So, you have a plane on a conveyor belt...

[Gargamel]

1 minute ago, p1t1o said:

I said "almost"

WELL, ALMOST ISN'T GOOD ENOUGH!

 

My god... it's contagious.... it's spreading to other arguments....

[Gordon Fecyk]

This could be an interesting exercise for KSP engineers though. Can we make conveyor belts or other moving platforms using Infernal Robotics?

[p1t1o]

9 minutes ago, Gordon Fecyk said:

This could be an interesting exercise for KSP engineers though. Can we make conveyor belts or other moving platforms using Infernal Robotics?

The problem is in most iterations of the problem, the conveyor needs to "think" - that is, to change its speed based on a sensor reading.

Is that possible in KSP? Can KOS handle that maybe?

[magnemoe]

8 hours ago, Gordon Fecyk said:

This could be an interesting exercise for KSP engineers though. Can we make conveyor belts or other moving platforms using Infernal Robotics?

You could make an giant rover running backward and an plane starting up on it. 
Problem is game physic, would even an car on top, work probably as some has tested that spin gravity work. 
 

[magnemoe]

8 hours ago, p1t1o said:

The problem is in most iterations of the problem, the conveyor needs to "think" - that is, to change its speed based on a sensor reading.

Is that possible in KSP? Can KOS handle that maybe?

More an problem KSP don't handle the physic accurate enough. 
Test, put an fighter jet on the belt, TWR at maximum trust is close to 1, and its no reason not to use maximum trust. 
To have the plane stand still the belt has to accelerate so fast all the engine power is used to accelerate the small wheels on an fighter jet and an bit of rolling and bearing friction but this is unlikely to be more than a few kilowatt and does not scale much with speed. 
Again where do the energy from the jet engine go? see nothing else for it to do than to spin three small wheels, they will spin up extremely fast. 

Now the treadmill has to spin up just as fast as the wheels to keep the plane in place, downside is that it mass is magnitudes larger tan the wheels, so you require magnitudes more power just to spin it up as its so much heavier than the small wheels, you also has to add force directly.
Some can calculate the power requirements for various treadmill designs but it would be idiotic high. 
If you assume the wheels will break down same rule apply to the treadmill and its again way more complex than wheels designed for high speed takeoff and bumpy landings. 
 

[Wjolcz]

I ain't reading all 17 pages of this nonsense.

Wings need airspeed to generate lift. If the conveyor belt makes the plane stand still relative to the air then it's not going to fly.

That's it. Can we now let this thread sink to the bottom and be forgotten forever?

[Gordon Fecyk]

OK how about this to solve the treadmill acceleration problem?

Build the platform to use jet engines and get roughly the same mass and TWR as the aircraft resting on it. Detach and stage both at once. The platform won't have wings so it should just stay on the ground; maybe angle the engines down or use spoilers to keep it grounded. Throttle up, then stage. Aim the camera at a stationary object apart from this contraption to observe.

[DAL59]

9 hours ago, p1t1o said:

The problem is in most iterations of the problem, the conveyor needs to "think" - that is, to change its speed based on a sensor reading.

Is that possible in KSP? Can KOS handle that maybe?

No, because you need an ideal treadmill.  

[mikegarrison]

7 minutes ago, Wjolcz said:

I ain't reading all 17 pages of this nonsense.

Wings need airspeed to generate lift. If the conveyor belt makes the plane stand still relative to the air then it's not going to fly.

That's it. Can we now let this thread sink to the bottom and be forgotten forever?

And just how would a conveyor belt make an airplane stand still relative to the air?

I agree this whole thread is nonsense, but not for the reason you state. It's an old puzzle with a known correct answer -- that the plane ignores the conveyor belt and takes off.

[DAL59]

11 minutes ago, mikegarrison said:

And just how would a conveyor belt make an airplane stand still relative to the air?

Treadmill prevents the wheels from moving forward.

[mikegarrison]

2 minutes ago, DAL59 said:

Treadmill prevents the wheels from moving forward.

How?

[Superfluous J]

21 minutes ago, mikegarrison said:

How?

That's for the guys in engineering to figure out. I'm an ideas man!

[Wjolcz]

28 minutes ago, mikegarrison said:

How?

Magic.

I really want to derail this thread and get it locked.

[mikegarrison]

Brakes are off, right? Because if the brakes are on, then you don't need a treadmill. An airplane on a regular runway with the brakes on is moving nowhere even at full engine thrust.

I assume you have "ideal" wheels that spin completely freely. Obviously these are not possible in the real world, but neither is a runway-sized treadmill that can move at 200+ kts. So ideal wheels.

Now sit the airplane on the treadmill and turn the treadmill on. What happens? The airplane just sits there. No matter how fast the treadmill runs, the wheels spin the same speed and no force is transmitted to the plane.

Thus you see that the treadmill can not transmit any force to the plane. Period. So it doesn't matter how fast or slow the treadmill is turning. The only force being transmitted to the plane comes from the engines. Therefore the plane accelerates as normal, and takes off when the airspeed gets up to Vr.

[Reactordrone]

1 hour ago, Wjolcz said:

I ain't reading all 17 pages of this nonsense.

Wings need airspeed to generate lift. If the conveyor belt makes the plane stand still relative to the air then it's not going to fly.

That's it. Can we now let this thread sink to the bottom and be forgotten forever?

That's the point of the thought experiment. Can the belt make the plane stand still despite the plane's propulsion being independent of the belt. A remote controlled car can be held in place because the wheels drive the car and that is being countered by the belt. If the drive is external, say a person standing in front of the conveyor belt pulling the  car with a rope, can the belt apply enough force to overcome the force applied by the rope (or the aircraft's engines).

[55delta]

2 hours ago, Wjolcz said:

That's it. Can we now let this thread sink to the bottom and be forgotten forever?

Hahaha...You can't just kill a thread by demanding it die.

Threads die when no one posts, when everyone has lost interest. Apparently, even though we know the plane takes off and the majority agree why it takes off, we aren't done batting the subject around some more.

Just be patient and so more worthy intellectual challenge will come along to draw all the posters away.

[Brotoro]

11 minutes ago, 55delta said:

Hahaha...You can't just kill a thread by demanding it die.

Threads die when no one posts, when everyone has lost interest. Apparently, even though we know the plane takes off and the majority agree why it takes off, we aren't done batting the subject around some more.

Just be patient and so more worthy intellectual challenge will come along to draw all the posters away.

But what if he demands that it's an 'ideal' thread that always sinks faster than new messages can cause it to bump back up? Huh? Huh?

[55delta]

1 hour ago, Brotoro said:

But what if he demands that it's an 'ideal' thread that always sinks faster than new messages can cause it to bump back up? Huh? Huh?

Then logic dictates that he would bump up other threads instead of posting in this thread. This follows the rule that the thread with the latest post is on the top of the sub-forum.

But like the plane taking off, this concept is not always thought out on an individual level.

[Gordon Fecyk]

All right...

I just finished filming several takes of the same experiment using a treadmill and aircraft I constructed in the Spaceplane Hangar. I used rockets instead of jets to ensure a consistent thrust-to-weight ratio and to take "smart treadmills" out of the equation. Both craft have near-identical TWR and acceleration once staged and separated.

Praise the mighty Cubic Octagonal Strut for enabling this whacky construction. Other considerations include using Ferram Aerospace to ensure realistic drag and lift.

I'll make the craft file available. The current version of the Kerbal X download only uses stock parts and tries to lighten the aircraft mass as much as possible. I'll edit this post as soon as the video is up on YouTube and the craft is available on Kerbal X or some place.

 

To address @Gargamel and potentially others, I published an updated version of the craft to Kerbal X that can, barely, lift off at the end of the treadmill. This was a pretty lousy aircraft to use for the experiment, as the stall speed exceeded 65 m/s and its (v4) treadmill departure speed was 50 m/s. I'd need to try better wings or maybe launch the craft on a world with thicker air.

 

Edited June 1, 2018 by Gordon Fecyk
Added follow-up video

[Gargamel]

55 minutes ago, Gordon Fecyk said:

All right...

Love it!

Glad you did the real speed test, and the stationary test, as the slow speed was messing with my perception of the plane v treadmill. 

Now we have some practical examples, and a virtual one.    But cue the "But KSP physics..." and the "The treadmill wasn't long enough, it didn't take flight!" arguments!

I would imagine with hangar extender, and flipping them both around, you could get flight while on the treadmill. 

[p1t1o]

Did...did this just start all over again? Are we back to the beginning?

OH MY GOD WE'RE ON A CONVEYOR BELT

[DAL59]

[razark]

4 hours ago, p1t1o said:

OH MY GOD WE'RE ON A CONVEYOR BELT

And I think we have quite thoroughly demonstrated that a thread on a conveyor belt both takes off, and yet goes nowhere.

[Gordon Fecyk]

We have hundreds or maybe thousands of people in this forum with access to a very versatile and inexpensive physics simulator. Yet we have this three-year-old thread and, as of today, only two people who have attempted to reproduce the problem in said physics simulator.

I find this disturbing.

With that, here's a shout to @Shpaget for being the other brave fool. And here's an attempt that uses a very large part. Finally, a use for Ubiozur Welding.

 

 

Edited June 2, 2018 by Gordon Fecyk
It's 'Ubiozur.'

[Mad Rocket Scientist]

51 minutes ago, Gordon Fecyk said:

We have hundreds or maybe thousands of people in this forum with access to a very versatile and inexpensive physics simulator. Yet we have this three-year-old thread and, as of today, only two people who have attempted to reproduce the problem in said physics simulator.

I find this disturbing.

With that, here's a shout to @Shpaget for being the other brave fool. And here's an attempt that uses a very large part. Finally, a use for Ubizor Welding.

[...]

I hate to do this, but the gravity of Kerbin (or any planet in KSP) pulls aircraft towards the center of the runway, as it is perfectly flat. That may be the rolling visible with the unpowered aircraft. It's not super likely, but starting the runway in the other direction (or even better, on the properly curved area the KSC is on) would eliminate all possible errors.