Question about Car Horsepower vs. Torque

In car engines, the specs are often listed as both horsepower and torque. Because I understand that the two are related, what are the real world applications?

Would a vehicle with high torque and low HP be more responsive than one with high HP and low torque?

Would an engine with a low RPM peak torque be more able to perform what daring feats traffic situations require, or would it’s higher RPM ranges merely be unuseful?

Is it true (as stated by Saab in an add) that horsepower is meaningless compared to torque when comparing engines?

Any answers folks?

Torque is what gets the car moving, the more torque you have, the faster you can get up and go, also you need lots of torque for towing. Horsepower is what maked the car go fast, the more you have the faster top speed you will have.

Most cars have a fairly close torque and horsepower rating,

Diesel trucks usually have more torque than horsepower, as they tow heavy crap.

Torque (lb-ft) is the actual output of the engine. Horsepower is derived from torque (1hp=550lb-ft/sec).

The torque rating itself is not so important as the rpm associated with it. A hypothetical car with a torque peak of 400 lb-ft @8500rpm, is worthless, but with the same torque @3000rpm the car will be a monster.

Also of considerable importance is the shape of the torque curve (torque vs rpm). A car with a high narrow torque curve will drive like a pig at rpms outside that narrow band, while a long, flat curve will make the car a whole lot of fun.

Incidentally, the hp and torque curves for a gasoline engine always cross at 5252 rpm, but I can’t remember why. Somebody will be along in a minute to explain it though.

Here’s a site that explains it much better.


Torque vs. Horsepower

There are a few other threads here that deal with this; this one might explain some things the best tho.

The “How Stuff Works” article is technically correct, but IMO too simplistic.

Overall - read what I wrote in the other thread about torque and horsepower.

And pmh? Read your post - you answered your own question about why the torque and horsepower curves cross at 5252 rpm (which they do for any reciprocating or rotary engine, not just gasoline or even IC ones).

And F1 drivers would argue very much about the uselessness of a car with 400 lb-ft or torque at 8500 rpm.

The SAAB ad is essentially correct, although again a journalistic oversimplification. Sigh.

That’s a little misleading. Torque is the output we usually measure, since that’s what a dynamometer measures, but horsepower is also an “actual output of the engine”.

Sometimes it’s easier to understand if you substitute “work” for “torque”. Torque is one kind of work. By the time the torque works its way through your drive train it is equivalent to how much force your tires exert against the ground. That’s why you squeal your tires when you rev the engine and pop the clutch: you’ve pushed the engine into a high-torque part of the curve and, with the low gearing, you’re delivering more force to the road than the rubber can handle. Not much power, but a lot of force.

That’s the reason your car has gears – to match the torque requirements and power requirements of driving. Just starting out you need a lot of torque but only a little power so you want to climb the torque curve quickly. Low gears do that. At highway speeds you need the peak torque when you are at or near, peak horsepower. Your top gear is usually designed to do just that.

The ads about having your peak horsepower outside the useful range are essentially correct. Most (European) race cars are designed to run at very high RPMs and in top gear they’re designed to go at speeds that are unattainable outside the autobahns. Driving your Ferarri on the freeway is like getting a drink of water from a firehose, but that’s not why you drive a car like that anyway, right?

FWIW, the throttle on your typical spark-ignited engine functions essentially like a torque control. (In contrast, the throttle on a diesel engine controls RPM

You get maximum acceleration at the torque peak. The higher up the rpm scale the torque peak is, the faster your rate of acceleration.

No, it is not. Torque has the same units as work, but it is not work. This is a dangerous thing to say. Work is, at it’s basics a Force applied over a Distance. Torque tricks a lot of young engineers because it has the same units as force times distance, but it is not work.

Um…care to explain this one for me?


Think of it this way, measure horsepower and torque at the drive wheels instead of the engine (To see what is actually happening.
Take a pick-up truck with an automatic transmission and chain it to a wall (One that won’t fall apart when you pull)
Lets disregard burning out.
Hitting the gas applies torque to the wheels, but since the wheels aren’t moving (0 rpm), acceleration is also zero.
In a car with a stick shift and a torque peak with, say 3000 rpm, in each gear you acelerate through the rpm range, you feel lateral G’s increase approaching 3000 rpm, then slacking off past that rpm.
Hey, it’s physics, it’s what happens.

OK, my fault, I didn’t ask a very exact question when I flippantly said “Why?”.

Why do you say that the “higher up the rpm scale the torque peak is, the faster your rate of acceleration”? There are too many other variables here to say this in this manner.

Okay, I can’t explain it any better. You need an engineer or someone who really knows classical physics, one equipped with the proper mathematical language, to really explain it better than me.

One can keep dissecting terms with “Why is that?” for only so long, before you get to a fundamental, irreducable law of nature which cannot be explained further, and must only be accepted as true, as is.

The truth IS…why is irrelevant.

So you are saying it is a fundamental truth that the “higher up the rpm scale the torque peak is, the faster your rate of acceleration”, and that is simply that?

I am an engineer, BTW, and do know classical physics, and am equipped with the proper mathematical language. Did you see my link I posted here, BTW? I talk quite a bit about this in there.

No, I haven’t got to the fundamental truth, yet, only to the limits of my incomplete knowledge

Stepping in where I know I wasn’t invited
Anthracite, I think enolan means something like; The higher up the rpm scale the torque peak is, the higher the speed at which maximum acceleration occurs.

That way, you’d feel gs of acceleration for larger sweeps of a tach needle. Which might feel sporty, but wouldn’t necesarily mean you’d win the drag race.

Short answer: Generally speaking, (not always, but generally)
when it comes to gasoline engines, horsepower measures how well an engine
retains its torque (which you can think of as acceleration ability) as
the engine speed increases. A high horsepower car will pull hard and keep
pulling hard for a long time when you step on the gas. A low HP car will
not pull as hard, and the pulling will fall off very rapidly as engine
speed rises.

As for Saab’s advertising, it’s true that a 3000 HP engine that
produced almost no torque would be worthless. However, gasoline engines
simply don’t work that way. Torque and HP come as a package deal in a
gasoline engine. More powerful engines produce more of both. Less powerful
engines produce less of both.

Certanly, having more torque at lower engine speeds is nice. Most
people drive at low engine speeds so as to minimize the noise produced
by the engine. With more torque at lower engine RPMs, the car will feel
more tractable and be easier to drive in traffic.
But there’s no such thing as a free lunch. More torque at low speeds
is obtained by stealing from torque at high speeds, or vice-versa. So a
car that’s great around town may be a dog on the highway, and vice versa.
The only way to increase torque at all speeds is to add more horsepower.
Long Answer: This is going to be a looooong and highly detailed
explanation. I hope you’re interested in this subject in more than just
a superficial way.

There are two issues we need to cover here - horsepower in general,
and horsepower as it applies to cars.

First of all, what is horsepower in general? Well, technically speaking,
horsepower is torque divided by RPM. The formula for calculating horsepower

HP = (torque * rotational_speed) / 5252

This formula only holds when torque is measured in ft-lbs and
rotational speed is measured in revolutions per minute. If you measure
them with some other units (say, newton-meters and radians per second)
then the 5252 correction number has to be changed to account for it.

Rotational Speed is probably pretty self explanatory, but you may not
have a mathematically rigorous understanding of torque.

Okay, so what’s torque? Torque is twisting force. You calculate it by
multiplying a force time a distance. For example, suppose you have a
bolt in the wall that you want to screw in. You also have a wrench that
is about a foot long. If you put the wrench on the bolt perfectly
horizontally, then weld a one-pound weight to the handle of the wrench
exactly one foot from the center of the bolt, then the wrench is exerting
one foot-pound (ft-lb) of twisting force, or torque, on the bolt.

Of course, as soon as the bolt starts turning, the wrench will turn
too, and some of the weight of the weight the handle won’t be working to
turn the bolt any more, it’ll be working to slip the wrench off the bolt.
At some point the wrench turns enough to slip completely off the bolt, and
falls on your foot. Ow!
But now think about hooking up an electric motor to the bolt. This
motor is able to constantly exert one ft-lb of torque on the bolt no matter
what angle the bolt is twisted around at. How many horsepower is that
electric motor producing?

Ha, gotcha, that’s a trick question! You can’t know how many HP the
motor is producing because I didn’t tell you how fast the bolt is
turning! Remember the HP formula - To calculate HP you need to know
both torque and RPM. So until I tell you that the electric motor turns
at, say, 60 RPM (one full turn every second), you can’t compute horsepower.

But since I did tell you the rotational speed, we can calculate horsepower.
Working the formula, we have:

      HP = ( 1 * 60 ) / 5252
      HP = about .01

To make an analogy to the physics concept of work, which is force
but corrected for distance, horsepower is torque, but corrected for
rotational speed.

With that basic understanding in place, let’s move on to how HP
and torque apply to cars…
The first thing you need to understand is that the horsepower numbers
published in car ads are what’s called “peak” horsepower. You remember
that in order to calculate HP, you need to know two things - one,
how fast the engine is spinning, and two, how much twisting force it
can provide at that speed. Now, suppose you had an engine that produced
500 horsepower between 9000 and 10,000 RPMs, but could only product 17
horsepower from 0-9000 RPMs. The average horsepower of the engine would
only be (.9 * 17 + .1 * 500) = 65 horsepower. This car would be very,
very slow. But do you know what they’d advertise the car as? Yup, 500
horsepower! (Fortunatly, real engines generally don’t work this way.)

Similiarly, the torque number quoted is also “peak” torque. That
is, if you measured the torque at all engine RPMs, and then took the
best one out of all of them, that’s what they advertise the torque as.

Because of the obvious flaws of depending on only a single peak
HP or torque value, more saavy car folks generally like to see a
graph that plots engine RPM and HP, or engine RPM and torque. Or
sometimes even both on the same graph. By looking at this graph, you
can get some idea how the engine behaves in general. For instance, if
the torque curve is very smooth and relatively flat, the engine has a
similiar amount of torque at all RPMs, and the “peak” torque value they
quote will be a much more realistic estimate of the average torque value
of the engine. If, on the other hand, the torque chart looks like a
mountain peak, the peak torque will be much higher than the average
torque, and the engine will pull very hard at certain speeds but be
much weaker at others.

HP charts are harder to read, because HP depends on both engine
RPM and torque. Generally, horsepower will rise as torque and RPM
do, and when torque starts to fall, so will HP, though less steeply.
Since HP is the product of torque and RPM, the HP curve will not
start to trend down until the torque falls off so badly that the
continually increasing RPM can’t make up for it.

These graphs are sometimes called “dyno charts” because the way
to make them is to hook your car up to a large machine called a
“dynomometer.” The dynomometer measures the torque and RPM at the
drive wheels of the car, and uses that information to calculate
the torque and HP graph.

You will notice that ads for high-performance cars (or motorcycles)
virtually never include these dyno charts. Apparently the manufacturers
are afraid their customers might discover that most performance engines
are tuned for high peak power at the cost of low-end power, instead of
less peak power but more consistent average power. To make matters
worse, it’s often the case that the peaks in the HP and torque
curves must come at very high RPMs. Meaning you have to rev the
living crap out of the engine all the time, which most people don’t
enjoy. Also read

However, having a relatively peaky torque curve isn’t necessarily a
bad thing. As long as the torque is reasonable through most of the
RPM range, the peak can be an exhilirating burst of acceleration,
and make the car fun to drive. The problem comes in when the height
of the peak is so extreme compared to the rest of the torque curve
that the car has no power except for just a second. The same is
generally true for HP curves. The steeper the HP curve, the more
exciting the car will be to drive, but the more the engine will
feel weak during lower-speed driving. Also, a peakier torque curve
will generally require a closer ratio transmission to keep the
engine running in that narrow RPM band where it produces good
power. This in turn will mean a lot of shifting. Again, this
can be thought of as “more fun.” Or it can just be annoying
depending on how you like to drive.

Lastly… remember the peak torque and HP numbers we talked about
earlier? Those are measured at the output shaft (crankshaft) of the
engine. But the engine isn’t hooked directly to the wheels. First
it goes through the transmission, and then in most cars it goes
through a differential, before it gets to the wheels. These
intermediate steps are important for two reasons.

First, the transmission trades RPM for torque. So when your engine
is spinning at, say, 1000 RPM, your wheels are only spinning at,
say, 500 RPM. However, because you’re trading RPM for torque, you
gain torque at the wheels by using a transmission! Different gears
trade different amounts. There is a “perfect” amount of torque at
the rear wheels. This is enough torque to make the car accelerate
as fast as possible without breaking the wheels loose. When the
wheels break loose, you actually get slightly slower acceleration.

Oddly enough, the transmission does not create or destroy horsepower.
If you think about this, it makes sense. Remember that HP is torque
times RPM. So if you only trade torque for RPM in equal amounts, the
product of torque and RPM remains constant.

Well, actually, I’m lying. The transmission doesn’t create HP, but
it does destroy a little bit. The gears of the transmission (and
in almost all cars, the gears in the differential) are swimming in
lubricating oil. The friction between gears and the drag of the
gears turning through oil wastes some of the engine’s energy. That
energy can’t go to the rear wheels. This “drivetrain loss” can be
significant. On most cars, anywhere from 10-20% of the engine’s
power is lost going through the transmission, driveshaft and
differential. This is significant. My Nissan 300ZX twin-turbo has
300 horsepower at the engine… but by the time it gets to the rear
wheels, I only have 260 horsepower left to actually make the car go.

But remember those HP numbers in the car ads? Those are measured right
at the engine, not at the wheels. So they’re a double scam. The advertisers
never publish rear-wheel HP numbers, not even peak numbers, because they’re
so much lower. Never mind that they’re the only numbers that really count.
You can have 9 million horsepower at the crank, but if only 17 makes it
to the rear wheels, your car will be dog-slow. Moral of this story, don’t
trust the car ads.

Phew… okay, I’m done. Correction and comments welcome, especially
from Anthracite who seems to know about this stuff too.


Well Ben, I don’t see any need to correct, and my only comment is that it explains it all pretty well.

Have you seen my link above, wherein I explain how the engine actually produces the power, and relate power, torque, bmep, and physical engine design? I’d be interested in knowing what you think about that as well.

*Have you seen my link above, wherein I explain how the
engine actually produces the power, and relate
power, torque, bmep, and physical engine design? I’d be
interested in knowing what you think about that as well.

Still haven’t gotten the hang of the quote feature here…

I went back and read it. It’s good reading. I’d never
heard of BMEP specifically before, but I’ve heard tuners
say “average cylinder pressure is what creates torque.”
Which actually makes a lot of sense, if you think about it.
It also nicely explains why a higher compression ratio
(either by stroke length or forced induction) will produce more torque.

As far as engine design stuff goes, the “long throw
crank, short rod” idea is a neat one. But in practice I
wonder if the vibrations caused by all that mass moving
around such large distances wouldn’t tear the engine up.
Really, speaking as a total engineering nazi, all
reciprocating mass has got to go. We should all be
running jet turbines in our cars, since they have
only rotating parts, not reciprocating ones. Until
then, I think someone once said that there’s a
happy medium ratio you can strike between stroke
length and piston diameter. I can’t remember what
it was. 1.6:1, maybe.

I was raised (mechanically speaking) on relatively small
displacement Japanese engines with high redlines, and
I much prefer sky-high revving, forced-induction engines
that just keep pulling and pulling to the more traditional
low-RPM, peaky, large displacement American philosophy.
But I’m quite aware that’s a matter of taste, so don’t
take my jabs at American cars and their engines too
seriously. ;]

Speaking of, let’s talk volumetric efficiency. The
first obvious step in having decent volumetric efficiency
is a good intake and exhaust system that provides as
little restriction as can reasonably be had. DOHC
head designs are a major component of this design
philosophy, and it’s a pity Detroit seems to so set on
keeping their head designs in the stone ages. I noticed
the words “side valve” applied to a V8 engine in the
owner’s manual in my grandparents Caddy. I almost
puked on the spot.

Unfortunatly at some point doing all the stuff
necessary to make the intake and exhaust paths as
efficient as possible hits a major case of decreasing
returns. It’s just no longer worth the expense to get
that last 1%. At this point forced induction is what
you have to do if you want to keep the engine
breathing well. I’m a big believer in forced induction,
as noted above. It’s a pity nobody’s come up with a
simple and cheap way to do forced induction in
gasoline engines. Turbos are tempermental and
expensive. Superchargers are too parasitic on the
engine to appeal to the engineer in me. This is
another reason I think we should all be running jet
turbines - they have built in forced induction.
(Unfortunatly it comes at the expense of supercharger-
like lag! Maybe a viscous clutch between the
compressor blades and the exhaust blades…)

Have you seen Wild stuff!

Well, a lot of very successful engines have gone by the “long throw crank/short rod” philosophy. Here is a table of some useful numbers (or not so useful):

Where all dimensions are in inches, and B/S is “Bore/Stroke ratio”, and R/S is “Rod/Stroke ratio”, and piston speed is the peak, in feet per minute, at 5000 rpm:

Engine       Bore   Stroke  Rod    B/S   R/S   Piston Speed
302 Chevy    4.00   3.00    5.70   1.33  1.90  2500
305 Chevy    3.736  3.48    5.70   1.07  1.64  2900
307 Chevy    3.875  3.25    5.70   1.19  1.75  2708
350 Chevy    4.00   3.48    5.70   1.15  1.64  2900
383 Chevy    4.03   3.75    5.565  1.07  1.48  3125
400 Chevy    4.125  3.75    5.565  1.10  1.48  3125
454 Chevy    4.25   4.00    6.135  1.06  1.53  3333
455 Pontiac  4.152  4.21    6.625  0.99  1.57  3508
455 Olds     4.125  4.25    6.735  0.97  1.58  3542
455 Buick    4.3125 3.90    6.60   1.11  1.69  3250
460 Ford     4.362  3.85    6.605  1.13  1.72  3208
440 Mopar    4.32   3.75    6.768  1.15  1.81  3125

(yes, I did type this myself, not C&P)

Note the 454 Chevy - it has some pretty good numbers for providing the best opportunity to modify for power, in this limited example here.

Eh…“parasitic” is often applied to superchargers, but the exhaust restriction of the turbo is also parasitic.