herbal space program

That got me thinking. If I really want to put my boosters where my mouth is, maybe the best thing for me to build would be the most tightly packed core of bundled tanks possible that fits under an inflated 10m shield, with just enough wiggle room to avoid incineration due to small deviations from prograde. The lower, soupy phase of the ascent could be done with the shield deflated, but at the altitude where heating becomes a big problem and drag is much less of one, I could inflate it and keep boosting away at high TWR until I reach orbit. Staging would all be vertical, so the rocket is skinny down low and stubbier up high. I might just have to give that a try...

FWIW, I was working on the 5m scale-up of roughly the equivalent stack last night, with 19 Vectors on the bottom of the core, and I was able to get to a very promising ascent profile in terms of acceleration, until I exploded. And exploded again. And exploded again. Which is to say, I think at least for me the biggest challenge about showing that bigger is better is going to be heat management. I'm actually kind of curious what you are using for that on your own most recent ship, because a pointy fairing with a 2.5m heat shield right behind it still does not seem to be sufficient to keep my upper stage from vaporizing. It may just be that I'm still flying it by hand, and the oscillations of the Vector even when it's locked to prograde cause too much exposure of the sides. If that turns out to be the case, then I guess I'll have to abandon that approach and just see what McJeb can do with it.

The disagreement was about your assertion that ONLY the 1.25m tanks are suitable for making efficient Eve lifters, because "the large engines and tanks are too draggy", which is exactly what you said upthread. Whether or not a pointy rocket is better than a stubby one was never at issue. Of course a pointy one is better, but so is a BIGGER one. But anyway, I agree that if you have not accepted that fact by now, you are not going to, so there's no point in belaboring it further, at least not on this thread.

No, I think it's pretty clear what the issue is for you. It looks like you're trying to manually land the kind of ship that might be competitive in the dV contest we were talking about. I'm pretty sure none of us actually did that. We just used Hyperedit. If you actually want to land that thing, you need to adopt a wider stance. Set hardpoints and dummy tanks out, buttressed by struts, to mount sturdy lander legs as far away as possible from your CoG, You can then use the hardpoints to stage all that off when you're ascending. Or you could just make a wider lander. The tall-and-narrow thing is really about minimizing dV to orbit. 4X asparagus is easy to make so that it's a stable landing platform and is still plenty good enough to get you on and off.

Naah, this debate (at least the one I'm engaged in) is all about the edge of the envelope. You should mostly ignore all that for your own purposes and just try to create the most robust design you can. The craft we are talking about are going to be next to impossible to fly because of the extreme speed, aerodynamic forces and heating involved, so that even a very experienced pilot will likely have no choice but to let McJeb do it for them. That's not at all what you're trying to do, so it's not so helpful for you. AFAICT, you haven't actually posted a picture of your craft/attempt for a while. If you still haven't made it, perhaps doing so would elicit some helpful responses.

IDK, way back in the day they used to be ignored by the physics engine entirely, so you could make all sorts of exploitative craft out of them. Not sure how they behave now. And the inevitable debate among self-styled experts aside, I think OP has received plenty of helpful advice on this thread. And this is also an interesting question!

Awkward? This is pretty typical banter for this forum in my experience. You've just gotten a bunch of people who think they know what they're doing jettisoning some liquid fuel about it. I'm sure it will all be settled quite civilly.

How about you can make your rocket any way you like so long as you use only Stock/MH parts and you're not doing tricksy drag-and-heat-avoidance exploits like upside-down, clipped-in fairings? Also, any stable orbit should be acceptable, as should any takeoff site that is under say 150m. Why does it have to be so exact? The question we're trying to answer here is if you don't cheat the drag model, are wider or narrower tanks better for making the lowest-dV ascent possible from Eve? You and a few others insist that only the narrowest tanks are competitive, even though in RL physics it is most definitely the other way around, as explained by @Starhawk above, because drag scales as the square of linear dimension while mass scales as the cube. If you look at the equation for the Reynolds number on Wikipedia, you'll see that the term for linear dimension is on the top, implying that inertial vs. viscous forces in a given fluid will predominate more and more as objects get bigger, i.e. the more massive your rocket is for a given shape, the less it will feel drag. If it didn't actually work that way in KSP, then our aerodynamic model would be no better than what we had pre-1.0. Anyway, I'm interested to see what you can come up with at 1.25m that will beat the equivalent 5m rocket in terms of shape and TWR.

Now that's what I'm talking about! I think maybe the point of contention here is about Pure Stock vs. MH. In Pure Stock, there really isn't any good way to put enough boom behind those massive tanks to allow them to have the same TWR for a tall, skinny rocket as for a tall, skinny 1.25m stack with a single Vector. But with those nifty engine plates and that 5m Saturn V multi-adapter available, that's no longer a barrier. Based on your result, I would predict that a 5m rocket with the same shape and TWR could do even better, and I mean to see if that's the case, although you've set quite a benchmark there!

You can talk about testing hundreds of craft, but upthread we just did the experiment now! If you drop a 2.5m stack upside down from 10km, it only reaches a TV around 200 m.s, and if you drop a 5m stack, it reaches over 320 m/s. QED! The bigger your rocket is for the same shape, attitude, and TWR, the less it will be affected by drag, period. And like I said before, I'm relieved to discover that because if it weren't so then our aerodynamic model would be complete rubbish. I think your problem in trying to design a larger efficient lifter was probably just that you didn't have enough TWR, because the standard engines and mounts don't allow you to build stacks with enough TWR to take advantage of their better aerodynamics. For my 5m stack, I need to be going over 300m/s as soon as I get off the ground as possible, then I need to drop all those extra boosters and proceed at a TWR that follows the TV curve as much as possible through a perfect gravity turn. With the stack I dropped upthread, I reached orbit for right around 6.5km/s flying it by hand, and I haven't even gotten close to the needed TWR values to make it optimal. When I've done that, if you like we can do a head-to-head test with whichever of your hundreds of smaller craft you want to pull out. I'll even install McJeb for the occasion!

So I decided to do a little Hyperedit drop test from 10 km with my own, only slightly pointier but much bigger lifter, and I discovered to my gratification that my TV looks to have been around 330 m/s at SL: Note that the speed barely changes between 1.7 km and impact, indicating this free-falling craft is at/near TV. And I didn't even bother to stage off the side stacks. That tells me two things: 1) That bigger ships do indeed have less drag/mass than smaller ones in this game, i.e. the big tanks really are better, you just have to get creative about spamming engines onto their undersides, and 2) If I want to get off Eve for the least dV, I need to take this Moar Boosters thing to a whole new level!

What I found was important with those back-end shields was to mount them pointy side-up, because they really like to flip if they are the other way around, even if they are at the back end of your ship.

Wow, I thought that would have been closer to TV on Kerbin at those altitudes. I just did a quick check on my rocket, and assuming it's roughly like yours TV-wise, I'm getting close with mine at some points but not actually exceeding it anywhere. Who knew? Moar boosters it is then!

OK, perhaps I didn't say that quite right. I guess what I meant to say that the slope fro Eve is higher overall, which means that TV increases more quickly with altitude there than on Kerbin, and that this is especially felt in that 20-30 km band on Eve, where TV increases quite sharply from something within your likely range of velocities to something beyond it. IOW, to strictly maintain TV through that band, you would have to accelerate faster than the likely TWR of your rocket. So if you're making a practical rocket, the best way to offset that is to exceed TV earlier in your profile so that you can stay closer to it when it starts to rise really fast. I think that's it anyway. How else do you explain the good dV-to-orbit performance of these TWR >3 rockets?

I feel like based on my experience that this doesn't really work out to be true in the game, but maybe I'm underestimating TV on Eve for the type of vehicle that is competitive. In challenges, the ships that make it to orbit with the least vacuum dV expended seem to be the skinniest possible rockets with the highest TWRs, that basically tear a narrow hole in the atmosphere locked dead to prograde pretty much the whole way. You can see a couple of examples from myself and @Laie above. Mine made it to LEO from sea level for under 6.7km/s, and could do significantly better than that if I resorted to McJeb. Setting aside the TV question, I can say with some confidence that to stay strictly prograde for so much of your profile, you need to start your gravity turn much earlier than 20km. On my best ascents, I started mine around 4km and was already pitched to 45 degrees not far above 20km. Another thing to bear in mind is that Eve's atmo stops at only 90km but weighs 5 times as much as Kerbin's, so its pressure drops off much more steeply above 25km, causing TV to rise sharply up there and making it more economical maybe to eat some drag losses down lower so that you don't pay even more dV back in gravity losses up high. IOW, I think it's complicated!

Have you built a space plane yet? It's not so hard to land on Laythe and return with a Rapier/Nerv-based SSTO. Duna is a bit harder to do that way, but well within reach. For me, designing and flying planes added huge play value to the game.

I'll actually never do a ballistic re-entry on Eve again. For my own recently posted Eve challenge, I built modular units that link up to make an amphibious extracting/refining crawler that can dock up with the lifter: Full album: https://imgur.com/a/K4IPP84 The two units link up via front and rear docking ports, and the one on the bottom is capable of tipping itself to vertical using strategically placed LT-2 landing struts, which allows it to dock itself to the underside of the lifter.

If you can shave off the first 800 m/s or so of your orbital velocity, re-entry is really not so bad. It's trying to do it with a fully fueled 500t vehicle using nothing but shields and chutes that's super tricky.

Don't I know it! Remarkable how this exercise converged to the same general idea for both of us: https://imgur.com/a/7jFsTG2 Figuring out how stay stay behind all those heat shields and stage them off without exploding my ship was one of the toughest things I ever did for sure. The rest of it was fairly plug-and-chug

That was easy. Thanks!

Well the landing part is pretty awesome, but the taking off part, not so much. In addition to not having the OP airbreathing Rapiers at your disposal, you have to contend with the combination of a 1000 m/s higher orbital velocity and an atmosphere so thick that any kind of efficient winged ascent profile will cause massive heating. I guess the ideal solution would be to land your craft as a glider on an up-slope, stage off the wings, and then take off up the hill, turning vertical and staging off your wheels as you take off. ....So on my own weekend I was playing around with large-diameter low-dV Eve lifters, to see if it's really true that the big tanks are too draggy: While in pure stock the Mammoth on a 3.75m tank is no substitute for a bundle of Vectors on smaller ones, the MH 5m tank, multi-engine adapter, and the 3.75m engine plate collectively allow you to totally spam the underside of that tank with engines. The version shown has 12 Vectors on the bottom and 8 more short-burning ones around the outside, to provide an extra kick on takeoff, with a single Vector on the Mk3 upper stage. So far I've able to get this to Eve orbit from 100m for around 6.6 km/s flying it by hand, which is no easy task. I have not yet come close to staying fully locked to prograde the whole way through an optimal gravity turn, and any deviation from that attitude is costly down low and deadly up high. To avoid burning up and maintain some semblance of aerodynamics, I had to put a sacrificial shielded docking port on top, with a 1.25m heat shield behind it, as well as another one right behind the capsule. Even that level of protection requires staying completely prograde through all the upper parts. I am tempted to finally break down and install McJeb just to see what it can do with zero pilot error, but even without doing so I think I can say that bigger tanks are not necessarily worse than smaller ones if you can put enough boom behind them. I'm kind of relieved to find that out, because in actual physics, bigger tanks should have a significant advantage dragwise. BTW, could some kind soul please let me know how we embed Imgur albums these days? "<imgur>(URL)</imgur>" doesn't seem to be working for me. Answered, thanks!

Impressed as I am, for my challenges, that craft would be disallowed because it's basically using both extreme part offsetting and structural/backwards part clipping to cheat the heating and drag models. That's the only way you can reach such absurdly high velocities so low in Eve's atmo. That and given the extreme aerodynamic forces and flipping potential, there's also no way you could fly that thing except with McJeb, which I always disallow as well. But it still is pretty remarkable. I'd like to see how it performs in the current version, just so we know how it really compares. It looks like you're not using asparagus staging there. Unless you're going to do that, it's better to put everything in one stack.

I'm rather busy with my own challenge right now, but it would be interesting to see what people come up with. If it's like similar Kerbin-based challenges I've participated in, I would expect that Mammoth-based designs would do the best, since that engine has the best TWR on Eve, and the larger diameter stack should have a more favorable mass/drag ratio. It seemed to boil down to getting as much vertical velocity as quickly as possible out of the gate, maintaining a pretty aggressive rate of ascent relative to TV early, then making an early gravity turn at a high TWR, skirting the edge of exploding behind the lowest-diameter heat shield you can use.

If you "don't worry about the maths", then on what do you base the claim that your craft made orbit from Eve SL with only 5.8 km/s dV? That's the claim I'm taking issue with, not the one that your craft, whatever it was, made orbit per se. The dV tool we have in the game now is far from infallible, as I demonstrated above.

There is no way a sea level Eve ascent is possible with only 5.8km of vacuum dV. Even as actual atmosphere-adjusted dV that claim seems pretty iffy to me. That's because if Eve had no atmosphere, you would need to expend at least around 4 or so km/s of vacuum dV just to get going fast enough to stay in orbit. And that's only if you take off horizontally in vacuum, accelerating to that speed instantaneously, i.e. that is a hard theoretical limit and not any kind of practical one. And even setting that aside, there is no way you are pushing anything that size through all that soup for just an extra 1.8km/s over that limit. If however what we are talking about what the dV indicator says when you are sitting on the ground, then that sounds about right. For an example, here's a sea-level Eve lifter I made that can survive a ballistic re-entry fully fueled and then take off to orbit again, which was one of the hardest design exercises I've ever taken on: https://imgur.com/a/7jFsTG2 You can see that the dV indicator reads 5.3 km/s while it's descending through the upper atmo, but reads only 1.4km/s when it's near the ground (OK, that particular time it actually went in the Drink, but that's beside the point!). Its actual vacuum dV when staged as intended is something like 12km/s, and configured as shown it makes Eve orbit from sea level with around 2km/s dV to spare in low TWR orbital maneuvering ability. ....So I've now gone on to test the minimum suicide burn dV myself by hacking the Eve atmo with HyperEdit. At the lowest normally safe orbit of 91km, I was going 3215 m/s. To drop my PE to 1600m, I had to lose exactly 100m/s of velocity. When I hit terrain at 3km, about 90 seconds before PE, I was going 3554 m/s. At the rate I was accelerating, I estimate that if I had made it down to 1000m I would have been going around 3580 m/s. Adding it up, I find that to attain the lowest-energy safe orbit possible from the surface of Eve, an absolute minimum of 3680m/s is required.