Brikoleur's Simple Plane Race [closed]

[Guest]

<snip>

Edited June 20, 2021 by Guest

[QF9E]

Optimized for part count only. You must be able to land reliably so I included a full 3-point landing gear and full 3-axis control surfaces. I've tried to save one more part by using a V-tail but that was not successful - the plane was almost impossible to land.

9 parts, 7256 kg, $10353

KSC runway, prior to start

Wheelstop on the island runway

Wheelstop on the KSC runway, end of race. Race time: 6 min 26.

Edited April 20, 2021 by QF9E

[Guest]

Nice job! Anyone think they can beat that overall score?

[swjr-swis]

12 hours ago, Brikoleur said:

Anyone think they can beat that overall score?

Well....

I'll enter the Brik-SPR-1 (craft file here).

• Cost: 7894
• Parts: 13
• Launch mass: 2.945t
• Performance: 5m04s  KSC-Island-KSC.
• Score: 90  (25+15+25+25)

Spoiler

Craft in SPH, cost, part count, and mass visible. Command pod reaction wheel torque has been disabled.

Craft at launch, with throttle and speed at 0, world time visible

Craft on Island Airfield runway, with speed at 0,  world time visible

Craft back on KSC runway, with speed at 0, world time visible

Showing no damage. Jeb took it up to 656 m/s and made some 16.5G turns on this run. All in a day's work.

 

The Goliath is the only engine with built-in intake - that's one less part required than any other solution. But any other Goliath based design is basically going to be a 'modification on someone else's entry' now (other wing part, other control surfaces.... that's about it), so that pretty much makes it impossible to beat @QF9E on part count.

I'll just have to be content with taking all other categories and getting the now-highest overall score.

 

P.S.: racing an aircraft on stock keyboard controls only is... not fun. Not sure why the no SAS rule, but I suspect it will prevent a lot of people from participating.

 

[JB182]

hmmmmm seems interesting, im in! and... if i had screenshot with mechjeb, i swear i dont use the mecjeb. and... SAS is not allowed?! ok!

and... can i use propeller?

[Guest]

Congratulations on taking the top spot @swjr-swis. Cool-looking craft too.

1. The distinguishing innovation in @QF9E's craft is the combination of a single Goliath with a 1-part wing that also carries fuel. I won't allow other craft built around that specific combination of characteristics, even if you swap out the FAT wing for some other fuel-containing wing. I will allow other designs with a single Goliath or single fuel-carrying wing part, however, just not that specific combination. Being first has its rewards too!
2. The purpose of the SAS off requirement is to reward/encourage designs that function better as planes. A plane that's stable but not a lawn dart, and has an elevator trim control (bind elevator deployment angle to up/down or forward/back) isn't hard to fly without SAS, you just climb, descend, and land with the trim control rather than the stick, and give it a little nudge every now and again to keep it level – that would be annoying in a long haul but for a 5-minute dash it's not much of a chore. (In fact SAS off makes certain manoeuvres easier – coordinated turns, for example, well-designed planes do those naturally just by banking and adjusting trim.)
3. @JB182 I won't disqualify you just for having MJ installed as long as you're not using it. And yes, you certainly can use propellers – in fact I'm hoping there will be some prop planes entered as I think they'll do well in the mass and cost categories at least, although they will be handicapped with part count and speed.

@QF9E you now have until 6 am UTC, Apr 22, 2021 to update your entry, since you have been overtaken by @swjr-swis. 

Note: I'm counting the time window to update your entry from the next full hour after I've approved an entry and updated the OP. That means you may have a bit more than 24 hours as I'm not at the computer 24/7. However, you may also submit updates pre-emptively, as soon as the one threatening your position is posted, rather than waiting for me to approve it. If that happens I'll adjudicate when I see them; if I disqualify the entry you responded to, then I just won't approve your update either (although of course you can re-submit it if someone else overtakes you later).

[QF9E]

3 hours ago, swjr-swis said:

so that pretty much makes it impossible to beat @QF9E on part count.

It's almost as if I put some thought in it ;).

Problem with the Goliath is that it is a bloody big engine, so my design is never going to be competitive on cost or mass. My time is pretty conservative, and I took my time lining up for the landings, so I might be able to improve my time (and just *might* be able to beat your time, although the gap is quite big) by flying more aggressively. I also didn't think of using the Goliath's thrust reverser when I did my run, so that may make landing quicker.

3 hours ago, swjr-swis said:

P.S.: racing an aircraft on stock keyboard controls only is... not fun. Not sure why the no SAS rule, but I suspect it will prevent a lot of people from participating.

I've tried to bind my flight sim gear to KSP but no luck: I could not get rid of the large dead zone that KSP by default seems to use for joystick controls, and that makes accurate flying impossible. As to SAS: I almost always fly with SAS off, as planes tend to react much more naturally. 

[Pds314]

9 hours ago, swjr-swis said:

Well....

I'll enter the Brik-SPR-1 (craft file here).

• Cost: 7894
• Parts: 13
• Launch mass: 2.945t
• Performance: 5m04s  KSC-Island-KSC.
• Score: 90  (25+15+25+25)

  Reveal hidden contents

Craft in SPH, cost, part count, and mass visible. Command pod reaction wheel torque has been disabled.

Craft at launch, with throttle and speed at 0, world time visible

Craft on Island Airfield runway, with speed at 0,  world time visible

Craft back on KSC runway, with speed at 0, world time visible

Showing no damage. Jeb took it up to 656 m/s and made some 16.5G turns on this run. All in a day's work.

 

The Goliath is the only engine with built-in intake - that's one less part required than any other solution. But any other Goliath based design is basically going to be a 'modification on someone else's entry' now (other wing part, other control surfaces.... that's about it), so that pretty much makes it impossible to beat @QF9E on part count.

I'll just have to be content with taking all other categories and getting the now-highest overall score.

 

P.S.: racing an aircraft on stock keyboard controls only is... not fun. Not sure why the no SAS rule, but I suspect it will prevent a lot of people from participating.

 

Impossible on partcount? We shall see about that.

[Pds314]

There is more than one way to combine part functions, and more than one way to eliminate the need for them.

Edited April 21, 2021 by Pds314

[Guest]

6 hours ago, Pds314 said:

There is more than one way to combine part functions, and more than one way to eliminate the need for them.

Indeed. There are other part combinations that are similarly efficient as the Goliath + wet wing combo. Some of them are competitive in mass, price, and performance as well.

[QF9E]

Since my Goliath-based plane is a dead end as far as cost and mass are concerned I decided to go a different route: rather than use a capsule I used a command chair inside a fairing, as these are much cheaper and have less mass, at the cost of additional parts. However, I succeeded in designing a plane with combined ailerons / elevators so that keeps part count down a bit. My new submission turns out to be a pain to fly and land, here's a few screenshots from one of only two successful (out of many) attempts:

Spoiler

Cost $5927, Mass 2224 kg, 13 parts

Start at the KSC runway

On the island runway. This plane has a tendency to spin during landing, this time I was fortunate enough to keep it on the runway. Some valuable time was lost turning the plane around

Back at the KSC. Time: 3 min 36 seconds. The plane carries only just enough fuel: this was a deadstick landing as the engine had flamed out by this point.

 

And in case you'd like to see me do barrel rolls on final approach to slow down, here's a video of the entire run:

 

Edited April 21, 2021 by QF9E

[Guest]

Wow, that is some seriously impressive flying. I haven't gotten close to 3'36", best I've managed so far is a hair under 4 minutes.

Congratulations on taking the lead in all categories (complexity shared 1st place). @swjr-swis, you have until 2021-04-22T20:00:00.000Z to beat @QF9E's numbers. There is still room for improvement, I'm sure...!

Edit: haha those landings were tense!

Edited April 21, 2021 by Guest

[QF9E]

12 minutes ago, Brikoleur said:

Wow, that is some seriously impressive flying.

Thanks! I found that the key to a fast time is to be able to rapidly transition from race mode to landing mode. I found that a sustained barrel roll works best for that: lots of deceleration while you keep more or less on course. I also tried the Sarajevo approach (is fun to do on the island runway: you can dip below the runway there before landing) but mach 2 is simply too fast for that. A run and break works as well but takes too long.

With a bit more practice, flying past the island, then making a sharp turn and land could also be a very good tactic for a fast time: this means you do not have to do the sharp turn on the return leg.

Edited April 21, 2021 by QF9E

[Pds314]

Well, now beating them at complexity should be much easier.

[camacju]

This was a fun challenge! I'm not used to designing tiny planes so this was something new and interesting.

I aimed to optimize for cost, mass, and part count, but not time.

Craft in VAB. 2375 funds, 9 parts, 688 kilograms mass. The parts are three elevons, a Juno, an air intake, a basic fin, a command seat, a liquid fuel tank, and a small hardpoint to act as a landing skid.

This craft is actually pretty fun to fly. I'm not used to flying without SAS so it was a bit annoying at first.

Cruise at 200 m/s. Kerbal drag is really really high so I couldn't really go any faster.

EVA parachute deployed to slow down. The landing skid has a high impact tolerance so I can come down pretty hard.

Landed at the island airfield. The craft landed at a tilt so Bill had to get out and push it back to vertical, then push the nose around.

Approaching KSC

Chute deployed. I don't need to worry about how I land here as long as it's in one piece.

Landed at 6:32

Possible part count improvements: I can't think of any. Maybe removing the basic fin and using a V tail but not sure how good that would be. Alternatively you could remove the landing skid but that would make landings a lot harder.

Possible mass and cost improvements: Switching out the fuel tank and/or changing the fuel fill level. All other parts are pretty much as optimized as they can be.

The only possible major improvement over this design would be to forego jet engines entirely for a Spider or Spark powered plane, but I'm not sure about range or fuel efficiency.

Edited April 21, 2021 by camacju

[JB182]

having touble with small planes

[QF9E]

6 hours ago, camacju said:

Craft in VAB. 2375 funds, 9 parts, 688 kilograms mass.

Impressive! And pretty hard to beat as well. I totally forgot that Kerbals can deploy their personal parachute when seated in a command chair.

[Guest]

What a terrific, creative solution @camacju! At one point I was wondering if I ought to ban 'chutes and now I'm really glad I didn't, this is just the kind of out-of-the-box thinking I was hoping to see. Where did you hide the air intake? I can't see it in the screenshots although by the part count it's clearly there.

– I'm docking you 10 seconds in your flight time because you didn't provide a screenshot stationary on the runway prior to engine-on, and the time you're displaying is mission time rather than world time, and mission time starts counting from lift-off. It doesn't affect your ranking in any case, but the rules are the rules.

@swjr-swis and @QF9E, you have until 2021-04-23T06:00:00.000Z to update your entries. Can you beat that score? This one isn't going to be easy!

Point of order: I recognise two original innovations in @camacju's entry, and will disqualify entries from other contestants that make use of them:

• Use of the pilot's personal 'chute – Explicitly released for use by other contestants by the inventor, see below
• Use of the small hardpoint as a landing skid

These (plus @QF9E's Goliath + wet wing combination) are now listed in the OP.

I will allow other landing skid designs, and of course you can still use command chairs, just not the pilot's 'chute.

5 hours ago, JB182 said:

having touble with small planes

What kind of trouble? 

Edited April 22, 2021 by Guest

[camacju]

@Brikoleur I would like to un-claim paraglider technology and let any other participant use it, in order to see what direction other people take it.

As for the air intake, it's flipped around and clipped inside the Juno engine. This produces less drag but it wasn't really necessary.

One thing to note, that I didn't actually take advantage of, is that deploying a Kerbal's chute actually reduces their drag a good amount. However the lift from the chute makes the plane have a tendency to nose up.

Edited April 22, 2021 by camacju

[Guest]

2 minutes ago, camacju said:

@Brikoleur I would like to un-claim paraglider technology and let any other participant use it, in order to see what direction other people take it.

As for the air intake, it's flipped around and clipped inside the Juno engine. This produces less drag but it wasn't really necessary.

One thing to note, that I didn't actually take advantage of, is that deploying a Kerbal's chute actually reduces their drag a good amount. However the lift from the chute makes the plane have a tendency to nose up.

Most sporting of you. I will update the OP accordingly. 

So, other contestants – feel free to one-up @camacju with a better powered paraglider!

[QF9E]

59 minutes ago, Brikoleur said:

feel free to one-up @camacju with a better powered paraglider!

I've been doing so but it has been frustrating so far. I've tried making a paraglider that uses the Kerbal's parachute as a wing, but haven't been able to land it.

This craft has only 8 parts and is controllable - sort of. However, I haven't been able to land it in one piece.

I've also tried to steer by offsetting the CoM, by having 3 fuel tanks and pumping fuel around, and that sort-of works but I don't know: is it even possible to bind fuel transfer between specific tanks to the flight controls? I've also tried making a propeller-driven paraglider but I can't get my head around propeller physics in KSP - it seems to be doing stuff at random. Any tips on that would be welcome.

[Guest]

27 minutes ago, QF9E said:

I can't get my head around propeller physics in KSP - it seems to be doing stuff at random. Any tips on that would be welcome.

The main issue with props is torque; the craft will want to twist in the opposite direction of your motor, Newton being the annoying git that he is. So it's going to be pretty hard to make any kind of prop craft that you can fly comfortably without SAS and reaction wheels if it's single-engine. KSP's flight controls just don't let you counter the torque precisely enough. 

With that caveat, props are ultimately fairly simple, although they're a bit clunky to build because of symmetry issues. Here's how to make a basic twin-prop plane; you can work your way up from there:

• Start with a twin Juno plane, where the Junos are in nacelles that combine the intake and the engine (and maybe the fuel).
• Replace the Juno nacelles with small nosecones, flat face pointing forward.
• Stick a light-duty rotor on the port nosecone (symmetry off).
• Stick a pair of small prop blades or small ducted fan blades on the rotor, taking care that they attach to the nodes on it.
• Set the prop blades to Deployed, and deployment angle to 0. Verify that the blades are "flat" (not angled), with the striped side facing forward.
• Bind rotor torque to main throttle, and blade deployment angle to the Up/Down axis.
• Copy the motor + blade group and place the copy on the starboard nosecone.
• On the starboard rotor, set the rotation direction to Counterclockwise, and the prop blade variant to Counterclockwise also.

To fly:

• Give it a bit of throttle and verify that the props rotate in opposite directions.
• Nudge the Up control to increase pitch on the prop blades. If you did everything right, the plane will start to move forward in a straight line.
  • If it moves backwards, reverse the axis control
  • If it starts to spin, verify that you have the Counterclockwise blades on the motor spinning counterclockwise
  • If it pulls to one side or the other, verify that the blades in zero position are flat and not angled at all.
• Nudge Up as it accelerates, until you reach take-off speed, then rotate and take off normally; climb and nudge Up until you've reached cruise speed.
• Nudge Down to decelerate
• When landing, nudge Down until you're at your desired landing airspeed; cut the motor only after you've touched tarmac.
• To save power, keep the throttle at the minimum needed to maintain maximum RPM, unless you're producing so much EC that you don't need to care.

The crucial part here is that prop thrust is a function of motor RPM, forward airspeed, and blade angle. The faster you go, the steeper your blade angle needs to be. This means that take-offs and landings with prop planes are  super easy: if you want to slow down from top speed (say 300 m/s) to landing speed (say 40 m/s), you just reduce the blade angle, and your plane will slow down like it hit a wall and then stay at the speed determined by the blade angle. Similarly, they accelerate from zero to take-off speed like a bat out of Hell. 

I sometimes use KALs to program in blade settings for various speeds, and if not, fine-tune my props so that minimum angle corresponds to my desired take-off/landing airspeed and maximum angle corresponds to top speed. This way you only need to watch out for blade stalls caused by increasing angle faster than the plane accelerates, or loss of airspeed near the top speed, e.g. caused by hard manoeuvring.

Edited April 22, 2021 by Guest

[Klapaucius]

K.R.A.S.S.H. proudly presents our entry, Elena.  Named for Italian author Elena Ferrante, whose books I have just been reading after watching the series "My Brilliant Friend."  (I always go with the author theme.)  https://kerbalx.com/Klapaucius/Elena-Cheap-and-Fun-plane

Anyway:

832 kg

2268 funds

11 parts

Who need landing gear? It is costly and bounces a lot.

 

I did do a few quicksaves, but did not need them. I managed to do this in one take.  Stupidly, I forgot I was racing for time, and spent 1:20 planting a flag on the Island Runway.

So total time was 8:29 - 1:20 is 7:09. 

Here is the flight (with the boring bits speeded up a lot).  The plane is a hoot now that I have trimmed it out.  In the spoiler is a video of a Barnstorming flight. I got cocky and tried to land with the UI turned off.  Not so successful...

 

Spoiler

 

Edited April 22, 2021 by Klapaucius
typo

[Guest]

How wonderfully in-character of you to forget entirely about the race and plant a flag instead. Not going to penalise you for that! Cool little plane, with which you take the second spot in the rankings. This also pushes the deadline for @swjr-swis and @QF9E to update their entries to 2021-04-23T11:00:00.000Z.

Incidentally, I'm surprised landing skids are so practical. I would have thought that taking off on them would have been too difficult but it seems to work great!

[Klapaucius]

8 minutes ago, Brikoleur said:

Incidentally, I'm surprised landing skids are so practical. I would have thought that taking off on them would have been too difficult but it seems to work great!

Well...

 

Actually, it is rather random. The thing wants to skid sideways. I found that if I give it full throttle it will go at about a 45 degree angle to the left.  It does not always get up to speed before going off the runway and burying its nose. So, I discovered that if I slowly up the power, it will turn right (why? I have no idea...)  So I let it nudge right a bit and then gun it.  If the angle is not too severe I can get to takeoff speed before a crash becomes inevitable.

As for flying: I've got it set so the action group 1 locks the front wing once it has taken off. It is too twitchy otherwise. I also, as you could see, lower the authority limiter to about 5% on the elevators once I have taken off. It does not exactly level yet. Needs a bit of very fine tuning, but it works pretty well.

The first attempt I ran out of fuel because the rear tank was not feeding engine!  Got within a few hundred meters of the KSC runway and wound up in the water.