Streetwind

Members
  • Content count

    5760
  • Joined

  • Last visited

Community Reputation

3240 Excellent

3 Followers

About Streetwind

  • Rank
    Talks To Boosters
  1. I found KAC to be reasonably reliable, but not exact. It is the right time to leave, but not down to the hour or minute. You can definitely get an encounter when the mod claims you can, but for Moho specifically you might want a little more precision. Try alexmoon's launch window planner, or the ingame mod adaption, Transfer Window Planner. It's a bit more technical, but also lets you set up an extremely precise node. The ingame mod has a function to show you the spot where the node should go; for the out of game website, I've previously written up a guide here. (Note: this planner gives you single transfer orbits. What you're doing with the apoapsis departure is essentially a delayed bi-elliptic transfer, which means two transfers chained back to back, and the planner can't give you that in one go. It could give you each of the two burns individually, but that would require you setting up your interim solar orbit as a custom vessel in the planner, which is more work than the result is worth. You can fly the apoapsis departure without pre-planning.)
  2. It's weird, isn't it? But turns out, when you do the math, landing and returning from Moho is the second most dV-hungry planetary exploration task you can do in the stock solar system. Only landing and returning from Eve is harder, and that is quite obviously a special case. Among airless worlds, Moho is the king of fuel thirst. It just barely scrapes out ahead of Tylo, although the actual landing process itself is easier. Then, you add to that the fact that the difference in orbital period between you and your target, the actual speed of your target, and the distance to it, determines the difficulty of hitting the transfer just right. It's easy to hit Duna, because its orbital period isn't that much different from Kerbin. It's also easy to hit Eeloo, because it is just that far away and moving so slowly. Just burn sometime in the correct week, in the general direction of prograde, and you can have your encounter served on a silver plate. But Moho? Not as far away as you think, drastically different orbital period, and screaming along its path at a huge speed. Trying to hit that with a Hohmann transfer is like a minigolf trick shot by comparison. Being an hour off the optimal transfer window already incurs noticable extra cost in dV. Because of this high cost and low margin for error, people have devised an alternative way to visit Moho, which trades increased travel time for an easier and slightly cheaper trip. Make your ejection burn when Kerbin's position lines up with Moho's apoapsis, which conveniently also corresponds to the (I believe) ascending node. Put your periapsis on top of Moho's orbit, make sure the inclination ia matched, then time warp forward and loop around once or twice in solar orbit. Then you will be able to make a maneuver node at periapsis that finds an encounter with Moho without much trouble. (Of course, that method usually doesn't work all that well when you have a life support mod installed, but I don't see any in your screenshot.)
  3. That was true in versions .90 and older, before they completely replaced the aerodynamics model. Nowadays, aerodynamic losses are insignificantly small for normal-shaped rockets even if you are on fire going up. That said, if your ascent profile works for you, then that's okay. High TWR is more difficult to fly, because it's hard to steer when your vessel really bites into the air.
  4. I too like myself some SRBs. I also like to employ the obscure little Variable Thrust Limiter plugin. In real life, a solid motor's thrust is not always constant - the factory can manufacture it with a pre-defined thrust profile that's practically baked into the grain. This plugin lets you do the same in the VAB, in a way. The simple interpolation isn't perfect, but it's definitely a big improvement.
  5. @Phelan @DStaal NF Propulsion has a cryogenic gas separator that produces both xenon and argon. Also an atmospheric spectrometer to detect these resources. I can't advise how to use it, I have not gotten around to it myself yet...
  6. No, we don't have a habitat for the trip. And I don't mean in the sense of "there's not enough funding to pull it off". I mean in the sense of "nobody told anyone to start until now, so even if they did start now, it'll never make it in time". A few studies and isolated life support trials don't make a full integrated spacecraft anytime soon. (This has been one of the main points of criticism towards ventures like Mars One, by the way: the "no hardware exists" argument. You can't buy anything, it just does not exist. Until a prospective venture can show hardware they developed themselves that fills the gaps of what doesn't currently exist, their premise is unviable by default.)
  7. I have not seen evidence either for or against.
  8. They cannot EVA, no. But I assume that is hardcoded, and that there's no config setting for this. After all, why bother exposing a setting for something that makes no sense to change?
  9. @theJesuit - those numbers are per-level modifiers. For instance: a 1-star pilot gets a x1.30 multiplier on their G-tolerance. After advancing to 2 stars, they get a x1.45 multiplier instead. Tourists, being untrained, always have a 25% tolerance penalty regardless of what kind of level they might accidentally achieve during their trip.
  10. Does TweakScale perhaps work on the stock 10m inflatable heatshield?
  11. Mister Everyday Astronaut himself reported on Reddit that it is getting a much more physically impressive payload adapter. No hard performance numbers, but well, we can assume those to be under NDA until the Falcon Heavy User's Guide is made public.
  12. That's much too optimistic, I'm afraid... for multiple reasons. First, SpaceX has stated that first stage recovery inflicts a roughly 30% performance penalty on the Falcon 9. Currently, SpaceX advertises a theoretical 22.8 metric tons to LEO (where "LEO" is defined as a useless 200x200km short term parking orbit at the inclination of Kennedy Space Center). So it follows that first stage reuse drops the maximum payload to ~16 tons already. Next, you need to make assumptions about second stage recovery. The Falcon 9 as a whole is weighted more towards the second stage than other rockets would be in terms of dV budget, so that the first stage doesn't go too fast and has an easier time recovering. With stage 2 fielding the lion's share of dV to orbit, asking it to reserve fuel or carry additional dry mass for recovery hurts it more than the first stage. Unfortunately, we don't know the method of return that stage 2 will be using, which has a massive influence on how much investment will be required for recovery. But, let's guess at two cases. The optimistic guess says that it takes another 30% hit to payload capacity to return stage 2; the pessimistic guess says that it takes a 50% hit to payload capacity. Optimistic: 11.2 tons Pessimistic: 8 tons However, there's another factor in play here, namely the fact that there's a hard limit on how much actual mass you can bolt onto the beefiest available payload adapter for the F9. And that is 10.8 metric tons. This doesn't affect the pessimistic guess, but it does affect the optimistic guess. So, as a result: Optimistic: 10.8 tons Pessimistic: 8 tons
  13. TMRO had a great interview with Lisa Tatge, a subsystems engineer and spacecraft operator on the Cassini mission. They go over the mission's past achievements, as well as (in great detail) the grand finale. Among other things, they explain how Cassini goes through multiple periapsis altitude changes over the course of its last 22 orbits through interactions with Titan, despite already being completely ballistic. The final maneuver was just completed on Saturday morning, now there's no propulsion left whatsoever. Only attitude control remains. No matter what happens now... it'll just have to happen. Interview begins at 24:30.
  14. It's in Nuclear Power, actually. We moved it there a few versions ago because people kept missing it in the other node and kept petitioning for a smaller-than-400 Ec/s reactor, remember? I think there's room for cost to go up, perhaps to something like x1.5 the cost of the best solars. I definitely would want people to at least go "hmmm, that's expensive" when choosing to go for a reactor over solar. They are strictly better for all but two planets, so they should come with a noticable hit to the wallet. I'd argue that for a probe, it's grossly overpowered. A RTG, or some 50 kg of solar, would be a much better choice if you have a passive consumption of under 0.1 Ec/s and only an ultra-rare instance of active Ec consumption through science transmission. Also, keep in mind: the chief energy consumer that these reactors are encountering ingame are Near Future's own electric engines. 90% of all reactors used by players, at minimum, will be going towards powering those. And early electric engines take small amounts of power, increasing as you unlock higher tech levels. Your suggestion of moving the KerboPower up in the tech tree would result in a situation where people unlock electric engines requiring 15-30 Ec/s, with the first reactor offering 400 Ec/s. And then later, they unlock engines requiring 200 to 400 Ec/s, and their next reactor unlock gives them 60 Ec/s. That's just completely backwards. Plus, it's neither a mistake nor a problem that there exist situations in which a part is the optimal part for the job. For example, when you need one Kerbal on a spacecraft, the Mk1 Command Pod is often the best choice. The Mk1 Lander Can unlocks later in the tech tree and is lighter, making it ideal for certain situations; the External Command Seat is even lighter again by a large margin. But the basic pod is still more practical in nearly every case when there's an atmosphere and/or a need for reaction wheels involved. That is despite the pod being literally given to you for free in the starter tech node. So yes: there definitely is a certain size of spacecraft with a certain amount of power consumption for which the KerboPower absolutely is the best possible choice you can make. But just because such a configuration exists doesn't mean it cannot be the first reactor to be unlocked. One reactor must be the first, and all the others have their own optimal niches too.
  15. I'm not sure I follow you here, unfortunately... 1.) The KerboPower reactor has the worst specific power rating, the lowest efficiency, costs the most funds per Ec produced, and pulls the least EC out of a unit of EnrichedUranium. The only time you would use it is if no other reactor can do the job, because all other reactors are better in all game-relevant performance metrics. The comparison with the RTGs isn't really as straightforward either, because RTGs don't run out of fuel (unless you mod them to do so). Their high price reflects the fact that they are an infinite energy source. NFE's optional decaying RTGs patch will actually slash the price down a fair amount in coming versions. 2.) I don't know what numbers you're referring to here. Can you explain in more detail? As ar as Ec output per unit of fuel goes, our spreadsheet puts the Excalibur at 345 million and the Hermes at 358 million. This is suitably in line with the progression of the other models. 3.) Some numbers in the config file don't work as you think they do. Don't judge reactors by their config file numbers. Judge them by their ingame performance. 3a.) Efficiency for reactors goes from 28.57% (KerboPower) to 48% (Hermes). I don't know where you got 10% or 20% from. You must be using "efficiency" in a different context than us, so please explain what you mean. We define efficiency as the percentage of total reactor power that is Ec output (with the remainder being waste heat). For the Hermes, 6,000 Ec/s (AKA 6,000 kW electricity) is 48% of the total reactor power, so that total reactor power is 12,500 kW and waste heat comes out to 6,500 kW. But you will never find the number 12,500 anywhere in the config files, because as mentioned, config file numbers are not always that straightforward.