Welcome to Jester's Trek.
I'm your host, Jester. I've been an EVE Online player for about six years. One of my four mains is Ripard Teg, pictured at left. Sadly, I've succumbed to "bittervet" disease, but I'm wandering the New Eden landscape (and from time to time, the MMO landscape) in search of a cure.
You can follow along, if you want...

Tuesday, February 4, 2014

Failed success

In his book Lost Moon, astronaut Jim Lovell describes the Apollo 13 mission as a successful failure because it did not achieve any of its mission objectives but that his crew returned safely from the moon. By a similar measure, this is a story about a failed success. Yep, time for another KSP update, and it's a long one!

In my last Kerbal Space Program update, I expressed my intent to build a big-ass planetary transfer vehicle to be named Enterprise. In between writing and editing Winter Summit sessions on Saturday, I built and launched her:

That's a total of nine X200-32 series fuel tanks, three "Skipper" engines, and six in-line docking ports down the sides. I topped it with a pair of big RCS fuel tanks, a science lab, and a three-man crew cabin and large size docking port. RCS thrusters down the entire length ensure reasonable maneuverability for such a large ship. And power is provided by two XL solar arrays feeding four medium-size batteries. At the top, a pair of Science Jrs and a pair of Mystery Goo containers plus various other science instruments. As you can see, no parachutes and only the loosest nod toward any further separations. Enterprise was built for space.

My intent building her was to have a heavy transfer vehicle to which I could attach landers and then send out to Duna and later, to Jool. My space station was built to refuel a big beast like this and Freedom served her purpose perfectly because Enterprise was a really thirsty ship!

Docking went surprisingly smoothly and I all but emptied Freedom's two big fuel tanks into Enterprise's three, resulting in full tanks when I undocked.

My original plan was six landers but on review of how big Enterprise turned out to be, I revised this plan down to four. Instead, I decided to attach a pair of refueling shuttles to the central docking ports. My intent was that if my landers could make Munar orbit but were having a hard time reaching Enterprise's parking orbit (which I intended to be polar), then the shuttles could go down and add that little bit of extra fuel to make the difference. Knowing that the G-forces would stress the docking ports, I thrust-limited all three engines to 33% and off Enterprise went to the Mun! She started at about 160T but by the time she settled into a nice 1 degree polar orbit over the Mun, she was down to about 140T.

With the science lab attached to Enterprise, that meant I could run all sorts of science as I orbited the Mun, using the lab to boost and clean the experiments, transmit the data, repeat repeat repeat as I passed over each biome. In the process of this, I racked up about 200 or so science. Not bad at all! So far, all was going to plan!

That meant it was time to launch a whole series of new landers to the Mun. This time, I aimed for four more difficult-to-hit targets: the north polar crater, the north polar highlands, the northern basin, and the southwest crater. This new series of missions I confidently called the Orion series. Each one of them had to either land in the dark or at least in deep twilight. Thank Heaven for the landing radar on the landing cans or these landings would never had worked...

Orion 4's landing spot turned out to be particularly spectacular, right on the rim of an enormous crater. That tiny dot is the lander:

I was really pleased with my new lander design. I was aiming for something that can get off Duna with a full science package. Given that all four landers set down with about 70% fuel, I think I'm pretty close! On Duna, the landers will be able to rely on parachutes to bleed a little bit of speed so I'm currently feeling pretty good about my heavy lander design. Missions complete, it was time to send the landers up! Enterprise was in her 1 degree polar orbit and I launched each lander on that track one by one as Enterprise passed overhead. In each case, I lifted off, immediately made a gravity turn, and accelerated to 325m/s. That got me to 30km and from there it was only a matter of rendezvous and docking, something I've mastered...

I encountered one nasty glitch with one of the missions. On the fourth lander, I followed my standard procedure for a launch which proceeded without incident. Then I checked the orbital track expecting a 30km or so ballistic trajectory marker on the map. Three out of four landers did that no problem. The fourth refused to show to show any track at all, and claimed instead that the spacecraft was still LANDED:

While in this situation, the game wouldn't let me switch spacecraft, accelerate time, or do anything else. It claimed I was "moving over terrain." Shutting down the game and restarting put the lander back on the Mun but launching again repeated the problem. This went down three or four times until I gave up for a couple of hours and came back to the game later. At that point, I was able to get a normal launch out of the lander and it docked, though still with some really odd graphical glitches (note the G-force bar at the bottom and the time acceleration bar at the top).

Still, mission (close to) success! All four landers down, all four landers back up, all with juicy juicy SCIENCE on board! I also got to entertain myself a bit because apparently I set one of the landers down a little too hard, so one of my more brave kerbals had to go outside and repair three of the legs. In the process, he got a pretty spectacular view out his window:

Repairs proceeded without issue, all craft docked, I set the thrust limit on the three main engines down to 15% and it was time to head for home! Module lag was bright yellow, of course, but I expected that. What I didn't expect was bright yellow module lag plus thrust limiting plus a long injection burn meant about 20 minutes of tedium as I was only able to adjust delta-v by about one-half meter per second per second... and I had about 500 meters/second of delta-v to change. Then once Enterprise was ready to settle back into her parking orbit, I needed another 15 minutes of tedium to circularize the orbit. Ouch!

All I could do was sit there and watch the spacecraft lighten... at one point I was burning about 3.5T of fuel per minute. Those two burns plus waiting out maneuver points and slllooowwwwwly pitching Enterprise around to each new burn trajectory ate about an hour of some of the worst tedium I've ever experienced in a video game. Still, in due course, everyone made it back to Kerbin orbit safely and it was time to undock and go home:

Each lander performed flawlessly from this point onward. One of the tricks that I used to try to make Enterprise more stable and reduce the stress on those docking collars was to empty the fuel in the landers once they were docked. Once Enterprise was back in Kerbin orbit, I sent over just enough fuel to fill one of the four tanks, then spread it around evenly to the other three. That turned out to be the right strategy and I brought a minimum of fuel back down to the ground. Usually about 2/3 of this was enough for the deorbit burn and I used the rest to aim for specific landing sites I wanted to hit.

Final score, each mission brought home 333 Science, for a grand total of about 1530 Science total from the Enterprise mission to the Mun! Score!

But of course the downside is that module lag and stresses on the docking collars. Unfortunately, I don't think I can use Enterprise as she was intended for Duna. While the ship is capable of the mission, the game and its player really aren't. I calculate I'd have to sit and baby-sit the burn to Duna for more than an hour and that's just not something that I'm willing to do. So all in all, Enterprise completed all of her mission objectives... but unfortunately I have to rate her as a technological dead end. I might still send her to Duna or Jool but if I do it will be with light-weight probes docked to her ports, not landers.

Still, fun stuff and over all, I continue to be enjoying myself immensely with KSP! Does it show? ;-) And the landers themselves have proven themselves. I think I can use those for Duna and Jool just fine. Back to the drawing board for a way to get them there (he said with glee)!

EDIT (4/Feb/2014): Overnight, I thought of a way to have my cake and eat it too. Enterprise is going to Duna! More in next week's update.


  1. You should build interplanetary craft as a long "chain" of docked modules, with a drive section, preferably consisting of nukes, on the front and everything else on the back. A puller-design will automatically enhance the stability of the craft.

  2. I can't really tell if you know about this or not based on what you wrote, but you can reduce burn (real-)times with physics warp, which you activate by holding shift and pressing period (or whatever you have mapped for time warp). Depending on the severity of your part lag and the integrity of your vessel it can take you all the way up to 4x speed. On really laggy ships it can at least make game-time run at about the same speed as real-time.

  3. What about sending Enterprise ahead of the main mission to use as a refueling point for the return journey?

  4. Nuclear engines are your best friend for interplanetary transfers.

    You can press the ALT key + time acceleration keys to physical time warp (up to 4x) in space. Helps with long burns, but causes phantom stresses as the physics system attempts to keep up. You'll feel this the worst with lateral docking ports such as on the enterprise; the landers would bend towards your retro burn vector.

    I'd recommend an in-line system - like a space train or big rig - as that is the most stable (make sure you are PULLING and not pushing the cargo or you will jackknife it!) under acceleration.

  5. try: http://kerbalspaceport.com/quantum-strut/ for securing those docking ports.
    also possibly: http://forum.kerbalspaceprogram.com/threads/55657-0-23-Kerbal-Joint-Reinforcement-v1-7-1-18-14-v2-0x2-experimental-available!

    i know you're running stock. but there are lots of problems like this in stock

  6. Re your weird glitches, do a quick verify game files in steam. That's the kind of bug I'd expect to see if someone had a mod installed wrong or something similar, but you don't seem to use mods. Possibly some file corruption going on.

  7. Be sure to practice your aerocapture before you head out, and don't count on your chutes slowing you too much on your Duna lander. Just saying... ;-)

  8. I agree with Mabrick. Try to come in as shallow as you can you can, at as low an altitude as you can. If your chutes deploy with too much speed on them they tend to tear chunks from your landers as well so you may need to burn retrograde a little before they deploy.

    1. Yeah, I landed a probe on Duna a couple of weeks ago. I used a combination of aerobraking and thrust to slow it to 3000m and 100m/s, then used parachutes from there.


Note: Only a member of this blog may post a comment.