There are times I wish my laptop could handle both KSP and a video capture program. · 7:11am Jan 26th, 2017

Today I flew what has to be my most kerbal flight in Kerbal Space Program. (And, by the way, it had nothing to do with the CSP story- this is a separate game session.)

So, in this game session I managed to get three different missions at once that all required a Mun landing. I also had four tourists who needed an orbital flight above Kerbin, but not a ride all the way out to the Mun.

"Okay," I thought, "I'll unlock the three-man capsule and go from there."

Problem: when I went to build the rocket, I realized I had unlocked the capsule but not the heat shield sized for it. Thus, I could -launch- the three-man capsule, and -maybe- it might survive return from orbit, but no way would it survive a return from Mun.

So, I thought, I need to send one kerbal to Mun, but where do I put some tourists so I can get part of that contract cleared in the same flight?

And then I had the Bad Idea. (KSP inspires a lot of them.)

See this picture.

The center capsule and stack, plus the engines and the bottom tanks on either side, land on the Mun and return. Everything on the sides above about the midpoint of the center stack decouples, releasing two, count 'em, TWO three-kerbal spacecraft (nearly identical to the landing vehicles you might see in Missions 13/14 of CSP), each with independent engines and fuel for orbital maneuvering and deorbit burn. All told, this thing launches with seven kerbals inside.

Of course, this big ugly thing would never get a green light in real life, but in real life a mission of this sort would get its own bespoke rocket stack and lift system, not a passel of off-the-peg parts to be thrown together like Lego bricks. This is a lot of the fun of KSP.

But the thing is, this, as it sits, wouldn't get off the pad. It needs a rocket underneath it to get to orbit so those tourist pods can decouple and the rest of the ship fly on. In fact, the rest of the ship needs an extra engine and fuel for the Mun Transfer Orbit burn. And since the payload to orbit is already huge, the rocket underneath needs to be big, too.

Did I mention I hadn't unlocked all the parts I really needed to do this yet? 'Cause I hadn't.

This is what the final assembly looked like on the pad:

This is how underpowered for its weight this ship is; see those six auxiliary engines in a ring around the bottom of the second stage? Those had to fire at the SAME TIME AS both the first stage and the expendable SRBs for the thing to even get off the ground. In the first test flight I failed to do this, lost a couple of seconds and nearly dropped the whole mess on the pad before the extra engines lit, and ended up failing to achieve orbit.

So, revert to launch and say, "That was the simulation- THIS is the real flight!"

You're not getting any pictures from here on because the two above were made after the fact, for this blog entry. I didn't actually take screen caps during the mission. You're about to find out why.

So, the second flight got to orbit, and I decoupled the two passenger pods. This brought the second design flaw into play (the first one being the lack of the properly sized engine for the liftoff stages).

Once I decoupled the tourist ships, I had three spaceships, right?

Problem: I had only two pilots in the program. Rather than hire a third pilot (who would be green anyway), I decided I could perform a manual landing with Scientist Bob Kerman in the mun lander, while Jebediah and Valentina, the experienced pilots, brought the civilians down safely.

Thing is, if your ship doesn't have either a pilot or a probe core, you don't have SAS- even if there's a non-pilot kerbal in the command seat. There's no automatic system to keep you pointed straight. I would be flying the Mun landing entirely manually from this point on.

But that comes later, because at this point I discovered the third design flaw- one I had no clue existed, and which surprised the hell out of me.

My plan was to move the tourist ships a bit apart, get the Mun ship on its way to its destination, and then de-orbit the tourist pods one at a time. So I shifted control from the Mun ship to one of the tourist pods.

At this point I could no longer revert the flight to launch.

OK, Valentina in control, staging looks right, all set for a test burn of the engine, right? I hit the staging button to activate the engine.

And the decoupler fires, jettisoning the engine and its full fuel tank, leaving Valentina and her passengers adrift in space.

I turned off the game and went away for a bit, considering deleting that game session altogether, since I'd either just killed two or three kerbals- because tourists can't make space walks to transfer from one ship to another, and I didn't have docking ports yet in this gameplay.

As I did other things, my backbrain provided answers. Look at that first pic up above again. I built the center stack first, then attached two fuel tanks to either side, then engines underneath those; that was my Mun lander. I then put decouplers on top of the tanks, engines on top of the decouplers, fuel tanks on top of the engines, then heat shield, then passenger pod, and finally capsule.

This is important because, when I decoupled those ships, the game treated the oldest piece of each ship as "root"- the place controlling everything else. And that oldest piece remaining was the engine. So, when I hit the staging button, the game thought, "Wait, I'm already the engine, so there's no need to activate me. On to the next task in the list- which is decoupling. Bye-bye useless capsule and passenger compartment!"

And then my brain said: So you've got one ship with no engine. But it still has working reaction wheels and SAS, so it'll hold a heading. That means you can use the other perfectly good ship less than ten meters away from it to push it down.

Eureka! The mission continues!

Back to the game, switch to the second passenger ship, click on the engine and activate it outside the staging sequence...

... and realize Design Flaw Number Four.

I had put solar panels on the Mun ship. Thinking the passenger ships would have a safe, dull trip, I hadn't put any solar panels on them. That meant they had a limited battery life- and when the power ran out, not only would they not hold attitude, but the parachutes wouldn't fire, resulting in a slight kerbal-colored spatter among the inevitable explosion on impact.

I couldn't use the Mun lander because it was too big and heavy for me to turn and maneuver quickly enough to bring into the right position to push the crippled capsule home. My only options were to use Jebediah's ship or else launch another one and make a rendezvous- and Jeb's ship was, after all, right there.

Long story short, I wasted a LOT of electricity- moving myself into position, then letting both ships go around in orbit to the point of maximum burn efficiency, then finding they'd drifted sixty meters apart in the course of the orbit, re-positioning the ships, pushing, realizing I was pushing in the wrong place and RAISING instead of lowering periapsis, re-positioning again, and finally getting in the right place and executing the burn.

The plan: get Valentina's ship into just enough atmosphere to slowly decay her orbit without instantly bringing her down, then bring Jebediah down, then steer Valentina down, while letting the Mun lander orbit safely above.

By the time Valentina's trajectory was skirting into upper atmosphere, Jebediah's battery was down to under 20% charge. I was convinced only a miracle could save him.

But hey! This is Jebediah BadS Kerman! If he has ten seconds to save a craft, his only thought is, "What will I do with the other five seconds?"

And the fact that I was playing this in 1.2.2 (instead of the 1.1 build I'm using for CSP) made it possible, because the newest version of KSP vastly improved the performance of SAS. Before, an SAS on prograde or retrograde would constantly overcorrect trying to find its vector, burning electricity like Pinkie Pie going through confetti. In 1.2.2, SAS is as efficient on retrograde as it is on stability mode. I could have brought the ship down on -four- percent battery charge.

So, with Jebediah and two happy, clueless tourists home safe, I went back to Valentina. It turned out the shallow entry and skip-off was good for only one extra orbit; I got back to her ship just in time to bring her down through her final descent, which turned out quite a bit steeper than Jebediah's (one hazard of aerobraking) but still safe.

So. Tourists down. Back to non-pilot Bob and his first time behind the stick ever, and he's not only going to fly to the Mun, he's going to land.

Without going into detail, I learned just how difficult it is to hold attitude on a KSP rocket without SAS. (Which should not be at all surprising. Real-life spacecraft give constant computer assistance to pilots even on what is called "manual override.") And as fuel tanks got jettisoned and the ship got lighter, it also got twitchier and more unstable.

"Unstable" is not an adjective you want to add to "moon lander."

Simply put, at low speeds I proved unable to keep the vector in one place. I had too much lateral drift and a poor sense of how close the surface was, and... well...

... it wasn't a crash so much as a bounce followed by a belly flop. The ship stayed intact and functional, but on its side. I discovered that the cargo bay doors were strong enough that, when they opened, they actually flung the ship into the lack-of-air for a backflip. I tried a couple of times to use this to get the ship on its landing legs, but failed, and on the third attempt one of the legs broke.

But- the cargo doors weren't broken, and I could use them to get airborne long enough to light the engines and get some altitude- I hoped.

So, walk on Mun, plant flag, collect science, get my scientist back in his ship, and fling open the cargo doors one more time.

I've seen many bottle rockets fly straighter than that moon lander's ascent.

And it got worse. I mentioned one of the landing legs broke, right? Well, here's the thing; if I'd had one central engine, likely there would have been no trouble, or it wouldn't have been so bad. But I had two, and one side of my spacecraft was now slightly less massive than the other, the difference being one landing leg.

I couldn't go above one-quarter throttle without the ship becoming completely uncontrollable. Balance is that damn important.

Somehow I got the lander back into Mun orbit and then on escape trajectory back to Kerbin, and then a correction burn to get the ship's periapsis into atmosphere. My plan was to burn most, but not all, of my remaining fuel on my first pass through Kerbin's atmosphere to slow down the ship, then use the last dregs of fuel to get on a proper, comparatively slow, safe re-entry vector.

I came pretty close to burning up the ship in that aerobraking-plus burn, because the ship would not hold vector. No SAS, engine burning as hard as I dared, an imbalanced ship, and just enough air resistance to make the ship want to audition for Riverdance.

But- Bob survived and came out into space with an orbit much, much lower than I'd had any reason to hope for. One last burn, and then I jettisoned everything except capsule, heat shield and parachute. At this point, I thought, my worries are over- the capsule is aerodynamically stable for re-entry, so I can take my hands off the controls until it's time to pop the parachute.

And so it proved- until I saw I was going to come down on mountains.

Well, as Cherry Berry might say, buck.

I held off on the parachutes until I was coming down pretty much vertically, hoping I'd drift over a patch of flat highlands. Fortunately, though what I landed on wasn't flat, it was flat enough. The capsule flopped over, then stopped on its side, and Bob Kerman, the seventh and final victim of my idiotic flight plan, was back on his home world, safe and sound.

I'm only sorry I don't have the video... but maybe it's for the best. Scott Manley would ask what the hell I was thinking, Danny would call me a piker, and RoninPawn would say, "Hold my beer."

Report Kris Overstreet · 445 views · #Kerbal Space Program #Changeling Space Program #not story related

Comments ( 8 )

Viewing 1 - 50 of 8
- Newest First
- Oldest First

Razzy

Razzy #1 · Jan 26th, 2017 · 2 · ·

... And here I thought some of CSP's attempts were rinky dink. This sounds like something Dragonfly would design.

FanOfMostEverything

FanOfMostEverything #2 · Jan 26th, 2017 · · ·

That is magnificent, all the more so because you actually made it work. Poor, poor Bob.

Kris Overstreet

Kris Overstreet #3 · Jan 26th, 2017 · 1 · ·

4397034 Yeah, there were a few moments in the ascent from Mun where Bob would have felt like he was on the inside of a clothes dryer with the drum on gimbals.

Circut Breaker