What's the dumbest thing you did that actually worked?

[sturmhauke]

I finally got around to updating to 1.12 the other day. I keep several installs around for mod reasons; I had an old career mode save from when I started playing KSP way back in 1.3, but sadly it wouldn't work in 1.12 without a lot of iffy save file editing, due to too many parts being deprecated and removed. So I wrapped up a couple missions in a 1.8 install as a sort of farewell, and then started a new career save in 1.12 with some new mods (the biggest being Strategia, OPM, and USI Life Support).

So anyway, I'm still early in a new career, with new stuff to worry about, and I tried to do a Mun mission. Nothing too fancy, just plant a flag and return. But it's been a while since I started a brand new career, and I still only have short little noodle tanks and underpowered engines. I had to make do with a tall, spindly lander. Oh, and did I mention I have only very weak reaction wheels and no RCS thrusters? I think you know where this is going.

I built a somewhat questionable rocket, which tumbled when trying to do a gravity turn. Revert to VAB, rebuild, relaunch, repeat. After a few cycles of that, I had a rocket with somewhat longer strap-on boosters and comically large fins, but that was enough to get the job done. After a few more cycles, I got the launch, ascent, and Munar transfer correct. And then it came time for the landing. Picked a spot, burnt the last of the transfer stage to start the descent, jettisoned it, came in nice and gentle... and then tipped over because I misjudged the ground slope and the lander couldn't correct for it.

I had Jeb and Bob get out, plant a flag, do some science, and then hop back in. I've launched a fallen lander before, no big deal, I thought. Wrong. Those other times, I had stronger control systems, tougher parts, and/or miscellaneous tricks like using a service bay door to help right the lander. I didn't have any of that here. I tried a few times to just scrape along while trying to pitch up, but I couldn't get going fast enough to clear the ground and rotate without blowing up parts.

Then I realized I was near the rim of a large crater. If I could get to the edge and launch over the cliff, I might be able to regain enough vertical speed before crashing. So I crawled the lander up the slope under low thrust, so as not to break stuff again. After several long minutes and some backsliding. I managed to get into position. Here goes nothing. I fired up the engine, hit full throttle and rotated as soon as I cleared the cliff, dropped more than I really liked... and then started rising again. Success! I cleared the far side of the crater easily and got back into Munar orbit.

So what are your tales of the absurd?

[swjr-swis]

5 hours ago, sturmhauke said:

So what are your tales of the absurd?

I once babysat a 1.10 instance of the game for 7.5 hrs, because the long-range aircraft I had created for a circumnavigation challenge kept rolling to one side and needed to be manually kept on track by tapping 'e' every 2-3 seconds. Pure stock game, no mods to autopilot for me, physics warp made it even worse and required constant tapping that I could not keep up with, and after a lot of tweaking and searching I couldn't find the source of the phantom roll. Worked perfectly stable in 1.3.1, but the challenge required 1.10, so if I wanted an official entry to the challenge, this was the only way. 7.5 hrs... tap tap tap.

It worked - held the top entry for most laps around Kerbin for a while. But not ever doing that again.

[Gargamel]

9 hours ago, swjr-swis said:

I once babysat a 1.10 instance of the game for 7.5 hrs, because the long-range aircraft I had created for a circumnavigation challenge kept rolling to one side and needed to be manually kept on track by tapping 'e' every 2-3 seconds. Pure stock game, no mods to autopilot for me, physics warp made it even worse and required constant tapping that I could not keep up with, and after a lot of tweaking and searching I couldn't find the source of the phantom roll. Worked perfectly stable in 1.3.1, but the challenge required 1.10, so if I wanted an official entry to the challenge, this was the only way. 7.5 hrs... tap tap tap.

It worked - held the top entry for most laps around Kerbin for a while. But not ever doing that again.

Ya know..... they make macro editors that will automate that tapping for ya.     You'd still need to sit there to monitor, but you'd at least be able to do other things....

[Lisias]

10 hours ago, swjr-swis said:

I once babysat a 1.10 instance of the game for 7.5 hrs, because the long-range aircraft I had created for a circumnavigation challenge kept rolling to one side and needed to be manually kept on track by tapping 'e' every 2-3 seconds.

I learnt one from a Scot Manley video recently that may help you.

Take a docking port, place it on the craft's ceiling, select Control from it and then SAS to Radial Out.

You will have to add some 'brute force trimms on your craft (disabling all axis, deploying them and then manually setting the Deployment Authority), but at least in theory it works.

(Tried by first time on a extremely assymetrical craft on 1.12 without sucess though. Didn't had the time to try it again yet)

[Just Jim]

Watch Chapter 5... to this day I'm amazed it actually worked... lol
 

 

[sturmhauke]

Nice? How much did that thing weigh?

[Just Jim]

8 minutes ago, sturmhauke said:

Nice? How much did that thing weigh?

I don't remember off hand... but now I'm curious, because I made it before we added in all the dV information. I need to find the craft file and load it into 1.12.1 and see how much dV and TWR this beast actually had

[sturmhauke]

I did a large space station in two launches - one for the main body, and another for some additional modules. It was quite a beast, but it only went to LKO.

Spoiler

[Just Jim]

13 minutes ago, sturmhauke said:

I did a large space station in two launches - one for the main body, and another for some additional modules. It was quite a beast, but it only went to LKO.

  Hide contents

Very nice!!!

[swjr-swis]

3 hours ago, Gargamel said:

Ya know..... they make macro editors that will automate that tapping for ya.     You'd still need to sit there to monitor, but you'd at least be able to do other things....

I was doing other things, even if with half an eye. I may be dumb, but not crazy. I'm not sure a macro would've done the job though; it wasn't a consistent rol, otherwise even a bit of trim could've solved it.

 

3 hours ago, Lisias said:

Take a docking port, place it on the craft's ceiling, select Control from it and then SAS to Radial Out.

I do that regularly, mostly to fly VTOL type craft. I suspect if I'd tried that in this case, the plane wouldn't have kept its heading, it would've just translated the self-corrected roll into yaw.

 

Still, either one of those suggestions might've at least made it less of an ordeal, but by that time I was already well underway and I wanted to have it done and over with, not start all over again. I can be a bit stubborn at times, if you hadn't noticed.

 

 

[DiomedesDK]

I once made a xenon powered minmus lander for three kerbals and a lot of science. It had such a low TWR that it actually was more controlled crashes than actually landings... I could only go back go orbit when had deployed all my science

[jimmymcgoochie]

When my first Earth orbit rocket design didn’t quite have the thrust off the pad I needed, I did what you’re never ever ever supposed to do in RO/RP-1, and added MOAR BOOSTERS! Six copies of the upper stages to be precise, which looked ridiculous, probably cost a lot more than necessary and was rather prone to engine failures using so many moderately unreliable little engines, but it worked.

 

And then a bit later on I did something pretty similar, taking a rocket with two boosters and just adding four more boosters to increase its total payload capacity.

And as if things couldn’t get any worse…

Putting boosters on the boosters for EVEN MOAR BOOSTERS! 

You may notice that the last rocket is slightly bigger than the others. It weighs about 12,000 tons and can put ~650t into LEO in a single launch. In comparison the rocket above weighs 350t on the pad; this thing could almost put two of those into orbit in one go!

Bonus extra- Mun lander fell over due to an engine failure just above the surface and a slight slope on the ground. I scraped it around to point uphill and then did this:

Landing leg pogo bounced it up just enough to gain enough speed and altitude to just clear the top of the ridge, then it was full speed ahead and set course for Kerbin. Minus the legs though, they got dumped before circularising around the Mun so crashed into the surface.

Edited September 21, 2021 by jimmymcgoochie
Added another picture which didn't work well on tablet

[sturmhauke]

Remember kerblings, if you think you have enough boosters you should put a couple more on just for good measure.

[Superfluous J]

Easiest question to answer in the world for me.

This.

Edited September 21, 2021 by Superfluous J

[king of nowhere]

well, i believe for me it was starting to carry out large complex missions with ludicrously large ships.

(notice: the ship may look smaller than it is because i enlarged the engines and convert-o-trons in an attempt to save part count. the mining drills, large reaction wheels or ra-100 antennas are good references. Do also notice the time stamp)

I started this trend last december; since then, i performed a grand total of 2 and a half missions. yeah, they take that long.

I don't know the most time-consuming part; the lag caused by the huge motherships, having to stop time warp regularly to harvest food from the greenouses, or having to stop at longer intervals to manually check up every single component of the ship to try and avoid malfunctions.

I've had lots of fun designing ships that could carry out such missions, getting to use them for extended times and really appreciate them, solving whatever weird problem the game threw at me. but still a huge commitment.

I'd count it as the dumbest thing I do because a lot of people called me crazy over this

 

1 hour ago, jimmymcgoochie said:

When my first Earth orbit rocket design didn’t quite have the thrust off the pad I needed, I did what you’re never ever ever supposed to do in RO/RP-1, and added MOAR BOOSTERS!

never understood that part. i mean, ok, i understand about structural problems and all, and you can't just brute force things. but isn't MOAR BOOSTERS what we do in real life too? we really use boosters to increase payloads.

 

 

[jimmymcgoochie]

8 hours ago, king of nowhere said:

never understood that part. i mean, ok, i understand about structural problems and all, and you can't just brute force things. but isn't MOAR BOOSTERS what we do in real life too? we really use boosters to increase payloads.

It’s not just about structural issues or payload capacity- 1950s and early 60s rocket engines weren’t particularly reliable so clustering and adding boosters just increases your chances of failure. This isn’t such a big deal when the boosters are fairly small (as in the first rocket above where the Aerobees were ~5% of the thrust of the main engine), but I’ve lost a few 350-ton rockets when engines failed and seen two or even three boosters fail in the same launch. 

Those same engines tended to have limited or non-existent control systems (either throttle or gimbal) so the ability to compensate for a failed booster engine is greatly reduced over stock KSP where basically every engine has basically unlimited throttle and a decent gimbal.

[sturmhauke]

Also, in KSP you can put boosters on boosters on boosters to a ludicrous degree and still make it work. As far as I'm aware, real rockets have a practical limit of 6 strap-on boosters.

Edited September 21, 2021 by sturmhauke

[Lisias]

I built a deep space probe, to be launched by a Mün Shuttle Mission hitchhiking it to take advantage of a sparing cargo space.

(shuttles going to Mün is dumb by itself but, hell, is way more fun!)

So I launched the damned shuttle using a reusable booster, then inserted the shuttle into a transmünar trajectory where the external fuel tank was also a probe, able to come back to Kerbin, do a controlled reentry , splash down on chutes and be recovered. It took some days of playing until I get everything right to do it properly without cheating.

Then I decoupled the probe after the transmünar injection burn, put it on a prograde attitude, hit the Ion Drive and gone for dinner.

Nearly an hour later, I came back and to my horror I realised that I had attached the Ion Engine backwards! So by accelerating into a prograde orientation, I was essentially retrograding! But the burn looks normal, I was going towards a Kerbin escape trajectory… 

Then I realised that I also had attached the Remote Guidance Unity backwards too… 

So the blunder kinda worked nevertheless….

[Entropian]

I once created an SSTO that flew by the Mun to reach Minmus.  I decided to do a close flyby (you probably know what's coming) at about 5000 meters AGL.  Because I did the correction burn so far out, a small (accidental) RCS rotation burn ended up pushing my periapsis a lot lower than I intended, causing me to fly by about 5 meters or so above the ground.    But hey, I got my gravity assist!

[Corona688]

Caused the mother of all staging errors.  The smoke finally parted after a sixteen-second slideshow of atomic flame, revealing the uppermost stage just ... hovering there on 1.0 TWR.

Landed it.  I did not revert that flight.

Edited September 22, 2021 by Corona688

[sturmhauke]

19 hours ago, Lisias said:

I built a deep space probe, to be launched by a Mün Shuttle Mission hitchhiking it to take advantage of a sparing cargo space.

(shuttles going to Mün is dumb by itself but, hell, is way more fun!)

You should join us on the Shuttle Challenge thread. We go to way dumber places than the Mun.

[Staticalliam7]

I wedged a rocket in between 2 shuttle looking tanks/srbs because i couldn't get enough dv to the mun..

 

 

it worked

[qromodynmc]

i dont want to go all over it again so here's what happened. it was pretty crazy stuff.

[Randox]

I once landed on Kerbin with a Kerbal riding shotgun on the exterior ladder. 

 

It was a result of my first manned orbital rendezvous. I had a small purpose built rocket that went up first, equipped with a single occupant command pod and an RCS system (since this was to be the craft to make the final rendezvous), and the second rocket was carrying my Minmus lander, using a 3 man pod that was fully crewed. After circularizing the larger rocket a couple km away from the smaller craft, I switched control and immediately advanced the staging, jettisoning the command pod from the rest of the craft and stranding myself in orbit. Hearing epic rescue music in my head (and perhaps doubting my ability to make a rendezvous again), I was able to maneouver the larger rocket into a closer parking orbit and then EVA the stranded Kerbal over to it so he could hand on to the ladder.

This was frankly a pretty stupid idea not just because the 4th Kerbal would have to ride on the outside, but because that lander wasn't designed for a powered landing at all. It was supposed to jettison the command pod to land, but most of the ladder was on the fuel tank so I couldn't detach it. I did a slow burn of the engine on the way in to keep my speeds down, and reconfigured the parachutes to deploy as high as possible to try and avoid ripping the Kerbal riding on the ladder off the craft when they opened (it worked!).

I opted for a water landing to try and avoid blowing up on impact, then decided that was still probably going to end poorly and so took control of the 4th kerbal right before landing and had him jump. It worked. The running rocket engine didn't cook him, though the entire rocket very nearly landed on him. The rocket did break up on impact a bit but nothing (important) exploded, and everyone survived. Yay team!

[splashboom]

I took a rover to the Mun for an observation mission and the way-points were on a huge and steep slope. I had to remember tricks from down hill skiing navigating the slope from side to side. The last report I did while sliding upside-down after I tumbled when entering the area.