LBP Showcase / Reviews / Recommendations #1
"Aerodynamic" properties of wheels
Archive: 88 posts
| This may have already been discovered and posted, but I've never seen it anywhere, so here I go. I believe it's well established that lbp doesn't use aerodynamics, so what I'm describing is more likely something to do with inertia. Spinning round objects do interesting things in real life: gyroscopes, precession, curve balls, and if you rev up a motorcycle and spin the back wheel during a jump, it causes the back end of the bike to drop faster than the front, so I set out to do some experiments to see if any of that applies in lbp. Did a bit of reading and found out that the curve ball and motorcycle wheel thing are aerodynamic properties (something about the properties of a fluid with relation to spinning objects: like the air molecules tend to stick to the surface a bit. If you hold your hand like a blade and run water down the back of your hand, instead of dropping right off the bottom, it will actually flow a few millimeters up the other side before pooling into a drop and dripping off and air follows that same principle). But I decided to experiment anyway. I made a platform of wood, loosely bolted in the middle to dark matter with metal wheels attached to either end. I kept it all symmetrical using the grid, of course. When neither wheel spins, it sits motionless and the same holds true when both wheels spin (opposite eachother of course). But when one wheel spins, it causes the platform to tilt: a counterclockwise wheel on the left side will cause that side to rise, and a clockwise wheel will cause it to lower... just like the motorcycle wheel. My initial experiment used pink floaty for everything, but I moved on to heavier materials with the intention of discovering whether this could be used for a vehicle stabilizer. I set up a similar rig with a wooden platform and rubber wheels. I slapped a tilt sensor on it using speed switches to control the wheels. Then I placed my sackboy on the platform and ran to one end to throw off the balance. It did work somewhat to stabilize the platform: it only tilted about 10 degrees before I couldn't make it go down any more with my weight (the platform is wood 19 small grid units across with metal wheels 6 1/2 units in diameter: each set to lift their end of the platform with the motor bolts set to a speed of 45). I did get it to drop another 5 degrees by jumping on it but that was all. The balance was such that I could make it tilt about 70 degrees with the wheels turned off. So I would define that experiment as a moderate success: it proved the principle but sackboy's relatively low weight of 50 little big pounds (lb2) (http://forums.littlebigworkshop.com/lbp/board/message?board.id=creationgen&thread.id=2803&view=by_date_ascending&page=1) compared to the rig's weight of 945 lb2 and the rig's inability to fully compensate for his weight (it wouldn't tilt very far, but it wouldn't right itself either) means that it's not a practical design for a vehicle stabilizer (smaller wheels placed further from the center would probably be as effective or moreso than the bigger ones relatively close to it, though, but I still figure it wouldn't be effective enough for a practical use). On to the curve ball. It was a complete flop. So I figured the only way to avoid any variance with gravity was to shoot it straight up. I made a quick extending piston rig with a lifter made of dissolve. The piston extended from 7 to 58 in .3 seconds but before reaching full extension triggered a switch that dissolved it. Then I motorbolted a wheel to it. With no spin at all, the wheel launched and dropped in exactly the same spot it launched from, so I figured it was accurate enough for my test. I cranked the motor to 60 and fired it.... and it still landed exactly where it took off from. So yeah, you can't throw a curve ball in lbp (well, I suppose you can by using the emitter's rotation, but that's exactly the same thing). Also, it turns out there's no real fluid dynamics either: the same curve ball launcher applied under water yielded the exact same results: the wheel drops exactly where it was launched from. So anyway, you can use a spinning wheel to provide a small amount of lift or down force to an object, but it can't do a whole lot: rockets are still more effective. | 2010-03-12 10:01:00 Author: Sehven  Posts: 2188 |
Cool, I never thought of that.  | 2010-03-12 10:41:00 Author: Jazve  Posts: 341 |
| But I decided to experiment anyway. I made a platform of wood, loosely bolted in the middle to dark matter with metal wheels attached to either end. I kept it all symmetrical using the grid, of course. When neither wheel spins, it sits motionless and the same holds true when both wheels spin (opposite eachother of course). But when one wheel spins, it causes the platform to tilt: a counterclockwise wheel on the left side will cause that side to rise, and a clockwise wheel will cause it to lower... just like the motorcycle wheel. Sounds like an application of air resistance and leverage. The movement of the platform is opposite to the movement of the outside edge of the wheel (outside being the furthers edge from your pivot), because whilst the air resistance acting on the wheel is equal all around, the resistance force acting on the outside edge will create a larger moment due to increased distance. You should probably be able to get a greater overall force by increasing the ratio of the size of the wheel : distance from platform pivot. That's what I'm assuming is causing this, correct me if I've misunderstood anything  If that is what is going on, then you could probably apply slight stabalising forces using a speed switch from your "piece-of-floaty-on-a-bolt-tilt-sensor" (see, clear namings ), as you are literally generating a rotational force. It might well be easier to control than rocket stabailsers (don't you vehicle makers have issues with them overshooting, if you aren't careful?), although as you mention it might simply be too weak to actually achieve anything.Is certainly an interesting discovery. | 2010-03-12 11:02:00 Author: rtm223  Posts: 6497 |
| I don't think it's anything to do with air... but I'm not positive. If it were because of air, then my curve ball should've worked. Then again, in a complete vacuum, pink floaty moving at any kind of speed would remain in motion until it hit something, but it doesn't--it decelerates pretty quickly. So there must be something approximating air even if it's some kind of simple decleration formula. At any rate, I think it has more to do with inertia. From my extremely limited understanding of physics, I'm guessing that a clockwise spinning wheel bolted on the right side of a medium to large mass exerts upward pressure on that side because the right side of the wheel, which is further from the objects axis or rotation is moving down while the left side, which is closer to the axis, is moving up. Since the right side is further from the axis, it exerts a greater amount of force on the object than the left side giving a greater net upward force (due to that whole equal and opposite reaction thing). I'll have to do some more experiments tomorrow using a pink floaty rig that's free floating rather than bolted at its center to dark matter. I'm guessing the axis of rotation will move somewhat opposite of the spinning wheel (I don't fully understand why--at least not well enough that I can explain it--but that seems like the logical result) so that it will spin around a point between the center of the object and the left end, or possibly to the left of the entire object. I'm also curious what would happen if both wheels spin on the free-floating object. I'm guessing the forces will cancel each other out if the wheels spin opposite of each other, but I'm not sure what will happen if both spin the same direction--I imagine it'll spin around its center. Then again, it could be simpler: two objects connected by a motor bolt will both be acted on by the bolt, spinning in opposite directions: whichever has less mass will spin more... like a helicopter without a rear stabilizer will spin opposite of the propeller. Yeah, that's probably what it is. Yeah, rockets are tricky to use for stabilizers, like you said. I forgot to mention it, but in the test I mentioned where I was trying to throw it off with my weight (the semi-successful one) I was using a tilt sensor with speed switches, just like you said. It would tilt until the wheel was near its max speed (the key was really close to the switch) and then it would hold its position there. Hopping on the edge pushed it down a bit more on the lower wheel speeds, but at 60 it would raise enough while I was in the air to offset the extra downward force. A platform with both wheels spinning opposite each other at full speed was slightly more stable than one with neither spinning, but only by a tiny margin. I could probably use some tricks to increase its effectiveness: increasing the wheel's size/mass in relation to the load (size would probably be more effective than mass due to the sides being further away from each other but both would have a significant effect) moving the wheel further from the center of gravity, and using double bolting to get a wheel with twice the speed (which would more than double its output, due to the combined mass of the two wheels). Still, I doubt anything could make it as effective as a well tuned rocket or series of rockets. | 2010-03-12 12:48:00 Author: Sehven Posts: 2188 |
| I don't think it's anything to do with air... but I'm not positive. If it were because of air, then my curve ball should've worked. No, because a curveball would rely on surface friction on the object. LBP doesn't have this. It does have air resistance based (at least partly, not entirely sure) on volume: I'm guessing that a clockwise spinning wheel bolted on the right side of a medium to large mass exerts upward pressure on that side because the right side of the wheel, which is further from the objects axis or rotation is moving down while the left side, which is closer to the axis, is moving up. This is close to what I was talking about, I'm still pretty convinced it's air resistance to the motion of the ball and not specifically inertia (although I suppose in a way it is conceptually inertia). If it was inertia / momentum-based then you would see the most effect when starting / stopping the bolt (where change in momentum actually occurs), not while it is running smoothly. The air resistance applies a force in order to slow the spinning of the ball, so that would operate based upon the speed of movement. Since the right side is further from the axis, it exerts a greater amount of force on the object than the left side giving a greater net upward force (due to that whole equal and opposite reaction thing). This I agree with. I'm not sure about air resistance, but I think one of the main factors is volume of material. If this is the case, you may actually see improved performance if you reduce weight of the spinning wheels. The same volume of material could give similar forces, but the overall mass of the platform is reduced. But increasing the size and the ratio of inner edge to outer edge distance fromt he center of the platform should definately increase the effectiveness Pure speculation though The stuff about the pink floaty rig will mess things up though, as you will apply rotational force from the motor bolt to the platform at the point of the motor bolt, where as the dark matter connected bolt only had lateral forces effectively acting on the platform. | 2010-03-12 13:05:00 Author: rtm223 Posts: 6497 |
| Another thing I'll have to try is using the tv glitch, err... I mean the "stack multiple objects in the same space and glue them together, which causes a wierd flickering effect" glitch. If I stack a dozen or so pink floaty circles, I bet the game engine will count each of their mass/volume, making a driver essentially 12 times as powerful (I doubt it would be exactly 12 times as powerful: there's probably some kind of curve or something, but eh). Combine that with a double bolt to get a speed of 120, and the thing might actually crank out enough power to do something. Of course, all that pink floaty will increase drag on the vehicle which may cancel out any stabilizing benefits it has (I would want to use it on a mech, and adding drag to a mech is very bad for balance: it results in the feet trying to walk out from under the body). Yeah, I'm not sure what'll happen when I try the free floating rig. If it's air resistance, then I would suspect the axis of rotation would move further from the wheel, but if it's the helicopter thing I mentioned in my last post, it would move closer to the motor bolt.... I think--my understanding of all this stuff is pretty fuzzy. | 2010-03-12 13:39:00 Author: Sehven Posts: 2188 |
| ...what I'm describing is more likely something to do with inertia. Sounds like an application of air resistance and leverage. Actually nothing to do with air resistance. It's actually to do with torque (http://en.wikipedia.org/wiki/Torque), which is like a rotational variant of inertia. What's happening is that as the wheel's rotation accelerates, it applies several forces on the wooden platform, which due to inertia, are opposite to the direction of the wheel's rotation at every point along its circumference. Now, as the wooden platform is fixed by a fulcrum, the principle of torque suggests that whichever of these forces is furthest from the fulcrum causes more acceleration than the others, therefore the point at which the wheel is attached to the platform will accelerate in the opposite direction from the direction that the wheel is moving at the furthest point from the fulcrum. So in this example, the platform will always rotate in the opposite direction to either wheel. There's a slight oddity, in that it should only occur when the wheel accelerates, and stop when the wheel stops accelerating. If you attach a three-way set to speed to the motor bolt, and gradually increase the speed, the effect is much less than using a three-way set to direction, and making it accelerate suddenly. But in the directional case, the force should be a one-off 'push' when you first pull the switch, and that's all, as long as the wheel has a constant rotational velocity. Edit: Looks as if you figured all this out in the time it took me to write all that. But to clarify, air resistance is not the cause. The same thing would happen in a vacuum. | 2010-03-12 13:52:00 Author: Aya042  Posts: 2870 |
| It might be that the wheel is actually applying some sort of torque to the block. That would explain the fact that the bigger the torque and the longer the lever arm (distance to the pivot) makes it tilt more. It's weird, cause I thought that materials weren't subjected to internal forces here in LBP.. Edit: ^^^ geez.. he beat me to it.... | 2010-03-12 14:03:00 Author: javi haguse  Posts: 744 |
| Edit: ^^^ geez.. he beat me to it.... That eye patch is slowing you down!!  | 2010-03-12 14:47:00 Author: jwwphotos  Posts: 11383 |
| Thanks Aya. That's kinda' what I figured, but, as I said, my understanding of physics is a little fuzzy. Most of what I know is from my own casual reading and/or discovery channel or whatnot + whatever they teach in high school. Never did go to college or take any physics classes or anything. At any rate, I'm going to do s'more experiments in a little while. I'll try increasing volume without increasing mass, and increasing mass without volume and see what I get. Also, while I was trying to fall asleep last night, I had some ideas that might make for a better control pod, so I'll have to try that as well. I also wanna' try testing for air resistance a bit more to see if it's calculated by mass, volume, or surface area (I'm betting it's just a generic resistance applied by mass, but I guess I'll see). | 2010-03-12 19:47:00 Author: Sehven Posts: 2188 |
| That's kinda' what I figured, but, as I said, my understanding of physics is a little fuzzy. Frankly, I had to look it up to remind myself. I finished the UK equivalent of high school 15 years ago, so I barely remember any of it, and, yes, that makes me an old man, but... Same applies for all the stuff I did in electronics which is relevant to LBP. Fortunately I came across this rather useful website (http://tams-www.informatik.uni-hamburg.de/applets/hades/webdemos/toc.html) with a bunch of interactive Java applets, and over time it all came back to me. | 2010-03-12 20:37:00 Author: Aya042 Posts: 2870 |
| Ok, experiment results. First the free floating pink floaty rig. This one surprised me. I unbolted it from its dark matter backplate and then turned on one of the wheels, and it spun exactly like it did when it was bolted. Like it spun around its central axis. Also, I let it run for a while and it showed no signs of slowing down, which is contrary to the expected behavior based on the whole torque thing. Wierd. It actually took about 5-10 seconds to accelerate to its full speed which it maintained constantly after that. Next, a simple mass test. Using the same pink floaty rig (but back to bolting it to a dm backplate) I cut triangular holes into the left wheel: 4 holes, making the whole thing symmetrical like a wheel with spokes. Unsurprisingly, the solid wheel easily overpowered the spoked one. Obviously it ran slower due to the opposition of the spoked wheel, but it still moved at about 6 rpms with motor bolts set to a speed of 20 (the one with the other wheel turned off ran about 10 rpms). The second stage of the mass experiment was to use the tv glitch to stack multiple wheels into the same space (who said two objects cannot occupy the same space at the same time?) I created a pink floaty wheel that was actually five pink floaty wheels and set it to oppose the single wheel on the opposite end. Just like with the spoked one, the wheel with more mass easily over powered the other and the resultant speed was about 20 rpms. My final test was to detach them all from their dark matter back plates and watch what happened. They all spun about a point which was closer to the higher massed wheel: it's center of gravity, I suppose. The "spit in Isaac Newton's face" quintuple wheel actually exceeded the mass of the entire rig, which resulted in the rig spinning around the wheel, though not perfectly of course: the axis of rotation was near the interior edge of the wheel. At this point, I was pretty happy with the results of the experiments. I didn't test for equal mass but greater volume/surface area because I really didn't feel like doing the math to calculate two different sized wheels with the same mass, or setting up a scale to weigh them on. It is my conclusion that the wheel's mass and therefore its inertia are responsible for the torque effect just as Aya said. I was somewhat surprised that the rigs spun around their center of gravity; that the spinning motion of the wheel didn't cause the axis of rotation to move. I suppose if it did, it would have been a result of friction or something, but I'm not sure. I wonder how a real world rig such as this would behave. Since there's no such thing as pink floaty, it would have to be done in space to avoid gravity and friction throwing off the results. Would it spin around the center of gravity or would centrifugal force or something cause it to spin around a different point? Would it decelerate as Aya said? Probably not due to the lack of air resistance and that whole "object in motion tends to stay in motion" thing. Anyway, I'm done with my seemingly useless experiments (I'd be interested to know if anybody uses any of this stuff to make something that actually does something). On to my control pod experiments. [edit] Stumbled upon a sorta' practical use for this. I threw together a quick three wheel drive rover to run my control pod experiments on and discovered that if I try to drive forward (to the right) during a jump, the right side of the rover will pitch up, and driving left will cause it to pitch down. With accurate enough controls and quick reflexes, a player can recover from from an end over end flip (if they do it before the rover actually pitches more than 35-40 degrees and the flip isn't too strong). Maybe it's too early to abandon this idea for use in a stabilizer. | 2010-03-12 20:54:00 Author: Sehven Posts: 2188 |
| A found this out a while ago actually when a creator brilliantly made a sack-tracking missile with two glitched-carboard floaty wheels and an mgs rocket. The proportions were unbelievably precise so that it basically floated, spinning in circles in mid-air and flying forward when it hit the angle of its target. Great stuff... | 2010-03-13 02:00:00 Author: Incinerator22  Posts: 3251 |
| I wonder how a real world rig such as this would behave. Since there's no such thing as pink floaty, it would have to be done in space to avoid gravity and friction throwing off the results. Actually, there kinda is such a thing in the real world. Bear in mind that pink floaty doesn't have zero mass, but just has a density, relative to air, such that it would stay exactly where you place it. Kinda like a balloon filled with a some ratio of, say, helium and nitrogen such that its upward force exactly counteracts gravity. When you detach it from the DM, then its implict fulcrum would simply be its center of mass. Friction is another issue, and is accounted for in LBP. There's just something wrong with the simulation when the wheels are powered by a directional switch. When you use a speed switch, the behaviour is very much like it would be in the real world, i.e. if you gradually increase its angular velocity, then the forces applied are small enough to be cancelled out by atmospheric resistance, but if you suddenly increase it, then the force is greater than that resistance can cancel, so the body will tend to rotate around the center of mass, which as you discovered, depends on the relative masses of the two attached wheels. On the subject of your loosely bolted pink floaty wheel, it's possible that you could get a similar technique to work underwater if you found a material whose density is such that it would stay exactly where you place it underwater, which is clearly not the case with pink floaty in LBP. I'm now wondering if that's the case with that material used in that shark that someone mentioned to you on LBW which looks like dissolve, but isn't. That might well be the underwater equivalent of pink floaty. | 2010-03-13 02:18:00 Author: Aya042 Posts: 2870 |
| That's pretty cool. It's actually quite a bit easier to make a sack tracking missile, though: stick a glass rail on the back and set up two emitters to spit out short sections of dark matter track for the rail to slide along. Have each track emitter angled differently (one makes the missile turn right and other makes it turn left) and hook a 180 degree sensor switch to each. To make it multiplayer compatible, set one of the sensors to require all. As for wheeled vehicles, I can't believe I never noticed this effect before. If I make my rover jump straight up from a standstill and then try to go left or right, it'll pitch so bad that it can actually land on its nose and be unable to recover before flipping over. If I jump to the right and then run the wheels left, it'll go end over end before it even hits the ground. Also, if I'm driving right and jump and then quit spinning the wheels, it'll immediately do a nose-dive: the same principle that works during acceleration works inversely during deceleration. | 2010-03-13 02:24:00 Author: Sehven Posts: 2188 |
| It's times like this when I badly, badly wish MM just gave us this information that they've stored pages and pages of on their computers. GAH! | 2010-03-13 04:50:00 Author: Incinerator22 Posts: 3251 |
| Aya, I don't think the shark was made of a special material. I suspect it simply had the right portions of material (wood and metal/stone) to give it a neutral ballast in water (of course, I've never looked into it 'cuz the shark doesn't interest me, so I don't know for sure). It's been done for submarines before. I've never bothered experimenting 'cuz I don't really like lbp water very much. Also, while your helium explanation would surely apply in real world physics, I think Mm simply turned off the "affected by gravity" variable in pink floaty, rather than setting up a realistic method. It obviously has mass, though: make a small block of pink floaty and a really big one, and then drop equal weights (such as a sackboy) on each: the smaller one will easily be pushed along (or pulled if sackboy is grabbing it) but the bigger one will go slower due to its inertia and/or drag. I believe it has roughly the same mass as peach floaty, which is why a tilt sensor can work with a bit of peach in the top half of the circle, but further testing would need to be done to decide conclusively. Enigma, I'd be curious to know how such a test comes out, but I seriously doubt a wing would generate lift in lbp. If the aerodynamic properties existed to properly act on a wing, then those same properties should have worked on my curveball launcher. Of course, there is still the principle of drag (which I have to fight with in mech building: the legs tend to walk out from under the body if it's not compensated for). If you increase the mass above the thrust line (a straight line from the rocket that's pushing the plane) it'll cause more drag, which will cause the plane to pitch up, unless its weight is stronger than the upward force caused by the drag. Pink floaty would be the best choice for such a "wing." So it should be possible to make a wierd approximation of airplane flight, but I'm pretty sure it would be tough to really get a handle on it well enough to make a workable vehicle. | 2010-03-13 06:12:00 Author: Sehven Posts: 2188 |
| To achieve this, the wings are shaped in such a way the the air traveling under the wing is traveling faster than the air above the wing. I've tried this before, but it doesn't work, presumably because the laws of aerodynamics don't apply in LBP. Aya, I don't think the shark was made of a special material. Well, it looks like dissolve, sounds like dissolve, but has the density of stone or metal. After a closer look, it looks to simply be a piece of stone or metal "cow glitched" to look like dissolve. So you're right - there is no single material which has neutral bouyancy in water. Also, while your helium explanation would surely apply in real world physics, I think Mm simply turned off the "affected by gravity" variable in pink floaty, rather than setting up a realistic method. Well, there might be a way to detemine which is the case, but since there's no way to change the amount of gravity in LBP, it could be kinda tricky. I'm leaning torwards the density theory based on the relative rates of ascent of materials in water. Of course, there is still the principle of drag (which I have to fight with in mech building: the legs tend to walk out from under the body if it's not compensated for). Not quite sure what you mean by "walk out from under the body". | 2010-03-13 14:49:00 Author: Aya042 Posts: 2870 |
| Not quite sure what you mean by "walk out from under the body". He's talking about how when a biped walks, it leans slightly forward to compensate for the tendency of the upper-body to swing backwards. In that light, I'll answer for you, and say that it is an inertial issue, not a drag issue. Again, the same would happen in a vaccuum. | 2010-03-13 14:55:00 Author: comphermc  Posts: 5338 |
| In that light, I'll answer for you, and say that it is an inertial issue, not a drag issue. Again, the same would happen in a vaccuum. I see. Yeah, that's torque again. You can cancel it out by making the legs stick out by an equal length above and below the fulcrum point (assuming they're of uniform density), or by making a short part stick out above the fulcrum, but just make it much heavier than the part which sticks out below. Might not be practical, of course, but meh... | 2010-03-13 15:40:00 Author: Aya042 Posts: 2870 |
| You can cancel it out by making the legs stick out by an equal length above and below the fulcrum point (assuming they're of uniform density), or by making a short part stick out above the fulcrum, but just make it much heavier than the part which sticks out below.) Hm... there might be something to that. By using invisible material, I can make balancers that stick up opposite the legs. I could make them theck and sackboy wouldn't run into them, and I could use the tv glitch to make them extra dense so they don't need to stick up very far. Would it be possible to acheive the same effect in a piston-cornerstone design where the leg segments don't actually rotate? Piston-cornerstone means you've got a horizontal piston connected to a piece of material that acts as a cornerstone, and a vertical piston that connects the cornerstone to the actual load: in this case, the mech's foot. It's the same principle used in 4 direction sack trackers. Yeah, it's inertia, but there does seem to be drag as well, if I used rockets to boost the upper body to the same speed the feet are moving and then turned the rockets off, it would start to lean back again, even if the walk cycle were perfectly smooth (no starting/stopping). | 2010-03-13 19:30:00 Author: Sehven Posts: 2188 |
| you guys are all nuts... but in an amazing sorta way... | 2010-03-13 19:36:00 Author: Kern  Posts: 5078 |
| Hm... there might be something to that. Bear in mind, I haven't actually tested it - I'm only basing it on Newtonian mechanics. Would it be possible to acheive the same effect in a piston-cornerstone design where the leg segments don't actually rotate? Not sure - probably depends on the exact configuration. Are you referring to you Mech Mark IIa design? Yeah, it's inertia, but there does seem to be drag as well, if I used rockets to boost the upper body to the same speed the feet are moving and then turned the rockets off, it would start to lean back again, even if the walk cycle were perfectly smooth (no starting/stopping). Again, that sounds like torque/inertia, not drag. You'll scarely notice any drag effect unless you're dealing with very small masses. | 2010-03-13 19:57:00 Author: Aya042 Posts: 2870 |
| Yeah, the mkII and the VT-04 both use corner stones. On the 4, I hid them behind the feet (originally, they were inside but stiff pistons tend to bend, so the cornerstones would hit the wall of the foot and break) but they sometimes become visible when the pistons bend. I've decided that I absolutely HATE wobble bolts for mech legs. To get them to work right, you want to calculate the angles they'll use so that it gives you a good walk cycle, but due to lbp's kind of weak connectors, the actual results never seem to match the theoretical (strength of connectors is based on what they connect, but we don't have any exact figures to use in calculations). Yeah, I s'pose it's inertia. I've tried off-setting weight in mechs with lots of peach floaty at the top, but the floaty has inertia and tends to drag on the the top of the mech. I suppose you'd get the same results on a real life robot full of helium (hydrogen would better: only one proton and no neutrons as opposed to helium's two protons and one neutron=1/3 the weight). [edit] Y'know, I once tested balloons vs peach floaty to see which was more buoyant. Can't remember which is was, but now I'm wondering if there's a difference in their inertia too, and if that difference is proportional to their buoyancy. | 2010-03-13 21:05:00 Author: Sehven Posts: 2188 |
| There is, actually, drag force in LBP. Try launching a piece of pink floaty at reasonably low speeds. It will come to rest much sooner than you might think. The problem is that it's simulated... and not necessarily simulated by real physics. | 2010-03-14 05:01:00 Author: comphermc Posts: 5338 |
| Yeah, I figure my mech balance problems are a combination of inertia/torque and drag. Since it doesn't move forward at a constant speed (it takes steps, which kind of make it lurch forward, stop, lurch, stop...) it's constantly dealing with the acceleration and stopping torque. I suppose I could try to make it more graceful, but it's easier to find other ways to overcome the problem. Finally started working on the new mech. It's going to use every freakin' trick that I know... except the one this thread is about. I'm not sure how you could test for the difference between inertia and drag. I suppose a vehicle could be built with a mast sticking up above it and a pink floaty "drag chute" behind it. Once it gets up to full speed, any leaning back that it does would be the result of drag. Not sure how to get it up to speed and still leave it susceptible to drag pulling the top back (if drag can cause it to tilt, then inertia during acceleration will easily cause it to tip). I suppose it could be fired out of an emitter so that it's already going at its cruising speed. Not sure if that would work... or work well enough to give accurate results. At any rate, I can't imagine there's any way to get exact numbers for drag and inertia, so I guess it doesn't matter that much. I guess it's just one of those things you have to feel out when building. | 2010-03-14 07:37:00 Author: Sehven Posts: 2188 |
| Yeah, the mkII and the VT-04 both use corner stones. It's difficult to tell on the VT-04, since it looks as if you've hidden the pistons which move the feet. For the Mark IIa, it looks as if you already figured out that placing the horizontal pistons about halfway up, rather than at the bottom, works better. The closer it is to the vehicle's center of mass, the less torque it will produce, tending towards zero when it's exacly along the line of the center of mass. You will still get linear inertia, which can be compensated for in a number of ways. When the horizontal piston moves forward... ...have another piston holding a similar amount mass move backwards from the same point. ...have a rocket fire backwards from the same point. ...emit a piece of invisible DM behind it, whose lifetime is equal to the time spent moving forward. The easiest method is just to ensure that the mass of the moving parts (i.e. the feet) have a much smaller mass relative to the main body. Then the force pushing back on the body becomes more marginal as the relative mass of the body increases. There is, actually, drag force in LBP... The problem is that it's simulated... and not necessarily simulated by real physics. Indeed. In real physics, drag and aerodynamics are caused by the same reason, which is that air is made up of lots of tiny particles, which act like a fluid. In LBP drag is just a simple constant force deceleration calculation, hence why a fan works in the real world, but not in LBP. | 2010-03-14 15:45:00 Author: Aya042 Posts: 2870 |
| For the Mark IIa, it looks as if you already figured out that placing the horizontal pistons about halfway up, rather than at the bottom, works better. The closer it is to the vehicle's center of mass, the less torque it will produce, tending towards zero when it's exacly along the line of the center of mass. That's why my new one (VT-05) balances so well. I started on it last night (by the time it's done, it's going to use just about every lbp trick I know) and when I started making it walk, my jaw dropped: the balance was awesome. I didn't really know why, but I wasn't about to complain. On the mkII I made the corner stones as small as I could without compromising their ability to lift the feet. On this one I'm using invisible cornerstones, so I didn't bother trying to shrink them (now that they're already built, I'm kinda' wishing I'd made them theck: same weight and piston power, but sackboy can't run into them and I could make pivoting turrets in the thack layer that can pass through the same area as the corner stones--can't believe I didn't think of that until right just now). So yeah, bigger cornerstones gives it better balance. That also explains why the VT-04 gets such bad traction: the cornerstone and the foot are in the same place, so they're putting equal (nearly) force on each other. Yeah, I'm using the short lifetime dark matter all over the mech. It's got a tilt sensor connected to the two main dm emitters (one front and one back) and then I'm in the process of rigging dm emitters for the feet (to improve traction). Aya, you wouldn't happen to have any ideas on a better control pod than the one I'm using, would you? The whole running on top of a wheel (https://lbpcentral.lbp-hub.com/index.php?t=22150-Vehicle-tools-Tilt-and-motion-sensor-and-control-pod-%28Copyable%29) thing. I'm pretty happy with it, but it does have a kind of slow response time (right around half a second between when the player pushes right and the mech starts moving, though some of that delay is in the mech's control logic so it might not be possible to improve it); if it were possible to process the player's inputs faster, it would be nice. Also my jump switch (https://lbpcentral.lbp-hub.com/index.php?t=23843-Trying-to-make-a-better-control-pod) is actually working worse in the mech than it did in the test car, so I might have to abandon that idea. | 2010-03-14 18:29:00 Author: Sehven Posts: 2188 |
| Aya, you wouldn't happen to have any ideas on a better control pod than the one I'm using, would you? I had put some thought into it yesterday, but I couldn't think of anything which would work very well for a moving control pod. Have you had any success with a jetpack-based control pod? The advantage of that would be being able to have something happen when the player moves the left stick up and down in addition to left and right. | 2010-03-14 18:46:00 Author: Aya042 Posts: 2870 |
| I briefly looked into jetpack control (meaning actual controls, not like my starfighter where you just carry the ship using a jetpack), but found that the force required to push sackboy back into the center "dead stick" zone when you let go of the stick was too hard to pull off. Also, it would mean dedicating a bigger area to the control pod, and they're (there's two in my mechs) already pretty big as it is. Besides that, I'm still hoping to build a better working jump switch and that wouldn't really work with a jetpack: you'd have to press the stick up, and with the up/down movements of the mech over uneven terrain and/or the rocking it does while walking, the up/down controls would accidently trigger too often. A jetpack controller would work great for an external control pod that doesn't get knocked around like an internal one does. I suppose what I should try is making the shaft above the wheel (where sackboy stands) narrower, so it doesn't take quite as much movement to get the wheel spinning. Doing that would decrease its tolerance to tilting, though, but it seems the jump switch won't let me get away with tilting anyway, so that's not much of a sacrifice. Maybe I should also experiment with the materials in the pod. Right now it uses a cardboard wheel with a thin layer metal gear. I tried rubber, but it seems like it had too much inertia and was difficult for sackboy to spin. It can't be grabbable, so that basically only leaves cardboard. Changing the gear to glass would cut its mass in half, so that might help make it easier to spin, but it could cause problems with its effectiveness. Guess I'll just have to try it and see. [EDIT] The closer it is to the vehicle's center of mass, the less torque it will produce, tending towards zero when it's exacly along the line of the center of mass. Just wanted to led you know that this comment just solved my traction problem. I've been experimenting with anchoring the feet with invisible emitted dark matter behind the feet (sensors detect the foot is on the ground and turn on the emitter) but the results were disastrous. But when I moved the anchors to the cornerstones, it worked great. It now has pretty good hill climbing ability. | 2010-03-14 20:47:00 Author: Sehven Posts: 2188 |
| you could have a "traditional" control pod with emitted jet pack assist for certain functions. a hybrid control system | 2010-03-14 21:12:00 Author: IStwisted  Posts: 428 |
| Actually nothing to do with air resistance. It's actually to do with torque (http://en.wikipedia.org/wiki/Torque), which is like a rotational variant of inertia. Welll... I did say leverage, which is the same principle as torque There's a slight oddity, in that it should only occur when the wheel accelerates, and stop when the wheel stops accelerating. If you attach a three-way set to speed to the motor bolt, and gradually increase the speed, the effect is much less than using a three-way set to direction, and making it accelerate suddenly. But in the directional case, the force should be a one-off 'push' when you first pull the switch, and that's all, as long as the wheel has a constant rotational velocity. But to clarify, air resistance is not the cause. The same thing would happen in a vacuum. Did we come up with an explanation for the oddity above (I did skim through the pages, but wasn't sure, seems like people just ignored it...). It seems like the accelleration test using a speed switch verifies the concept inertia is involved, but the "oddity" still suggests to me that air resistance might have some impact. I don't really know of any other force that exists in LBP that would create the continuing force over time.... | 2010-03-14 22:37:00 Author: rtm223 Posts: 6497 |
| Did we come up with an explanation for the oddity above... Personally, I didn't bother to investigate any further. | 2010-03-14 23:26:00 Author: Aya042 Posts: 2870 |
| I'm not sure if it should only occur when the wheel accelerates. The wheel is pushing against the other object constantly. That whole equal and opposite reaction thing comes into play so both objects keep spinning, with the object with lesser mass doing more of the spinning. Just as Aya said, stopping the wheel suddenly does create reverse torque. In my test vehicle, if I was in the air with my wheels spinning (driving forward) and maintaining a level pitch, letting the wheels stop suddenly caused the vehicle to pitch down pretty hard. I'm sure if I practiced enough, I could get pretty good at controlling the car's pitch in mid-air, but I wouldn't want to bother... and nobody playing a level with a vehicle like that would want to bother either. It was more of an interesting observation than something that would be practical (that vehicle was for testing the jumping control, not for testing pitch control). Also the tilt sensor/dm emitter is a much more reliable stabilizer for most of my vehicle work. A pink floaty wheel stabilizer might be useful on something really lightweight, though, like a flying vehicle. you could have a "traditional" control pod with emitted jet pack assist for certain functions. a hybrid control system That's an interesting idea, and I'll have to give it s'more thought, but the two just ideas just don't seem to be compatible. The run-on-wheel control pod requires sackboy to be constrained in a really small space, while a jetpack pod would require a bit more space. Also, a jetpack pod requires buttons on all sides, or for sackboy to be in a spring box or something, which wouldn't really fit in a run-on-wheel pod... especially since sackboy needs to be actively held down onto the wheel. | 2010-03-15 00:03:00 Author: Sehven Posts: 2188 |
| That's an interesting idea, and I'll have to give it s'more thought, but the two just ideas just don't seem to be compatible. The run-on-wheel control pod requires sackboy to be constrained in a really small space, while a jetpack pod would require a bit more space. Also, a jetpack pod requires buttons on all sides, or for sackboy to be in a spring box or something, which wouldn't really fit in a run-on-wheel pod... especially since sackboy needs to be actively held down onto the wheel. If your primary control pod is wheel based you wouldn't need a full jet pack control system. Just an assist. For example: say for your jump button. player jumps to make the mech jump. tiny jet pack on piston catches him and holds him off of the wheel until the mech lands. maybe some search light switches detect the position of the player and spin the balance wheel accordingly. mech lands and emitted X looses the jet pack I dunno... just throwing ideas out there. Concerning the oddity: I think both forces are at play. the one off push and the weaker constant force exerted by the spin of the bolt on both objects. directional activation gets you the push plus the constant force. using a speed switch effectively eliminates the one off push put not the constant force of the spin. Also, it turns out there's no real fluid dynamics either: the same curve ball launcher applied under water yielded the exact same results: the wheel drops exactly where it was launched from. Does water effect how many rotations the ball completes before landing? side note: after initially reading this thread I made a sub that uses no jets. It has hidden metal wheel right in the center with a 3 way directional switch for turning and emitted rubber for acceleration. A very simple implementation but that wheel makes for awesome control. I'll publish it if anyone wants to play with it. | 2010-03-15 15:52:00 Author: IStwisted Posts: 428 |
| ...using a speed switch effectively eliminates the one off push put not the constant force of the spin. It's actually the other way around. With the speed switch, you get the one-off push, but not the constant force, which seems consistent with basic Newtonian mechanics. Remember that F=ma, so a constant force produces a constant acceleration. Once the wheel reaches a constant velocity, it's no longer accelerating, so it shouldn't produde any forces. Now if the object rotating wasn't a uniform circle, then it could be attributed to centripetal force (http://en.wikipedia.org/wiki/Centripetal_force), although in that case, the platform would tend to vibrate rather than rotate, since the direction of the force would be continually changing direction. However, that still doesn't explain the disparity between the directional and speed switches - their behaviour should be identical when pulled from their minimum to maximum extents in the same period of time. For now, I'm happy to attribute it to a bug in the game's physics engine. | 2010-03-15 17:06:00 Author: Aya042 Posts: 2870 |
| I still don't think it classifies as an oddity. In applying the second law (F=ma) the "a" would represent the acceleration of the platform (not the wheel) and the "F" would represent the constant force generated by the spin. As long as the bolt is spinning force is being applied. The acceleration of the platform acts accordingly. As long as force is being applied the platform will spin. | 2010-03-15 17:55:00 Author: IStwisted Posts: 428 |
| As long as the bolt is spinning force is being applied. So what force is being applied? 'Cause it ain't torque, inertia or centripetal force. | 2010-03-15 18:57:00 Author: Aya042 Posts: 2870 |
| So what force is being applied? 'Cause it ain't torque, inertia or centripetal force. Air resistance. Or whatever you choose to call it. Not true air resistance but a simple dampening force on motion if you like. A force that opposes the direction of movement. Much like the way that water creates a dampening (no pun intended) force, which is what makes paddle boats move. It's just that water hasa much stronger dampening effect than air. I don't know what you were saying about the disparity between speed and directional, but from what I've seen of LBP physics, there is definitely a force that acts to oppose motion, it definitely affects spinning objects as well as those moving in a linear fashion, so would be applicable here and would, logically, apply a constant overall torque to the platform (as per my early posts in this thread) when the wheel is moving at a constant speed. I don't see any logical reason why this "air resistance" / dampening force would not be aplpicable here, when it is everywhere else in the game. Plus, taking into account dampening forces, there are definitely forces involved to keep an object moving at constant velocity (angular or linear). So just because overall acceleration is 0, it doesn't mean there aren't forces acting on a body and it certainly doesn't mean that their is no force being produced by the known driving mechanism (else my fuel economy would be greatly improved ).Essentially this whole thing is about reactive forces, those that oppose change and those that oppose motion, working in combination. | 2010-03-16 01:38:00 Author: rtm223 Posts: 6497 |
| For the control pod, I want to maintain directional control during jumps, so lifting the player off the wheel is the opposite of what I'm going for. I think my jump switch works in theory: I had it working decently on a car (though it worked better to the left than the right). I'll just need to tweak the heck out of it until it does what I want. The sub thing sounds pretty cool. Glad this discovery did something for somebody. As for the debate between Aya and IsTwisted, I'm lost. It seems to me, though, that when you have two free-floating objects connected by a motor, they are always both going to spin. Equal and opposite reaction. Just like a helicopter: the rotor spins one way and the helicopter spins (much slower due to the difference in mass) in the opposite direction (that's why helicopters have tail rotors: the tail pulls against the rotation of the helicopter). A helicopter would exhibit the same spinning motion in a zero gravity vacuum (though the tail rotor would obviously be useless in that case). Aya, the vibrating platform thing you're describing assumes that the spinning force of the off center wheel isn't being generated by a motor connecting the two objects. You could set up a zero strength bolt to connect two objects, and put rockets on the smaller object to make it spin, and that could cause the larger object to vibrate (though experience with tilt sensor experiments has shown that it would probably rotate at least a little). | 2010-03-16 03:03:00 Author: Sehven Posts: 2188 |
| So what force is being applied? 'Cause it ain't torque, inertia or centripetal force. I'm gonna have to go with all of the above. and a little of this Air resistance. For the control pod, I want to maintain directional control during jumps, so lifting the player off the wheel is the opposite of what I'm going for. maybe you could use a tethered jet pack and winch to pull the player into the wheel. kinda like a seat belt. I haven't been able to see the difference between using a speed switch and directional. actually the only difference I've seen has to do with releasing the lever. On the speed setting inverse force is applied the moment the lever bounces back past the center position but with directional you gotta pull it all the way to activate. so with directional the platform continues to spin but speed causes it to stop. I rigged up winches to do the pulling for me (to avoid the bounce back) and they both look the same. | 2010-03-16 04:34:00 Author: IStwisted Posts: 428 |
| maybe you could use a tethered jet pack and winch to pull the player into the wheel. kinda like a seat belt. The problem is I want to have my cake and eat it too. I want sackboy to be able to trigger a switch by jumping, but I also want him held down on the wheel so he keeps good traction with it when the mech is rocking. I think I've just about got it, though. I still have no idea why I get different results when running and jumping to the left or right (left is 100% successful, but right is about 60-70% which is better than the 40% I had before), but through a lot of trial and error I'm getting there, and the traction seems to be perfect, or at least good enough that there's no interruptions in the mech's walk cycle. Didn't really mean for this thread to be all about my control pod: I had another thread in the help section for that. I appreciate the suggestions, though. I haven't ruled out the idea of a hybrid control pod, but I'm going to stick with this method until I conclude that there's no way for it to succeed. | 2010-03-16 11:52:00 Author: Sehven Posts: 2188 |
| i published the sub and my test platforms. level name = resistance is tactile (copyable) Aya is right in that if you did this in the "real" world you wouldn't end up with the perpetual motion. But I propose that if you somehow got your hands on some real dark matter and real weightless motors, and could somehow nullify the 3rd demension you would get similar results. I'm convinced that the oddity is caused by fiction. | 2010-03-16 13:34:00 Author: IStwisted Posts: 428 |
| Air resistance. Or whatever you choose to call it. Not true air resistance but a simple dampening force on motion if you like. A force that opposes the direction of movement. Much like the way that water creates a dampening (no pun intended) force, which is what makes paddle boats move. It's just that water hasa much stronger dampening effect than air. Hmm. We need to clarify here whether we're takling about real physics or LittleBigPhysics. In real physics, a rotating wheel is dampened only by the friction in its parts, and drag (which for a uniform circle would only be skin friction (http://en.wikipedia.org/wiki/Parasitic_drag#Skin_friction)), so theoretically a wheel rotating on a frictionless bolt in a vacuum will spin forever. However, even with a practical example, these resistive forces are very low for a wheel of sufficient mass, hence the existence of flywheels (http://en.wikipedia.org/wiki/Flywheel). In LittleBigPhysics, it seems that there is a dampening effect applied to both linear and angular velocity, in the form of a constant force opposing the direction of movement. Now because F=ma, then it's much less noticable as you increase the mass of the wheel. e.g. for the rotational variant, if you attach a very heavy circle to a frictionless bolt, it will remain spinning for a very long time. It now occurs to me that the reason that the platform continues to rotate in the directional case, is not that there are any forces being applied when the wheel has a constant angular velocity, but simply because the bolt is frictionless. Once the initial push occurs, there's very little stopping the platform rotating indefinately, and it's only when you release the switch that the reverse torque causes the platform to stop. That also explains the discrepancy between slow and sudden acceleration, i.e. that if you slowly increase the angular velocity, then there's never enough force produced to overcome the inertia of the platform. Simply increasing the strength of the bolt seems to prove that this is all there is to it. | 2010-03-16 15:05:00 Author: Aya042 Posts: 2870 |
| interesting side notes (well, interesting to me) I changed every thing to pink floaty and deleted the bolt all together and it still rotated around it's axis as if it were still bolted. it never moved. Pink floaty wheels also created the most spin on a wooded platform by removing all weight from the ends. | 2010-03-16 15:25:00 Author: IStwisted Posts: 428 |
| Little big physics. No point in talking about real world air / water resistance here because it's irrelevant - it will just lead you to make false assumptions, like the assumption that a paddle wheel on a boat will generate more force than a solid wheel in LBP. It doesn't, in fact the opposite is true. It now occurs to me that the reason that the platform continues to rotate in the directional case, is not that there are any forces being applied when the wheel has a constant angular velocity, but simply because the bolt is frictionless. Once the initial push occurs, there's very little stopping the platform rotating indefinately, and it's only when you release the switch that the reverse torque causes the platform to stop. That also explains the discrepancy between slow and sudden acceleration, i.e. that if you slowly increase the angular velocity, then there's never enough force produced to overcome the inertia of the platform. Simply increasing the strength of the bolt seems to prove that this is all there is to it. I can debunk that with the test I ran last night. Make the entire thing out ouf pink floaty, with the center loose bolt on the platform stuck to dark matter. IF, as you say, we would expect to see the platform stopping quite quickly, as pink floaty can't move or spin for long, without the dampening force from stopping it. Further, you can tweak the center bolt to strength 10. This will stop the platform from rotating. Then start the motors running, so that the initial kick of intertia is actually prevented from having any effect. Leave these wheel spinning so that, by your definition, there should be no forces acting anywhere. Loosen the centre bolt back to 0. The result is that the platform continues to move because some force acts upon it. With no better explanation for this effect than "air resistance", and no logical reason why air resistance wouldn't have an effect, I'm sticking with the theory that both inertia and air resistance are playing a part here. Of course, as you increase the mass, inertia will have more effect (to the point that you won't see the air resistance being strong enough to create any noticable push), and as you reduce it (like using pink floaty) air resistance will have more effect. | 2010-03-16 15:29:00 Author: rtm223 Posts: 6497 |
| With no better explanation for this effect than "air resistance", and no logical reason why air resistance wouldn't have an effect, I'm sticking with the theory that both inertia and air resistance are playing a part here. I still think the biggest driving force is torque. When the pink floaty platform (without the dark matter) continued to spin in place I initially thought it disproved the air resistance but after thinking about it the air resistance would be applied uniformly and therefore counteracted uniformly. I was thinking of coating one side with glass and the other with peach floaty to cancel out the weight but that made my brain hurt. torque is applied as long as there is angular momentum. granted more is applied during acceleration but when I use a pipe wrench to fix my sink I'm applying torque for as long as I'm applying a constant force. | 2010-03-16 15:42:00 Author: IStwisted Posts: 428 |
| Poth the intertial effect and the air resistance will apply a torque to both the wheels and also the platform.... Torque isn't the source of the forces, it defines the way the forces affect a rotating body. Whichever becomes the dominant force out of the intertial effect (inertia isn't actually a force, btw) and the air reisistance will be defined by the weight of the wheels that are powered. | 2010-03-16 15:54:00 Author: rtm223 Posts: 6497 |
| Little big physics. No point in talking about real world air / water resistance here because it's irrelevant - it will just lead you to make false assumptions, like the assumption that a paddle wheel on a boat will generate more force than a solid wheel in LBP. It doesn't, in fact the opposite is true. Well, if the purpose of the discussion is to work out how LittleBigPhysics works, then it's easier to assume that LittleBigPhysics matches real physics in most cases, and then just list the counterexamples, rather than listing all the forces which apply to LittleBigPhysics. Further, you can tweak the center bolt to strength 10. This will stop the platform from rotating. Then start the motors running, so that the initial kick of intertia is actually prevented from having any effect. Leave these wheel spinning so that, by your definition, there should be no forces acting anywhere. Loosen the centre bolt back to 0. The result is that the platform continues to move because some force acts upon it. So the implication here is that LittleBigPhysics does indeed have more aerodynamics (and hydrodynamics) than I originially assumed, since there seemingly is some simulation of skin friction in place, although to what extent it matches the same effect in real physics is another matter. In fact your assertion that the shape of the wheel in the paddleboat example affects the amount of friction, suggests that the shape of the object might have some effect in these calculations, although it's clearly not realistic, otherwise the paddle wheel would generate more friction than a uniform circle. It might be as simple as the mass of the rotating object which determines the amount of skin friction. | 2010-03-16 16:28:00 Author: Aya042 Posts: 2870 |
| In fact your assertion that the shape of the wheel in the paddleboat example affects the amount of friction, suggests that the shape of the object might have some effect in these calculations, although it's clearly not realistic, otherwise the paddle wheel would generate more friction than a uniform circle. It might be as simple as the mass of the rotating object which determines the amount of skin friction. I doubt it's mass. At least not with air. If we assume that the force due to gravity is F=mg, then if the air resistance is also proportional to gravity, all objects should fall at the same rate. They don't. Also, you can very easilly see that polystyrene being slowed by air resistance and metal being slowed by air resistance are affected differently, by supporting them on 0-strength pistons. If the air resistance is proportional to mass then they would decellerate at the same rate (F=ma leads to a situation where m is cancelled out if F is proportional to m). Volume seems to be a strong contributory factor, as changing a given object to a different shape but same volume seems to allow it to fall at the same rate (so we certainly aren't seeing friction on the surface, or build up of pressure exerting a force, hence why I prefer to use the highly ambiguous term "air resistance" - which is more an effect, rather than implying any particular cause). However, that would lead us to a scenario where any object of a given density falls equally, and if I remember correctly, tiny LEDs fall slower than big ones. Although, that might be a lie. Still, volume is a big contributory factor, hence why you can distinctly see an effect of air resistance from pink floaty in my experiment, but not when you use metal. | 2010-03-16 16:51:00 Author: rtm223 Posts: 6497 |
| I think it's surface area. Firstly, what exactly do you mean by surface area? Because the surface area of a paddle wheel is greater than that of a solid wheel yet it is affected less by resistance forces. If you mean cross-sectional area, then you would expect a double-thickness paddlewheel to produce the same force as a single thickness one, and they don't. Actually, as far as I can tell it's not directly linked to any one property of the object. I was right that smaller objects fall slower, but I was wrong when I said that two objects of different shapes but the same volume and material fall the same. A 20x1 block of polystyrene falls faster than a 4x5. Volume and mass are equal and total surface area is greater on the faster object. Note: orientation of the thin piece makes no odds. I very much doubt that we can accurately quantify any of this, but it does exist. | 2010-03-16 20:21:00 Author: rtm223 Posts: 6497 |
| ... Right now it uses a cardboard wheel with a thin layer metal gear... METAL GEAR?!?! :eek: Heh.. I just realized, with the jump switch, did you try using a sensor switch? Maybe one attached to a material sackboy can pass through (if that is possible) attached to the original design (on the piston)? Or maybe just a sensor attached to the wall of the vehicle and the other part completely separate from the vehicle. I remember seeing how the tilt of the vehicle would cause the original magnetic key to activate, making it jump when it didn't need to. I may be mixed up, I am reading this thread from newest to oldest. Heh. | 2010-03-16 20:27:00 Author: RobotCrash Posts: 121 |
| A 20x1 block of polystyrene falls faster than a 20x5. Umm. Are you sure that's what you meant? | 2010-03-16 20:32:00 Author: Aya042 Posts: 2870 |
| ummmm, mayyybeeee? :blush: | 2010-03-16 20:35:00 Author: rtm223 Posts: 6497 |
| A 20x1 block of polystyrene falls faster than a 4x5. Interesting. According to the terminal velocity equation... http://upload.wikimedia.org/math/6/e/3/6e306f943fc864e7ee41a1b3a7f16172.png ...if you remove all those things which are constant (including the drag coefficient, since the orientation makes no odds), you end up with the terminal velocity being proportional to mass divided by the projected area, i.e. the amount of the surface which is pushing against the air. What if they're just using the smallest surface of the object for that figure, regardless of orientation - that would account for the 20x1 vs. 4x5 results. | 2010-03-16 20:50:00 Author: Aya042 Posts: 2870 |
| Yes, but he said that orientation doesn't matter. So, tipping the 20x1 on its side would still fall faster according to rtm's assertion. I'm not sure if it's true, but maybe they use the smallest surface as the deciding drag area? I 'unno. I'll log on now and check it out. | 2010-03-16 21:16:00 Author: comphermc Posts: 5338 |
| ...since the orientation makes no odds Yes, but he said that orientation doesn't matter. What if they're just using the smallest surface of the object for that figure, regardless of orientation... maybe they use the smallest surface as the deciding drag area? I hear an echo. | 2010-03-16 21:22:00 Author: Aya042 Posts: 2870 |
| what exactly do you mean by surface area? I think, at the time i was thinking number of corners but then realized I should prolly wait and test before just blurting out randomness. I deleted the post but apparently not in time to save face.:blush: I very much doubt that we can accurately quantify any of this, but it does exist. there is some specific formula used by the game to make the calculations so it's already been quantified. | 2010-03-16 21:27:00 Author: IStwisted Posts: 428 |
| What if they're just using the smallest surface of the object for that figure... Nope. Metal is twice as heavy as wood, so a 4x4 block of wood should fall at the same rate as a metal right-angled triangle whose two smaller sides are both length 4, but the triangle hits the ground first... so, maybe it takes all the edges into account somehow. | 2010-03-16 21:54:00 Author: Aya042 Posts: 2870 |
| Haha, sorry Aya. I was in a rush to try and catch up! :blush: | 2010-03-16 22:10:00 Author: comphermc Posts: 5338 |
| You guys are smart. I'm st00p1d. http://i7.photobucket.com/albums/y282/anon359/Durr%20Hurr%20Hurr/DurrHurrBasset1.jpg | 2010-03-16 23:36:00 Author: qrtda235566 Posts: 3664 |
| Robotcrash, I didn't really understand what you were saying here, but you can show me the next time we're in oc. The discussion is getting kind of interesting. I'd pretty much written off aerodynamic forces in lbp as being more generic: just based on mass. It's interesting that different shapes seems to react differently. I changed every thing to pink floaty and deleted the bolt all together and it still rotated around it's axis as if it were still bolted. it never moved. I already said that. Make the entire thing out ouf pink floaty, with the center loose bolt on the platform stuck to dark matter. IF, as you say, we would expect to see the platform stopping quite quickly, as pink floaty can't move or spin for long, without the dampening force from stopping it. I said that too. My original tests were with pink floaty loosely bolted to dm, and they spun forever. They also behaved exactly the same, rotating around the center, without the dm and bolt. Pink floaty wheels also created the most spin on a wooded platform by removing all weight from the ends. I had started to notice that, but hadn't pursued it much because most of my experiments just used pink floaty anyway (I only used metal wheels once in a quick experiment). I'm starting to think that spinning wheels may be more practical for stailizers than I had originally thought. I already mentioned the way that my test car could control pitch in midair by spinning its wheels. And just a few minutes ago I did a test with one of my starfighter ships: I took off the piston stabilizer and attached a tilt sensor and a pink floaty wheel. The result was that it would jitter between upright and leaning slightly: the wheel would spin to counteract the tilt, but once it was upright, the wheel would suddenly stop, which would generate torque to throw the ship backwards, at which point the wheel would kick in again... It may be possible to get better results by lowering the strength of the motor bolt (so it takes a bit longer to stop) and/or rigging logic to cause it to stop more slowly than it starts. At any rate, it kept the very back-heavy ship from falling backwards, so I consider it at least a partial success. | 2010-03-17 02:52:00 Author: Sehven Posts: 2188 |
| I said that too. Well, not really. You ran the same test, and found the same result, but inferred different conclustions from it. I reran the test to ensure that I understood the mechanics properly and that my hypothesis/conclusions from your initial test were accurate, then extended the test case to further validate my air resistance theory, which you didn't believe from your initial tests so clearly you didn't say the same thing at all | 2010-03-17 10:39:00 Author: rtm223 Posts: 6497 |
| Oh, ok. I.... almost .... kinda' said that. Or something kinda' like it. Err.... yeah. So I was just checking my thin gas thread and I'm starting to notice a pattern. Every thread I start seems to spark a debate between Rtm and Aya. That's kinda' funny. I suppose in a little while Rtm will point out the numerous threads I've started that didn't feature a debate, just to prove me wrong again. | 2010-03-17 19:50:00 Author: Sehven Posts: 2188 |
| Maybe this is slightly off topic but I figured it would be impossible to figure out the air resistance of a wheel with out first quantifying the air resistance of the material itself. since card board and poly have the same mass but obviously different textures I figured they would be the best materials to start with. I made this thing to launch rectangles of various combinations of the materials. 14192 What's weird is both of the combo pieces showed more air interaction than the solid "control" piece. The one with poly as the leading surface just went faster but pretty mush stayed straight. The one with poly as the trailing surface showed all kinds of lift potential. 14193 Although it seems that texture does interact with the air resistance, the lift in this example seems more related to corners than the material. I say this because when I changed all material to glass the one with more corners on the leading edge still showed more lift potential. BTW here's a simple gravity test with wheels attached. The back layer was just a square of the material. the middle layer had motor bolt set to 1 and the front had motor bolt set to 60. I don't know what the implications are really. I just think it's a cool pic. The fast wheels seem to make metal fall faster but not wood??? 14191 | 2010-03-18 19:05:00 Author: IStwisted Posts: 428 |
| So I figured the only way to avoid any variance with gravity was to shoot it straight up. I made a quick extending piston rig with a lifter made of dissolve. The piston extended from 7 to 58 in .3 seconds but before reaching full extension triggered a switch that dissolved it. Then I motorbolted a wheel to it. With no spin at all, the wheel launched and dropped in exactly the same spot it launched from, so I figured it was accurate enough for my test. I cranked the motor to 60 and fired it.... and it still landed exactly where it took off from. Okay, few flaws here. first of all, a baseball has the stitches on it which are often raised further than the surface of the ball creating disruptions in air current around it, causing more air resistance on one side than the other causing the ball to curve! Try your test again, but add notches (try lots of little notches or 4 big ones, best to do both though). Also, if you've ever used an emitter, you should know that you can use angular velocity to create spin on an object. so try an emitter instead too with the angular velocity cranked up to the MAX. If you A: tried this already or B: don't really feel like doing it, fine but at least try the notches thing! | 2010-03-18 21:12:00 Author: TJP12409  Posts: 10 |
| from nasa (http://www.grc.nasa.gov/WWW/K-12/airplane/ffall.html) free falling objects = no air resistance. From my tests it seems that the same holds true in LBP. Try emitting 2 balls (sans notches) horizontally with the same linear velocity. One with angular velocity maxed clockwise. The other counter-clockwise. The idea is to see how the spin counteracts the gravity. as for the wheels not effecting the path of the ball... that seems to contradict what I saw (see the bottem pic from earlier). But I was using squares dropped from dissolve and not launched golf balls. What material were you using for the wheels in that test? *As proven with your initial platform test the motorbolt would induce an opposing spin on both pieces of your assembly. | 2010-03-18 21:27:00 Author: IStwisted Posts: 428 |
| as for the wheels not effecting the path of the ball... that seems to contradict what I saw (see the bottem pic from earlier). But I was using squares dropped from dissolve and not launched golf balls. What material were you using for the wheels in that test? Um... what? There were no wheels in my golfball test. The golfballs were the wheels. The idea was to see if a fast moving, spinning object would have a curved trajectory. In real world physics, a spinning ball will curve (that's how baseball pitchers throw curve balls) due to something to do with air friction or something. My golfball launcher tested whether that was true in lbp: it wasn't. If there were a wheel attached to the golfball, then it probably would have a curved trajectory, but that's not what I was testing for. | 2010-03-18 22:10:00 Author: Sehven Posts: 2188 |
| free falling objects = no air resistance. From my tests it seems that the same holds true in LBP. No, and no. What you have there is freefalling object calculations with the assumption of no air resistance. Hence all objects fall the same in a vaccum. All objects do not fall the same in LBP. | 2010-03-18 22:22:00 Author: rtm223 Posts: 6497 |
| I was referring to the quote in TJP12409's post. It doesn't actually say who he's quoting. how did your golf ball launcher induce spin on the ball? @RTM my point was it's impossible (for me at least) to differentiate between the gravity effect and the air resistance with free falling objects. Yes, different materials fall at different rates. but some of that variance is due to mass and some must be due to air resistance. So by "the same holds true in LBP" I meant you can use the same assumption to conduct experiments in LBP maybe | 2010-03-18 22:23:00 Author: IStwisted Posts: 428 |
| I quoted the guy who made this thread in the first place! First post he made! Don't know why it didn't put his name there though... | 2010-03-18 22:32:00 Author: TJP12409 Posts: 10 |
| how did your golf ball launcher induce spin on the ball? The golf ball was motor bolted to a block of dissolve. The dissolve was connected to dark matter with a piston. It would extend rapidly, with a mag switch triggering during the extension, which killed the dissolve, allowing the ball to continue in a perfectly straight trajectory while retaining its spin. Then I waited for the ball to drop. It came down in exactly the same place it was launched from (it didn't move left or right at all). When it landed, it would roll in the direction it was spinning (it was still spinning when it landed, so it wasn't a problem with the spinning motion being messed up by the launcher mechanism). | 2010-03-18 22:52:00 Author: Sehven Posts: 2188 |
| Yes, different materials fall at different rates. but some of that variance is due to mass and some must be due to air resistance. It's all due to air resistance (drag). All objects fall at the same rate in a vacuum regardless of mass. The difficult thing is to determine how mass, volume, and surface area, determine the terminal velocity of an object in LBP, since it clearly doesn't match real physics. | 2010-03-19 00:09:00 Author: Aya042 Posts: 2870 |
| It should be easier to figure out than real physics. | 2010-03-19 00:35:00 Author: IStwisted Posts: 428 |
| the golf ball may appear dimpled, but sometimes game designers just get lazy and make the thing 'perfectly round'. also, if the piston is rigid when sending the golfball up, you are not allowing it to curve on the way up too (which it might be more likely to). so, again, try actual notches, not just LBP's simulated objects! | 2010-03-20 03:24:00 Author: TJP12409 Posts: 10 |
| What are you talking about? The golfball's dimples have nothing to do with anything: I just picked it because it was round. I could just as easily made a wheel of any non-floaty material, but it didn't matter. Why would I want notches? And if the piston let it curve on the way up, it would ruin the test. The whole point was to see if the spinning motion would cause a straight moving ball to curve, like it would in real life. In real life, it's because or friction with the air, but lbp physics don't seem to be that complicated: rather than actual friction, it's just a generic resistance force that's applied to moving objects and that causes them to eventually slow down and come to a stop. The golfball test is conclusive. If you don't believe the results, run your own test and come to your own conclusions. To accurately test for a "curveball" effect, first, you need to ensure the ball's trajectory isn't affected by gravity--a ball thrown to the side would obviously curve down because gravity pulls it down--so it needs to go straight up (in real life, you can thrown it any direction and test for a curve to either side, but we only have one axis of rotation in lbp, so that's the only one we can use to test with). Second, you need the ball to be going in a straight direction when it's launched: if it's already on a curved trajectory, then it will continue on a curved trajectory, proving nothing. | 2010-03-20 05:52:00 Author: Sehven Posts: 2188 |
| The golfball test is conclusive. If you don't believe the results, run your own test and come to your own conclusions. To accurately test for a "curveball" effect, first, you need to ensure the ball's trajectory isn't affected by gravity--a ball thrown to the side would obviously curve down because gravity pulls it down--so it needs to go straight up (in real life, you can thrown it any direction and test for a curve to either side, but we only have one axis of rotation in lbp, so that's the only one we can use to test with). Second, you need the ball to be going in a straight direction when it's launched: if it's already on a curved trajectory, then it will continue on a curved trajectory, proving nothing. I started thinking that maybe the golf ball just doesn't have enough surface friction. So I tried this: 14253 I motor bolted a dissolve circle to a DM square. Then motor bolted the golf ball to that. Set both bolts to 60 and glued a "control" golf ball to a piece of dissolve then attached a button to the dissolve components. In the pics the control ball is the one with the sticker on it. The spinning ball def shows some curve now. 14254 | 2010-03-20 14:46:00 Author: IStwisted Posts: 428 |
| Does the spinning ball still have the dissolve attached to it by a motor bolt? That will throw off the test. The idea was to test a single spinning object without having the torque between it and another piece of material. That's why I used dissolve: it could be removed and allow the ball to keep spinning and flying on a straight trajectory. | 2010-03-20 22:51:00 Author: Sehven Posts: 2188 |
| Does the spinning ball still have the dissolve attached to it by a motor bolt? That will throw off the test. The idea was to test a single spinning object without having the torque between it and another piece of material. That's why I used dissolve: it could be removed and allow the ball to keep spinning and flying on a straight trajectory. right. I recreated your test. just double bolted and without the piston. to maximize speed and minimize vibration. I just used a button to trigger the dissolve. | 2010-03-21 00:31:00 Author: IStwisted Posts: 428 |
| Without the piston? How did you launch it? | 2010-03-21 08:14:00 Author: Sehven Posts: 2188 |
| I didn't. I just dropped them. one spinning and one not. and observed the difference. | 2010-03-21 13:42:00 Author: IStwisted Posts: 428 |
| What part was double bolted? | 2010-03-21 18:47:00 Author: Sehven Posts: 2188 |
| Dissolve circle m-bolted to DM (SP=60). Golf ball m-bolted to said dissolve circle (SP=60). so It (just the golf ball) is spinning @120 when you trigger the dissolve. | 2010-03-21 21:42:00 Author: IStwisted Posts: 428 |
| The proportions were unbelievably precise so that it basically floated I remember I did that with a cardboard box and got it to float perfectly using pink floaty and orange floaty proportioned perfectly and without the grid too. | 2010-03-24 01:17:00 Author: singularik  Posts: 130 |
newtons first law: an object in motion will want to stay in motion. im actually studying this is physics right now (i only got a 72% on the test  ) any way i have a dirt bike that uses the same principle, if you go off a jump and the bikes nose is too high, i can let off the gas and make the wheel stop moving which in turn makes the bike tilt forward, however if i apply the throttle again while still in the air, the bike will tilt back up. the funny thing is if the wheel is turning at a constant speed the object it is attached to wont move. however it is only when the wheel changes its speed (accelerate or negative accelerate {deccelerate isn't actually a word} ) that it will affect the body it is attached to. also i would just like to say that i think lbp has relatively good physics, discounting bugs and the lack of aerodynamics/ hydrodynamics. also, sorry if anyone already stated this but i read the article and saw there where 5 pages of comments and didn't want to read them all so sorry. | 2010-03-31 04:42:00 Author: horwitzer  Posts: 255 |
| Yeah, there were kind of a lot of comments--it's nice that this topic got so much attention and I learned some interesting things from everybody. I'll have to do s'more tests one of these days to eliminate the acceleration factor and see what happens. What I mean is that in my experiments, I started with a block of pink floaty connected to a wheel of pink floaty and then started the spinning, so obviously, acceleration affected both bodies. What I should try next is to lock the block (glue or bolt it to dissolve), set the wheel spinning, and then trigger the dissolve. I'm betting the block will still spin opposite the wheel (equal and opposite reaction and all that), and that it will settle at the same speed of rotation as the one without dissolve, but it will take quite a bit longer to get there. It's like how a helicopter has to have the tail rotor constantly fighting the spinning motion of the copter. | 2010-03-31 08:48:00 Author: Sehven Posts: 2188 |
| I made this golf ball launcher. Metal wheel motor-bolted to dark mater. dissolve wheel motor bolted to metal. golf ball motor bolted to dissolve. A mag key on the metal wheel triggers the dissolve. The direction of spin on the third bolt definitely has an effect on the distance traveled. I placed rulers and numbered them all the way to the edge of the map. 1468814689 Tried it last night with glass and wood to see if glass had less "air resistance". Both materials flew exactly the same. | 2010-03-31 14:48:00 Author: IStwisted Posts: 428 |
| Yeah, there were kind of a lot of comments--it's nice that this topic got so much attention and I learned some interesting things from everybody. I'll have to do s'more tests one of these days to eliminate the acceleration factor and see what happens. What I mean is that in my experiments, I started with a block of pink floaty connected to a wheel of pink floaty and then started the spinning, so obviously, acceleration affected both bodies. What I should try next is to lock the block (glue or bolt it to dissolve), set the wheel spinning, and then trigger the dissolve. I'm betting the block will still spin opposite the wheel (equal and opposite reaction and all that), and that it will settle at the same speed of rotation as the one without dissolve, but it will take quite a bit longer to get there. It's like how a helicopter has to have the tail rotor constantly fighting the spinning motion of the copter. i agree with you except about the helicopter rotor part, that doesn't use inertia, rather than just a propeller pushing air. also what i mean by accelerating is if the wheel is rotating at a constant speed, the body wont move, but if the rotation speed changes, then the body will move but as soon as the wheel is turning at a constant speed again, the body will stop moving..... in theory. | 2010-04-01 02:10:00 Author: horwitzer Posts: 255 |
LBPCentral Archive Statistics
Posts: 1077139
Threads: 69970
Members: 9661
Archive-Date: 2019-01-19
Datenschutz
Aus dem Archiv wurden alle persönlichen Daten wie Name, Anschrift, Email etc. - aber auch sämtliche Inhalte wie z.B. persönliche Nachrichten - entfernt.
Die Nutzung dieser Webseite erfolgt ohne Speicherung personenbezogener Daten. Es werden keinerlei Cookies, Logs, 3rd-Party-Plugins etc. verwendet.
Die Nutzung dieser Webseite erfolgt ohne Speicherung personenbezogener Daten. Es werden keinerlei Cookies, Logs, 3rd-Party-Plugins etc. verwendet.