Jump to content
  • Who needs a level runway anyway?


    JPLRepo

    Do these look familiar:

    PoFhk8C.jpgarh5M8l.jpg

     

    Runways are indeed not flat. And KSP’s runways are no different.

    In fact one might argue that KSP’s runways are too flat! But that’s not what you the fans (or the planes in KSP) wanted. So I was given the task of looking into the Runways.

     

    Well hey I thought.. Can’t be that bad. They don’t look like this do they? Or maybe they do…

     

    gfRmYzT.jpgTQAFxHK.png

     

    Let's Begin!

    First impressions were this should be a simple job. Just align the models in Unity. Why hasn’t anyone done this? Well, I can tell you, it’s not that easy at all. The issue begins when you look at the size of the runway models and small inconsistencies that seem to creep in when you import the models into the Unity engine.


    Unity has some tools that support object positioning. But the best tool for the job is the vertex snapping tool and this works best when your mesh (the actual triangles that make up the model) has a physics collider attached to it and these have straight, square edges. This is where the fun begins because the Runways in KSP are made up of separate sections which do not have square edges or lend themselves very well to having physics colliders added to them easily because of their shape.The ends and bases of them are rounded and extend into the ground and Unity changed the way physics colliders act from when these models were first created for KSP and the current versions of Unity KSP is running under now.

     

    Level 3 Runway

    Recreating the models was just not an option given the time that would take. So the process I came up with was to take the existing model, create cut down duplicates of the model meshes that were square so they could be used as colliders (but also had edges that matched the existing seams), line it all up in the modelling software and export.

     

    Next we import to Unity. Using the new meshes I began adding unity physics colliders and then vertex snapping the sections together. Once this was done the Unity objects were put together into a new Runway prefab (a term in unity which creates a prefabricated object which can then be instantiated when the game is running).

     

    Finally all the references to the new prefab must be re-generated to hook up all the code that supports destructible and upgradable facilities.

     

    What could go wrong? Well there are a lot of these references scattered throughout the game setup. You can easily miss one or two and then wonder, “Where'd my shiny runway I spent so many hours recreating go?”

    KpHIiJY.png

    Once the runway was in the right position there were still some small gaps, and while not letting my OCD get in the way too much, I began painstakingly doing fine tuning adjustments to the positioning of the runway sections.

     

    A bit later and finally everything is right and you get the new and improved runway ready for testing.

     

    KOOPZMe.png

     

    Rinse and Repeat

    … And so the process then gets repeated with the Level 2 runway.

    nnWfb9j.png

     

    Luckily I've now got a workflow happening and the Level 2 runway is done in half the time.

    With that done… Now what should we do with that lumpy bumpy Level 1 runway?

     

    But that's a story for another day.

    Edited by JPLRepo


    User Feedback

    Recommended Comments



    8 hours ago, NemesisBosseret said:

    lol i second this.   stock wheels are death on the bumpy run ways lol i used to get off the lvl 1 runway and take off from grass because it was impossible to take off from the dirt one.

    Tip: position (shift+position for nice precision without losing snap) the wheels in the way so that your planes are initially angled with nose up. Just make front wheels a bit lower and rear heels higher placed. That way upor plane starts with the nose slightly raised over horizon. If u dont screw the rest it should start on its own after 60-100 m/s speed.

    Link to comment
    Share on other sites

    I'm just curious about something that seems related. Recently I've been driving rovers around kerbin and occasionally I hit a "crack" in the surface similar to this. Most of the time the rover just bounces a little, but occasionally it acts like it just hit a brick wall and explodes. This leads to obsessive quick-saving, and I kind of like just setting the rover on autopilot with mechjeb and doing other stuff (taking care of my 1-year old, for instance). 

    Are the planet surfaces as a whole subject to the same unity issue raised here? If so, I'm not optimistic about the prospects of that being fixed, based on how hard it sounds currently. 

    Link to comment
    Share on other sites

    4 hours ago, drhay53 said:

    I'm just curious about something that seems related. Recently I've been driving rovers around kerbin and occasionally I hit a "crack" in the surface similar to this. Most of the time the rover just bounces a little, but occasionally it acts like it just hit a brick wall and explodes. This leads to obsessive quick-saving, and I kind of like just setting the rover on autopilot with mechjeb and doing other stuff (taking care of my 1-year old, for instance). 

    Are the planet surfaces as a whole subject to the same unity issue raised here? If so, I'm not optimistic about the prospects of that being fixed, based on how hard it sounds currently. 

    The planet seams are part of the PQS system which is a name given to the system which generates the planet's quads (Sections that make up the surface of the planet).
    It has been talked about in the KSP Weekly and we have been working on fixes for this issue. We know the issue stems from a change in Unity but have been unable to identify exactly what Unity changed at this stage.
    In the meantime we have several work-arounds that we have been working on and hopefully will find an acceptable one for the next release.

    Link to comment
    Share on other sites

    8 hours ago, drhay53 said:

    I'm just curious about something that seems related. Recently I've been driving rovers around kerbin and occasionally I hit a "crack" in the surface similar to this. Most of the time the rover just bounces a little, but occasionally it acts like it just hit a brick wall and explodes. This leads to obsessive quick-saving, and I kind of like just setting the rover on autopilot with mechjeb and doing other stuff (taking care of my 1-year old, for instance). 

    Are the planet surfaces as a whole subject to the same unity issue raised here? If so, I'm not optimistic about the prospects of that being fixed, based on how hard it sounds currently. 

    wheels+low mass center do half of the trick

    my science loaded Akita rover is able to travel easy on warp 3 up to 4 on flat land while any other construction from stock parts hardly manage to keep alive for a couple of minutes with warp 2.

    Link to comment
    Share on other sites

    16 hours ago, Rybikson said:

    Tip: position (shift+position for nice precision without losing snap) the wheels in the way so that your planes are initially angled with nose up. Just make front wheels a bit lower and rear heels higher placed. That way upor plane starts with the nose slightly raised over horizon. If u dont screw the rest it should start on its own after 60-100 m/s speed.

    Current plane i have made has two jet engines using the 6 man prop plane cockpit.... Rotate speed is 45 to 55 m/s.    Very forgiving flight mechanics.   Very tough plane even tho got all G and pressure difficulty switches on with max heat damage aka 110 or 120%     it takes off on any terrian.    Ive made two variants that carries either 2 or 4 rekt escape pods so i can odst some kerbals to specific locations with out landing.  Travel up to 12,000m and get up to 440m/s and go into a 75' degree dive.   Line up with drop zone and release at aprox 5k to 6k and hit my target within 50 to 100 m.   Lol.   I know how to make some sick planes but thank you.   I know about the shift trick and how to change from semitry to radial and angle lock.     The real issue wasnt the rolling up and down the runway it was the kick to the left and right bumps in the run way that would kill you.   You would flip and stall

    9 hours ago, drhay53 said:

    I'm just curious about something that seems related. Recently I've been driving rovers around kerbin and occasionally I hit a "crack" in the surface similar to this. Most of the time the rover just bounces a little, but occasionally it acts like it just hit a brick wall and explodes. This leads to obsessive quick-saving, and I kind of like just setting the rover on autopilot with mechjeb and doing other stuff (taking care of my 1-year old, for instance). 

    Are the planet surfaces as a whole subject to the same unity issue raised here? If so, I'm not optimistic about the prospects of that being fixed, based on how hard it sounds currently. 

    Sir on close scientific examination of ur statement i have to confirm that this "issue" ur experiancing has a specific name.    Its what us old ksp players call the kraken.   Ur paranoia with the quick save button is actually a sign that you are learning the game mechanics very quickly lol.      Sir i advise you never abandon ur quick saving paranoia for the kraken strikes suddenly and makes absolutely no sense when it does lol

    Link to comment
    Share on other sites

    On 2/23/2017 at 7:42 AM, Rybikson said:

    Tip: position (shift+position for nice precision without losing snap) the wheels in the way so that your planes are initially angled with nose up. Just make front wheels a bit lower and rear heels higher placed. That way upor plane starts with the nose slightly raised over horizon. If u dont screw the rest it should start on its own after 60-100 m/s speed.

    That's all great until you want to land and slam the tail of your plane into the ground because the rear wheels are so low.

    Link to comment
    Share on other sites

    5 hours ago, Alshain said:

    That's all great until you want to land and slam the tail of your plane into the ground because the rear wheels are so low.

    so u got to use some science stored in your head to put the rear wheels in such line with front wheels so any other tail part stick out of the "landing line" - and we mean the smooth 5 degree angle not 30degres - even the stock straight tail can handle this if u position the wheels far enough (preferrably on the plane/tail curve)

    moreover - if u position your wheels straight and got bad designed tail you will crash this tail at the starting takeoff when ur nose goes up, so ... protecting the tail is a separate plan, and wheel positioning for nice takeoff is another :)

    Link to comment
    Share on other sites

    If the game supports it, my suggestion for the lower level runways would be to play with the wheel grip and friction behaviour.

    All runways are flat. Maybe L1 and L2 have the odd pothole or two, but generally flat.

    The level 1 runway gives high resistance when you're rolling. On the other hand when you brake the wheels are likely to skid. It's made of gravel after all! On the other hand it still ought to give better grip than the grass. The level 3 runway lets wheels spin easily when rolling and get loads of grip when braking. L2 somewhere between.

    The lower level runways can also be shorter, of course.

    Link to comment
    Share on other sites

    11 hours ago, Rybikson said:

    so u got to use some science stored in your head to put the rear wheels in such line with front wheels so any other tail part stick out of the "landing line" - and we mean the smooth 5 degree angle not 30degres - even the stock straight tail can handle this if u position the wheels far enough (preferrably on the plane/tail curve)

    moreover - if u position your wheels straight and got bad designed tail you will crash this tail at the starting takeoff when ur nose goes up, so ... protecting the tail is a separate plan, and wheel positioning for nice takeoff is another :)

    That would be contradictory to what you said about putting the nose up while it sits on the runway.  Most real world planes sit level on the runway for very good reasons, and the tail isn't the only reason either.  Landing with the nose pointed up is going to create lift when you want downforce.

    Edited by Alshain
    Link to comment
    Share on other sites

    30 minutes ago, Alshain said:

    That would be contradictory to what you said about putting the nose up while it sits on the runway.  Most real world planes sit level on the runway for very good reasons, and the tail isn't the only reason either.  Landing with the nose pointed up is going to create lift when you want downforce.

    1. contradictory? wrongo

    http://imgur.com/a/VB1Bj
    Just try and dont argue...


    2.Most planes sit level? :D wrongo

    Every small sport plane is initially raised by the nose

    3. Big planes

    a) got enough aero-force generated by the wing profile to takeoff

    b) are also raised - just the angle is smaller

    c) use flaps, slats or other oversized control surfaces (small planes usually dont have enough place and need to reduce weight much more)

    d) lower their rear wheels during takeoff, we just see them usually in the passenger-comfy-leveled position

    4. U dont need extra downforce on landing - gravity makes it for you - just use the front-slip (u travel in same direction just increase drag by raising the nose while your speed is not enough to make all plane turn up)

    * this last bit of lift only decreases the wheel stress so its useful as well

    5. Its always safer to land the rear wheels first with raised nose - u still got the front wheels and u greatly reduce speed in this raised-nose position

    Link to comment
    Share on other sites

    6 minutes ago, Rybikson said:

    1. contradictory? wrongo

    http://imgur.com/a/VB1Bj
    Just try and dont argue...


    2.Most planes sit level? :D wrongo

    Every small sport plane is initially raised by the nose

    3. Big planes

    a) got enough aero-force generated by the wing profile to takeoff

    b) are also raised - just the angle is smaller

    c) use flaps, slats or other oversized control surfaces (small planes usually dont have enough place and need to reduce weight much more)

    d) lower their rear wheels during takeoff, we just see them usually in the passenger-comfy-leveled position

    4. U dont need extra downforce on landing - gravity makes it for you - just use the front-slip (u travel in same direction just increase drag by raising the nose while your speed is not enough to make all plane turn up)

    * this last bit of lift only decreases the wheel stress so its useful as well

    5. Its always safer to land the rear wheels first with raised nose - u still got the front wheels and u greatly reduce speed in this raised-nose position

     

    1. Yes, contradictory.   Both of those designs in that image are a danger of impacting the tail while trying to land.  Remember, you land with the nose lifted.  That tail will impact.

    2. First, small sport planes do not qualify as "most planes".  Second, not every sport plane has a lifted nose.  Taildraggers do, but that's another design and landing profile entirely.  Most tricycle landing gear will sit level.

    Edited by Alshain
    Link to comment
    Share on other sites

    Every plane generates the danger of tail crash ... if u land angled by 30 degrees.

    Im not going to continue that argument, got too many everyday experience with this stuff and tired of explaining the obvious. In this game every of my planes is preangled and i manage to sit 260tons in the mountains. Or short-takeoff the early game straight-tailers thanks to this simple fix. This would be obsolete in larger planes if the game calculated the aeroforces better on well-profiled wings (not flat plank-shaped rectangles). If u need to have the last word, feel welcome, I'm out. If u manage to pi$$ me off enough i'll make some videos ;P

    Link to comment
    Share on other sites

    1 hour ago, Alshain said:

    Most real world planes sit level on the runway for very good reasons

    Yes, the very good reason is that most real world planes are Cessna 172's :-P

    Link to comment
    Share on other sites

    Desktop%2002.25.2017%20-%2021.19.33.01.p

    Green = Good

    Red = Bad

    Regardless of how bumpy the runway is, one rule is that it must be possible to roll down it in a straight line.

    When the player simply taxis onto the grass next to the runway to take off, as I do on the level 1 runway, why bother having it?

    Link to comment
    Share on other sites

    Darn it is it really a rocket science? take the center of mass in the middle ad put rear wheels a bit higher. You can always agle your wings if u are too scary to try pre-angled planes. Most of such planes just takeoff by themselves and keep straight 1-3 degrees ascending path line due to the small frontslip.

    Link to comment
    Share on other sites

    6 hours ago, roarke said:

    The runaway are not flat by choice.

    A runway that extended all the way around the planet would be a circle. You either have the ends raised because the runway is flat but placed on a sphere meaning you roll towards the middle or you have it follow the curve of the planet, so it is not flat, but you do not roll.

    I`d rather not have a runway that requires brakes or I roll along it. A perfectly flat runway is a bad idea. Unless you define `flat` as `equidistant from the equator` in which case flat is good.

    Link to comment
    Share on other sites

    18 hours ago, Rybikson said:

    Darn it is it really a rocket science? take the center of mass in the middle ad put rear wheels a bit higher. You can always agle your wings if u are too scary to try pre-angled planes. Most of such planes just takeoff by themselves and keep straight 1-3 degrees ascending path line due to the small frontslip.

    OR:

    You could put your control surfaces far enough away from your COM to provide enough lever arm to rotate the aircraft on it's rear wheels, and not risk a tail strike on takeoff or landing.

    Then you mount your rear wheels close enough to your COM to provide a fulcrum around which the aircraft rotates while it's on the ground, allowing for the change in COM between fuel tanks being full (Takeoff), or empty (Landing).

    OR:

    Strap an SRB to your airplane and do a vertical takeoff, and use parachutes to land, not bothering to use the runway at all...

    Link to comment
    Share on other sites

    2 minutes ago, MailletC said:

    OR:

    You could put your control surfaces far enough away from your COM to provide enough lever arm to rotate the aircraft on it's rear wheels, and not risk a tail strike on takeoff or landing.

    Then you mount your rear wheels close enough to your COM to provide a fulcrum around which the aircraft rotates while it's on the ground, allowing for the change in COM between fuel tanks being full (Takeoff), or empty (Landing).

    OR:

    Strap an SRB to your airplane and do a vertical takeoff, and use parachutes to land, not bothering to use the runway at all...

    basic rules:
    small-to-zero-preangle ->no rotation, the plane takes off and lands without any extra pitch control

    - takeoff -> speed is enough to make aeroforce and u just start moving slowly up

    - landing -> speed is not enough and u just start falling slowly (while ur nose is still pointed on the horizon or above - depends on construction and ammount of speed u need to drop with frontslip)

    mid-to-hard-pre-angle (for the high thrust planes, probably with low aero surfaces), the plane is initially hard-angled and need this angle to takeoof and landing:

    - takeoff -> u start to go up fast after short runway (attack angle 10-20 degrees) but still no rotation u just follow this steep ascending path

    - landing -> u make steep fronslip with 0 thrust and finish below stall speed with nose up losing rest of speed, while ur rear wheels touch the ground, then aerobrake + inverse thrust to lose rest of speed

    finally:

    if there is no point sticking out of your front-rear wheels line and u dont make any heavy rotation, u may position wheels as u like

     

    why rotation is bad?

    1. If u raise nose fast and dont have TWR over 1.8 you lose your tangent speed making your control surfaces useless and u cant drop nose back as a result

    2. any unwanted rotation is to be correct by the control surfaces which makes u flu slower coz of increased steering drag

     

    if u cant manage fuel mass well try to use external fuel tanks on decouplers and keep the dry mass center near the full mass center

    final touch: use fuel balancer

    extra control: deploy/retract ur control surfaces to counter the unwanted pitch

    here, look at my oldtimer and try to repeat your oppinion about wheel position being bad :D this plane uses like 80m runway to takeoff and never got problems with landing - it just follows the line its angled

    http://imgur.com/a/OeLpJ
    if u need any special craft desing ill help / if u want to argue without scientifical background or experience just leave me alone ;P

    p.s. use parachutes to final slow speed after landing heavy derp or just land on the water, vertical takeoff is rather a pain

    Link to comment
    Share on other sites

    4 hours ago, Rybikson said:

    here, look at my oldtimer and try to repeat your oppinion about wheel position being bad :D this plane uses like 80m runway to takeoff and never got problems with landing - it just follows the line its angled


    http://imgur.com/a/OeLpJ

    That's a taildragger, of course it's angled back... it sits on it's tail and becomes level as it gains speed and lifts forward on the runway.  It's a completely different scenario from either the image you posted or the one MailletC posted, which are a tricycle configuration.  You don't want a tricycle configuration with the rear wheels 'shorter' than the front, it both takes off and lands completely different from a taildragger.

    Edited by Alshain
    Link to comment
    Share on other sites

    his tricycle could also takeoff angled with those rear and front wheels if he dont plan any rotation over rear wheels and balance it. Got one 240t cargo plane with tricycle wheels, preangled as permanent frontslipper. Its second config uses back ramp so the angle allows the cargo-rovers and drilling rigs to drive out smooth without ramp-bend-fail.
    u still trying to prove that 5-10degree main line angle is dangerous to the tail as a general rule? u admited u need downforce when u land ur planes so , who knows, maybe there is some alternative avionics science in the universe...

    Link to comment
    Share on other sites

    Yes. It is a danger to the tail, and yes you do need to reduce lift and have aerodynamic grip to land.  If your wings are angled back too far as a result of touchdown of the rear wheels being too high, lift will be your enemy and you will have to force the plane's nose down after setting down the rear wheels.  Even your tail draggers land level and then set the rear wheel down only after they have slowed down enough to not create lift that would make them unstable.  Of course that is assuming you don't destroy your tail in the process. 

    Watch a plane land

    https://www.youtube.com/watch?v=xrfjJ3f8KI4

    Now, if the rear wheels had been higher mounted than the front, those pilots would have to pull the nose even higher just to get the rear wheels on the ground before the front wheel.  That would cause a tail strike or at best make it hard to get the nose down.

    Edited by Alshain
    Link to comment
    Share on other sites

    never seen the plane landing movie or infographics about the aerodynamic downforce... maybe because in common sense of physics that would mean:

    - ur vessel is heavier than its mass determined only by the gravity

    - if theres downforce = there is no upforce = nothing prevents the vessel from smashing on the ground with its free fall velocity + downforce velocity

    - if u land with lowered nose = your front wheels touch the ground first taking all the weight (mass + downforce)

    - there is no graduation u just smash on the ground and u cant prevent the tail drop (coz u cant lower the nose more) and the distance between wheels multiplied by about 50% of the mass is put straight to the rear wheels when it touch the ground

    - u got no ability to make an emergency takeoff

    I prefer my way (nose up) - no sense to explain it, just reverse all of the statements from this list :D  (aaaand im not going to argue with your klingon avionics anymore, have a a nice ... landings)

    Edited by Rybikson
    Link to comment
    Share on other sites

    Lol, so you don't understand how planes work and that makes it "Klingon avionics".   I'm sorry but you clearly don't want to learn, so I'm done with you.  Enjoy your unstable planes.

    Edited by Alshain
    Link to comment
    Share on other sites

    7 hours ago, Rybikson said:

    [...]

    That is utter and total BS and Alshain done well to call you out on this.

    The lift is determinated by the shape, size, angle of attack, and wind speed (aka. speed of the plane in KSP ... because there is no wind... in RL that would be the wind speed + speed of plane) over the wing.

    If your lift is smaller than your mass (aka. total down force) you have a effective down force that makes your plane sink.
    Having a down force doesn't mean that your plane has to smash into the ground with several times its own mass. (Where did you get your number from btw? Did you pull them out of a Klingon south end? The lower one I mean)

    Only pointing your wings (and or engines) down creates additional down force since your lift the doesn't point upwards anymore.

    ------------------------------------------------

    I design my planes with very level gears. If any only a few degrees (1-2) upwards (and an aileron at the nose to give additional control). The front gear as much forward as possible and the back gear just behind all CoM (CoM and dCoM) which gives me enough force to lift the nose or press it to the ground if I need to. (the latter if I am coming in hot)

    But I am a really bad pilot ... and it has shown that a girder (they can take quite a punch) under the tail can be a real life saver in case I pull up too much (when launching) or my angle of attack is just a little too high when landing (you don't know how many engines / tails I lost with stock planes)... Though, on my last high speed plane I added a set of level 1 front wheel ... looks rather funny)

    Since I also tend to fat finger keys in high stress situation ... I usually add a set of "training" wheels under the wingtips (also helps with bumpy terrain). Since my last plane needed the ability to go over 19km ... top height is ~22km (before my engines run out of air) I need to add in an extra pod to the wingtips ... which now carry my training wheels ... (I call them training wheel, cause in normal conditions they don't touch the ground ... only when things go south they prevent important parts to kiss the ground) but in this case they are lower than usual and almost touch the ground.

    Link to comment
    Share on other sites



×
×
  • Create New...