Making Terrestrial Buildings Fly...

[Hotel26]

[This is a Work-In-Progress and a Request-For-Consultation.  I'll put a link in Work In Progress to this topic, but don't wish to otherwise clutter that thread.  Moderators will let me know is this topic doesn't belong in Spacecraft Exchange, but I'm looking for an exchange of ideas.]

The story starts with this crazy thang ["Boom Vang"] which is launched atop a booster:

It's 2x Mk3 crew cabins with a cupola and I use it to mark a "regional airport" and it's also a compact "town" for 32 Kerbals.  After re-entry, it uses my "Cruise" system to fly (slowly) to pin-point the target, at which point it jettisons the paraphernalia of flight and settles to Kerbin.  (I have a lunar equivalent of this, too.)

I also have this crazy thang ["Silo"], as yet unpublished:

It's the same deal, except it can produce fuel.  I use this to denote a "hub airport".  It's a kind of terminal building.  It only has a single Mk3 crew cabin.  (And there's a lunar version, too.)

Key in this is that you can see Gryphon III docked to the Silo and it provides a conduit for transfer of passengers and fuel between the terminal and visiting aircraft.  The Gryphon is also able to execute a "push-back" for aircraft that need it.

                                                        

OK, so here is where I am going next (code name: Luxor):

Don't ask me why this works like this on water (a swimmer in the water can even reboard this!).  But, yeah, water port!

Making this fly is beyond my current abilities, I think.  I am willing to sacrifice the (sub-)orbital launch capability and just fly it however painfully slowly to its final destination.  (There aren't going to be more than about a half dozen of these.)

Incidentally, I wouldn't mind removing most of the central bulk and reduce it down to 2 cupolas, back-to-back, say.  It's main purpose is simply to root the 3 columns, anyway.  I also have this idea about placing the columns far enough apart to place ground girders between them in a triangle, each with an upwardly-vertical claw placed, so that a client aircraft has clearance to nose in and retract the nose gear to 'kneel' on the claw to dock.  I've found that this is a super flexible method.  Then using an "uphill approach" to one of those 3 docks would allow a client craft to simple release its brakes and roll backward to depart.  No ancillary equipment required.

                                                        

Fundamentally, I don't really care how the objectives are achieved, nor what the result looks like.  Those objectives are:

1. large footprint so that it's not likely to topple over on an unlucky landing
2. accommodate at least 50 Kerbals
3. be able to fly from KSC to anywhere on Kerbin (i.e. 180 degrees) or else be able to land for refueling on the way
4. be equipped to jettison as much of the wing/engine/control surface paraphernalia as possible to most resemble a stationary building once landed
5. maneuverable to a pin-point landing (within 200 meters, say, with skilled pilotage) made finally, vertically, under chutes
6. store some fuel that can be deposited by a tanker for refueling operations
7. provide direct passenger and fuel transfer via 1 or more inbuilt claws, pointed upward and with sufficient clearance to accommodate most conventional aircraft of widely varying sizes

Does anyone have any interest and/or ideas/suggestions for collaboration on this?

 

Edited May 11, 2019 by Hotel26

[Guest]

I made a quick experiment based on your design idea and I don't think it's that hard to get this to orbit and back. Just put a bunch of lifter stages under each of the three pods and "hat" with sufficient parachutes on the central module and/or the pods. My rough experiment was designed to your spec, and it needed 3 Vectors on each of the pods to make orbit with plenty of TWR to spare. It went up, re-entered, and landed nicely, although touchdown was a bit of a bump at over 7 m/s. It's a ballistic craft so precise landing is what it is; it's no easier or harder than with any other rocket with retro rockets and about 400 m/s dV to play with.

It should certainly be possible to give it more precise landing capability by adding jettisonable control surfaces to fine-tune the vector during the last stage of the descent, prior to deploying the chutes.

I will likely return to this idea once Breaking Ground drops as I love bases and want to go nuts with the new capabilities.

Edited May 11, 2019 by Guest

[fruitsbar]

Just heard about the Breaking Ground DLC and I am hyped! I plan on getting it after I have Stormworks.

Anyway, enough of DLCs, that's an amazing creations! I always try finding smart ways to build bases.

[Hotel26]

11 hours ago, Brikoleur said:

It should certainly be possible to give it more precise landing capability by adding jettisonable control surfaces to fine-tune the vector during the last stage of the descent, prior to deploying the chutes.

Sure.  Take a look at the screenshots on Boom Vang, which show the three stages of flight: orbit, flight, landing.

And then take a look at this precursor, which was the origin of Cruise[TM] technology.

 

At one time, I was lobbing Aquila Cruise at Gilligans Island, 130km to the NE of KSC and grouping them with about 20m accuracy.  It's quite fun and I learned eventually to do a bomb-dive at the target to kill horizontal speed and throw the chutes at the last possible moment.

So the broader scope is that I am finding long flat (no kinks) stretches of grass to be used as commercial airports.  After landing, fatigued pilots do not wish to taxi a long distance to the terminal.  So I really want to be placing them with at least 100m accuracy.  That's why the "flight phase" is so important.  In addition, Cruise has a range of around 1,000km (painfully slow) but it really adds tolerance to a badly-placed re-entry.  At the equator, this isn't such a problem, but when timing a re-entry for a higher latitude in which one must lead the target (as well as time the orbit to commence), having some latitude to navigate to the target becomes quite crucial.

Really appreciate you taking the time, @Brikoleur to prototype this and offer advice!

For those who want to go hands-on with this project, I recommend experimenting with Boom Vang and Aquila Cruise.  I'll therefore soon post Silo to KerbalX as well.

For anyone else, as silly as these vehicles may seem in concept, they are really good fun and they'll exercise some unusual piloting skills.

Edited May 12, 2019 by Hotel26

[Hotel26]

I gave the Vectors a try and they are great.  I think it's reasonable to scale back to 3x Mk3 crew cabins, too, as that totals 48 Kerbals.    I think one claw should be sufficient too, since they add drag.

I guess it's a good idea to build the 3 sides separately -- in my original, I built them symmetrically which meant an unnecessary "dorsal wing".  And build it as a runway-launched vehicle to flight test it.  Then add the rocket paraphernalia at the end of the project...

[Guest]

I made another attempt which failed in an interesting way: I think that would lead to a pretty nicely guidable solution. 

I made it a conventional two-stage, with about 3000 m/s on the lifter stage leaving about 1400 on the second stage. The trick with the second stage was that I had thrusters strapped radially to the pods. I had it tuned so that it that the second stage had just about 1.2 TWR empty. 

The failure was that it flipped on re-entry, wanting to fly nose first. It survived, slowed down, and went back to tail-down once I deployed the 'chutes and made a very nice soft landing, but obviously I couldn't control my landing zone at all accurately, hard to do when tumbling wildly.

The improvement would be to add some aerodynamic lifting and control surfaces to the thrusters -- not enough to turn it into an airplane, but enough to make it a guided missile. Re-enter like a spaceplane, then guide it aerodynamically to the target zone, point it upwards, deploy 'chutes, float down. Since it has a TWR > 1, I could even make small hops for fine-tuning on the ground before jettisoning the thrusters. If I wanted really precise positioning, I'd swap the landing struts for wheels and drive it into position.

Edit: tried it, flew like a lawn dart. It would need a fair bit of aero tuning to make it genuinely controllable. I'm quite confident it's doable however. I also streamlined the design somewhat, putting the retro rockets inline hidden inside structural tubes.

I don't do Kerbin base building myself at this point anyway so I don't know if I'll ever get around to fine-tuning the aero for this unless I decide to send something like this to Laythe one day. However I really like this three-pod, three-claw base design -- I was able to get an ISRU setup, science lab, comms, power, and space for 10 kerbals into it and it flew to orbit beautifully with about 1300 m/s to spare. With a drop tank attached to the centre module I could easily send this almost anywhere in the Kerbin system other than Eve. It's a really neat base design @Hotel26 and highly adaptable for just about any purpose!

Edited May 12, 2019 by Guest

[panzer1b]

I dont really build much of anything for kerbin these days, but i have made my share of flying buildings capable of landing on planets.

Was a bit lazy when i launched this and had it preassembled (each module is technically designed to deploy itself or be skycraned down), i can always detach modules and or add more modules as needed atop the 2 currently on the base

This is far from my only building that is self landing (with a parachute module it could theoretically land on kerbin too), but its one i finished this weekend.  While most people build their ground bases and then attach em together horizontally via docking ports on the sides, for some reason i prefer vertical stacking, more challenging to land it, but easier to attach together.

And yeah, nomatter how bad this is with regards to part count, i always like to make my own 100% custom interior for my bases, beats the subpar IVAs the game ships with (only mod that ive seen with truly amazing IVAs is the spacestation pack with those centrifuges in it, and even that doesnt come close to the coolness of custom IVAs).

[Guest]

All right I got a bit bored so I made a flying building. You're right @Hotel26, it is fun to deliver it. I didn't get it quite as precisely as you did but with a bit of practice it ought to be doable.

Note that it flies decently (for a building) at full thrust, but when you cut the throttle it goes into lawn dart mode.

Craft: https://kerbalx.com/Brikoleur/Flying-Building

Note: I still think ballistic delivery with only limited "guided missile" style guidance is the way to go. I haven't tuned the aero on my previous effort to get quite that far but I'm convinced it's possible. Another advantage is that the craft can be fully three-way symmetrical. It's not going to be a very effective rocket because you will need lifting surfaces on the lifter stage as well to counteract the ones in the re-entry stage, since they will otherwise cause the launch to flip over.

Edited May 14, 2019 by Guest

[Hotel26]

2 hours ago, Brikoleur said:

All right I got a bit bored so I made a flying building.

Amazing job!  I'm very keen to fly this.  The end result looks very decent, too.  I am truthfully quite staggered you made this fly.  It's the surface equivalent of a space station!  I am already imagining the planes parked around it.  (Downhill side reserved for airplanes with no reversers!  )  Thank you, @Brikoleur!!  I'll post some feedback...

[Hotel26]

Thinking about your comment about launching versus gliding, a long-range idea (should it ever become necessary), is to launch the building to space vertically; then launch a winged "glide-bed" to fly up and rendez-vous, adding wings, controls, etc.  Then de-orbit, survive re-entry, and glide/fly (some power) to the exact destination and deploy under chutes.  This is probably the most scalable route?

More after I've flown your building!

[Guest]

It's not that hard to make things fly if budget is no object. Just add wings and engines. 

I did some further experiments on ballistic delivery. Got it to the proof of concept stage. I'm not sure how you'll like it though, but it does have the range. Personally I prefer this as I would be much more likely to use this somewhere other than Kerbin and Laythe is the only other body where a glider/flyer would be preferable to a rockeet.

I put some heavy engines (nicely hidden away) under each of the three towers, and a ring of airbrakes around the centre (they look rather decorative when not deployed). The re-entry module's centre of mass is behind the centre of drag, which means it returns tail first. The powerful retro rockets make it possible to tune your return trajectory through the entire re-entry. In theory you ought to be able to plant this pretty precisely:

(1) Pick a re-entry corridor that passes over your landing zone as precisely as possible

(2) Tune the port/starboard direction as early as possible -- this will not be possible during the hardest aerobraking period because aero forces will keep you pointed retrograde

(3) Pull the landing position closer by judicious use of the airbrakes and retro thrust as needed

(4) When falling near-vertically, fine-tune the landing position by thrusting laterally

Parachutes are a safety feature in case you run out of dV tuning the landing, the retro thrusters are powerful enough for a safe landing on their own.

My first attempt was to return to KSC and I missed by a couple of km; I'm pretty confident than practice and an improved design would let me land it right on the launchpad.

An easier-to-deploy solution would be to make the re-entry module a glider, but that would be a good deal harder to design and would require lofting a much bigger module to orbit. I think I'll give it a try at some point anyway.

[Hotel26]

Yes, thinking about it today, I had come around to the idea of ballistic.  Anywhere to the other side of the planet is directly sub-orbital and maneuverable.  Rocket flight in the upper upper atmosphere has the advantages of less energy for re-entry. 

I'm flying this now and have been able to touch-down horizontally at a very low vertical rate (for the purpose of refueling) but have almost given up on it not breaking up on touch-down, even after some auto-strutting.  Stall seems somewhere around 140m/s.  Possibly switching to drag chutes would stop it settling to the ground so hard.  Maybe landing on fumes would help this situation, too...?

Ceiling is somewhere around 2.5km but I'm still exploring that.  More shortly.  The goal looks feasible!

I took the oxidizer out and loaded as much fuel as I dared...  Will see how far it can go.   Has range with 1/3 fuel for 1,200km and need 1,800, and has plenty of oomph still on the runway, so should be able to carry more...

Incidentally, one of my earlier attempts had 3 winged surfaces symmetrically (because I am crap in the VAB).  It's extra drag but it launches better.  And it flies (in a glide) better, even if less efficient.  And if it's not landing (with wings), it really doesn't matter either if they are not well-placed with respect to the ground...  to glide better, more wing may help and require less power if potential energy is being traded.  Less fuel.  So, I think you're on the right track.

(Three wings on lift-off with the dorsal wing being jettisoned before apogee...  Wings purely for glide; not extended flight.)

All of your above could be great fun.  No wings at all?  For this exercise, making several attempts to nail a landing site within several hundred meters would just make it all that more satisfying.  One would recover the losing attempts. 

[Well, I made it across the Atlantic to Buenos Aires!    It looks very futuristic, in the middle of nowhere, like something out of Woody Allen's Sleeper...!]

Flies with 5,200 kals in it.  One only needs 1/4 turn around Kerbin, if one can land it one time on a polar cap to refuel.  2 flights to anywhere, for just 900km range.

"Ice Station Zebra, this is FlyBing 17 orbiting HOTEL, burning off fuel prior to landing; expect us inbound in 3 hours 20; FlyBing 17...".

Edited May 15, 2019 by Hotel26

[Hotel26]

[Guest]

You're not supposed to land it! Just shed the plane-ey stuff and parachute into position. Nice job managing it though, it must've been a handful because of the way it noses down when you cut throttle.

If you want to make it go higher and faster just add engines.

[Hotel26]

6 hours ago, Brikoleur said:

it must've been a handful because of the way it noses down when you cut throttle.

  I flew it onto the deck at 180 m/s full throttle, after establishing a very gentle descent way out, and only chopped the throttle and threw the chutes when the wheels were rolling on firm ice.    The fuel trucks will need onboard GPS to find where that machine parked up and will have a trek to get there.  But the landing strips are pretty big at the caps!

I haven't fully loaded it with fuel to see what it's max range is.  I fully loaded the aft tanks and the wings, everything jettisonable.  I am completely satisfied with this aircraft: it's fit for purpose!!  Thank you.

Edit: "building"

Edited May 16, 2019 by Hotel26

[Hotel26]

Super excited by this whole topic and progress so far!  What a lot of fun...  Brikoleur has shown the way.

So imagine this, but without the wings...

It's an all-symmetrical build.  3 Vectors on the tail and one under the cupola.  Tons of TWR and arrives in orbit with enough dV remaining to make it to the Mun.  It's the lunar equivalent of our Kerbin model.

The drag and mass of the wings do not stop it making orbit and so it can be re-entered (have not yet tried).  I'm still doing low-altitude flight tests.  The tanks are set up for rocket flight , which means the tail tanks drain first and, thus, the CoM is constantly shifting forward.  As a result, so far, it has that "lawn dart" behavior, but if the fuel priorities are reset for level flight -- and with a bit of play with the final position of the wings -- I think @Brikoleur's scheme of bringing it down in the target zone and then making a final, powered push toward the marker can work.

We'll see.

Edited May 18, 2019 by Hotel26

[Hotel26]

This is Eclipse and I've shared a link to the craft file [v1.6.1+]:

It can be operated in any of 3 mission profiles:

1. suborbital to descend into the target zone and then glide/fly to the target
2. go to orbit, refuel and then de-orbit [my original plan , but actually,much more complicated than simple sub-orbital]
3. go to orbit, refuel and then proceed to another planet/moon.

When approaching a target on the surface of an atmospheric body (e.g. Kerbin), dive-bombing the target from pretty much above gives the best accuracy.  You can also make a low, high-speed pass and swoop up, possibly tumbling deliberately to wash off speed and then pop the trunk and the chutes.

If going to vacuum for good, the wings can be jettisoned either on the pad or at about 8 km, after which they don't serve so much purpose.  KP9 disables the 3 aft Vector gimbals, which I found was useful for circularization.

The craft file above is set up to drain the aft tanks first, so that you can jettison the tail engines, empennage, tanks and dead weight.  For sub-orbital or re-entry, the fuel priorities of all tanks should be set identically, because you'll be keeping the empennage .

To play with this machine, I was firing it vertically off the pad but pretty quickly turning north.  (I use Atmospheric Autopilot a lot, so no guarantees about stability in all phases of flight.)  Land one 100+ km somewhere north and mark it with a name such as "Eclipse target" so that you can select that particular one as target for subsequent flights.  Then simply start throwing darts at that first target and practice landing as close as you can.  My last try was a flukey 36.2 meters albeit with a structural failure ensuing at touchdown due to a new tank that had not been strutted.

And, don't be surprised: it's built upside down.  The rudders hang downward...

Have fun and let me know how you get on and what improvements you might suggest!

Edited May 19, 2019 by Hotel26

[Hotel26]

Enthused about all of the above, I went back and revisited the unfortunately-named Boom Vang.

I came up with Bullseye:

      

  

Bullseye is sub-orbital only but can make it to the other side of Kerbin.  It's quite docile in maneuvering to pin-point the landing and has long legs, gliding.

It's more or less done, except that I am experimenting with ugly bolt-ons to give it a claw on the ground extended on a boom for airplane docking.  It might be possible to put a rover contraption into a compartment behind the Mk3 cabin that would come down separately under chutes and then drive over and dock.  (Or maybe this is a job for "robotics"?!)  But, at this point, I like the handling of Bullseye so much that I refuse to mar it.

Bullseye gives you a "town" anywhere you want on Kerbin, which then gives you a reason to fly all the beautiful planes that we Kerbals have at our disposal.  What a wonderful world!

Edited May 22, 2019 by Hotel26

[Guest]

I made another iteration. This iteration, unlike the first one, is entirely controllable even at zero throttle. 

It also has rocket-assisted take-off which will fling it to 10 km altitude, from where it will be able to cruise to its destination. I have no doubt it would be possible to get this to orbit simply by strapping on bigger boosters.

I think my preferred version would be ballistically delivered and only have glide capability post re-entry. That would take a bit more experimentation, also I'm not sure this would survive re-entry as some of the wings are a bit fragile.

https://kerbalx.com/Brikoleur/BASEF2

Edited May 24, 2019 by Guest

[Hotel26]

50 minutes ago, Brikoleur said:

I think my preferred version would be ballistically delivered and only have glide capability post re-entry.

Take a look then at Eclipse and see what ideas it stimulates...?  Eclipse can go lunar, too.
  

Additionally, I have a single building variant, called Bullseye.  (Recall that my plan is to build aviation networks: cities, towns...)
  

I actually mentioned Eclipse above, I think, but this will be the latest version.

P.S. @Brikoleur I have been flying your Tern a lot lately.  I have to say it is one of my favorite craft of all time.  (I have to confess I've clipped 6 Junos into just 2 bays so that it can access landing sites at up to about 5,200m and this version (Tern-R) is retrac as well.)  I've been flying it so much (debris investigation and clean-up missions) that I feel comfortable I've mastered the transitions between HORZ and VERT...  Great machine!

Edited May 24, 2019 by Hotel26

[Guest]

All right, circuit around Kerbin and pinpoint delivery.

Craft: https://kerbalx.com/Brikoleur/BASEF-B

Edited May 24, 2019 by Guest

[Hotel26]

7 hours ago, Brikoleur said:

All right, circuit around Kerbin and pinpoint delivery.

I love it.  It's such a thrill to lob these things just where you want!!  Well done, sir.

         -------------

In a seemingly complete non sequitur, I've realized that a Mk-3 cargo bay, either open or with a ramp attached, could act as a very nice submarine dock (I problem I had not yet solved).  Just drive the sub in to trap it; propel forward to meet a claw; open the bay doors to release the sub...

Edited May 25, 2019 by Hotel26

[Guest]

I want to make a Mk 3 cargo bay variant of this base that includes drills, ISRU, power, and science lab. Hard to package all the Gigantors without ruining the clearance for the craft to dock.

[Hotel26]

1 hour ago, Brikoleur said:

I want to make a Mk 3 cargo bay variant of this base that includes drills, ISRU, power

I have this: it's code-named 'Silo'.  It's a Mk3 cabin with fuel, drills, ore tank, ISRU, radiators, power.  It was in the publication list but hasn't made it yet. 

Naturally, also a lunar version: https://drive.google.com/open?id=1QgOP47KAw9GE5rxwB6JdQnJFLzfhJns8

Although I typically use KerbalX Goblin for the latter.

I've built 3 planes this morning.    The latest is on-topic, called Swallow.

It's a plane that comes apart in mid-flight to deposit a Hab + LanderCan.  (The pilot has to bail out herself and land under a separate brolly.)

Update: actually, the lander can has command control, so she simply has to transfer back there and then flip the Red Switch...

Edited May 25, 2019 by Hotel26

[Guest]

I really love this three-tower design of yours. Got everything packaged into a base that works pretty nicely. Still have to test it with an engineer to see how the power/thermal holds up with the increased efficiency. 

I deploy one of the chutes early to yank the base into the proper attitude without too much force.

Note the improved design of the Claw-holders for much better clearance. The base is now effectively hexagonal.

It could use a few more 'chutes, landing was a bit on the rough side at about 9 m/s.

Edited May 25, 2019 by Guest