HeadHunter67

I understand. I can imagine that it's a more challenging proposition than a docking trainer. But I know it will help a lot of new players - I cannot be the only one who's had so much trouble with it, even after all the videos I've watched and threads I've read.

That ring is sick. Yet another awesome part I'll "need" for my space station.

This is precisely the issue - and everyone that ignores the practicality to focus on the "theory" is thinking too hard and wasting their time. The answer to the "theoretical" question is: DUH! Why would you need a heat shield if you were only travelling at terminal velocity? The whole reason heat accumulates in re-entry is because a ship is going faster than terminal velocity and atmosphere is accumulating under the craft. It's friction and oxidation. Anyone who needs a "theoretical" answer doesn't understand terminal velocity in the first place. So, then, it's clear that the only problem worth considering is the practical application. And it's equally clear that the rocket needed would have to be more than twice the size of the rocket that launched just the initial payload (in fact, considerably more so). Given that a majority of efficient lifters only have a 10-16% payload capacity to LKO, a 10-ton payload would need about a 65-ton lifter. So if you were planning to deliver that 75-ton assembly to another atmosphere, it would need a lifter of about 470 tons. (Even launching just the return de-lifter in Sethnizzle's theory would take 406 tons, not much of a savings eh?) You might think "but I need less fuel coming down, because I'm assisted by gravity, rather than fighting it!" Well, that's what causes re-entry heating, as I've already explained. If you want to come down no faster than terminal velocity, you will be fighting gravity all the way. It matters not whether it's offsetting 1G on the way up, or the way down, it takes the same amount of energy. BINGO. Ralathon gets it. That's why it's done the way it is, since the days of Mercury.

Hooray! That is the best news I've heard for KSP! I learned so much and so well from your docking mission - especially from playing around with docking other parts of the pods to other ports to practice translating. I got it down pretty well, and in no time at all. If you can build a tutorial mission that teaches people how to accomplish a rendezvous in the first place, I am confident that I will get the hang of that quickly too! I eagerly await.

It's nice to see someone tackling this. I had basically done something similar by hand - there's no need for three sets of decoupler, or nose cones, or fuel tanks etc. when they share the same model in different scales - so I simply placed the .cfg files for the 1m, 2m and 3m parts (naming them accordingly part1m.cfg, part2m.cfg and part3m.cfg) into the same folder with one set of model files. Saved a MASSIVE amount of memory footprint just doing that. It allowed me to put the parts I wanted into an already nearly-full build. This method might not work for someone who's already got .craft files or existing ships using these parts (though it should), but for a new build it's perfect.

Also, on a slightly unrelated note... after undocking, I tried to land them both at KSC. Without any mechanical assistance, I was ablee to get one craft to land in the ocean 43km East of the pad, and one in the mountains 55km West of the pad. I underestimated on one and overestimated on the other... but refining my landing technique is a topic for another thread, I guess.

Thanks, Steve! That was kind of the whole point behind my Evil Plan. I've already constructed a modest station (two Jumbo fuel modules, a hab core and power/RCS) but that was done using MJ's rendezvous autopilot (and sometimes docking autopilot, though now I'm actually better at manually docking than MJ is anyhow). Now I just need to come up with some better station designs for my new project - there's lots of inspiration here on the forums. Thanks to all of you for being far more patient with me than I would have, with all my newbie questions and failure to comprehend. I think I got the basics. Only problem I still have is due to the SAS changes in .21. I find that even when SAS is engaged, my marker sometimes slides away during orbital burn maneuvers. The smaller consequence is that it makes my orbits a little eccentric or off-declination if it happens when circularizing... but if it happens when trying to kill RVEL, it can spoil a rendezvous entirely. I'm wondering what's ultimately changed - they touted these changes as a replacement for ASAS, allegedly now you don't need it because SAS should function the same, except you can maneuver without disabling it... but the reality is, that very difference can lead to the problem I have. So how do I get around it? Adding more reaction wheels? The problem I then face is that sufficient torque for early maneuvers (ascent, orbital insertion, circularization) means the smaller craft that's left at rendezvous has too much torque and the slightest feathering of controls throws me way off whack (not what I need when fine-tuning a rendezvous!). I suppose I could toggle off the unneeded torque at that point, but honestly, my hands are already pretty full and time is of the essence by that point. How do you effectively keep the ship from drifting off that blue maneuver node target?

The nice thing about using lights of any kind for docking is this: If you place them in the same fashion when constructing both vessels (for instance, 4x symmetry on nadir/zenith/port/starboard, or 45 degrees off of each), then it's simple to use them for making sure you are properly aligned on the roll axis when docking (of course, the wonderful Docking Alignment Indicator plugin has a handy roll indicator anyhow). And lights are helpful for more than just the docking itself - I find they help in the near rendezvous immediately prior to initiating docking. It's hard to set a port as a target (thus activating your docking plugin of choice) if you can't see the bloody target craft, right?

For someone who doesn't want to struggle with mathematics while having fun, the Zenith lifters are a godsend. Not the prettiest launchers, but VERY reliable - and reasonable part count for what they do. Using the SelectRoot plugin, I added all of these lifters in Subassembly Manager so I can slap them on underneath my chosen payload. However, even with Temstar's recommended technique for saving the subassemblies (and, in fact, using his subassembly files), it seems sometimes the struts and fuel lines get borked on the larger lifters. A small price to pay for power.

Thanks for the pictures. It helps to have a visual like that.

So you never actually did the burn from the maneuver node? Or was that left out?

Thanks, how do I view those large enough to make them out? Clicking each gives me about a 700x400 image. Can they be viewed larger? I know I can zoom by clicking on them but then I can't seem to view the annotations or scroll through. I think I follow you up to the "approaching the target" image. Adding 115 m/s should surely cause your orbit to completely pass by something 20km away. Also, how do you "pull" the prograde marker onto the target marker (or "push" the retrograde onto the anti-target?) Please explain the method by which I can reliably align. I must agree, however, that the Docking Alignment Indicator is a wonderful plugin. I don't strictly need it at that range, but it can sometimes be helpful in knowing which way to translate (those needles are the best addition to the mod yet). This time, when I tried, I mostly used the navball and the eyeball within 100-ish meters. I thank you, and all the others who've offered advice, for your time and considerable patience. I think my knowledge and my technique are improving. I wouldn't dream of saying I've got it down like docking, but I am getting better. The most challenging thing about the close approach was because that entire portion happened on the night side. I had only the little target circle and my headlights to guide me. Once we got within about 20 meters, we crossed the terminator and thankfully, completed the docking in the light for a nice photo op. Here's my second-ever successful attempt at manual docking. I used SmartASS a tiny bit (mostly once to align the docking target to the approach ship, and again to figure out what I needed to do when the distance started increasing again no matter what I tried). Next stop... learning to land these bad boys at KSC!

OK, so I managed to get a nice intercept - looked like 0.6km until I finished the burn, turned out to be 0.4 when done. Turned to begin a very slow braking when the target got within 2.5km, managed to close the distance until about 660m... then distance started increasing. No matter what icon I burned against, the distance increased. What do I do to close again? Even pointing straight at the target and burning only increased the distance. This is exactly what I meant at the beginning - I can get close enough to wave before my target sails by, forever lost to me. I could use Smart ASS to TGT+ or RVEL- but the whole point of the exercise is to do this myself. I'm so close but as they say, "A miss is as good as a mile"... or in the case of orbital maneuvers, a million meters.

Here's an example of what I was trying to say earlier. The interface makes it impossible for me to plan what should be a simple maneuver - even if we ignore for a moment that it can't distinguish between a planned maneuver that should already have been performed (target time T+) or one yet to come (target time T-). Two ships - one at an 80km orbit, about to catch up to one at a 100km orbit. If I zoom out farther, there will be no differentiation between those rings; if I zoom in any further, I'll lose sight of the smaller orbit because of the planet. I can't work with this!. I don't see a way that I can plot anything useful in this screen - and going back to space view is like flying blind. How am I supposed to plan a maneuver or burn like this?

When I try to explain reaction wheels, I ask if someone's ever seen the thing where a person sits on a spinning stool and hold a bike tire by its axle. Put some spin on that tire and when you change the angle of the axle, the stool spins as a result. Now imagine that you have three machines on your craft that do something just like that. Those are your reaction wheels - and that's how your craft can turn without any apparent external input. It only becomes clear when you stop thinking of the craft's motion as the internal subject of a cause, rather than the external result of the effect.

My question here is, how do you know if the target is "ahead" or "behind" if you haven't even established your orbit yet? (and a related question is: "What's 'ahead' or 'behind' when launching from the ground to a rendezvous? Where's the cutoff?) I've learned enough from watching MJ that I'm willing to try it myself - but honestly, I'd submit MJ as the best way not to perform a task the "right" way. When I've tried rendezvous in MJ, it does some outrageous things - my docking target is at a 100km orbit, so it propels the docker to a 218km orbit or something crazy to intercept. And when killing RVEL, it kind of flops back and forth sometimes too quickly for me to understand what's going on. It does, however, have the advantage of being "hands-on" for me - I can alter camera views, switch between screens, etc - things I can't do when watching a video. What I really need is a tutorial scenario for rendezvous - Blizzy78 made an excellent docking trainer that taught me how to dock in no time. I can dock at a port on the other side of a station, using the port on the back of my docking pod... no problem! But as for the getting there, I'm reading and seeing a lot of information that I'm having trouble rectifying into a coherent method.

Please feel free to elaborate if I am wrong, but all that guarantees is that you will arrive at the same orbit as the target - but I am failing to see how it guarantees your target will be right there at that moment - especially if you're not even using maneuver nodes to see encounters. How are you seeing an encounter in this situation? Or are you just flying around and around that ellipse until the serendipitous moment where the clockwork lines up that one time (which may mean a long, long time of watching)? I'm hearing a lot of "seat of the pants" advice, but what I need is something reliable - something that can be replicated reliably and if done properly, will work in a predictable way.

OK, the image helps a little (though it's kind of busy and the part you describe is hard for these old eyes to make out clearly), but I'm still not seeing how there's any guarantee the target will be there when you get there. What's to say it won't be on the other side of the world then? In which case, all you'll be doing is chasing each other around the orbit. Am I missing something here that's obvious to veteran players? It sure would help, thanks - providing that the information I need is pretty clear to make out amidst all the rest. I wish there was a way to hide all the other orbital stuff that doesn't pertain to the target (other ships, debris orbits, etc).

I guess my problem is, I just can't visualize this part right here. To my understanding, your new apoapsis and periapsis will never intersect the target's orbit, though it should cross at two points somewhere along the line. Whether your target will be at either of those points at the same time as you, seems like a matter of luck to me. Hang on here? Am I understanding that the "best practice" for rendezvous is a blind guess? How in the world am I supposed to kow where the target is going to be by just blindly changing my velocity? Maybe "eyeballing it" works for those who already can do it, but I can't imagine how one can learn to do it that way without a lot of frustration and aggravating trial and error. Let me be clear that I'm not willing to give up fun to be good at a game - seems to defeat the whole purpose. I've already invested a lot of hours of annoyance at trying to prefect this, to the point where I've almost given up and said "%#$& rendezvous, let the robot do it!" Also, now that you can't just undo a mission, I'm not keen on killing dozens of Kerbals to get to that point. I guess I'm not the kind of guy that learns well by reading something and being asked to imagine something I don't understand. Seems clear to anyone who's done it, but try explaining a waffle to someone who has no concept of a pancake. As for the video tutorials I've watched, it seems they've taught me incorrectly, or else I'd already be able to do it.

As for speeds? I'm talking, for instance, about going from a stable orbit at 80-100km (where one might refuel or wait for a window), to an orbit of 120-150km where one might want to dock with or set up a station. Maybe my mistake was trying the method I've seen where the orbits are significantly apart (like 100:300km) so my RVEL is about 300 m/s. The problem I have is, when I tried to practice as you advised a pair of orbits that are only maybe 10km apart at most, I cannot manipulate the map screen or maneuver nodes effectively. If I zoom in close enough to see the difference between the orbits, I lose focus on what the planned maneuver will do. If I scroll out far enough to see its effects as I plan it (and you kind of need to), I can't actually manipulate the node, because as I said, if your pointer's off by a pixel it will skew the maneuver too much. That's why I was trying orbits that were farther apart - because I could see what's going on. If I wait until 30 meters, it's too late - the target sails by and keeps going. I mean, how much later could one wait to start killing RVEL? And I don't see how those two statements can be rectified to one another - at 30 meters, if you have 1% of the distance in velocity, you're only going 0.3 m/s. In other words, you were with the target all along. Still not getting a "when" answer that would seem to work or make sense. I'm not trying to be obstinate.

Meaning, an orbit at which an intercept can never occur. I guess I'm not understanding how this helps - it seems that would defeat the whole purpose of going to the effort of matching the orbit, for the sake of one shot when your orbits intersect again. No sir. I am talking about encounters at 400 meters or less. I will reiterate that my problem is not an issue of getting close enough to the target - it's a matter of killing the RVEL at the right time so as not to widen the distance (if too soon) or watch the target sail by into the distance (if too late). All the toying around with orbital arcs isn't answering the one sticking point for me in this whole thing. But I thank you for your assistance.

It's a bit hard to pull off even with the maneuver nodes - since they're so finicky that if your mouse pointer's off by a pixel when you slide them, it deforms the thrust vector in an undesired way - or closes the node... neither of which will help when time is tight. I understand the basics of a Hohmann transfer - accelerating to raise your orbit to a higher one or decelerating to lower it, that little arc that connects your previous orbit to the new one essentially. That's not the problem. The problem, as I've said before, is knowing when to shed the sudden difference in relative velocity that got you to the new orbit in the first place. This is why Apollo13 and I are having the same issue - too soon and we can never close the gap (because chasing the target would alter the orbit), too late and it sails right by, scant meters away, never to be seen again. When. When?

Well, that's a lot of food for thought, thank you. I want to practice rendezvous at an altitude of at least 120km (as that seems the lowest sensible altitude at which to establish a station), but if 300km isn't going to have any noticeable difference, I'll stick to the lower orbits.

Wait a second... you mean relative to the target? So perhaps my issue is that I've been misunderstanding the info I've been given. I was taught to burn off my RVEL by the purple retrograde marker. If that's not the case, it's no wonder I sail by. So, tell me then, what are the purple icons good for in target mode? And in that mode, what shows my orbital pro- and retro-grades if I need them?

I was with you right up until this point. Could you elaborate? Because I'm not understanding what you mean by "raise" or "lower" when you speak of the "same orbit". ...Aaaaand, this is how you wind up with an "orbit" that terminates below atmosphere. I managed to turn a 300km circular orbit into a sub-atmo orbit this way yesterday. I call this process "jousting" - it's horribly wasteful of fuel and is guaranteed to drop your orbit below the target. Docking's not a problem for me - trying to rectify all the (sometimes contradictory) advice I'm given is what's hanging me up. I'm sure this all seems clear to people who think rendezvous is easy (just as I think docking is easy)... but there has got to be a more sensible and fuel-efficient way to cancel relative velocity and still end up close enough to the target to finish via RCS. I know where to point and what to do - the problem seems to be in the "when", and nothing I've seen anywhere has answered that question for me.