Kerbodyne Challenge Series: Spaceplane Tankers

[Wanderfound]

Okay, this is another revival of an old challenge.

It's open to both FAR and stock, but the thing that got me interested is the recent changes in FAR. The new upper atmosphere skin drag makes it a lot harder to get a big tanker up with a worthwhile amount of fuel, and I've been struggling to design something useful in that class.

So, that's the challenge: build a spaceplane tanker, take it to orbit, deliver fuel to whatever space station target you feel like, land it back at KSC. Two leaderboards for each of FAR and stock: one for total fuel delivered, one for fuel/√ flight cost after recovery. Fuel delivery must be a balanced LF/O mix.

To keep it interesting, and put the focus on efficiency and good design rather than CPU capacity, there's a maximum part count limit of 150. Must be crewed, must be a HOTOL spaceplane, stock parts only. If you want to use a mod that you're unsure about, ask.

I'll get an entry up as soon as I build something worth flying.

Leaderboard

Stock

juzeris, 2,416 units delivered, √0.5533/unit

Starhawk, 2,100 units delivered, √0.63/unit

GoSlash27, 2,100 units delivered, √0.63/unit

juzeris, 14,400 units delivered, √0.816/unit.

FlipNascar, 5,548 units delivered, √0.88/unit.

FAR

UnusualAttitude, 12,485 units delivered, √0.915/unit.

Wanderfound, 8,288 units delivered, √1.02/unit.

Edited April 3, 2015 by Wanderfound

[UnusualAttitude]

Excellent challenge, count me in!

Good choice for the part-count limit.

Would Kerbal Joint Reinforcement be acceptable? If not, I will temporarily resort to strutting...

- - - Updated - - -

Also Wander, for a level playing field, shouldn't there be a reference orbit for the target station? 100x100km, 250x250km?

Edited March 28, 2015 by UnusualAttitude

[HoloYolo]

I can use the debug menu! I win.

Please state mods, whether or not we can use the debug menu, and if we can exploit glitches to oblivion.

[Wanderfound]

Obvious hacks and cheats barred by Wheaton's Law. As per the OP, if in doubt, ask.

If you can, keep it stock. It'd be nice to have a collection of ships at the end that people can use, and the more stock it is, the more people can use it.

The target station can be as low as 70x70.

[GoSlash27]

Darn it!

I built an entry for this only to realize belatedly that it needs to be crewed.

If anyone wants a practical stock drone SSTO spaceplane tanker, here ya go:

http://wikisend.com/download/570898/PeaceFrogII.craft

Starting over...

-Slashy

Edited March 28, 2015 by GoSlash27

[DancesWithSquirrels]

Durn it! I read this earlier today, before the challenge was edited to require HOTOL or a crew.

Oh well, here goes my VTOL SSTO (non)entry, anyway. (Thanks to Torquimedes for the hints on embedding an album)

Javascript is disabled. View full album

Spaceplanes are fun, but the stock aerodynamics are horrible (I'm a pilot), the wing attachment vs. center of lift issue drives me nuts, and the time required to pilot up/down is more than I care to deal with just for a launch system. So I tend towards big-a** VTOL SSTO rocket-only launchers like this one: 87 ton payload, 1140 tons at launch, 173 tons returning, vernors for RCS and three Mainsails for an OMS. Can land on parachutes only, although I try to reserve a bit of fuel for a softer touchdown.

87 tons to a 82km orbit with about 400 m/s to play with. Total of 880 tons of fuel used, so it's not efficient (10% payload for fuel expended) but it's fast and simple (only 136 parts).

Edited March 29, 2015 by DancesWithSquirrels

[juzeris]

Well, here at Unreasonable Contraptions the engineers decided it would be cool to satisfy the challenge requirements (flying Stock though), but admittedly did leave out everything else. Where corners could be cut, they were cut, where they couldn't - a hammer was applied liberally. Seeing the part count limitation, they unanimously agreed to use the biggest, baddest parts they could find lying around, and then see if the result can be thrown out to space.

The result is the Poofy Jaarthus, already affectionately nicknamed 'Pain' by the pilots. It does not do anything properly - it want to tilt sideways on the runway, flip over while in the air, lose altitude while in the stratosphere, run out of electricity while in space, doesn't want to do anything at all while docking, plummet straight down while landing and drive into buildings when finally landed. This and more are, of course, the reasonable result of it being light on wings, having no control surfaces, only one battery and reactor and no solar panels, no RCS, just two struts and, well, somebody forgot to install brakes on the landing gear for the virgin flight, but at least the last one is now fixed (probably).

On the other hand - on the virgin flight it delivered 14400 units of fuel (one full Kerbodyne S3-14400 tank equivalent) to LKO (80x80) for a measly 11747 funds, of which 6610 funds were for the fuel itself, so fuel in orbit now costs only 0.816 fund per unit, which is only 77% more than on the ground!

Operating manual - action groups:

1 - Rocket engine

2 - A few turbojet engines (4 probably)

3 - The remaining 16 turbojet engines

4 - The majority of air intakes, with them closed it should fly just fine to maybe 12km or so

Staging - this will immediately disable the rocket engine, so feel free to launch the craft by staging

Operating manual - flight instructions:

1 - While taking off, keep pitching up, otherwise the rear wheels might lose grip and the plane fall over sideways.

2 - While low in the souposphere autopilot will most likely be inadequate, but it behaves mostly fine under manual control, your mileage may vary.

3 - Try to hit at least 29 km altitude with a reasonable amount of vertical velocity and at least 1700m/s of orbital velocity, it should be capable of reaching that without (relatively) too much trouble before engaging the rockets. Then run the rocket and some jet engines in parallel.

4 - Docking is a bit harder with just one rocket engine, don't forget to keep changing between control points

5 - It has a big battery, but with engines off electricity production is minimal, conserve it

6 - Landing - it should survive a landing at a vertical velocity of -5 m/s, -8 is pushing it. Even empty it needs a high angle of attack to maintain altitude and at that AoA it loses velocity very quickly and plummets to the ground. You'll have to land with engines hot, probably running at least group 2 at half thrust. Quicksave before attempting a landing!

Javascript is disabled. View full album

No mods are used in the craft, built-in MechJeb was used for flying it, but mostly the simple parts of it. Admittedly we did hyperedit the benchmark station into a 80x80 orbit at about 50 km distance from the spaceplane, but now that we know the ascent time (from the screenshots!) an orbital ascent with a 50 km margin of error (that's ~21 second) does not sound completely unreasonable. Also we finished with a little bit of space fuel leftover, which could reasonably be used to achieve a more distant rendezvous. Also no Kerbals were sitting in the craft, but it does have a cockpit for two and it would not present any physics changes, so I hope that's fine.

While it does look a bit loaded with air intakes, it also has lots of jet engines, resulting in only 2.3 intake per jet engine, which would look completely normal on a more reasonably sized craft.

[Starhawk]

If anyone wants a practical stock drone SSTO spaceplane tanker, here ya go:

http://wikisend.com/download/570898/PeaceFrogII.craft

Nice craft, Slashy! I especially love the rear gear placement.

Since I saw your links in a recent post to the wing part and intake analysis charts for 0.90, all my new spaceplanes use mostly strakes for wings. And no more shock cones, either.

What do you get out of using the shorter Mk2 LFO tanks rather than the longer Mk2 ones?

Happy landings!

[GoSlash27]

Nice craft, Slashy! I especially love the rear gear placement.

Since I saw your links in a recent post to the wing part and intake analysis charts for 0.90, all my new spaceplanes use mostly strakes for wings. And no more shock cones, either.

What do you get out of using the shorter Mk2 LFO tanks rather than the longer Mk2 ones?

Happy landings!

Thanks, I'm glad you like it!

It turns out that Pork jet made an error with the configuration files and scaled the drag coefficients for half sized parts. 4 half sized tanks make the same drag as 1 full sized tank.

Best,

-Slashy

[Starhawk]

It turns out that Pork jet made an error with the configuration files and scaled the drag coefficients for half sized parts. 4 half sized tanks make the same drag as 1 full sized tank.

I knew there had to be a reason. Thanks!

Reverts flight to spaceplane hangar.

Happy landings!

p.s. Just so you're warned. Imitation is the sincerest form of flattery.

Edited March 28, 2015 by Starhawk

[GoSlash27]

I knew there had to be a reason. Thanks!

Reverts flight to spaceplane hangar.

Happy landings!

p.s. Just so you're warned. Imitation is the sincerest form of flattery.

Knowledge is meant to be used, so release the hounds

Best,

-Slashy

[Wanderfound]

Got an entry, but it's not the sort of thing I'd put up on the Kerbodyne thread. It flies fine, but it'll instantly explode if you touch time acceleration during the ascent, and requires a careful pilot during reentry. But it does deliver a lot of fuel...

Javascript is disabled. View full album

Fuel delivered:3,730 LF and 4,609 O.

9/11 ratio required, so only 4558 units of the O count. So, total effective delivery is 3,730 LF and 4,558 O, equalling 8,288 units of fuel in total.

Total fuel bill including delivery: 8,443.47 units of LF and 9,397 units of O.

LF costs √0.8/unit, while O costs √0.18/unit. So that's √6,755 for LF and √1,692 for O, making a total fuel bill of √8,447. Divide √8,447 by 8,288 units of fuel and we get √1.02/unit.

Craft file at https://www.dropbox.com/s/hi3nwcn64yvzw94/Kerbodyne%20Titan%20Fueller.craft?dl=0

[UnusualAttitude]

Nice...

Here's my first entry, The Jug. It's a quick and dirty reconversion of my standard Mk3 cargo and got about two orange tanks worth to 75x75 km. 130-something parts.

It delivered:

LF: 5,696

O: 6,867

LF corrected for O equivalent: 5,618

Total fuel usage: 11,360/12,972

Total cost in funds: 9,088/2,335 - 11,422

12,485 for 11,422 funds or 0.915 funds per unit. Please check my maths.

I'll be working on a dedicated tanker. So far, my fuel fraction is 0.68 and I think this can be improved with a Mk2/Mk3 hybrid or fuel tanks with a better fuel fraction than the spaceplane parts. The limit will probably be the thrust of the KR-2L engine if we are limited to 150 parts...

Full flight:

http://imgur.com/a/uOL2K

Download:

https://www.dropbox.com/s/cy1fha51k1ol051/TheJug.craft?dl=0

Edited March 29, 2015 by UnusualAttitude
Correct Imgur link

[GoSlash27]

My entry is up.

The idea was to see how much I could get away with using just 1 turbojet.

http://s52.photobucket.com/user/GoSlash27/slideshow/KSP/PeacefrogIII

http://wikisend.com/download/261372/PeaceFrogIII.craft

It is extremely docile and stable to fly with no major vices.

I delivered 13.5 tonnes of fuel and oxidizer ($964 worth) to a 103x103km orbit while expending $362 in supplies.

1.88 units of monopropellant ( $2.25)

417 units of fuel ( $333.60)

and

144 units of oxidizer ( $26.18)

This works out to $26.81 per tonne. Not sure how you want to calculate that, so hit me up if you have any questions.

*edit*

fuel units delivered= 2,100

Flight expense= $362

Markup would be $0.17 per unit on top of the base $0.46 per unit on the ground, so fuel in orbit is $0.63 per unit ( a 37% markup).

Best,

-Slashy

Edited March 29, 2015 by GoSlash27

[Starhawk]

Knowledge is meant to be used, so release the hounds

So I take a look at the craft Slashy posted as his first intended entry, and I'm intrigued. He's pushing the limits of what I've considered possible with spaceplanes. I take a look at his design, planning to shamelessly steal all his secrets.

I build my new creation. I release the hound! Scamper, whimper, plop down on grass... Well, that didn't work.

I expected a few difficulties because I was pushing the limits, so I figured I just hadn't figured out the proper ascent profile. After half a dozen tries (using up all my KSP time for the weekend), I stopped. I saw Slashy's entry today, and downloaded it. It flew like a charm!

I realized I had been beating my head against a wall. I stopped and carefully compared the designs. As soon as I started looking at the numbers I did a major facepalm! I was using way too little intake area. A quick redesign and I was back in business.

The new design with enough intakes worked beautifully and is now in orbit.

Mine isn't pretty like Slashy's, but it's beautiful in function.

More later...

Happy landings!

Edited March 31, 2015 by Starhawk

[GoSlash27]

Starhawk,

Great jorb! It's neat seeing such a big plane get into orbit with so little engine.

What's your OMS rocket arrangement? I can't really tell from this picture.

FWIW, you really shouldn't need that many intakes. It won't hurt your performance any, but once you get past "enough to do the job", they get a bit spammy. 0.0013m^2 intake area per tonne of aircraft should be sufficient.

My entry is using 5 radial intakes (4 was on the ragged edge but worked) and that's equivalent to roughly 2 ram intakes.

I'm looking forward to seeing your mission report!

Best,

-Slashy

[Mister Dilsby]

Not really an entry because it's not crewed, might have too many parts (I never counted) and is only mostly SSTO--but I thought you might enjoy my Spacecow Mk4 none the less.

170t at launch, 20 turbojets, selectable in two groups (12 in #1, 8 in #2). SRB-assisted takeoff. Brings 50t of fuel/O2 to LKO along with up to 750 units of monopropellant. Four standard ports ventral in an incredibly suboptimal "Space Udder" configuration, two ports dorsal, and one small port on an extended forward boom.

I think I could definitely squeeze more altitude/velocity out of her on jets and thus deliver more fuel to orbit. But I don't want to spend time optimizing a spaceplane in the last weeks of v0.90. It works, sort of, and it's what I'll be using to tank up my fleet for what will REALLY be my last v0.90 mission, an Eve Rocks attempt.

Javascript is disabled. View full album

Edited March 31, 2015 by Mister Dilsby

[Starhawk]

What's your OMS rocket arrangement? I can't really tell from this picture.

My standard configuration, and yours. Two 48-7S's.

FWIW, you really shouldn't need that many intakes. It won't hurt your performance any, but once you get past "enough to do the job", they get a bit spammy. 0.0013m^2 intake area per tonne of aircraft should be sufficient.

My entry is using 5 radial intakes (4 was on the ragged edge but worked) and that's equivalent to roughly 2 ram intakes.

Well that's exactly the thing. I'm used to limiting myself to 15 tonnes/turbojet and my favourite setup is 2 ram intakes/turbojet which is more than enough. So that's what I used for my first attempt. But with over 20 tonnes of mass, 2 rams is not enough. After flying the Peacefrog III, I ran the numbers. 20 tonnes needs .026 m^2, but two rams is only .020 m^2. It made all the difference. I did managed to cajole it into orbit eventually, but I needed lots of rocket burn to establish my apo above 70 and ate into my payload to do it.

Five radials is .030 m^2 which is exactly the intake area of three rams. Going from two to three made all the difference. I could have gotten away with two rams and a single radial according to the numbers.

Anyway, I hope to get the mission completed and posted later today. Real life keeps getting in the way.

Happy landings!

[GoSlash27]

My standard configuration, and yours. Two 48-7S's.

Well that's exactly the thing. I'm used to limiting myself to 15 tonnes/turbojet and my favourite setup is 2 ram intakes/turbojet which is more than enough. So that's what I used for my first attempt. But with over 20 tonnes of mass, 2 rams is not enough. After flying the Peacefrog III, I ran the numbers. 20 tonnes needs .026 m^2, but two rams is only .020 m^2. It made all the difference. I did managed to cajole it into orbit eventually, but I needed lots of rocket burn to establish my apo above 70 and ate into my payload to do it.

Five radials is .030 m^2 which is exactly the intake area of three rams. Going from two to three made all the difference. I could have gotten away with two rams and a single radial according to the numbers.

Anyway, I hope to get the mission completed and posted later today. Real life keeps getting in the way.

Happy landings!

Oh, I've gotcha! *facepalm*

That looked like 4 rams in the pic...

Looks like you've got it wired!

Best,

-Slashy

[UnusualAttitude]

Wow, this is hard. I've spent a couple of evenings trying to get anything above 250 tonnes up there and have so far failed miserably. Too much flex in the wings, causing my designs to roll, flip or spin out even if FAR says green. Testing is also time-consuming when you can't time warp, as anything this big will take at least 15 minutes to get up to speed. Drag also seems to increase exponentially with size, making performance sluggish even with a good TWR. I suppose the twin KR-2L spaceplane shall not be... Taking a break and starting again from scratch if I have sudden inspiration.

Any joy on scaling up, Wander or anyone else?

Some interesting designs from the stock aeronauts... I would be curious to see more valid large Mk3 entries, just to see how you get around the structural problems...

[GoSlash27]

Wow, this is hard. I've spent a couple of evenings trying to get anything above 250 tonnes up there and have so far failed miserably. Too much flex in the wings, causing my designs to roll, flip or spin out even if FAR says green. Testing is also time-consuming when you can't time warp, as anything this big will take at least 15 minutes to get up to speed. Drag also seems to increase exponentially with size, making performance sluggish even with a good TWR. I suppose the twin KR-2L spaceplane shall not be... Taking a break and starting again from scratch if I have sudden inspiration.

Any joy on scaling up, Wander or anyone else?

Some interesting designs from the stock aeronauts... I would be curious to see more valid large Mk3 entries, just to see how you get around the structural problems...

I think the big problem is going to be the part limit.

The biggest (and least efficient) wing panels only have a lift rating of 2. They will add up in a hurry.

By the time it's all said and done, it's probably going to wind up in a weight range where the old Mk2 parts are preferable.

I'll take a look at it in the next few days.

Best,

-Slashy

[Starhawk]

All right, I have two entries. Here's the first one. This is the one I previewed in my earlier post. I call it the SerenityNewt.

Javascript is disabled. View full album

Total costs were 1347 funds to deliver 2100 units of fuel to a 105 x 105 orbit. For 64 funds per unit.

I didn't fly the ascent particularly well, and it comes in slightly less efficient than Slashy's PeaceFrogIII.

Here's a link to the craft.

So I took another crack at it...

- - - Updated - - -

So here's another entry. This is the GrumpyToad.

Javascript is disabled. View full album

Total costs were 1326 to deliver 2100 units once again. Comes to 63 funds per unit.

Damn. Once again not a great ascent. I hit terminal velocity on the way up. I thought this one would be a bit more efficient, and it was. But not by very much.

The GrumpyToad is better looking though, I think.

Here's a link to the craft.

Happy landings!

Edited April 1, 2015 by Starhawk
added craft files

[GoSlash27]

*Snort* "Serenity Newt"

Yer killin' me, Smalls...

Either one of those is right in the ballpark of where I'm at in terms of efficiency.

Time to see what we can make happen with multiengine Mk. 3 designs.

I'm going to start conservative with a 90 tonne design; see if the low drag of the fuselage sections can make up for the god-awful wings.

Best,

-Slashy

[Starhawk]

Either one of those is right in the ballpark of where I'm at in terms of efficiency.

Yeah, I was hoping to edge out your score, but it's hard to get too much more efficient.

Time to see what we can make happen with multiengine Mk. 3 designs.

I haven't had any luck with my Mk3 designs so far. I'll have to take another look at one of my prototypes.

Happy landings!

[GoSlash27]

After playing with the Mk3 parts, I have concluded that they're just plain not worth it if you're trying to build an efficient SSTO tanker.

The tank sections are too draggy, the large wing panels are too inefficient, and the available control surfaces are too ineffective.

I can make a functional SSTO out of these parts within the part limit, but it's not going to put fuel in orbit as cheaply as the Mk2 can.

Best,

-Slashy