Kerbocracy

General KSP science question: does deltaV potential of a given rocket change depending on where you are in a gravity well? My intuition tells me that for a given rocket, it should have less dV in a bigger gravity well (e.g. Jool vs. Kerbin) and also it should have less dV when it is closer to the body than when it is farther away. For a fixed amount of energy in a rocket, the rocket should change it's velocity (e.g. delta V) more or less depending on the gravity acting on it. But I'm not sure if this is the case in KSP: I'm using Engineer Redux to estimate dV on a rocket I'm building, and it allows you to select different planets. The TWR changes from planet to planet, but the dV for the rocket system remains the same across planets. I find this counter-intuitive; shouldn't a rocket be able to change it's velocity less in a larger gravity well? I'm trying to build an Eve return craft so I'm scrounging for dV...

Hey all, anyone know the answer to the following question: when radial decouplers decouple, how much mass do they leave no the parent piece? They appear to leave little nubs. I wonder what fraction of the original decoupler mass these are. The same question might be applied to struts - when struts are broken, where does the mass go? I'm assuming it stays with the parent node...

Interesting; I've never tried landing on my heat shield...problem is that the extra weight it has will make for a faster landing. Doesn't the shock go up the structure and knock stuff loose (decouplers and such)? I really like the idea of putting the parachutes on a detachable lower half. Mass reduction is critical I'm finding. I had a system that had 11k dV last night, and I added struts for stability and it dropped to 9k:) I'm also going for the powered descent, so the lander will have a bottom stage that will remain, with science, parachutes, and the descent engine, taking as much off the ascent stage as possible. I still insist on the lander can though because conceptually I think the Kerbal needs 'storage' for the soil sample...

I've tried this and it worked (I had four engines pointing down on a Mun rover). It increased the traction a bit, but not enough to make it worthwhile overall... The ground doesn't seem to impose a lot of friction in KSP; Kerbals slide forever along the munar surface...

Sorry to hear about your Kerbal being attached to the wrong thing..that's rough. I installed MechJeb too look at my deltaV but found out that the mechjeb module has a weight of 6, which totally throws off my delta v estimate. I installed engineer redux, and their assembly module is virtually weightless, so I could get a read on my dV. The rocket I was about to send to Eve was about 9300 dV, which would be able to handle certain takeoffs - say 5000m and above. But seeing the incredible dV that yours has, I want to try for a smaller design now that I've got a read on dV as I build. So thanks for that. I do want to send a lander can though rather than a seat. I also like the idea of leaving the science gear there. But I can send an unmanned lander to do science; I just need the Kerbal there to get a soil sample. One question I have, is that currently my 9300 dV lander is about 130 tons. One thing I'm wondering about is how to slow the thing down as it lands. I don't want to waste fuel on the landing, but can't really fit any more parachutes on it. I used an online parachute calculator and found out I was coming up short. My landing trials on earth were controlled crashes (landing at about 13 m/s) and hence didn't allow for a viable re-takeoff. I was kinda hoping that on Eve with the thicker atmo my parachutes might be more effective. But then again the gravity is bigger. The parachute calculator indicats that I will still have a controlled crash at about 9 m/s given my current config. So. It's back to the drawing board for me. I want to build a lander (based on the light lander can) that has about 13k dV, so I can go wherever I want on Eve. I want to bring a rover. I want to bring some science gear but am prepared to leave it there. Perhaps it will go on the rover.. More work for me. My first three attempts at this have probably taken about 8 hours of building testing time:) But alas, it's fun. Oh and P.S. - I think you have to go through re-entry in the atmo, right? For this reason I was strapping a giant heat-shield on my lander that gets ejected after entry. But for realism sake I think this is part of the requirement...

If conduct test-launches using autonomous control (e.g no Kerbal pilot), make sure you clear the Kerbals out of the command seats during every launch from the VAB as they will be automatically populated there each time you edit the craft. How many times have I changed a single strut, done a launch, and at 13km in the air with a wonky launch, realize Bill is in the pod and there is no safe way home... If you are launching docked components, remember to 'Disable Crossfeed' unless you want to burn fuel from the component you are launching. When placing ladders, check to make sure it didn't automatically apply the symmetry mode of the underlying stack. When is it appropriate to have a ladder on every side of the craft? I often don't realize this 'bug' until I'm in orbit.

It's a bit sad that it's all CPU. I have an i7-3770k which is at factory clock (I forget what it is). I could try to overclock... But for gpu I have a gtx690, which would just love to use one of its cores for physx computations... The wasted bandwidth there just hurts... I get decent performance in KSP, but at high part counts (~1000) it gets quite laggy. So much for the massive space stations/cities I want to build. Hopefully they commit to finding a way to get gpus to carry the physics load (as they are apparently perfectly suited to that).

Nice lander! I like the rover/lander combined concept. But question for you - does it really have 11km/s dV? It doesn't look like it to me. Reason I'm asking is that I'm currently building an eve-return system and I've been struggling to produce something that I have confidence in for getting off Eve. I don't have any mods for dV estimation, so I'm going off feel. I know that getting off Eve is like 2.5x Kerbin roughly in dV. So with my vehicle I'm testing on Kerbin - getting Kerbin orbit and then some. The one I have now is like Kerbin escape + 2000 dV. I think that's adds up to close to 7000 m/s. But right now my ship is quite a bit bigger than yours. It has 5 stages, and 2 booster stages on top of that. The last stage is an asparagus burn for high atmosphere. Finally, I have a single ion-electric engine attached to the lander can for final orbit correction and return to the mothership. I heard that the Eve atmosphere goes out to 100km, so I imagine there is a ton of dV lost in air resistance and the gravity turn must be much later than on Kerbin. Anyway, how did you get the 11km/s dV estimate? Have you tried to take off from Eve in it? How does it do in taking off from Kerbin?

Ok makes sense. Though - if I separate from my stages _right_ before crossing into the new SOI, shouldn't both me and my debris be subject to the same changes upon crossing into the new SOI? Therefore both of our trajectories might change in similar non-intuitive ways?

I like! I also would like an alarm that sounds when your stack starts to encounter dangerous resonant oscillations. I think that in real life they have sensors for this and they cut engines based on it. I would just like to have the experience where an unstable build starts to oscillate and in the period prior to breaking up, alarms are going off. Other options might be: low power: remember 'Main Bus B Undervolt' from Apollo 13. too high pitch/roll/yaw rate (similar to stability loss above, but I'm thinking also in a vaccuum) - so when you are spinning out of control (e.g. after an explosion) an alarm sounds. Damage to parts. Like if a strut breaks, there could be a 'Master Alarm' that goes off that you have to press to deactivate. I always like that Master Alarm switch that they have to press in spacecraft:)

Yes! That's cool. I noticed that when you enter the Jool gravity well you get so many encounters with its moons. It feels like you would have to think carefully to produce a stable orbit without any spurious encounters. I love the concept that that could happen though. I'm going to run a couple experiments on this when I get home...

Ok thanks for that. Collisions clearly depend on the physics of the crafts being modeled, so if we're focused elswhere or outside of 2.5km, we have to live with a no-clipping reality. So the question then (still) is - do unattended objects have their trajectory updated by encounter with a gravity well... If so, that's really cool: it means that my debris in interplanetary space might eventually do some unpredictable stuff, or might dispose of itself:) If not, that's less cool. But more functionally, in the context of my original question, it means that if I want to dispose of my transfer tanks in interplanetary space using a gravity assist (as in my Mun approach), I need to detach from the transfer stage after the encounter with the gravity well in question; detaching before will result (likely, if detached debris > 2.5km away) result in the transfer stage not being gravity assisted and remaining in a kerbin orbit. I suspect that this is the case...

The reason I wonder about it is because when not focused on an item and if it's > 2.5km away the game stops updating the object's trajectory based on the physics model. This is when when you detach from stages in the atmosphere, once you separate by 2.5 km, those parts continue on rails and you continue to slow down because the game stops applying the air resistance to those parts. They only disappear from their 'rails' if their track takes them below 25000m in altitude. So, if you put debris on a decaying orbit (say Pe = 40km), the object's orbit will only decay while you are 'flying' it. Otherwise it continues on rails and ignores the atmosphere. By this logic, if unattended encounters with celestial bodies rely on the physics model, then I guess they won't happen. This would be consistent with the developer's choice to update the physics of only the object being 'flown'. But encounters are different than drag and might not take the physics model into account (thought they might take the object's mass into account, mind you the mass of our orbiting objects is negligble compared to the celestial bodies). Another related question is if orbital collisions can happen with unattended objects. For example, can your unattended space station orbiting Kerbin collide with space debris while you are doing an EVA on Duna? I think not. But we haven't had a definitive answer to these questions; everybody's got some kind of semi-related opinion. It would be nice to have a Dev chime in on this. I suppose I could go do the experimentation myself though:)

I think the anomalies should be randomly placed (with constraints so they look good) at each career start. Special parts should be unlocked for anomaly finds (e.g. parts that can't be unlocked otherwise).

I totally agree. Dumping fuel makes the craft lighter, so there could be uses for it. For example, if I build a generic lander that can land on bodies with an atmosphere or ones that have none, I will have a parachute assist in one case and not in the other. Therefore it might take more fuel in the case of the vacuum, and in the atmosphere case I might want to dump the unneeded fuel to lighten the load and reduce the likelihood of parachute and landing gear breaks. Anyway, I like the idea of complete control of that stuff just for cool micromanaging. I love transfering fuel. Wish there was a way to it more precisely (as in the above post). I would also like a way to automatically rebalance it for symmetry against the axis of lift, so when you shuffle stuff around it can then be exactly symmetrical again.

I think that would be kinda cool - but it would have to be a part you would add, right? The only thing I would worry about is that with all the smoke/exhaust/dust that would be below the craft as it landed it might obscure the view. Didn't the Apollo guys land via instruments? Taking the camera thing further,one possibility would be to have rover views restricted to a camera that they carry. It would be like an IVA but from a rover point of view.

Spurious encounters happen often close to Kerbin though. I've had spent stages that are on a wide Apoapsis around kerbin and had them cross the Mun or Minmus's path. When this happened I was excited because it meant that the object's orbit would change and just _maybe_ I wouldn't have to track it down and de-orbit it myself. In this case I focused on the debris object (flew it) and saw it through the encounter, which indeed slowed it down and crashed it into Kerbin. The question remains - if I had not 'flown' the debris, and had left it unattended, would it have had the encounter with the Mun and thus had its trajectory changed?

Perhaps you're right - I might have been at > 1x time accel. when I crossed the SoI boundary. Next time I dump my transfer stage I'll make sure to cross the SoI boundary in 1x. But the more basic question remains: will the orbits of a piece of debris be updated by an encounter with a celestial body? In the case of my debris that is orbiting Kerbol, if it happens to encounter Eve, say, when I'm not watching it, does it's trajectory get updated by the encounter or does it continue on 'rails' as if it never had the encounter?

Hi all, I have a garbage disposal question. I've been disposing of my spent stages in interplanetary and interstellar space. For example, on my way to the Mun, I'll plan a trajectory with my transfer burn that goes through the Mun's influence (behind it) and out the other side into a Kerbin Escape trajectory. Once I've achieved that trajectory, I dump my transfer stages and then my lander or probe does burn at periapsis to get into orbit. The other stages etc just sail on into interplanetary space. If I give them enough energy, apparently they escape Kerbol as well. I like this process because while there is still litter, it is over such a huge space that I'm guaranteed never to see it again. In doing this I have noticed some quirks that I wanted to mention. 1) Let's say I separate from my transfer stage prior to my Mun encounter. My transfer stages are drifting very slowly away from my lander. Upon Mun encounter, they suddenly accelerate away from my lander (to about 10km within 10 seconds or so). Why does this happen? It's as if my trajectory has changed upon Mun encounter and they have continued on my previous trajectory. If so, why does my trajectory suddenly change upon Mun encounter...? 2) A related question is: if I wait until I make my Mun encounter to jettison the transfer stages, they continue past periapsis and to Mun escape on the other side. Meanwhile I've burned into orbit at Pe. Do those spent stages then achieve Kerbin escape as well following the Mun escape? The reason I ask is that under 1), when I left those stages, I still had the trajectory plot showing Mun encounter, Mun Escape, and Kerbin escape, indicating [to me] that the spent stages would eventually achieve Kerbin escape. When I jettison them in the Mun's influence, they only have 'Mun Escape' listed, and I go on faith that they will also achieve Kerbin Escape. However, because I suspect a sudden trajectory change upon entering Munar influence (described in 1), I no longer know if the jettisoned stages will escape Kerbin. In testing, I've found that they do. But I've also found some jettisoned stages that I thought would have Kerbin escape doing a very wide Apoapsis orbit of Kerbin... 3) In examining my spent stages that have managed to escape Kerbin orbit, I find them listed as 'Orbiting kerbol' or 'on an escape trajectory from Kerbol'. The orbiting Kerbol ones make sense, but the escape trajectory ones seem to have defined Kerbol orbits. So I'm not sure why they are listed as escaping; their orbits are defined (and fall inside of Eeloo's orbit). 4) A related question: Are unmanned debris items' orbits updated by encounters? For example, for my debris that is orbiting Kerbol, if a piece happens to encounter Duna's influence when I'm not looking, will it get captured or have it's orbit altered? This relates to my question in 1) where I was wondering if I jettison debris just prior to an encounter, does that debris still have the encounter. I suspect based on my observation that the answer is 'no'. Thanks in advance for any feedback.

Hi all, Forgive me if this has already been said; I scanned the titles here and didn't see it. IMHO, I think that science in Kerbal should have a role in advancing our knowledge about the Kerbol system. More than just unlocking parts and providing humorous blurbs of text, the results of science experiments and measurements on other worlds should provide functional knowledge about those worlds. It would be neat to start the career in a state of ignorance about the Kerbol system and then acquire knowledge about 'what's out there' through missions and experiments. In essence, this would be real 'exploration'. I have to admit that when I started the game, I was a little disappointed to be able to pan around, zoomed out, in the map view, and focus on and examine the planets and their moons. I could see their basic appearance, get a sense of whether they had an atmosphere, and make decisions about which ones I wanted to land on. This was all by just paning around. On this basis, I discovered a terrestrial moon that became a priority for a manned visit. I feel that this access to knowledge about the Kerbol system was too free. We should have to earn it through science and missions. Perhaps this could be a 'Hard Mode' or 'Kerbel Prize mode'. I can imagine a starting situation where we have basic astronomical knowledge: we know the orbits of the major bodies in the solar system. But we don't know much about them except perhaps, the class (e.g. planet vs. gas giant). We then do missions to discover the moons and the particulars of the individual planets. Importantly, to see them, and to learn about them, you must go there. Information that I can envision being hidden initially and revealed through missions might include: a) atmosphere profile: whether or not there is an atmosphere, and it must be profiled by a descent probe. There would be a corresponding density graph to look at and use to plan future entries, fly-bys, gravity assists, aerobraking maneuvers. existence of moons; while perhaps a bit unrealistic because they can be found through astronomy, I would like to discover moons by visiting the main bodies. Along those lines: c) precise orbits of smaller objects. It would be neat if we didn't know the orbits exactly, and had to guess a bit and correct on the fly. The gravity meter might be used to determine the exact mass of the main body, and the moon, thus refining the player's understanding/model of their orbits. To take this further, the orbits could be defined as bands (e.g. confidence intervals) - a band that we know the body will pass through; The width of that band gets thinner as you know more about it. That way there would have to be some careful guessing/calculating to make safe orbits; this would be especially interesting if you don't know atmosphere composition. Uncertainty would be cool!!! Remember that it doesn't have be uncertainty in the underlying model, but rather just uncertainty portrayed to the player. d) restricted vision: focusing on, and zooming in and out of planetary bodies allows you to inspect their surface. I think surface inspection should have to be done by hand. Either have probes that orbit the bodies allowing you to inspect them in real time, or even better: take photographs that you can review. Imagine you are orbiting focused on a satellite, and you spot a feature of interest, you roll and photograph it and it tags the position and gives you a still shot that you can review. You can then make decisions about where to send a lander. I don't like being able to examine the surface at leisure in god's-eye-view and pick a landing spot. Similarly, when landed, I don't like being able to zoom out to half the planet's width and inspect the surface features in that zoomed out way. To look over that next ridge, I should have to go there (or fly over it). e) physical characteristics of the bodies; material composition, magnetism, volcanism, etc. (I know this might be geeking too hard). Please suggest any other knowledge that you think could be initially hidden and then revealed through science! I could imagine a Kerbolpedia that would be filled in as you accumulate knowledge in Career! (remember the UFOpedia.. P.S. this game is special: I spent ~100 hours playing it in the first two weeks that I had it. That must be a personal record...