Fearless Son

I am still waiting for the day that they make Kerbals consume monopropellant for their EVA packs. It will place an interesting (to me) aspect of resource management on EVA use. Sure, you might have enough Delta-V to do that EVA maneuver, but what happens when you can only do it so-many times?

So I had a contract to retrieve some science data from Minmus. I left a lander in Minmus orbit that I have been reusing on previous missions, I just send a crew module out to dock with and refuel it, then leave it there and return with the data in the crew module. Unfortunately, I forgot to put solar panels on my transfer module. Too bad, but I can manage it! Then I realized that I put too little monopropellant into it. Too bad, but I managed to just get right next to the science lander! So I EVAed a pilot to the lander, then docked the lander with the transfer module. Phew! Then came my first routine landing, right onto one of Minmus' slopes: Some good science there! So since it was just half a kilometer away and the gravity was low, I could afford the fuel to jump to that nerby flat (one of the smaller ones that counts as a generic "flats" biome): Yay, more science! Then it was time to return to orbit and dock with the transfer module to take the crew back to Kerbin. ... that was where things went a little haywire. The intercept happened on Minmus' night side, and unfortunately I realized too late that the lander was facing toward the transfer module and was going too fast. They collided, sending the transfer module into an uncontrolled spin and breaking the pair of deployable solar panels that the science lander relied on. Because the transfer module had no probe system, no monopropellant, and no way of autonomously generating electricity, the odds of stabilizing it from its wild spin were not good, and the loss of the solar panels on the lander meant that there was now a hard cap on how much electricity the mission had remaining. The mission commander got on the radio: "Kerbin, we have a problem..." The lander was light and had enough fuel to preform a transfer to elsewhere in Kerbin's sphere of influence, but it lacked the heat resistance, parachutes, and/or thrust-to-weight ratio to safely make a Kerbin landing. So I put it into a conservative aerobreaking course, and worked on getting a rescue mission together. I could just dock with the lander and transfer the crew and science, but it was so close to Kerbin and so small that I could probably just land the whole thing if I could shield it from the atmospheric compression and add appropriate atmospheric landing equipment. I was planning on retiring the craft soon anyway, so this seemed like a good excuse to bring her home. So that is exactly what I did, and here is the recovery module docked with the science lander: Some weird graphical artifacts on Kerbin's textures there for some reason. Anyway, I let this thing do a bit more aerobreaking passes and settle itself in. Even after the ablator from the heat shield was completely expended, it still served its purpose of absorbing the brunt of the atmospheric compression and making a detached shockwave that would protect the rather more delicate module docked to it. Incidentally, through the many passes I made I kept flying by a piece of space debris, sometimes less than twenty kilometers separation. The relative velocity was enough it might have done a lot of damage if it hit. Let this be a lesson about leaving junk in LKO, especially along the commonly traversed equatorial plane! Anyway, something blew up on descent, but nothing critical as far I could tell. Probably just a solar panel or an RCS thruster. I managed to splash the whole thing down on Kerbin's night side and got a not-bad amount of science from just a run to Minmus: Think I might be ready for a crewed Duna mission soon... On the other hand, that might make a very good sparkler to celebrate launches and homecomings!

If you find Mun landing challenging, I might recommend you learn to use the piloting assist system. Basically, as long as you have a modestly skilled pilot (or more advanced probe core) in your craft and you turn on SAS, then you can have the pilot keep the craft's nose pointed in a particular direction relative to the motion of the craft. This will not autopilot anything for you, but it can make things much easier to do efficiently. So if you are having trouble sticking a safe Mun landing, go in for a surface landing as normal, where you fall in a steep arc, but turn on the SAS and set the piloting assist to point the craft retrograde (that is to say, facing away from the direction the craft is going.) As you are falling toward the surface, throttle up the engines to reduce your speed. You do not need to do it too much, too high, just enough that you are confident you have enough thrust and time left to arrest your velocity by the time you hit the surface. The fact that the piloting assistance will keep your craft pointed straight away from the direction of its fall will ensure you are facing almost completely normal to the surface by the time you impact, which makes a safe touch-down much easier. Once you are near the surface, try to get the descent velocity down to below ten meters per second, preferably below five. Keeping the throttle opened just a tiny bit should work. Then cut all velocity as soon as the landing struts hit the regolith. Having enough fuel for all of this and a return will be the next big challenge. Think of making your lander as light as absolutely possible for this.

I tend to avoid the middle distance by treating anything above a kilometer as a variation on the "long distance". Suppose I am at the closest point in the current orbits, and there is a five kilometer separation between the vehicles. Okay, I go into the map mode, make sure the nav ball is up, then set another maneuver node. I tweak the values until there is an intercept on the other side of the planet that has a separation of less than one kilometer. If I cannot do that, then just try to get the separation as low as possible, and then do it again at that point. Repeat as many times as necessary. One method that helps is to set your velocity indicator to show velocity relative to the target, then when you get close you thrust toward the retrograde marker (which at this point is showing the difference between you and the target) until the relative velocity is very close to zero. That will give you an opportunity to plan your next maneuver to get even closer much more carefully. Oh, and when doing an approach from that middle distance, yes, thrusting toward the target will get you there, but your course will seem to "drift" away from it. Use the lateral RCS systems to re-center the prograde marker over the target marker, and keep yourself on course.

Putting these two together, and we get "The Mun is a Harsh Mistress", by Robert A. Kerman.

Built my first successful VTOL craft. Includes two Panther engines for lift and a couple of Juno's to provide additional forward thrust (toggled via action group.) I use it for collecting science from those hard to reach places.

A pair of Junos on some mk. 0 liquid fuel tanks, pressed into the mk2. fuselage to give them attachment points astride the docking port. My initial design called for a couple of Spark engines for vacuum maneuvering, but those proved to be too weak and fuel-hungry for atmospheric flight and they were redundant for vacuum flight with the transfer stage there. So I elected to go with the air-breathers instead because they gave me a lot more control and options for making a safe landing than a gliding descent would allow.

On Saturday, I did a couple of missions. First one was to establish a space station in Munar orbit capable of accommodating five Kerbals, generating power, and with an antenna. I was not prepared to build a station there yet (I have not unlocked the parts I want for stations) but all the other requirements could be completed easily enough with a single ship. So, trying something a little different, I launched this shuttle: Obviously it has bigger boosters than that to get to orbit, but that was taken as it completed its circularization burn on its transfer stage. The design proved sufficient to orbit the Mun and return, though I had to use most of the fuel in the shuttle's tanks to do so. Still, the shuttle had plenty of liquid fuel left for its landing needs, as it only uses a couple of small engines with a dedicated liquid fuel reserve sufficient to keep them going a good while. I did not want to risk flying all the way back to KSC (I overestimated the number of aerobreaking passes this thing would make before its return becomes terminal) so I brought it to a nice grassy touchdown: My second mission of the evening was a simple "Plant flag on Mun" mission. Not much on its own, but I still had a few biomes I could tap for science as long as I was headed there anyway. Hence, I went to the north pole, which is a difficult landing site. Still, my lander proved it could handle it easily enough, even if the tip of the thing does "wobble" a bit too much under thrust: The wide landing fins in this design makes it easy to stay upright, even with the excessive tallness of the lander. The crew modules have a separate return stage, taking off and abandoning the landing fins, larger engine, near-empty fuel tanks, and science equipment at the landing site and returning the crew with dead weight. So long as I can collect the science and return it to the cockpit, it works pretty well.

Something I have found as a "workaround" for the issue of seated Kerbals adding mass is to shift the seats from being on top of a vehicle to being on the sides of it. Admittedly, this is primarily something I do for rovers (keeps the center of mass low when the two seated Kerbals are beside the central axis instead of above it) but it could also work for rockets like this. That way, the center of mass and center of drag all stay balanced across the center of thrust, assuming you fill both seats.

All that is missing from these screenshots is a squadron of TIE Fighters and a "Don't get cocky, kid!"

Okay, so here is a perpetual problem I have when I play KSP. I like to make my vehicles cost-effective, and that means recovery of useful parts. I have gotten pretty good at getting them back down to Kerbin survivable, but the closer they are to KSC, the more the funds recovered from the craft. This is especially important with spaceplanes, since they often need a good runway to touchdown on. Unfortunately, my attempts to land near KSC from an equatorial LKO position have a bad habit of undershooting the target, often leaving my craft coming down a continent away. Even a spaceplane, that could theoretically make that flight back, often finds itself running out of fuel during the long return flight (fuel budgets are especially tight for such craft.) Can anyone share a best practice method for getting craft down near KSC, assuming equatorial LKO as a starting position? Things like what altitude range above the KSC I should aim for or how far around the globe I should begin my breaking burn would be helpful. I realize there will always be some approximation involved in this depending on the aerobreaking characteristics of the craft and its orientation, but anything that helps me better evaluate my optimal entry would be valuable.

That is a big part of what makes the Mun surprisingly challenging for such an easy to reach body. I am convinced the Mun's geology must consist entirely of rubber, broken down by various grades of elasticity and brittleness. I find the trick is to land somewhere extremely flat, to come down excessively straight, and to cut the throttle just before landing. If you have some RCS onboard (as you might if you have a transition module you dock with) you can use a little of that with SAS to help correct the tilt. But once your lander is relatively upright and not likely to bounce up any further, cut the SAS and RCS and let gravity settle it down.

Especially important coming back down! If you get most of your electricity from deployable solar panels, you are going to want to fold them back in before you make your descent since they are super-delicate and will break off in all but the absolute slightest amount of atmospheric drag. This means that you will probably not be generating much, if any, electricity on your reentry. Make sure you have enough batteries to last the critical phase of your descent since reaction torque will probably be the only thing you have going for you to guide your orientation. Try setting your periaposis a bit lower. Most players aim for 45K or thereabouts when aerobreaking around Kerbin. 65K will eventually get you down, but it will (as you have noted) take many passes to slow down to a terminal speed. An EVA while over 75KM from Kerbin's surface is perfectly safe. Anything above 70KM is safe, actually. In fact, you can probably manage it above 65KM (I know I have a few times, under admittedly desperate circumstances, think Sandra Bullock and the Chinese station in Gravity.) Any Kerbal who departs an orbiting craft will be going at exactly the same velocity as the orbiting craft unless a force acts to accelerate them apart. Atmospheric friction is one of those things, hence why high altitude and/or high speed atmospheric EVAs is dangerous, but any EVA with a periapsis above the atmosphere is safe enough so long as you do not get lost out there or run out of EVA propellant. See 3. I would actually encourage some orbital EVA, if only because you can gather EVA reports to return for science, and you can do a different EVA report for each biome you pass over. That adds up to a lot of science!

Assuming you are playing in science or career mode, there is benefit to running a mission crewed as opposed to a purely probe situation. Crew can conduct crew reports, EVA reports, and collect surface samples, all of which means more science than you could ever get with a purely uncrewed mission. That having been said, the only crew you need to maximize all that is a scientist (hence the one Kerbal missions.) However, there are benefits to bringing other Kerbals with different roles along. A good engineer is useful if you are using a rover (and need to repair a busted wheel) or are doing ore drilling (they speed the process up) or are doing multiple parachute uses (like if you are using them for landing on an atmospheric body and later using those same chutes for returning to Kerbin.) Likewise, a pilot might not seem necessary, but you need to be pretty far down the tech tree (and willing to shell out the funds) to get probe cores that do half of what a good pilot can do, and the presence of a good pilot gives you bonus turning torque and stability correction over what you might get with a purely electrical solution, even on top of whatever a core is giving you. You could argue that a well-designed craft will not need that extra stability, but it helps if the craft is ever damaged and the center of mass is shifted just a little off the center of thrust, or the atmospheric drag becomes asymmetric. A pilot will help you make it back down safely where a probe might not.

Ah yes, the kreepers, which explode your ship when you land complex geometry during physics warp. "Niiiiiiiice shiiiiip.... *BLAM!*"

Redundancy. If one person, for whatever reason, becomes incapable of fulfilling the mission, then the other person can take over for them. If you only have a crew of one, then that is a single critical point of failure for the mission. Humans in real life, unlike Kerbals, are usually the weakest link in a mission, while also being the most adaptable component of a ship. Having more than one on a crew substantially increases the odds of mission success.

I have never hit any space debris (that was not the result of an immediately prior jettison from the same vehicle that got hit) but my experience was heavily influenced by Planetes, so I try to avoid generating an excessive amount of it and removing what significant debris I do put up there.

Just be very careful about the oxygen content of the medium you are burning it in:

Not a bug, just an "unintended feature".

Oh, a few people reminded me of another I have: Quality of life is essential for our krew. Kerbalnauts going crazy from cabin fever and isolation reduces a mission's probability of success! Never send a Kerbal out into the void alone except for very short missions (a spaceplane deploying a probe in LKO before returning or a single pilot sent out on a Kerbin-system rescue mission are a couple of acceptable examples.) For any longer trip, go out with multiple krew so no one gets too lonely. Further, make sure any craft that goes out on a trip that lasts more than week has a well-furnished passenger module. Pilot seats might be comfy enough in zero-gravity, but a Kerbal needs a space to get up and stretch a little without having to go EVA. Plus those cabins are where all the big snack lockers are!

That is actually one of the exact situations I had in mind for this kind of thing. If you are using a single-engine landing craft you can kind of manage it by right-clicking and manually setting the slider, but you have to be quick. However, if you are using a multiple-engine lander (like the kind you might if you have a docking port on the bottom or a package you intend to drop on the surface to reduce ascent mass) then not only do you have to be quick with the right-click menu adjustments, you also need to be precise to get everything to the same thrust level lest you end up with asymmetric thrust and corkscrew in a way to make your suicide-burn literal. There might be something to be said for having multiple symmetric engines and disabling them in pairs while allowing the remaining ones to go full thrust, but that introduces its own problems. For example, you would need at least four engines (two to disable and two to keep on,) which might be a bit overkill for many landers. Not to mention that it introduces additional complications about trying to balance the fuel quantity per engine (if each is fed from its own tank.)

I love the way those probe looks like it has a tiny Kerbin mounted on top.

Always have a plan to bring home any Kerbals sent into space. We do not leave our krew behind! Minimize costs while maximizing returns (i.e. a more expensive mission is justified if it could accomplish what would otherwise take two missions worth of funds to do.) Space exploration represents a significant diversion of economic resources, so make it count! Minimize debris that stick in orbit, especially in Low Kerbin Orbit. Further out and lower density debris are a little more forgiving, as are debris that remain on planetary surfaces. Every piece of flying space junk is one more obstruction we put on the paths of those travelers who will walk the trails we are blazing!

Well, I had planned on doing something in KSP tonight, but I ended up having to stay at the office until at least 2:30 AM, which is where I am posting this from.

Okay, so the ability to set thrust limiters on engines is great, and with the exception of solid fuel engines, those thrust limits can be adjusted while in-flight. This is quite handy for things like reducing thrust output to make fine maneuvers without overshooting, or when using a VTOL craft with downward facing nozzles that need to be adjusted to balance the thrust across the mass. However, it does come with the limitation that trying to do this while in flight and (especially) under thrust is incredibly difficult. On a single engine craft this is kind of possible, on multi-engine craft this is a nightmare. Try limiting the thrust of four engines, one at a time, while also under power and you will see what I mean. Thus, I would like a way to adjust maximum thrust on-the-fly (ha!) using action groups, if I can previously specify some thrust limits in the VAB/SPH. This would allow things like an efficient transition from a VTOL mode to a forward-thrust mode, or the ability to steer a craft by controllably making the thrust asymmetric. Or even just having a multi-purpose lander that can switch modes to best suit the gravity of the body it is trying to land on. And (I hope) this would be a relatively easy thing to add.