Demetrious

I'm glad to see it's a known issue, at least. I have no idea why it seems to strike ships in kerbol orbit, though. I've had satellites that encountered the Mun and been thrown out of the Kerbol SOI; switching to them has never produced a problem. I wonder if it's related to the long duration of the flights, the fact they have a planetary intercept in the far future, or both. Running KSP 1.05, latest version, by the way. Extra info: -- The ship wasn't big; not a massive interplanetary vehicle assembled in orbit with an obscene number of parts. It was a small rover delivery vehicle. -- I'm running a Core I5-3570K with 16 gigs RAM and a Geforce 670; hardware shouldn't be an issue. -- I've only ever had this problem in the two instances cited - switching back to an interplanetary vessel after it's been in space for hundreds of days, while it's en-route for an encounter with the target planet in a transfer orbit.

A while back I launched a mission to Duna. I completed ejection from the Kerbin system, entered solar orbit, made a second burn to adjust my trajectory, and was pleased to see the patched conics report an intercept with Duna. Satisfied, I returned to the Space Center and flew many more missions here and there, occasionally checking on my rover mission in the Tracking Station - still on-course for intercept, days ticking down, with the ship's situation listed as "escaping the sun," even though it was simply on a simple Homann transfer orbit to Duna. So, at long last, it was only weeks away from its Duna encounter, so I clicked on the tracking station and switched to the ship, and lo and behold, what do I get? I get this. Trying to "quit" back to the Space Center did not actually show me the Space Center - it just showed me the starry skybox. Even quitting to menu and re-entering game just showed me the starry skybox instead of the KSP launch complex, even though the time acceleration bar was present at the bottom. I had to quit and restart the game to fix this, and naturally, when I checked back, my flight (which had been in progress for over 200 days) had vanished. The ship was dead and gone, as if it had never existed. This is the second time this has happened to me - the first time, (in an older version) it was a manned mission, and it auto-killed all the Kerbals aboard, too. I figured this would've been fixed by now. I figured wrong. I'd like to know what did this, because it is seriously liquiding me off. The only mod I have installed is Kerbal Engineer, version 1.0.18.0.

I stumbled across the 1.3 plugin on the Spaceport, installed it, and promptly removed it feeling very disappointed when it proved antiquated (.viz broken in 20.2.) On a whim I thought to search the forums, and when I found someone was actively working on new versions I squealed like a little girl. Just posting to say your hard work is greatly appreciated; I'm especially impressed that you're refining the code and working on effects-balancing at the same time. For a volunteer effort, that's above and beyond.

I've seen well-made bases built upon landing gears (which roll freely and don't provide much resistance) that are then pushed together by a dedicated "tractor," but that method of having the rover underneath the module, then having the module hike itself up on legs so the rover can detach and roll away, is really elegant. I've started assembling my own base and... my solutions have not been so elegant. Here's the Kerbin test prototype of what I wanted to do - add a vehicle service module to my Mun base core. I decided to do it by adding rover parts that I could then drop off once the module was deployed: This gave me several challenges - move something heavy, (all that fuel for re-fueling surface vehicles/ships) and move something bulky (those big girder-legs) and move something asymmetrical (lots of RCS to keep stability under thrust,) and somehow take that load and transition from the transport/landing configuration to a wheels-on-the-surface-configuration. Which is how I ended up with a ship looking like this: I let MechJeb handle the landing, because precision landings near other structures are a pain in the rear end otherwise: On the surface at last. Now, we need to deploy the vehicle, so Bufred Kerman has arrived to take control: His expression here is priceless. LETS RIDE THIS BRONCO! First, I decoupled the coupler holding the "nose" onto the truss structure of the landing rocket, then gave the lander (which had its own remote control pod) a brief jump of the throttle to lift it from over the rover and set it down on its side nearby. The "rover" was sitting on the flat base I made for it from a decoupler and a Rocomax adapter. Next, I decoupled that and gave it a brief forward jolt with the few RCS thrusters I'd placed on it, putting it wheels-down and ready for transit: Then Bufred carefully backed it up to the Mun base, lined it up very, very carefully, and blew the couplers holding the transport legs and extraneous equipment, dropping it into direct contact with the base: Docking is a little finicky; the game wants it "just so" in terms of distance before ships will come together. In orbit this isn't a problem, the natural drift of ships (helped by the "magnetic attraction" between docking ports versus the equal-and-opposite kickback when ports bump into each other) will mean they'll eventually drift into the "zone" and snap together. With a base, obviously, this won't happen as easily, but it takes very, very little to goose ports together: in my case I switched to the Mun Base core, which has an MK2 lander-can as its control module, and just hit the roll buttons a bit - even that tiny bit of wobble induced by the SAS was enough to nudge the modules into docking properly. This is very, very inefficient - as evidenced by the fact that my first "vehicle" the vehicle service module will service is a tractor being sent up there to drag away all the extraneous crap lying about from deployment stages - but it does work, at least. If you can build boosters, that is.

Which would be here'>http://kerbalspaceprogram.com/forum/index.php?topic=17393.0']here, in a thread I started about the exact same problems without, apparently, seeing that one already existed. Go me. There\'s a lot of good advice in very little space there, so give it a look. I learned, straightaway, that simple G-Forces are the problem, and that\'s because the new parts provide truly massive power. As others have pointed out, the new engines generate a few orders of magnitude more thrust then the old ones. My remaining, insurmountable obstacle was solved simply by throttling back a bit - just a wee bit - after takeoff. My rocket was basically having the top smashed into the bottom when G-forces were exceeded - crushed top-to-bottom like a beer can, with the point-of-failure being the oft-lamented new decoupler. Using the 'new' parts as a powerful main-stage and the old parts to make good boosters works fine, but after accounting for the terrific power of the new engines, I was able to launch a (properly-reinforced) rocket into orbit that was a carbon-copy of one of my oldest and best designs. It\'s possible - just watch that G-meter like a hawk and keep it in the green.

That\'s the conclusion I\'m coming to as well. I\'ve had some success using 'large' main stages and using 2-tank 'large' parts/engines as boosters - the immense power makes them fantastic for that - but even then it can be iffy. That would explain it. If there\'s a G-meter in the GUI I really need to start watching it - it\'s obvious I vastly underestimated the size and sheer power of these new parts. I moronically thought they were there to match the 'new' 3-man command pod, so, as the command pod had been scaled up, so had the rocket parts, and roughly the same relationships applied as with building with the 'small' parts. I can see I was mistaken. The way you cluster multiple 'small' boosters onto one decoupler is pretty clever; it solves the issue of simply not having enough space to stick boosters onto (which was driving me towards using the new parts as boosters, inevitably.) I\'m also glad to see I\'m not the only one who thinks the new radial engines are really, really nifty. Thanks for your help! EDIT: I especially like your use of fuel lines + the radial engines so that all three stages are firing from the get-go, but the final stage ends with full tanks.

Upon downloading 0.16 I skipped gleefully into the VAB to try out the new 'large' parts. After embracing this new era of more realistic scaling, I soon discovered the new parts have all the structural integrity of beer cans. It makes sense that, with the parts scaled up roughly 3X, their structural integrity would be lower. Since the weight of support struts wasn\'t scaled up, you can brace them sufficiently without making it less efficient then a 'small' rocket. That\'s logical, right? Except it doesn\'t work. You can watch those huge fuel tanks wiggle and wobble just sitting on the pad, as if little more then superglue on top and bottom keeps one attached to another. I\'ve tried bracing, and cross-bracing, and X-Bracing, and - on my latest rocket - I\'ve even resorted to running struts right down the tank-stacks, to try and hold them together. Sometimes, I can get the smaller ones into orbit, but any proper three-stage, Mun-capable rocket is a no-go. The latest one made it to 35K or so, solid as a rock, before it all went sideways. (Literally, after half the rocket exploded.) And leading the Event Log, in those same maddening, mocking tones: Structural Failure on linkage between Fuel Tank and Fuel Tank- Clearly I\'m missing something here. Anybody care to educate me? I\'ve built rockets that look like an explosion in a girder factory and I still can\'t put much payload on them (i.e. anything more then a minimalist Mun lander with the 3-man, 4 mass command pod) without them going mass asplode. EDIT: Okay, re-checked, apparently it\'s a bit more then 3X. I are smrat