Wikipedia:Reference desk/Archives/Science/2019 November 20

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

November 20

A horizontal force is applied to the top of a stationary ball on a frictionless horizontal surface

F=ma so the ball will move horizontally with acceleration a, but since the force is applied parallel to a line through the ball's center of mass, wouldn't it confer a moment and thus start the ball rolling? But I thought balls only roll with static friction? 104.162.197.70 (talk) 01:25, 20 November 2019 (UTC)

The force would start the ball moving and would start it rotating in the same direction as it would if it was rolling; but as you say, with no friction it wouldn't actually be rolling. It might be rotating faster or slower; I don't know how to determine that. --76.69.116.4 (talk) 02:44, 20 November 2019 (UTC)

Why would the ball rotate if the force is straight through the center and the surface is frictionless? How would it even know which direction to rotate in? 173.228.123.207 (talk) 03:06, 20 November 2019 (UTC)

If there is a net force acting on an object, it will accelerate. If there is a net torque on an object, it will rotationally accelerate. A line of force passing straight through the center of mass would not lead to acceleration, but the OP had specified a line of force parallel to the center of mass. The first answer contains an error, though - the ball could be rotating in either direction. At least when I was learning physics, this was referred to as "slipping" to distinguish it from rolling. Someguy1221 (talk) 03:15, 20 November 2019 (UTC)

The force isn't straight through the center as 173.228 says; it's on one edge. The ball could not rotate in either direction as Someguy says, because the force determines which way it moves. --76.69.116.4 (talk) 08:05, 20 November 2019 (UTC)

"the" force is inaccurate. I guess we consider the force of cohesion will keep the ball solid instead of breaking apart, but the main thing to NOT forget is the force of reaction (to weight and all other forces) that provide for the vertical compensation to keep the ball on the surface. Without the latter, the ball would go through the surface (just like, when hit on the right side, it jerks to the left side).

The standard analysis of the hitting force F is to decompose it into a pair of forces, one normal Fn to the surface and straight to the center of the ball (so: no torque) and the other Fs parallel to the surface (so : torque). You need to know the moment of inertia of the ball to calculate the spin (as accurately stated by Greglocock below, rolling being a special case of spin, and without friction we don't know if the ball will overspin, underspin, or spin exactly as needed to have it roll).

said otherwise, you solve Newton–Euler equations

Gem fr (talk) 09:02, 20 November 2019 (UTC)

I'm having trouble seeing this. How is such a force being applied horizontally to the top of the sphere? --jpgordon^{𝄢𝄆 𝄐𝄇} 03:53, 20 November 2019 (UTC)

I think this problem is analogous to pulling the string on a wound-up yo-yo that is stood up on edge. DMacks (talk) 05:01, 20 November 2019 (UTC)

Absent friction, we have a relaxed constraint - the linear velocity is not constrained to any specific relation with respect to the rotational velocity.

I think the OP's question is a variation on the theme of Example 7.8 (offset pulleys) or 7.9 (rolling disk) in Marion & Thornton's Classical Dynamics. One can solve these problems in many ways, but the easiest way is to write out the lagrangian. With friction present, we can relate the two velocities (linear and angular) with an equation of constraint, and solve by a relaxation method. Without friction, we can eliminate the constraint, and λ is trivially zero, yielding a trivially directly soluble system of equations.

If you're unfamiliar or unwilling to use these methods, you can try to directly compute the net torque and the net force. It ought to go without saying: when we suspend disbelief to permit the frictionless idealized system, we equally suspend disbelief to allow the set-up to impart an arbitrary force vector at an arbitrary location. How this is accomplished isn't important: we're dealing with idealizations. So, we can have a net torque due to the application of a single applied force that is applied to the object at a location offset from its center of mass. In other words, in this idealization, the ball can roll even without considering a second effective force due to friction. This problem, by construction, calls for a simplified and idealized model. If we want a more realistic model, then we're going to be solving a different problem.

Idealized models of balls will rotate any time they have angular velocity, which can be imparted by applying a net torque over time. Friction plays no specific role, except that friction could be a contributing factor to the net torque. Any other method of generating a net torque is equally acceptable, in this idealization.

Nimur (talk) 07:02, 20 November 2019 (UTC)

It'll spin rather than roll. The angular acceleration will be some proportion of the forward acceleration, which could be worked out by considering the moment of inertia of the sphere. This is known as English when playing billiards, I vaguely remember. Greglocock (talk) 07:50, 20 November 2019 (UTC)

English is side-spin. The question here is about top-spin (Glossary of cue sports terms). DMacks (talk) 07:51, 24 November 2019 (UTC)

• Any force can be broken down into two components, and for the sake of analysis, we can set up our reference frame in whatever way is most convenient, so for our purposes, you would set up your origin for your coordinate plane at the point of impact, with one axis pointed at the center of the ball (radial), and the other axis tangent to the sphere at the point of impact. You would then take the force vector and break it into component vectors along those two axes. Any part of the force that applies to the tangent will spin the ball, and part of the force that applies to the radial axis will move the ball. --Jayron32 13:55, 20 November 2019 (UTC)

The terms "hit" and "impact" that have been used in responses above suggest a scenario where something external transfers its Momentum to the ball but the question title states only that "A horizontal force is applied to the top of a stationary ball..". In Jayron32's reference frame we see a finite tangential component force and zero radial force. In a solid ball of a real-world material a shear wave spreads at the Speed of sound from the top. Answering what happens? is answering how the shear wave will dissipate. In a FDM analysis there is a distribution of momentums to the particles of the ball during the first time step. In subsequent time steps Spherical harmonic oscillations decay within the ball, constrained only by reflection of any radial extension at the ball-to-surface contact point and by the material's plasticity. To prolong the incidence of "horizontal force is applied to the top" is difficult because when the ball starts moving the location of its top changes, and we no longer have a "stationary ball". DroneB (talk) 18:40, 20 November 2019 (UTC)

but since the force is applied parallel to a line through the ball's center of mass ... Any line is parallel to some line through any given point (in a geometry that's not positively curved). —Tamfang (talk) 21:14, 22 November 2019 (UTC)

A late afterthought: a similar situation applies if you imagine a skater standing still on the ice, and someone else skates up, intending so glide past very close to her, but brushes against her left side, just hard enough to set her moving. Until she reacts, she's going to start rotating clockwise and also moving in something like the direction the second skater was going. Hope this helps. --76.69.116.4 (talk) 07:07, 24 November 2019 (UTC)

Fuel explosions

Why doesn’t diesel or petrol explode when it’s spilt around the place where it goes in even when the car is warm and it’s a hot day? Is it just because there’s not enough of it or because it’s not hot enough? 2A01:4C8:103A:8E11:30B0:E9BF:A9CA:5EAA (talk) 14:33, 20 November 2019 (UTC)

Petrol (gasoline) has a autoignition temperature of 247–280 °C, while diesel has an autoignition temperature of 210 °C. That would be a REALLY hot day, and if the weather were that hot, you've got greater things to worry about than spilled fuel. --Jayron32 14:45, 20 November 2019 (UTC)

Ok I see. So I guess a cigarette or lighter would be the biggest danger and hence why that’s banned in petrol stations.

You're also advised to ground yourself on your car's body before opening the gas cap, to help prevent any sort of spark once you begin fueling. ←Baseball Bugs ^{What's up, Doc?} carrots→ 15:29, 20 November 2019 (UTC)

a hot engine will have parts that are this hot, though. Which is why engine fires are not that infrequent, you just need some pipe leaking fuel at a wrong place. But a fire is not an explosion, you just need more than fire, you need first a fuel mix with air at a proper ratio, filling some space, then (and only then) an ignition. Vapor pressure is involved (I don't know enough to explain more), but the fact is this very seldom occurs (Not sure it can even occurs, in usual condition of enough air around: far less spectacular ignition will occur before) Gem fr (talk) 15:31, 20 November 2019 (UTC)

FYI it is not possible to light petrol with a cigarette. As a teenager we used to buy 2l (empty coke bottles) of petrol to start braais (don't try this at home) to the regret of many eyebrows. Despite our most foolish attempts no one was ever able to ignite the fire with a cigarette. Thanks Anton 81.131.40.58 (talk) 15:41, 20 November 2019 (UTC)

It's relatively safe to start a braai with petrol (but don't). How people get hurt is almost always when they start one without, then it doesn't burn fast enough, and then they pour petrol onto it. If you pour petrol onto an ignition source, you're pretty much guaranteed an explosion. You don't know where, but somewhere between the petrol and the braai, there will be just the right mixture to ignite. Andy Dingley (talk) 15:45, 20 November 2019 (UTC)

I think you (Andy Dingley) and I could get along just fine over a bottle of witblits and a braai. Anton 81.131.40.58 (talk) 16:12, 20 November 2019 (UTC)

What basis do you have, beyond personal experience, to assert that it's "not possible" to ignite gasoline with a burning cigarette. ←Baseball Bugs ^{What's up, Doc?} carrots→ 17:30, 20 November 2019 (UTC)

[1], [2], [3], [4]. The long-and-short of it is, under a highly controlled set-up, it is just barely possible to ignite a container or fuel tank of gasoline with a lit cigarette, and only slightly more possible to do so with an open puddle of gasoline, but basically, no. If you simply drop a lit cigarette into liquid gasoline, it does not ignite. --Jayron32 18:36, 20 November 2019 (UTC)

It is my experience, too, that a lit cigarette thrown into gasoline puddle does not ignite it, but I don't know how. By the color a cigarette burns at 700 degrees celsius. The fuel is there, there is plenty of oxygen in the air - and yet no ignition. Weird. אילן שמעוני (talk) 20:18, 20 November 2019 (UTC)

Not that weird. You need heat, not just temperature. A 700 C lit cigarette tip may have heat enough to rise the temperature of contacted gasoline by, say, 70 C or less (because the mass of gasoline is so much higher). And the geometry is bad, far worse than the geometry of a candle or oil lamp mesh. Gem fr (talk) 21:40, 20 November 2019 (UTC)

• Burning fuel needs an oxidiser to go with it - just oxygen in the air is enough. It won't burn or explode on its own, however you light it.

There are exceptions - things called monopropellants don't need a separate oxidiser. This also makes them quite hazardous, so they're really limited to just rocket propellants, and one called AVPIN which was used for starting jet engines. They also sold it as Thermolene for dragsters!

To ignite a fuel, you not only need to supply a heat source, so as to reach the ignition temperature - that's quite easy - but you also need to have a flammable mixture of fuel vapour and air. A small quantity of these fuels will easily reach the LEL (Lower explosive limit), where there's enough fuel to combust (if ignited). It needs to be warm enough to be above its flash point to do so. This is a commonly used term, nearly always used incorrectly - it's the lowest temperature at which it can make a flammable vapour. Not the point at which that flammable vapour will ignite (that's the autoignition temperature). The flash point for petrol (gasoline) is anything more than "cold day in Moscow", so it will light with an applied flame pretty much anywhere. However the flash point for diesel is more like "Summer day in Rome", so it's hard to set fire to diesel (a flame will do it, but a spark has trouble).

However as well as the LEL there's also a UEL or upper explosive limit (and a UFL or flammability limit too). This is the point when there's too much fuel vapour (and not enough oxygen) to make it either explode or burn. For petrol, this is easily reached inside the fuel tank. That's why some cars can have fuel pumps mounted inside the petrol tank (with sparks too) without exploding. Even inside a jerrycan of petrol with the cap off, it's over the limit.

A good example of this is backdraught in building fires. The closed room is full of flammable fumes, but doesn't burn until a door is opened, air rushes in and the mixture is diluted enough to drop below the UEL.

So if there's an accident and the tank splits, it doesn't explode. But if there are sparks nearby, or a hot exhaust catalyst, it can catch fire, and it will usually do so around the edges of the petrol spill, not from the centre. Diesel is vastly safer for that, which is why the military switched to using all-diesel some years ago. Andy Dingley (talk) 15:40, 20 November 2019 (UTC)

Here's a recent news story about a car fire triggered by a burning cigarette.[5] ←Baseball Bugs ^{What's up, Doc?} carrots→ 07:18, 22 November 2019 (UTC)

Carbon fiber tape for DIY heated clothing

I'm interested in DIYing some battery heated clothing using carbon fiber tape and this may be a stupid question but does it matter at all what carbon fiber tape I use as long as the resistance is right? One supplier said their tape is not suitable for this application but it was advertised as 100% carbon in which case I don't see a problem and I'm wondering whether they're just taking a precautionary stance or whether product selection is more complicated than I figured. There's a guy in Russia selling what he claims is suitable but I think the price is unreasonable so thought I could source it myself. Of course I want it thin enough to be able to sew. 3K and 240 gsm I suppose.

If you need a second opinion, it isn't suitable for this application.

While the question itself is not stupid, it is stupid to continue following through after experts and vendors tell you that your approach isn't suitable.

What you are proposing carries serious risks for fire, burning, electrocution, and other hazards.

Talk to more experts; consider taking some time to formally study textile engineering; and don't do something that can cause serious harm to yourself and to others. If you still don't "see the problem," you need to do some more homework before you go any farther.

Nimur (talk) 16:18, 20 November 2019 (UTC)

Per Nimur, electricity is ALWAYS more complicated than you will figure; it's why an electrician is so expensive to hire, and why they require years of training and licensure and carry lots of insurance in case they screw up. Resistance heaters (which is what you're basically building) are not the sort of thing you should homebrew, especially in clothing (which can catch fire) and your body (which can be burned if things go badly). Please don't. --Jayron32 16:28, 20 November 2019 (UTC)

• Yeah, super easy. Just web search. Personally I'd probably buy gloves, as heated inner gloves are cheap anyway and gloves are a PITA to sew. Most people make their own by taking a pair of either thin cotton gloves, or loose-weave Kevlar work gloves (for higher powers), lightly tacking the tape to the outside of those and then wearing them inside a loose-fitting pair of bike gloves. The tape needs to be the right stuff (conductive, not structural) but eBay's full of it. Generally don't try to heat any area which will be compressed - you don't want the tape to be getting wear and you don't want to have it pressed into your skin directly - a bit of distance through fabric avoids hot spots. About 5W-10W per glove is about right, depending on personal taste. For the 15mm tape, about 3-4 feet of tape is needed for each glove, depending on how far you wrap the fingers. Most designs go full length, from back to front of the finger, looping over the ends, but you might prefer to skip the last joint for more dexterity - depends how good your circulation is.

For this length of tape, 7.4V from a double LiPo pack is a reasonable supply, but (IMHO) it's worth adding a modern switched-mode power supply as a regulator. These are a few bucks now (eBay again), have voltage and current displays and are settable by pushbutton. Use a battery pack protector circuit (eBay again) and flat cell packs. 18650s are junk and their claimed ratings are all lies by a factor of 10×! Connectors are a PITA. Small ones break, big ones are lumpy. You don't want screw or latching contacts. I use the ubiquitous 5.5mm × 2.1mm barrel connectors. You have to decide if you use serial or parallel connection for your garments. Two matching gloves is easy, but gloves and jacket / trousers as well are a bit more awkward. Jackets probably need four cells and about 15V (or bike 12V), so you can run jacket, trousers, and a pair of series-connected gloves in parallel. Or wire the whole lot in series, go high voltage, and use a constant current circuit with a self-regulating output voltage.

As fractures used to be such a problem, I wouldn't use wire for heating for any of it any more. But I might use flex PCB element heaters in the back of gloves (I have good circulation and like dexterous fingers).

For joining wires to tape, some use silver-loaded paint. I just strip the (stranded) wire, fold it into a zig-zag, sandwich it between two layers of the tape and then oversew with metallic conductive thread (CPC).

Try it. It's all pretty easy. Andy Dingley (talk) 18:42, 20 November 2019 (UTC)

There are commercially available examples of electrically heated gloves, jackets, pants, socks and insoles that draw power from a motorcycle's 12 volt battery. Instead of DIY, seek a reputable supplier [6] [7] [8] and Caveat emptor. DroneB (talk) 19:11, 20 November 2019 (UTC)

Why not use flat resistors and avoid the entire problem? Flat resistors are cheap. It is a common practice among amateur astronomers to DIY heating for the secondary mirror in Newtonian telescope, and it works flawlessly. אילן שמעוני (talk) 20:25, 20 November 2019 (UTC)

Humans aren't flat. Also a resistor isn't long enough to go all the way round. As noted, flexible PCB resistive elements are easily available now and they'd do for backs of gloves or for kidney patches in a jacket. But otherwise, the flexible carbon tape is the way to go. Andy Dingley (talk) 20:39, 20 November 2019 (UTC)

Are you sure humans are not flat? Anyhow, the circumference of the mirror is not flat either. You use small enough resistors. The same approach, I think, will work with gloves. However an issue I didn't think before of is that the wires will be bent to and fro constantly. I don't think it will hold, so my suggestion goes down the drain either way. אילן שמעוני (talk) 20:52, 20 November 2019 (UTC)