Micro-Engineering Challenge

[SeventhArchitect]

To any who decide to take this challenge: I may change things around for the challenge itself. There are some geniuses out there who are phenomenal at this art and skill, and I'm one of them, and in the very beginning (before I even started on this challenge myself), I gave set rules that must be met to succeed at this challenge. Those set rules, as I have come to find out, were easily met.

As an overall point to be made, if you beat the original rules, that is easy mode, at least for this challenge.

All is subject to change, however, if things are changed to be harder, then I will say it is for that reason alone, and will give a harder version of the original challenge.

My weight limit WAS below 50 tonnes, and that was beaten 50 times over by one vessel, literally.

Before we start, I'll give an example. Or two. Can range from a drone thats simply 6 pieces to a full blown barge that has 256. Like the one I built. Sadly my barge didn't make it out of LKO, and I used some of it's fuel to make that orbit stable, it wasn't gonna make it out of Kerbin's influence. However, it is a step int he direction I want to go, and this is what I wanna bring to the Micro-engineering challenge, so attempt 1 failed.

Spoiler

Album OX3NlkY will appear when post is submitted

Here's the challenge. Make the smallest vessel you can that can go to any planet in the system and is reusable 100%

Requirements:

cannot use any big parts (or if you do, use less than 5) Any part bigger than the OKTO drone core is too big, but I'll allow a small limit of 5, ya know, for a science JR or some such, maybe even a nice powerful relay/antenna

cannot be over 30 tons. Previous original version (50t) was beaten, literally 50x over with 1 measly drone. Props to you, ya brilliant prick.

Must have a stable orbit when changing planets

(Optional harder version) Must be able to make a voyage between at least 2-3 planets in one trip

Unrequired:

Does not need to be an SSTO

Does not need to be the size of a penny

Going to and from the surface of celestial bodies

Having a kerbal pilot the craft

___________________________

I wanna see what you people will come up with. I will be taking this challenge up myself, but I'm gonna be doing it in career mode.

Extra props if you do it in career mode before I do

Extra challenge, if you do it in career mode, the end product cannot be over 50k

Because I don't know how much a xenon ghost engine costs, that pruce may be subject to change

I plan to use about 6-12 ghost engines, soooooo....yeah, probably subject to change. They sure sound expensive.

___________________________

Micro-engineering challenge 2.0 (Med)

Weight limit of 20t

has to be able to go to and from a planet's surface, or two. That does not mean you have to do all of this without refueling

Stabilized an orbit around every existing celestial body

___________________________

Micro-engineering challenge 3.0 (Hard)

Weight limit 50t

Must go to the surface and back to orbit of every celestial body in the game (except that one moon rock thing that's too small to have a gravitational pull)

Hell, if you can land something on that damn rock, you get the overachiever award

get something into kerbol's local atmosphere and fly it back out

Good Luck. You wanted a challenge, WELL HERE IT IS, YOU MASOCHISTIC [REDACTED]!!!

___________________________

The original concept was build something out of nothing but small parts. Yeah, you can have a 1000 piece object if it doesn't exceed 30 t for easy, 20t for medium, and 50t for hard, and it can even be as big as a space station for all I care. You're really just supposed to use nothing but small parts, but whats the fun in not using something bigger every once and a while? Majority part limit is 0.625m.

Edited February 16, 2018 by SeventhArchitect
finished the challenges

[vyznev]

Does "go to any planet" mean just doing a fly-by through the SOI, or does it need to be able to reach a stable orbit, or even land? And does "planet" exclude moons? Because "get to Jool SOI and back" is a very different challenge than "get to Tylo's surface and back". And does "reusable" mean every part has to be able to land safely back on Kerbin, or is it enough to just leave the craft in orbit (around Kerbin or elsewhere) where it could, in theory, be refueled and reused? Oh, and I assume unmanned is fine, but you might want to be explicit about that.

Also, what's a "ghost engine"? Do you mean one of these?

[SeventhArchitect]

1 hour ago, vyznev said:

Does "go to any planet" mean just doing a fly-by through the SOI, or does it need to be able to reach a stable orbit, or even land? And does "planet" exclude moons? Because "get to Jool SOI and back" is a very different challenge than "get to Tylo's surface and back". And does "reusable" mean every part has to be able to land safely back on Kerbin, or is it enough to just leave the craft in orbit (around Kerbin or elsewhere) where it could, in theory, be refueled and reused? Oh, and I assume unmanned is fine, but you might want to be explicit about that.

Also, what's a "ghost engine"? Do you mean one of these?

Simply make stable orbit, you can choose manned or drone, does not need to go to planet and back

Depending on how easy it is, I'll probly change things up

[vyznev]

OK, so by "reusable 100%" you just mean that any parts I stage off must be able to return to Kerbin? Or just achieve stable orbit? Or do you just mean that the design must be reusable?

Also, what counts as "big parts"? I assume any Tiny, Small or Mk1 parts are OK to use? What about Mk2? The Mk2 cargo bay would be awful convenient to have, since the only payload container in any smaller size class is the 1.25m service bay, which isn't really even big enough for an ion engine and a decoupler. Basically, trying to squeeze a functional probe into one of those becomes an exercise in creative part clipping. Or I suppose I could use a payload fairing instead, if those count as reusable. And are all radial parts OK, even things like Gigantor solar panels or Thud engines?

Oh, and what's with the "stable orbit" requirement? I assume it means I can't use a direct suborbital-to-escape ascent trajectory, but I'm not sure why that would need to be disallowed. If you can escape the SOI, boosting your periapsis into space is trivial.

[SeventhArchitect]

21 minutes ago, vyznev said:

OK, so by "reusable 100%" you just mean that any parts I stage off must be able to return to Kerbin? Or just achieve stable orbit? Or do you just mean that the design must be reusable?

Also, what counts as "big parts"? I assume any Tiny, Small or Mk1 parts are OK to use? What about Mk2? The Mk2 cargo bay would be awful convenient to have, since the only payload container in any smaller size class is the 1.25m service bay, which isn't really even big enough for an ion engine and a decoupler. Basically, trying to squeeze a functional probe into one of those becomes an exercise in creative part clipping. Or I suppose I could use a payload fairing instead, if those count as reusable. And are all radial parts OK, even things like Gigantor solar panels or Thud engines?

Oh, and what's with the "stable orbit" requirement? I assume it means I can't use a direct suborbital-to-escape ascent trajectory, but I'm not sure why that would need to be disallowed. If you can escape the SOI, boosting your periapsis into space is trivial.

It's just a rule of thumb to not use big parts, because, ya know, it's a micro engineering challenge

The smallest ship you can build that can be refueled, get almost anywhere it wants in the system, and it has to be as small as you can make it. Does not include your carrier rocket that brings it to orbit

Gigantor solars are what I'm gonna be rolling with, because insane power costs for those ghost engines

The one thing I don't know, is how much xenon they burn a second, and wiki just puts a battery symbol per second for the fuel consumption

Some people call those xenon engines ghost engines, which i prefer cause it sounds cool

Edited February 6, 2018 by SeventhArchitect

[vyznev]

Well, I made this thing:

It's a reusable SSTSO (single stage to suborbital; probably could make it all the way to orbit with more careful flying) spaceplane carrying a tiny ion probe that should have enough delta-v to get from LKO to anywhere, and probably even visit several planets in one mission. I haven't actually tested that yet, though. It's only got 200 units of charge and two 1x6 solar panels, so maneuvers may require patience and, especially for capture burns, careful planning. Still, it should be enough, although I guess sticking on a couple of extra radial batteries just for good measure wouldn't hurt the delta-v too much. Batteries are really quite surprisingly lightweight in KSP.

(Just to be clear, I'm not claiming to have completed the challenge yet, since I've only flown this thing up to LKO. Just wanted to post a work-in-progress update.)

Update: OK, so I decided to go back to the SPH and add some more xenon and batteries to my probe. It now weighs 793 kg (out of a total mass at launch of just a hair over 6 tonnes) and has over 8.5 km/s of delta-v. I still haven't flown it beyond LKO, but now I'm sure that it's got enough delta-v to go anywhere. Here's the craft file, if you want to try it out.

Spoiler

 

The launch vehicle is pretty easy to fly: just accelerate to about 100 m/s on the runway, pitch the nose up 45 degrees and hold that attitude until the Rapier engine flames out around 25 km or so, at which point your apoapsis should be a little above 100 km.

Then switch to closed cycle mode for the apoapsis burn. There isn't enough fuel to actually reach orbit, at least not with this ascent profile, but you'll get most of the way there and the ion engine on the probe can easily finish the job.

If you do the ascent exactly like I did, there should be a few units of spare fuel left in the plane when the oxidizer runs out. Even if not, I recommend stopping the orbital burn before the fuel runs out, since having a bit of fuel makes a controlled landing much easier.

With the airbrakes deployed, the plane is nice and stable on re-entry. You can retract them once the heat effects go away if you need to glide to a better landing site, although the roll control is kind of on a hair trigger until you get down to 15 km or so. The plane also likes to pitch down if you let it, but that's pretty easily controllable.

I recommend throttling up the engine just a notch or two before landing so you can maintain a nice shallow glide without slowing down too much and stalling. (Don't forget to switch back to air breathing mode.) As soon as you touch down, deploy the brakes and cut the engine. If you come down a bit too hard and bounce, don't panic; just stay calm and correct your roll and pitch before the next touchdown. 

And there you go. One probe safely deployed to LKO and ready to go anywhere you want.  

 

Edited February 7, 2018 by vyznev
update, craft file, mission report up to LKO, more pictures, typo

[sevenperforce]

Just so I'm clear...

Build a probe capable of going from LKO to a closed orbit around any other world, and back, without staging.

Can you specify the bonus (like 20% off your starting mass) for doing 2-3 other planets? And does it actually have to be another planet, or are moons okay? Any probe which could make LKO-Eeloo and back in one piece could EASILY make Minmus and Mun and back in one piece.

You could also specify bonuses for making it manned, or eschewing ions, or eschewing ions and nukes.

Also, is mass based on dry mass or wet mass?

[Martian Emigrant]

Hey there.

 

This is a blast from my past.

https://imgur.com/a/qkqjA

 

The album is a bit messed up but mostly in order.

24 tonne, 41 part.

Goes to orbit on it's own. Must do it's first refuel on Minmus.

One Kerbal.

16 worlds.
Sample pictures.

Spoiler

 

 

 

ME

[vyznev]

On 2/6/2018 at 12:38 PM, vyznev said:

OK, so I decided to go back to the SPH and add some more xenon and batteries to my probe. It now weighs 793 kg (out of a total mass at launch of just a hair over 6 tonnes) and has over 8.5 km/s of delta-v. I still haven't flown it beyond LKO, but now I'm sure that it's got enough delta-v to go anywhere. Here's the craft file, if you want to try it out.

Well, I decided to see how far the probe could go in one trip. So I went to Jool (with a Laythe fly-by). And then to Eeloo. And back to Jool again, just for a gravity assist to Dres. And then to Duna (with several Ike fly-bys), and finally to Eve with a Kerbin fly-by along the way.

Didn't quite have the delta-v to get to Moho and complete the grand tour in one go, although with a bit more planning and optimizing the transfers it might have been possible. Or I guess I could've just brought another tank of xenon. Still, not bad for a 6,010 kg launch mass.

Lotsa pix to prove it happened:

Spoiler

I was planning on going to Eeloo first, just to check that I really had enough batteries to make the electric propulsion practical out there. But while planning the transfer, I got a surprise Jool encounter, so I decided to go there first and hope that I could get a nice and cheap gravity assist from the Joolian moons off to Eeloo.

Getting from LKO to the Jool transfer orbit with the ion engine took a bit of work. There was no way to do it in a single burn, both because the burn would've lasted nearly half an orbit at full thrust and because I didn't have enough electric charge for such a long burn anyway. I ended up doing three periapsis burns on the trailing side of Kerbin (i.e. about 90 degrees before the marked node) to get my apoapsis up to the edge of Kerbin's SOI, and then one final long burn (with the engine throttled down to keep the batteries from draining) on the way up to gain the remaining velocity needed for Jool and to fine tune the ejection heading.

Tip of the day: you can physics warp in space by using the Alt key. Doing those long ion burns at 4x speed really makes them a lot more bearable.

A minor mid-flight correction burn gave me a free capture to Jool orbit with the help of a convenient Laythe encounter.

Then I spent way too long looking for the perfect multi-pass gravity assist for a Jool to Eeloo transfer before giving up and just deciding to brute force it.

Not the most elegant transfer burn ever, but whatever. It works.

Sorry Tylo, not going to visit you this time.

Burning downwards at Jool apoapsis to get the right inclination for an Eeloo transfer. If I'd planned my Jool transfer better, I probably could've saved most of this delta-v.

Fine tuning the Eeloo encounter. It takes such a long time to get from anywhere to anywhere out here.

Oh, hello, Eeloo. Nice to finally meet you.

And we have an orbit. A very elliptic orbit, to be sure, but an orbit nonetheless.

After boosting out of Eeloo's SOI again, it was time to head back inwards. I'd already been to Jool, so the next natural destination was Dres, but I decided to fly by Jool again for a gravity assist to save delta-v.

This segment ended up taking more than half of the total mission time. It also took a ridiculous amount of real time, even at max time warp. Fortunately I have KAC installed, so I could just grab a cut of coffee and do other stuff while waiting.

Approaching Jool again from a slightly unusual angle. Can you spot the big green planet in the screenshot above?

Oh, there it is. Anyway, I've been here already. Not stopping this time.

Unfortunately my carefully planned Jool to Dres transfer got messed up by some slight error in the earlier maneuvers, so I had to delete all the planned nodes and redo them from scratch.  

Hi Dres, long time no see!

We have achieved an orbit around Dres. Next stop, Duna.

No fancy gravity assists this time, just a simple Hohmann transfer burn.

Did I say no fancy gravity assists? I meant before getting to Duna.

I probably could've aerobraked to save some delta-v here, but decided to just play it safe and stay out of Duna's atmosphere.

Do you like Ike encounters? Because we've got Ike encounters.

This counts as a Duna orbit, right?

Tweaking the orbit just a little bit gives this nice S-shaped escape trajectory.

A slight retrograde burn at Ike periapsis is needed to avoid straight into Duna...

...followed by an inward radial burn around Duna to amplify the gravity assist.

Next up, a gravity assist off Kerbin and then onward to Eve.

This place looks familiar.

Another radial periapsis burn at Kerbin, this time with a slight unintentional dip into the upper atmosphere. Got a heat warning bar for a moment, but fortunately nothing actually blew up or fell off.

Mission control is excited about the successful Kerbin gravity assist maneuver. Some spoilsport does point out that we could've been at Eve more than 50 years earlier if we'd just launched a second probe there instead of sending this one off to Eeloo first, but this argument is generally dismissed as silly.

Not going to try aerobraking on Eve, that's for sure.

So purple.

Eve orbit achieved.

Unfortunately I found myself some 1 km/s or so short of the delta-v to reach Moho and complete the grand tour. However, I did have enough for one last stop.

It's been I while since I last went to Gilly. I'd forgotten quite how much landing here resembles docking in space, as opposed to actually landing on something with actual gravity.

Are we landed yet? It feels like we're not moving...

OK, are we now landed?

Yep, landed on Gilly. Sure took a while to get here. 

Ps. I figured I should fly a Moho mission, just to demonstrate that the probe can do it. So here goes:

Spoiler

I didn't bother launching a second probe, but just reloaded an old save with the first probe in LKO.

I used a gravity assist off Eve to save a bit of delta-v.

That's a pretty long burn. Fortunately, this close to the Sun, the solar panels supply more power than the ion engine can use, so I don't have to worry about draining the batteries.

Moho orbit achieved. Now what should I do with the remaining delta-v?

How about a game of Mohole golf?

Nope, doesn't look like it's going to be a soft landing.

 

Edited February 10, 2018 by vyznev
added moho mission

[SeventhArchitect]

I will, instead of quoting, just answer questions because I'm not super adept at the HTML/coding of posts

 

Martian, I will, unfortunately, have to say that a vessel that large does not count. Granted, it meets the requirements, and to be fair, thats a rather legit craft ya got there, but it doesn't really fit int he micro-engineering challenge.

To clarify, I'm hosting this challenge for people making objects the size of small probes. You gave me a standard SSTO space fairing vessel that can pretty much go anywhere. Which isn't that bad either.

 

Seven, Dry and Wet both, if you don't mind. Dry mass seems more applicable, but a probe could be twice as heavy with full tanks. As a concept, just label it, because 1/2 your weight being fuel definitely throws up tons different performance numbers than having 1/4 your mass in fuel, or 1/5. Better to label both. Plus, more info.

 

Vyznev, I have to applaud you for hitting it on the bullseye, but I also have to curse you for ruining everyone elses' day, because I'm gonna have to change the base rules soon.

Which will be done sometime soon after a bit of testing and flight. Or rather several hours from now (considering KSP's time consumption)

[SeventhArchitect]

Eh, couldn't get the imgur thing to work right, don't know how, an tried 5 different ways.

 

Update bumpdate, etcetera

[vyznev]

Enbedding imgur albums has been broken for ages, AFAIK. It says it's going to appear "when the post is submitted", but it never does. Embedding individual images from imgur works fine, though.

Also, I know I'm nitpicking, but "every celestial body in the game" technically includes Jool and Kerbol, but they don't have a surface to land on. You might want to exclude those. 

More seriously, you might want to consider defining the goalposts of this challenge a bit more concretely, so that it's easier to tell what counts as a valid solution and what doesn't. I mean, making the MicroShuttle was fun, and some of your new challenge goals look like they could be fun to try too, but trying to chase a moving target gets frustrating after a while. If you're not sure what level of challenge would be best, you can always create multiple tiers and maybe offer bonus points and/or honorary mentions for folks who go above and beyond what you've asked for. Or if that seems like taking it too seriously, just set a few basic requirements and say that anything that meets those is fine.

Also, your first post above looks kind of hard to read now, since it seems to have the old and the new challenge rules mixed together.  While I can't speak for everybody, my impression has been that most people who are looking to try a challenge really want to see two things written clearly and easily visible: 1) what's required for a minimum valid entry, and 2) how are entries ranked, i.e. what (if anything) makes one better than another? If those things aren't clearly visible, either people will just skip the challenge or they'll post something that doesn't actually meet the requirements.

Edited February 16, 2018 by vyznev
nitpick + addendum

[SeventhArchitect]

Hit a dry spot for KSP for a while, probly gonna get back into it soon, but might have to start from the ground up, because I lost my previous save where I did start from the ground up, and decided to just give myself all the science and resources to get to the same elvel, wasn't the saem after I did that. It was like, "Well, fak, what do?"

Alright, so, your words are being heeded, and I'm already thinking of ways to tweak the challenge. I've never posted a challenge before, to my knowledge/that I can remember, so, yeah, not the best at making them.

Of course, I did think about Jool and that other rock you couldn't land on (which's name I can't remember), and I actually did state that you don't have to land on that tiny rock, and I thought I did, but I didn't say anythign about Jool. I had intended to.

 

And last, this might be the beginning of a hiatus. I lost all drive for everything in this world yesterday, and it's now a weekend, and I have nothing to do, so it's not helping matters. Here's hopes that I come out of this stronger/better.

 

Deuces, Architect out, I'll hit this thread with an update once I get my next career up to speed. In fact. I think I'll try that right now.