reach its orbit. Anyway, all of my ships can get to a Jool encounter via Kerbin, but the earliest one to arrive there has such a rubbish encounter (>350dV to get down to Tylo's orbit) that I don't think it's even worth it to continue with that one. As it is KEKKJ was only saving me maybe 700dV over a direct Hohman transfer. Anyway, I believe that there is sufficient wiggle room on the first Kerbin encounter to allow you to place the second one where you need it, even if your Eve encounter is a few days off, for a fairly small price in dV. If you just put the second one in exactly the same spot on Kerbin's orbit as the first though, then exactly when you arrive at Eve really is crucial. --Just a quick update. After posting all that I figured out that the best way to correct my second Kerbin encounter is inbound, as close to PE as possible but before the approach direction starts to change. From there, boosting prograde or retrograde will change the ejection angle so that you can move your Jool encounter ~150 days for ~150 m/s. So far my best ship is headed for a survivable capture at Laythe with a total of 1289 m/s expended from a 70X75km LKO, my worst ship 1457. So if I can land it near deadstick, I should end up on the ground at Laythe for 1300-1500 m/s, with 3.3-3.6km/s dV left -- in all cases significantly cheaper than I could have via a direct transfer. As it turns out, setting up captures at Jool was pretty easy. For three of the ships I was able to set up Tylo encounters that slingshot me retrograde into Jool orbit for just a few m/s over my ~90 m/s plane change correction for Jool. With another ~3m/s, I can adjust each of these so it should culminate in a survivable Laythe aerocapture. For the fourth, I couldn't get the needed Tylo encounter, but I was able to set up a swing around Laythe into a Jool orbit that re-encounters Laythe at much lower speed 4 days (2 Laythe orbits) later. Anyway, this is starting to look pretty doable now...

Why yes, yes I did, for four separate ships, flying every maneuver manually . However, I should really not act too smug, as I haven't actually gotten all the way to Jool yet. The last Kerbin encounter requires me to get right down to the top of Kerbin's atmo, so we'll see if I can actually manage it without burning any more fuel. It's one thing to have the maneuver node editor telling you that you'll reach Jool, and another to actually arrive there. I'll certainly need to make another correction both right before and right after the second Kerbin encounter to dial in my Jool approach. As to making the Eve ejection adjustment while still in Kerbin orbit, I agree that would have made it cost significantly less, but the maneuver node editor is so wibbly-wobbly that far out that it can be really hard to tell where you are. Nonetheless I could no doubt do it better the next time, but given the many hours I've already sunk into this go, I've no stomach for starting over. Figuring out my Jool approach will be interesting. My initial plan is to try to set up a Tylo encounter that will slingshot me retrograde sufficiently to either get captured by Jool or alternately to send me into Laythe on an escape trajectory that is still slow enough that I wont burn up aerobraking there. I'm tempted to go through everybody's reports to see how they did it, but I want to try to figure it out myself first. I'm hoping I can maybe tag Pol or Bop on the way out too, but we'll see. My 0.90 Jool5 mission, which I unfortunately never posted before 1.0 came out, taught me quite a bit about slingshotting around the Jool system, so hopefully I can come up with something beyond just landing on Laythe.

Well I've been out of action for quite a while due to the pesky demands of meat life, but I'm going to finish this thing now. Congrats to everyone who's made it already. I've managed to get a fleet of 4 SSTO12s to Jool encounter trajectories using the famous PLAD K-E-K-K-J route. I went to school on Val and others' designs for the SSTO12, using all Mk1 fuselages, adding shock cones at both ends, and employing wet wings: The design uses 2 Nervas and 3 RAPIERs and weighs 56.1t fully fuelled, 20.1t dry. So far the best I've gotten it to LKO with is 4.8km/s dV, which is obviously not record territory but nonetheless respectable. The best I've done so far for the Jool multi-assist route is 1,231 dV all told, with all flying done manually. It's so hard to set up all the other parts that I never managed to work in a Munar assist, so my Eve ejection burns were around 1060 dV. Getting the orbital inclination at launch just right to avoid a plane correction later was also too much of a bother for me, so I ended up losing another 70m/s or so doing that right after leaving Kerbin orbit. The rest was just tiny little corrections this way and that to get each approach just right. It's hard for me to imagine how one could actually do a perfect launch into this route manually, as the stock interface just doesn't have the resolution. I did so many burns that were under 1m/s. anyway, I will post the whole story once I get finished. Hopefully the ~3.7km/s I have left on my initial Jool trajectory will be enough...

Nicely done. I suspected that with the new Mk1 fuselages making your plane out of only those would be the best move, and it looks like you've proven that to be correct. It looks also like having the 2 nukes instead of 1 offsets the extra weight by giving you enough vacuum TWR to burn the RAPIERs closed only a wee bit getting to LKO. I expect the 2-nuke design will also have a much better chance of getting you off Laythe without oxidizer than only 1 nuke would. I'm tempted to just scrap what I have and copy your design, but that wouldn't seem right.