Daddy, what does torque mean?

November 18, 2005 11:23 AM

Is this explanation of torque, horsepower, and their relationship correct?

I've always had some difficulty really grasping these concepts. But this explanation works for me; I 'get' it. However, before I commit myself to Mr. Largiader's revelation of this particular component of the Universe, I'm going to ask for a reality check from the more physics-minded MeFiers.

posted by mojohand to travel & transportation (11 comments total)

Looks like the correct explaination of torque to me!
posted by orlin at 11:40 AM on November 18, 2005

Yes, that looks basically right.

Here's another description.
posted by bshort at 11:41 AM on November 18, 2005

Another thumbs-up here. (Well, the physics bits anyway -- I have no idea about anything to do with automocars.)
posted by chrismear at 12:49 PM on November 18, 2005

Two comments:

- It's a good illustration of how insane the ft-lb-s system is for engineering units.

- It doesn't explain the point of the power vs torque duality very well. Either of the follow-on links at the bottom of the article do a better job. It's all about how quickly the engine can move to the top of it's power curve to provide maximum acceleration. The more torque at low engine revolutions, the faster the engine can reach maximum power.
posted by bonehead at 12:55 PM on November 18, 2005

It looks ok to me as well.

The way I like to simplify it is that torque is like a measure of twisting force, absent of any time-scale. Think of it like this: Suppose you had a wrench in your hand and a bolt. You might be able to apply a constant 100 ft-lbs of torque to that bolt if you were allowed to turn it at a very slow speed, say one revolution every 10 seconds. However, if you were asked to apply that same amount of twisting force but at 500 rpm, there is no way you'd be able to humanly do that task. In both cases you're talking about the same amount of torque, but the rate of energy being expended is very different. It's just like saying, "I can lift a 100 pound package 5 feet off the floor if you give me a few seconds to do it, but there's no way I can do that 10 times a second." That's the difference between force and power - the rate at which energy is being used.

Another way of thinking about it is that of (torque, rpm, power) there are only 2 independent variables: given any two you can find the third. So a power-vs-rpm curve gives the same exact same set of information as a torque-vs-rpm curve, just presented in a different way.
posted by Rhomboid at 1:19 PM on November 18, 2005

I don't blame you for being confused mojohand. I have wasted hours trying to show (otherwise knowledgeable) gearheads that power curves and torque curves are equivalent. Apparantly the idea goes against the common shop wisdom.

Next time you have an engine's curves in front of you, squint your eyes a bit - the power curve is just the torque curve stretched increasingly upwards as you go right.
posted by Popular Ethics at 3:27 PM on November 18, 2005

Yes. Also Uncle Cecil takes a crack at it here.

On the off chance his helps out more than the other, better explanations above here's another wording of what basically everybody else has already said:

Torque is how hard the engine pushes the bike. Power is the product of that push and how fast the engine is turning. Torque peaks before horsepower, though the details of that relationship depend on specifics like e.g., valve timing.

"Pushing harder" means higher acceleration at that point in time (F = ma.) "Higher power" means higher ultimate speed (mostly because of air resistance,) but it doesn't tell you how quickly you reach that speed. Tractor-trailer engines are rated for surprisingly low horsepower -- in the neighborhood of 500 r.p.m. -- but that's because they're designed to operate at low r.p.m. ranges; these engines have enough torque to get a 40-ton rig moving. Try that with your 500 hp muscle car and you'll be visiting the garage pronto.

"Torque vs. power" is just another way of saying "torque vs. rpm." Two engines might have equivalent peak horsepower but if they peak at different rpm then the one that peaks at the lower rpm has higher peak torque, and will result in a faster time off the line all other things being equal. This of course assumes the absence of torque-enhancing devices that might not kick in until after a certain engine speed (e.g., controlled supercharger) but even then unless the engine is rev-limited its torque peak with the device engaged will occur at a lower rpm than its horsepower peak.

Usually when you talk about a "torquey" car (or bike, I guess) its torque is relatively high at lower rpms or it's geared lower so that it's easier to keep it in the higher-torque regions of the curve. Note that wheel size affects effective gear ratio -- a bike with larger-diameter wheels is effectively pushing a higher gear.

Exercise for the reader: why did old steam locomotives have different-sized wheels depending on the application? Here are your hints:

1) With very few exceptions, steam locomotives lack a gearbox; our engine has to turn faster to make the train go faster.

2) The peak of a steam engine's torque curve is at zero r.p.m.

3) The typical freight locomotive train was much, much heavier than the typical passenger train.
posted by Opposite George at 8:20 PM on November 18, 2005

er, "in the neigborhood of 500 r.p.m." s/b "500 hp"

What is with me tonight?
posted by Opposite George at 8:23 PM on November 18, 2005

I like to think of 'torque' as the 'effectiveness' of the engine. Since

Power = Torque * RPM * (some constant number, different if you use hp and ft-lb)

you can think of the torque curve as what gives the power curve it's shape. When the torque curve begins to curve down, the engine is getting less and less effective and despite running it faster and faster you will get diminishing returns on power output until finally it chokes.

An interesting comparison to make sense of torque is to check out the torque curve of an electric motor - it spikes way up at very low RPMs, and diminishes with speed. What this makes for is a flat power curve, where the engine can make the same power at 500rpm as it can at 3600. This is why electric dragsters can go so damn fast and why diesel-electric locomotives are the thing.

But then again I'm an engineer so this might not be the most straightforward way to look at it.
posted by anthill at 10:21 PM on November 18, 2005

It's just like saying, "I can lift a 100 pound package 5 feet off the floor if you give me a few seconds to do it, but there's no way I can do that 10 times a second." That's the difference between force and power

That's a great explanation.
posted by Civil_Disobedient at 11:19 PM on November 18, 2005

something people haven't really mentioned, but which makes this all more complicated in practice, is that with a gearbox you can get any torque you want (ignoring friction and other mechanical limitations).

it's like the ancient greek guy who said "give me a lever long enough (and a place to rest it) and i can move the earth". the tradeoff is that the more torque you want, the lower the rpm you need to drop to (the bigger the gearbox ratio).

so the description on that page is fine, but is only really concerned with the power and torque that an engine produces at the crankshaft.

if you connect the crankshaft to a gearbox you can trade off rpm for torque. so, from that point of view, power is the important thing, because, assuming you always have the right gearbox available for any particular output rpm, the more powerful engine will give higher torque.

so if you have a harley engine and a high revving high performance sportsbike engine, and you add a gearbox to the sportsbike engine so that it is running at peak power, but the output from the gearbox has a low rpm like the harley, then the sportsbike will produce more torque.

in practice, of course, that seems wrong. in fact it's not wrong, but it is only for a single rpm. in practice, you care about accelerating over a range of rpm, and that's where you hit problems with a high performance engine.

the trouble is, that when you scale down rpm, you also scale down the power band. if you have a 10,000:1 gearbox (for an extreme example) and an engine that gives peak power between 20,000 and 30,000 rpm, then you get lots of torque between 2 and 3 rpm (dividing by 10,000 for the gearbox), but outside that range the engine is not very efficient.

so in practice an engine designed to run at low rpm tends to give better performance at low speeds because its effective powerband (after dividing by the gearbox ratio) is wider. if you could produce an efficient, continuously variable ratio gearbox then a high revving, high peak-power engine would always win out, because it would sit there, happily screaming away at peak power rpm, whatever the actual speed of the bike.

(which is why when you hear a sportsbike accelerate away from a standing start they change gear so soon).

(disclaimer: i know more about physics than motorbikes; i thought the page was really good)
posted by andrew cooke at 4:33 AM on November 20, 2005