Jump to content

Recommended Posts

WITHDRAWN. I just realised I used a part from SXT and another from AirplanePlus by accident! :P

I will fix the mishap and apply again.

Edited by Lisias
Whoopsy...
Link to post
Share on other sites

@Lt_Duckweed: the problem isn't braking as such, but how late and hard I can do it without putting too much heat on the nose cone. As it's already at the limit during the cruise, I have to be extra super double careful when initiating the descent.

@Mars-Bound HokieJust btw, I wonder if classing the entries by mass is really meaningful. Distinguishing between airbreathers and others would probably be more sensible (though even with airbreathers, it comes down to building ever more heat-resistant planes and brute-forcing them through thicker air).

Link to post
Share on other sites

Those are some good ideas.  I did not know that about MK2 parts.

I did manage to roughly tie your time (about 26:29) with a vertical ascent and landing craft and cruise maxing out at 3250.5m/s (at ~18:48) (with comfortable heating, just didn't have the fuel and thrust to go faster).  It lands on the tail, not quite on the runway but close.  I had shock cone intakes closer to the central axis of the craft which I believe greatly reduces the heat.  For the nuclear engines, I used shock cones and 2 ncs adapters cause of their heat tolerance.

There is another optimization which is to use ion engines to get just a little more thrust.  The nuclear engines provide enough electric charge to run some of them, but I didn't use this.

 

Link to post
Share on other sites
7 hours ago, nelso said:

I had shock cone intakes closer to the central axis of the craft which I believe greatly reduces the heat.

Mine are totally safe even though far from the centerline -- the only part seriously at risk is the nosecone proper. I'm not sure how this works, but it seems if the game figures out which part is the foremost, and applies some extra heat.

You're opening new dimensions in what one can get away with in terms of dive-bombing for the Space Center. My own attempts in that regard have been less successful:
wetoolow.jpg

Seriously. This thing accelerates well, can cruise at >3200m/s on a single nuke, but there's no end of control issues. It's fine on ascent where there's only tiny nudges to keep it it on the right track, but on descent I need to actually steer it, and that's where trouble starts.

Under 24 minutes, whoopee! Only that I don't think I could make it that fast with any kind of descent that counts as controlled. Valentina actually survived (a cockpit in a cargo bay is tough), but that kind of landing won't qualify.

 

Edited by Laie
Link to post
Share on other sites

Finally!

This is my (valid! :D) entry for the Challenge, on the Thunder Division.

Craft available on Kerbal X and the (full) Mission Report is here.

This craft flew 48 Passengers and 2 Crew at 1.3Km/Sec average around Kerbin in 54 Minutes from take off to full stop on landing.

0003.jpg

A hand-full, random, screenshots:

Due the touristic nature of the flight, no maneuverer over 5G was allowed. Think on the children! :sticktongue:

0007.jpg

0008.jpg

0012.jpg

0015.jpg

0021.jpg

0028.jpg

0036.jpg

0039.jpg

0042.jpg

0046.jpg

Damn.... Too high, too fast! :( 

0047.jpg

0048.jpg

But everything ended well. By a tad... :D

0051.jpg

0052.jpg

Technically, the craft stopped on the runway =P, just a tad after the paved part of the runway... :sticktongue: - Please tell me it's ok, this is the FOURTH time I do this Challenge with this vessel! (first one, I used parts from SXT by accident, the second I blew the craft due overheat - 1375 m/s is too much! ,  the third I run out of fuel but made a marvellous landing, but empty tanks is against the rules).


— — — DISCLAIMER — — — 

TweakScale was used on Wings, Control Surfaces and Landing Gears

-- -- -- MISSING PICTURE -- -- -- 

Forgot to include the SPH Picture here!

0001.jpg

Edited by Lisias
Disclaimer; SPH Pic
Link to post
Share on other sites

@Lisias, you're entry into the Thunder division has been accepted. 

 

6 hours ago, James Kerman said:

you have been awarded thread of the month for your cool really hot challenge

I am honored, @James Kerman. You were right about the "really hot" part, since everyone who participated in the challenge did so at great risk of blowing up mid-air due to overheating. 

Link to post
Share on other sites

Here is my compact entry in the Lightning division weighing only 10.6 tons (plus 6kg of snacks :)).

KTctwIm.png

Top speed was restricted to 2,670 m/s due to the limited heat tolerance of the Mk1 parts, but acceleration and braking was high due to the low mass.
The last minute pull-out of the dive when approaching the KSC was dramatic! Total elapsed time was 27:37.

A fun fact is that the entire journey was completely automated using a kOS script. If anyone would like to use the script and tweak it for their own entry then please feel free, source code is included in the spoiler below.

Spoiler

@lazyglobal off.

// Desired cruising altitude and speed
local altitudeSetpoint is 34700.
local airspeedSetpoint is 2670.

// Save runway location so vessel can navigate towards it during final approach
local kscRunway is ship:geoposition.

// Mutable vessel state
local upsideDown is false.
local approachingKsc is false.
local stoppedAtKsc is false.

// Increase roll control threshold and response
set steeringmanager:rollcontrolanglerange to 45.
set steeringmanager:rollpid:kp to 1.6.

// Initialise PID controllers with constants tuned to a specific craft
local maxYawAngle is 4.
local minYawAngle is -4.
local latitudePid is pidloop(0.004, 0.000007, 0.0001, minYawAngle, maxYawAngle).
set latitudePid:setpoint to 0.

local maxVerticalSpeed is 450.
local minVerticalSpeed is -700.
local altitudePid is pidloop(0.25, 0.006, 1.6, minVerticalSpeed, maxVerticalSpeed).
set altitudePid:setpoint to altitudeSetpoint.

local maxPitchAngle is 17.
local minPitchAngle is -47.
local verticalSpeedPid is pidloop(0.5, 0.0002, 0.1, minPitchAngle, maxPitchAngle).
set verticalSpeedPid:setpoint to 0.

local maxThrottle is 1.
local minThrottle is -1.
local airspeedPid is pidloop(0.05, 0.006, 0.006, minThrottle, maxThrottle).
set airspeedPid:setpoint to airspeedSetpoint.

// Start
sas off.
brakes off.
stage.

//Create ascent triggers
when ship:altitude > 100 then gear off.
when ship:altitude > 20000 then upsideDown on.

// Create descent triggers once craft is far enough away from KSC
when kscRunway:distance > 200000 then {
    when kscRunway:distance < 180000 then upsideDown off.
    when kscRunway:distance < 134000 then {
        // Vesssel responds differently when lighter and in thicker atmosphere so tweak altitude PID loop constants
        set altitudePid to pidloop(0.5, 0.0002, 0.4, minVerticalSpeed, maxVerticalSpeed).
        set altitudePid:setpoint to kscRunway:terrainheight + 10.
        set airspeedPid:setpoint to 0.
    }
    when kscRunway:distance < 30000 then approachingKsc on.
    when kscRunway:distance < 6800 then {
        gear on.
        chutes on.
        brakes on.
    }
    when ship:groundspeed < 0.1 then stoppedAtKsc on.
}

// Main control loop that keeps vessel at desired heading, altitude and speed
until stoppedAtKsc {
    // Yaw
    local lateralDistance is kerbin:radius * constant:degtorad * (ship:geoposition:lat - kscRunway:lat).
    local compassDelta is latitudePid:update(time:seconds, lateralDistance).
    local desiredCompassAngle is choose 270 + compassDelta if approachingKsc else 270.
    // Pitch
    set verticalSpeedPid:setpoint to altitudePid:update(time:seconds, ship:altitude).
    local desiredPitchAngle is verticalSpeedPid:update(time:seconds, ship:verticalspeed).
    // Roll
    local desiredRollAngle is choose 180 if upsideDown else 0.
    set steering to heading(desiredCompassAngle, desiredPitchAngle, desiredRollAngle).
    // Throttle
    set throttle to throttle + airspeedPid:update(time:seconds, ship:airspeed).
}

 

Edited by ManEatingApe
Link to post
Share on other sites

I gave it a try in my high-speed anomaly-search craft, the Rapier LR At 59.471 t it it an entry for the Thunder Division. It managed the circum-navigation in respectable 53 minutes, IMHO not too bad for a craft that wasn't optimized for this. (Also, I didn't use Mechjeb or so, but I did use physics-warp.)

Some impressions: The craft in the SPH:

B6kqhFw.png

Spoiler

The wheels have left the ground, we are officially airborne. Next Stop: KSC

49HyuI1.png

I'll get you, Mun! I'll get you!

d0CGy3x.png

Coming in hot.

ss0PqGA.png

 

 

 

Not my best landing ever. But nothing broke, so I call that a win. Once the cockpit has cooled down from "instant death" to "hothothothot", our intrepid crew can leave the plane.

BKUyedm.png

The full album can be found here: https://imgur.com/a/gRpFtSp

P.S. Of course I forgot to take a screenshot with the <F3> "Flight Results" after landing.:blush: Will you take my word that the landing is legit, or do I need to do this again?

Edited by AHHans
fixed typo
Link to post
Share on other sites
On 1/1/2020 at 7:07 AM, Laie said:

You're opening new dimensions in what one can get away with in terms of dive-bombing for the Space Center

In my experience the descent is really critical and can shave 1-2 minutes from the time.  It's also possible a reverse powered descent is optimal, which would mean the backward-moving COM isn't so bad.  It is a more complex maneuver and closer to what spacex wants to do with the starship.

I've also considered using an inflatable heat shield as a nose cone.. superior heat resistance (3500k) but don't know about the aerodynamics.  Also if inflated it can't be deflated.

 

16 hours ago, ManEatingApe said:

A fun fact is that the entire journey was completely automated using a kOS script

Very nice and congrats.  I was trying an automatic takeoff for my most recent huge craft, but I couldn't get the right attack angle in the 10000-15000m range to maximize airbreathing fuel usage, while also getting above that range when I reached an optimal speed.

Link to post
Share on other sites
5 hours ago, nelso said:

I've also considered using an inflatable heat shield as a nose cone.. superior heat resistance (3500k) but don't know about the aerodynamics.  Also if inflated it can't be deflated.

I think the inflatable heat shield as a nose cone is the way to go for maximum raw speed. With a quick and dirty test craft I managed to get it up to 3,500 m/s without even breaking a sweat heat-wise, so it could be pushed even further.

Link to post
Share on other sites
On 1/5/2020 at 1:07 PM, ManEatingApe said:

I think the inflatable heat shield as a nose cone is the way to go for maximum raw speed.

I've been thinking about it myself, but don't dare to get there. For one thing, I'm not convinced that it will truly help: I think that going much faster will be a problem for all parts, and higher speed will be of little use if you can't maintain it, but have to start decelerating ever so carefully from half a world away.

The other reason is that my current plane is giving me enough trouble as it is. I don't want to have that all over again but worse.

On 1/5/2020 at 7:40 AM, nelso said:

I was trying an automatic takeoff for my most recent huge craft, but I couldn't get the right attack angle in the 10000-15000m range to maximize airbreathing fuel usage,

I'm not directly looking at pitch, but am more interested in the climb rate. Early on it's defined by thrust & drag: such-and-such a percentage of excess power may go into climbing.

Once I'm supersonic, all these considerations quickly go out of the window as I try to keep the climb rate from getting too high. Then the other scheme takes over: limit the climb rate so I don't shoot out of the breathable air.

Either way, I'm looking at and controlling for climb rate, and adjust pitch to make the desired climb rate happen.

Works great.

On 1/5/2020 at 7:40 AM, nelso said:

In my experience the descent is really critical and can shave 1-2 minutes from the time. 

Yup, in my latest attempts I remain belly-up and pitch hard while flying past the space center, doing sort of a reverse looping that brings me down over the grassland west of the runway. This also keeps me about lined up with the runway. Altogether, it saves nearly a minute compared to the weaving descent I did before.

Sadly, the plane can't pitch as hard as I'd like it to.

EDIT:
Lest I forget, @ManEatingApe: a neat small plane you got there. Surprisingly small for what it's capable of (*thumbs up*). I'm a bit surprised to not see any noticable wing slant...?

Edited by Laie
Link to post
Share on other sites

25min122.jpg

Another fully automated flight, this time clocking in at 24:32. More pictures.

This plane has given me all kind of grief... the vertical M2 parts are supposed to give me the least possible drag, but come with side effects. The plane flies well on take-off and during the cruise, but when I want to pitch up beyond 15° for braking, everything goes haywire. I've constantly had to adjust the control authority of individual elevons depending on sideslip and which way I wanted to roll.

Edited by Laie
Link to post
Share on other sites
15 hours ago, Laie said:

this time clocking in at 24:32.

Congrats, 3000m/s at 3:48 is definitely a sign this is record breaking early on, and 24:32 is by far the best.  I would give you a recommendation for that.  Think it could do even better though, perhaps by 30 seconds by not doing the loop at the end.

I also did another run with a craft that can carry 32 passengers, in under 30 minutes.  Trying to optimize takeoffs and landings, and getting decent at quick landings.  This craft has all the requirements of OP but it's a little hard to board passengers from the ladder.

 

Link to post
Share on other sites
12 hours ago, Laie said:

This plane has given me all kind of grief... the vertical M2 parts are supposed to give me the least possible drag, but come with side effects. The plane flies well on take-off and during the cruise, but when I want to pitch up beyond 15° for braking, everything goes haywire. I've constantly had to adjust the control authority of individual elevons depending on sideslip and which way I wanted to roll.

I'm toying with this concept, but concluded that any gains are marginal and easily occluded by the problems.

The Mk2 parts are lifting bodies, they add some lift, so they add the equivalent parasitic drag. Oh well, you can easily compensate it by adding a smaller wing and taking advantage of the lift (as you will suffer the drag anyway).

Link to post
Share on other sites
5 hours ago, nelso said:

Think it could do even better though, perhaps by 30 seconds by not doing the loop at the end.

Thank you for the recommendation!

I must admit that I never even considered negative pitch for braking, but I also suspect that this plane's aerodynamics wouldn't allow it in any event. It can only pitch up, and not even very hard.... As it is, the loop is the fastest way of bringing it down.

This challenge made me learn a lot about vessel control. Among other things, I (that is, my script) can now loop through -90° pitch without messing up the roll, find the runway and line up with it in a matter of seconds, and pull out of a fierce dive just in the nick of time for a soft landing. These lessons are bound to come in handy on more ordinary missions, too. I'm grateful for that, though you'd find that hard to believe if you had heard me while struggling with it.

Going forward, I think the most important factor is being able to actually pitch hard and brake in mid-air, Looney Tunes style.That's what I'm going to try next. It might even be worth giving up a little cruising speed.

15 minutes ago, Lisias said:

The Mk2 parts are lifting bodies, they add some lift, so they add the equivalent parasitic drag.

As fair as i know, they add more than their fair share of parasitic drag. My first entry to this challenge was a pretty ordinary Mk2 plane, but in order to make the fuel last around the world, I had to carefully adjust the wing slant and trim the fuel so the fuselage would fly at next to no AoA. That's why the next one was Mk3 -- but that ended being altogether too bulky and heavy.

Link to post
Share on other sites
On 1/7/2020 at 3:44 PM, Laie said:

As fair as i know, they add more than their fair share of parasitic drag. My first entry to this challenge was a pretty ordinary Mk2 plane, but in order to make the fuel last around the world, I had to carefully adjust the wing slant and trim the fuel so the fuselage would fly at next to no AoA. That's why the next one was Mk3 -- but that ended being altogether too bulky and heavy.

Interesting. I'm getting good results on the Mk2 platform for heavy planes, but I agree that using them on small ones need some tweaking.

There're two "kinds of lift" on KSP, that dark blue one, and the cyan one. The blue are the lift factor some parts have, while the cyan is the lift applied to any part that is going fast enough, the kitchen's sink included. Perhaps, and this si something to be tested yet, your problem is that two different lifts generating two different "instances" of dragging?

Link to post
Share on other sites

@Lisias: that Mk2 parts suffer from extraordinary drag is  common lore and oft-repeated advice in the Gameplay Questions subforum. It's borne out by my own experience, too. However, I cannot give you any particulars.

@OHara seems to be our current specialist on everything that's odd about KSP errordynamics.

On 9/22/2019 at 9:24 AM, OHara said:

8) As the side-face of a (non-wing) part is tilted into the airflow, the force proportional to tilt angle is all counted as drag,  where we expect drag proportional to sine-squared of the tilt angle from standard aerodynamics.
  An Mk2 spaceplane in KSP1 at 3° angle of attack feels sin3° = 5.2% (where one would expect only sin²3° = 0.3%) of the of bellyflop drag of its bottom surface.

As I understand it, this affects all parts. Mk2 ones are hit especially hard because of their flat-bottomed shape, though.

Link to post
Share on other sites
24 minutes ago, Laie said:

@Lisias: that Mk2 parts suffer from extraordinary drag is  common lore and oft-repeated advice in the Gameplay Questions subforum. It's borne out by my own experience, too. However, I cannot give you any particulars.

I'm believing you. I'm trying to figure out why things happens this way.

Long term advices, by the way, should be taken with a grain of salt. Once I was advised to use control surfaces as wings as they are better - but empirically my designs proves and proves again that it doesn't matter at all - all what matters are the lift factor and the associated drag. Once you are used to scaling things up and down, that advice doesn't works anymore and then you start to pay attention to things like thermal convection, irradiation and conduction.

Long term advices also tells me to avoid current stock wheels and landing struts as the plague - but, again, I'm getting good results on them to a point that I'm not relying on KSPWheels anymore (except by the extra parts).

Things change, people make mistakes on inferring things, and so on. Always challenge all the knowledge by experimentation. :)

Link to post
Share on other sites

Quick question for folks participating on this challenge - what version of KSP are you using?

The reason I ask is that I'm having trouble pushing a fairing past 3,150 m/s without it exploding on version 1.8.1.

Wondering if it's something silly I'm doing or some changes in the aero-model.

Link to post
Share on other sites
15 minutes ago, ManEatingApe said:

Quick question for folks participating on this challenge - what version of KSP are you using?
 

KSP 1.7.3.

Half my crafts misbehave on 1.8, so I decided to stick on 1.7 and just go to 1.8 when my current gaming had run their curse. I didn't managed to migrate any of my old savegames, and this wasn't due Add'Ons - with a new Physics engine, apparently Squad didn't managed to replicate exactly the behaviour (not to mention new glitches and oversights, terribly common on such endeavours).

Link to post
Share on other sites

This sounds like a great competition and a good test of aircraft building skills.  Based upon what I have read so far I have a few points regarding competition rules etc.  Maybe it is too late in the day for this as entries have already been made, but comments are :  

1. Proof should be a video of circumnavigation and save scumming aught not be allowed.  Pictures alone are not sufficient, cheaters are unfortunately common, though in the (generally genuinely wonderful) KSP community they are less common, they may still exist, and question will always be in people's mind otherwise.

2. As a suggestion : A waypoint at opposite end of Kerbin should be passed within 1km, this is proof of circumnaviagation and deviation from a great circle just adds time then.  Otherwise small deviations from equator can remove seconds(minutes?) from time.

2.1 Deviation could vary from the 1km if it is agreed due to the difficult of controlling hypersonic craft.  Maybe 10km?

Did the maths, its not going to make any meaningful difference a few degrees above or below equator, especially for aircraft travelling at hypersonic speed, UNLESS the competition comes down to seconds(which I doubt).  About 5 degrees from equator should make about 10s difference or so at hypersonic speeds.  HOWEVER: Variation beyond this gives a steep difference(3770 * ( 1.0 - cos( abs(latitude) ) ) ).

3. Are there no standard rules for competitions like these?  I would just assume save scumming/deviation from stock parts would be part of a standard set of rules ?

4. Deviation from stock by tweakscale(or other mods that alter the fundamental properties of parts) aught not be allowed.  Resizing a part alters the drag cube and makes a big difference, to the point it is an unfair advantage over pure stock.  Should be stock parts only with cosmetics being only allowed (part altering) mods. 

4.1 I have never explored tweakscale myself, so if the crafts it creates are "pure stock" which I define as a craft file that can be loaded/run and works the same fundamental way in a pure stock ksp, then I may revisit my ideas on this point 4.  I dont know the mod.  Seems like this is something that aught not be allowed though.

5. This is about the aircraft not the pilot so autopilots such as kOS/Mechjeb etc should be encouraged as they allow a repeatable flight profile.  This is an "optional" of course but it is nice to load/learn from others' designs.  I have recently learned the best players of this game make flight manuals for all their "completed" craft, would be nice to submit manual/flight instructions as part of craft submission(optional only just a nice thing/idea).  Maybe I am just getting carried away as usual.

My 2 cents.  I am considering entering this competition.

Gavin786

Edited by Gavin786
Difference of opinion after doing the maths.
Link to post
Share on other sites
This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...