Kerbin Circumnavigation 1.0.4/1.0.5 - Aviator Challenge Continuation

[run1235]

would it be possible to essentially have f.a.r as a hard mode since it does make it harder to fly and such? i mean, building in stock just feels to easy. ^.^; first plane i've made has proven itself to be stable at 20k and without even making it half way around i can already tell it will make it all the way. probably wont set any records or anything yet as i'm just testing but yeah.... can there be a f.a.r category please?

[Xannari Ferrows]

I will consider removing the "no other aerodynamics" rule, but there will be no mention for those that use it, fair warning.

[Aetharan]

Curiosity that's been brought up, going back over some of the previous posts in this thread: Might there at some point be a category for vessels completing circumnavigation with either an inert payload, or with passenger counts? I've actually been strongly considering trying to build for a higher body-- erm, headcount than the previous passenger plane already shown. If such variables actually start to be counted, much as velocity is, then that would be all the more reason to work on that bit of engineering. It's your thread, and so your decision, but consider me to be in support of (and willing to compete for) "Workhorse" and "Passenger" Circumnavigator tags.

Edited August 14, 2015 by Aetharan
Minor correction of word choice.

[GraviTykillz]

Plane

Commin round

Final time

Bounced of runway.. but i would like to submit under the Close enough rule~

REally ..not sure how i can be faster..much respect to those under 40

Had to quickload once..dang..but well there it is

Final time 42.5

[Noname117]

I see you have some extra liquid fuel which may have been weighing you down.

[GDJ]

It's all in the take-offs and landings. Your speed is fine.

Get up to optimum altitude as quick as possible, and get back down as quick and as smooth as possible, and when you land, hit the brakes for all their worth and get stopped as quick as you can.

I'm confident you'll break the 40 minute barrier. Took me a few tries.

[Death Engineering]

Curiosity that's been brought up, going back over some of the previous posts in this thread: Might there at some point be a category for vessels completing circumnavigation with either an inert payload, or with passenger counts? I've actually been strongly considering trying to build for a higher body-- erm, headcount than the previous passenger plane already shown. If such variables actually start to be counted, much as velocity is, then that would be all the more reason to work on that bit of engineering. It's your thread, and so your decision, but consider me to be in support of (and willing to compete for) "Workhorse" and "Passenger" Circumnavigator tags.

Cool idea.. After the 52 kerb' flight, I started something just a bit bigger.

[Xannari Ferrows]

Cool idea.. After the 52 kerb' flight, I started something just a bit bigger.

Uh oh. This can always never mean good things.

[ExaltedDuck]

I took the rule change regarding tail cones as impetus to turn my focus back toward distance rather than speed. My late 1.01 craft with 12 turbojets and 45000+ fuel capacity would have been capable of 5 or 6 laps but as fate would have it, pilot error claimed its maiden voyage and the beating drum of progress (version 1.02, released the following morning) would mean mean there would be no take two, as its capabilities were diminished to where 3 laps would have been a stretch.

I never gave up on the dream. Sometime, I should put together a gallery of the cockamamie behemoths that have rolled out of my spaceplane hangar. Some actually flew pretty well. Many were an absolute handful. And most would explode in time acceleration.

I had a really strong contender last weekend. Carrying about 25000 fuel, it was able to reach a cruising speed around 1530-1540 on 8 rapiers and complete full speed laps with about 3000-4000 fuel. But the sheer weight along with the inevitable mid-flight balance changes made climbing difficult, requiring 2-3 speed dives during ascent through some very lawn-dart-ish flight characteristics, meaning that reaching supercruise took 9000-12000 fuel, leaving me with just enough enough for 4 laps in most cases.

So I looked back and I looked small. I took inspiration from the first plane I entered in this thread: Simplicity. NAmed for its design philosophy, it turned in a respectable 51 minute flight on a single turbojet. From it was born the Elaboration. I traded mk1 nosecone c-pit for a mk1 inline c-pit carrying a shock cone intake. The pre-cooler was swapped for a standard jet fuel fuselage, and an additional jet fuel fuselage section was added. The small gear were traded for a medium set on the hopes that the braking might be better. I took it out last night and fell about half a turn short. Realized I only need about 200 more fuel so a set of big s strakes along with a slight adjustment of the wing angle (aesthetics are important) and the addition of chutes and we have what I present below:

Javascript is disabled. View full album

5 full circumnavigations in 3 hours, 43 minutes sharp (or +2 seconds if you want to count the time I was in shock and nearly having a coronary when she rolled.)

Edited August 15, 2015 by ExaltedDuck
typos! typos everywhere! they're taking over the world!!!

[Xannari Ferrows]

Wha-what-what!!?? You can't be serious! 5!?

That is ridiculous man. It feels weird putting this up on the board. Oh well, I'll have to get over myself eventually.

[Mikki]

... 5 full circumnavigations in 3 hours, 43 minutes sharp (or +2 seconds if you want to count the time I was in shock and nearly having a coronary when she rolled.)

Hi ExaltedDuck,

Congrats! Have my tasty cookie, 5 times around must have been real hard... you are the king of "turning around"

[GraviTykillz]

So Close..i will get under 40..used just enough fuel to get around..

40.8

I will Join the under 40 Club Soon...I tweaked the back breaks to 25..that should get me under...

[ExaltedDuck]

Thanks Xan, Thanks Mikki. You two crack me up. (and are they those white chocolate and macadamia nut cookies from Costco? I friggin love those things)

I tweaked the back breaks to 25..that should get me under...

Parachutes? Radial drogues did very well for me. Mount them close to where your COM will be when your fuel is just about all burnt off. That way you won't lose as much control (seems a common habit is to put them far back which can make you go all lawn dart on the tarmac if you're still airborne when they're deployed). Then, come in low, level, and hot. Drogues can survive up to about 550 m/s. But it's hard to keep a clean approach much over 300 IMO, so the radial standard chutes work fine, too. They actually lend quite a bit more drag. Try to be less than about 10 m over the strip when they deploy. It can get rather dicey.

Edited August 17, 2015 by ExaltedDuck

[GraviTykillz]

I cant figure another way to slow my plane fast enough..Tried using an air brake but wasnt as effective ..Think im using droges..

Tweekin the back breaks is going to help also:huh: Im going to get there!

[ExaltedDuck]

airbrakes can be difficult. I followed Helmut's example of using landing for drag in the upper atmosphere then used hard pitching maneuvers if needed to get speed down more. I tried to get down to around 1000 m/s from cruising altitude to about 10000 m then tried to hold as much of that 1000 as possible, aiming my nose short of the runway to bleed down to 300-ish on the final nose up, the use the chutes instead of a typical flair maneuver to slow to touch. Like I said, it gets dicey.

- - - Updated - - -

also, you might be able to get a few more m/s while cruising if you can afford to swap the big s strakes out for another wing type. On some of my faster crafts, I used big s elevons and elevon 5's as primary lifting surfaces (often with smaller elevons as canards). They get a bit tricky to fly, though, since they pretty much go from neutral to full deflection with any input (any slight adjustment can easily turn into 15+ g aerobatics. =D). The drag reduction wasn't always woth it. Having a small wing surface plus an elevon doesn't seem to lose much speed versus having none and just an elevon and you can see on my latest velocity attempt, I opted for a rather full wing surface (C? maybe... it's the large rectangular wing connector). If I were to do it again, I'd probably try the small delta wings or maybe a swept wing variant. My gut feeling is that it's more important to minimize the level flight angle of attack at speed, and that takes a very careful balance of lift and mass.

Edited August 17, 2015 by ExaltedDuck

[lemon1324]

Well, It's done at last! Three full circumnavigations, all done using only basic jets.

Flown in the Icarus IV with Jeb commanding, and Bill in the co-pilot's seat. Total flight time 16hr 55m 52s!

 

A few notes on the flight:

 

• There was one quicksave, because I attempted to change the audio volume in-flight. This reset the screen to windowed, meaning I couldn't see the "accept" button to leave the settings dialog. I could, luckily, quicksave/force restart KSP/load to continue flying.
• Because most of this was flown overnight with autopilot and I took screenshots with AutoHotKey, the map-view screenshots (in this appendix album) aren't super-helpful. If you flip through them fast, you can see the planet spin around, but unfortunately it's always from the night side. To compensate, my kOS display displays the ship coordinates on-screen. Paging through the provided screenshots shows the MET and longitude stepping in half-hour and 30deg increments respectively. This should be enough data to verify that I did actually fly thrice around.
• On that note, kOS was used to give a live estimate of fuel efficiency eta = (surface speed) / (fuel burn rate) in meters traveled per fuel unit. kOS was also used to generate a data log at 15-sec intervals to plot the flight afterwards. No kOS shenanigans occurred.
• Speaking of shenanigans, all screenshots, scripts used, output data, and a craft file are available in this ZIP file on my Dropbox. Proof no kOS stuff was done other than listing data, and there's also the raw datalogs of the flight, containing lat-lng at 15 sec intervals if you need more proof. Unfortunately I accidentally turned kOS off about 200m up and on final for landing. Oops. It's not technically a complete log, but close enough.
• Pilot Assistant is the only mod used that affects flight controls, and honestly this is theoretically possible but practically impossible without it. MechJeb was used only to put a marker on the navball for landing. The actual landing was flown with pilot assistant, using keys to update setpoint. Telemachus was installed so I could check progress on my phone, but doesn't alter aerodynamics.

 

A few notes on the aircraft design:

 

• For range, I want my wing loading as low as possible. Since the big wings can hold fuel, max out on those first, and have two sets of them. Both wings are lifting, so no additional drag because of a stabilizer generating negative lift.
• All else being equal, with a turbofan (real-life too) to maximize range you want to go as fast as possible. Real commercial airplanes are limited by the drag-divergence Mach number. KSP doesn't model this in stock (FAR does!), but there's machLimit = 0.75 in the engine .cfg, and some velocity curve which I didn't actually plot. Experimentally, I found efficiency is maximized at about Mach 0.65. For a given speed, one engine is better (again, experimentally in KSP). Since I can't push enough fuel through the air at Mach 0.65 because drag, use two engines. Most of the time, I'm at around 90% throttle.
• Flying due east, I'm never going to turn. Use airbrakes as yaw spoilers to avoid having a vertical tail. I'm not totally sure this reduces drag, since I can't seem to get the debug menu to let me right-click on aerodynamic surfaces to get their aero forces, only on non-aero parts. On the other hand, it looks cool, and makes sense on a real airplane so it stays.
• The design choices did mean tuning PA was surprisingly tricky. At once point I managed to get the pitch controller to excite the bending mode of the fuselage, resulting in pretty crazy flutter on the control surfaces. Also, I didn't realize until pretty far into testing that the yaw became unstable at very low fuel and high altitude. Once tuned though, the airplane flies rock-solid.
• Tiny drag matters since I'm flying nearly three Kerbin days. Originally I flew this with medium landing gear and didn't make it; I had to land on the next continent to the west. Then I did a test flight and found that the small gear have less than half the drag! Switched gear, and made it with 266 fuel left. (of 5640). These end up being too short, so the rear ones are on the engines. Likewise, the Telemachus antenna was making a surprisingly large amount of drag, so I set just that part's drag model to none in the cfg. With the spirit of the rules, I'd argue, since it's doing nothing to actually make the airplane flyable, just means I don't have to be at my desk to know how it's going. Also, the ladder was making significant drag even when stowed, so I ditched it. Thus, no beauty shot with the pilots post-flight.
• The flight profile was actually optimized. I flew around for about an hour each at full and empty fuel, finding optimal altitude and speed, then calculated climb rate for the cruise-climb. Ideally the climb would be nonlinear, but since I set-and-forget for the entire cruise, it's linear between the two optimal points. It turns out good enough though.
• Side effect of using all these wings: when empty, the stall speed is about 30 m/s! It might be even lower still; that's just the speed I rotated when taking off for the empty-fuel test flight.
• Advanced canard winglets: originally to look cool, then adjusted their angle to trim the airplane for the cruise climb at neutral elevator (not how you'd do it on a real airplane, but I couldn't adjust wing incidence precisely enough).

 

Edited December 2, 2015 by lemon1324
Emphasize needing to restart KSP mid flight :0.0: / Update imgur embed

[Xannari Ferrows]

...

You... Have all of the respect.

Not 1, not 2, but 3!!?? How long did figuring out this design take? I'd go crazy before getting anything close to this.

This is crazy. noname is not gonna know what to say! I think you just made the history books with this.

Meanwhile, my link collider stopped working for some reason, so go ahead and grab you an Expedition badge if you want. Although there was no clear visual of the underside of your aircraft, I trust there were no offset parts down there.

[Noname117]

I've been sitting in front of my computer for the last 10 minutes just not knowing what to type. A triple circumnavigation on just basic jets?! That's been done now?! The Tiena Trophy has been won?! Holy flippin crud! I'm... still in shock to be perfectly honest.

Alright noname, get yourself together. You can type a full response to this, especially with that distracting centipede dead...

Alright, I think I can type a full, thought out response now. Firstly, congratulations lemon1324 for completing the first triple circumnavigation using just basic jets and for receiving the Tiena Trophy! This is truly an accomplishment to be proud of! Secondly, the amount of optimization which went into that aircraft is impressive. Optimizing the ascent, optimizing the drag, optimizing the engine throttle for best efficiency, optimizing the flight profile by using a mod... that's just impressive! I don't even think I'd be able to put up with doing all of that for the sake of a triple circumnavigation on basics. You really had some determination there lemon!

Anyways, I should definitely get the Tiena Trophy sent to you about now, whilst trying to think of what is next for basic circumnavigations. Faster single, double, and triple runs? Lowest altitude runs maybe? There is definitely still a future in basic jet circumnavigations, and I'm glad you can be part of it lemon!

[ExaltedDuck]

Yeah, color me impressed, too. I had a fun idea for a basic jet challenge: Zero displacement rally scoring. That is, take off, go the exact speed of kerbin's rotation, and land it 6 hours (exactly 1 Kerbin day) later. Effectively, it would be lifting off to stand still while the planet rotates below. . That would be an average surface speed of, what, 174.5? If my calculations are right...

Edited August 18, 2015 by ExaltedDuck

[MunGazer]

Yeah, color me impressed, too. I had a fun idea for a basic jet challenge: Zero displacement rally scoring. That is, take off, go the exact speed of kerbin's rotation, and land it 6 hours (exactly 1 Kerbin day) later. Effectively, it would be lifting off to stand still while the planet rotates below. . That would be an average surface speed of, what, 174.5? If my calculations are right...

That'd be interesting. If I'm not mistaken, you'd need to be on a 270 degree heading (due west) right? 3,770,000 meters / 21,600 seconds is about 174.5 so I think that's correct. Take off at midnight local time... land at midnight local time, the sun remaning constantly on the direct opposite side of Kerbin from you. Or take off at high noon and land at high noon, staying directly above your shadow on the ground. Also, chasing the sunset. Take off with the sun partially slipping behind the horizon, and keep it that way until you land. 6 hours of sunset

Edited August 18, 2015 by MunGazer

[Joshwoo70]

And... i finally finished my circumnavigation! yes!

so 1 hour 42 mins (inclusive of bob getting out)

mods used :

KER

KSP ARP (alternate resource panel)

Javascript is disabled. View full album

and a made a new page?! wat...

most of the trip is done in TOTAL DARKNESS (with the exception of landing lights)

[MunGazer]

And... i finally finished my circumnavigation! yes!

so 1 hour 42 mins (inclusive of bob getting out)

mods used :

KER

KSP ARP (alternate resource panel)

http://imgur.com/a/Oob5U

and a made a new page?! wat...

most of the trip is done in TOTAL DARKNESS (with the exception of landing lights)

Nice. Question: I'm new to mechjeb, I saw something like an "ascent path editor" in there. Is there a way to get your aircraft to automatically follow a pre-planned ascent route, like an autopilot?

[Imperial22]

Bi-Polar Entry, Bill Kerman has an unfathomed love for Kerbol-sets driving him to complete this challenge. He proposed a theory that if he traveled alongside the Kerbol-sets he would circumnavigate Kerbin, proving it is in fact spherical. Time of 1;17;23 is a bit slow, but heck we made it back, not on the runway but just 50km away on flat ground. Mechjeb installed but went unused as he is pathetic at controlling this particular craft!

Anyhow here we go:

http://imgur.com/share/a/zjNQA

[Xannari Ferrows]

And... i finally finished my circumnavigation! yes!

so 1 hour 42 mins (inclusive of bob getting out)

mods used :

KER

KSP ARP (alternate resource panel)

http://imgur.com/a/Oob5U

and a made a new page?! wat...

most of the trip is done in TOTAL DARKNESS (with the exception of landing lights)

Holy crap, that first picture nearly gave me a heart attack. To quote myself: "What the **** is this!?"

Honest to god, I think you have entered the realm of the absurd craft, but at least this time it worked.

And here'd proof of that.

http://www.datainterlock.com/Kerbal/circumnavigator%20resized.png

[Mikki]

And... i finally finished my circumnavigation! yes!

so 1 hour 42 mins (inclusive of bob getting out)...

...most of the trip is done in TOTAL DARKNESS (with the exception of landing lights)

WOOOOOWWWW! Real COOL Airplane!

...psst... your Bill is nuts...!