Here is a longish explanation of what torque and power really are, and (I hope)
what they mean to the driving experience (or at least to performance specs,
which frankly are not important to me except in competition or bench racing)...
I invite questions and non-argumentative discussion. I will start by saying
that, though I have no actual car-design experience, I *do* have impressive
credentials concerning physics and its applications, and I generally know
what I am talking about on this stuff! :-) [I am setting myself up for a
few potshots, I know...]
I invite any of you involved with club newsletters to consider reprinting any
of this, and at your request, I will try to make it prettier.
So without further ado (or adon't):
--------------------------------------------------------------
Quoted from Scott F., with references to a query by Larry S.:
>> How do Horse Power and Torque differ?
>Horsepower is torque over time. That is, Torque is what an engine
>makes. Horsepower indicates how fast an engine makes torque.
Not quite. In fact, it is technically incorrect [sorry, Scott, but I can't
let this go]. It *is* qualitatively, descriptively correct, almost, sorta'.
Torque produces acceleration but is only half of the story. Power (in units
of horsepower or any other, it doesn't matter) is not torque over time, but
*work* (a.k.a. energy) over time. Work itself is torque times a "distance"
traversed, in this case a rotational "distance", i.e. an angle rotated.
>It [torque] can be measured on a chassis dyno, which is a kind of brake;
>the dyno measures how much force the engine applies to the dynamometer
>wheel. That is where the term "brake horsepower" comes from.
>Brake horsepower is a mathematical abstraction that
>represents how much and how well an engine produces torque.
Not to be picky, but the key word here is "brake". The "horsepower" part is
still to come, and as you say, is not measured directly. We could argue the
semantics of whether power is an abstraction, but it is more specific than
just "how well an engine produces torque". It is, as Scott said, the product
of (engine speed) x (the torque produced at that speed):
> BHP = (torque x rpm) / 5252
[ The divisor 5252 is there to convert units, presumably starting with torque
in ft-lb (or if you prefer, lb-ft), and producing horsepower, defined as
1 horsepower = 550 ft-lb / sec. I'll accept that the number is right. If
you don't have a dyno of your own, it doesn't matter anyway! ]
>For your purposes, ignore the difference between torque and horsepower.
>There isn't any. They're just different ways of expressing how the engine
>produces results... Most people who talk about "horsepower" and "torque"
>talk as if they're two different things, like downforce and camber. That's
>nonsense. They're two ways of quantifying how an engine produces results.
Quite incorrect, even for his purposes. While both may be a facet of "going
fast", they are very different, and a proper understanding of the issue
requires that we recognize the differences. The missing ingredient is the
gearbox. In fact, that is needed to answer Larry's real question:
>> I guess the question boils down to: What does an increase/decrease
>> in HP or Torque do to the 'seat-of-the-pants' feel of driving? How
>> will it effect the off-the-line speed, passing speed, etc?
Work is force x distance, in this case rotational. Power is work (or the
ability to *do* work) per time, or F x D / T. So if you have so much power
available, then in a given amount of time you can "push" hard for a short
distance, or push easy for a great distance, or any combination in between.
It is the pushing (i.e. torque) that causes acceleration, but the distance
the wheels spin during that time must match the car's speed. The gearbox
is just a tool to let you trade off speed for torque. In other words, you
can get all the torque you could ever wish to have (enough to spin the tires
and any more won't help) just by gearing lower, *BUT* you do so at the expense
of how fast you can go while producing that torque. You can get high torque
at a low speed or lower torque at a higher speed, but for a given power, the
product of the two is limited. That is the definition of power, after all.
You might think that it would be better to have both torque *and* speed be
higher, but it doesn't matter (!) since the gearbox lets you pick your own
tradeoff between them anyway (subject to the fact that you have a finite
number of gears to choose from, which suggests that a variable-ratio gearbox
might be nice, but that is another story, and so we settle for 4 of 5 or 6).
What this means in practice depends on the car and the demands. Let's consider
three situations: (1) standing-start acceleration; (2) lower-speed-but-moving
acceleration; and (3) top speed.
(1) Standing start. This is a trick question, thrown in to help me illustrate
a point. You could put a Mk4 GT-40 427 c.i. engine in one car, and a Briggs &
Stratton lawnmower engine in another, and get the same acceleration from both
if you geared the B&S engine low enough! (Actually there *is* a limitation,
i.e. when the inertial load of the engine equals that of the rest of the car,
but let's suppose that the B&S has no mass to speak of. See point 2 below.
Of course, it has no torque or power to speak of either, but this is all
hypothetical anyway!) The difference is that the B&S car might be able to
do it only until it has reach .02 mph and gone .3 feet! While you are
standing, your speed is 0, and that implies infinite torque. What matters
is what happens when your speed becomes non-zero, and for the B&S, that
breathing space runs out real fast. (Actually an internal combusiton engine
generate no torque or power at 0 rpm, but the clutch can fudge it for you.)
The moral here is that acceleration is the desired parameter, but it doesn't
mean doodley-squat if you can't do it at a useful speed. Getting good torque
(and hence acceleration) *at a useful speed* requires power.
(2) Moving acceleration. This is a key point. In principle, if two engines
produce the same maximum power, then they can be geared to produce the same
torque at the same at-the-wheels speed, so in principle they should give
the same performance. Several things get in the way however. The mass of the
car and the moment of inertia of all rotating parts after the gearbox would
present the same intertial load, but the flywheel and engine parts would not.
The high-torque, low-rpm engine would be spinning slower (actually the critical
factor is accelerating less), and thus would be a lower inertial load. Hence
it should give greater acceleration than an equal-power lower-torque engine
spinning faster. This is more critical in a lower gear when the engine speed
is higher for a given road speed, and when the wind drag load is lower (see
point 3). It is more critical too when the mass of the car is less compared
to the flywheel etc. Finally, unless you have a variable-ratio gearbox, you
are actually running right at max power at only as many road speeds as you have
forward gears, and the rest of the time you are off the peak. Thus it is
better to have a wider power band, or more precisely, a power band that covers
a wider road-speed range. This usually comes about with higher torque at lower
rpms. (When you tune an engine's power peak higher, as long as the power band
width increases in proportion to the rpm rise, the useable power range does not
suffer.) Thus for equal power, the engine that gets it via more torque instead
of more rpms is usually a better choice for both acceleration and driveability.
For these reasons, a slight tradeoff of less maximum power for lower-rpm torque
can be beneficial even though it really is *power*, not torque, (and actually
power per weight but we have been assuming equal weights) that you need.
(3) Top Speed. The power consumed by wind drag varies directly with the cube
of the velocity. (Drag force varies with v**2, but we want to consider power,
not force.) At low speed, drag doesn't matter as much to acceleration, but at
high speed (i.e. cruising), it subtracts lots of power. Whatever is left over
after that has been "used" pushing the wind is what is available to accelerate
you. Also, at the higher gearing of cruising, the engine acceleration is less
of an inertial load compared to the mass of the car. So acceleration at high
speed is considerably less affected by the rpm effects described in point 2.
In fact, near top speed, acceleration capacity decreases because the drag
increases, so you can approach your top end only asymptotically. And you can
really max-out only if the gear has been chosen so that the engine is running
right at power peak when the road-speed-drag power requirement equals the max
engine power. Whether the gear is chosen for max speed or not, max power and
higher-speed acceleration depend on the power available from the engine at the
rpm you are running at that time (as determined by the road speed and the gear
chosen). In this case, two equal-power engines will give the same performance
regardless of the rpms they have to use. The only difference will be whether
one has a broader power band than the other, letting you have more power even
when you aren't at peak.
Obviously the "best" engine will vary with application. With more gears, you
can afford a narrower power band, so it depends on what gearbox you have.
With a heavier flywheel/car ratio, you'd prefer lower-rpm, higher torque, at
least for lower-speed acceleration. If high speed is more important (as when
running the Mulsanne) then more power is probably the most important factor
for low lap times, though the tradeoff of useful power over a range of speeds
must still be considered. For brute-acceleration (1/4 mile and 0-60 times, or
for autocrosses) then lower rpms (actually higher running gears) is desirable,
so equal power but obtained at higher rpms is not as good.
A final question is what happens when we limit displacement? You can "always"
make a car faster by using a bigger engine (if you can accept the added weight
and fuel consumption). However racing classes (and in some cases, tax classes)
often limit engine displacement "arbitrarily", and the practicality of existing
blocks and other components often limit displacement pragmatically. Given
two engines with equivalently state-of-the-art breathing, torque is directly a
function of displacement, so with a limited displacement, you have a limited
torque. So to go faster and to get better acceleration, especially when top
speed is important as in a race, manufacturers typically have no choice but to
move the torque peak up to higher revs, getting more power. As long as the
fractional increase in rpm is more than any fractional loss of torque, the
power will go up. (Or at least the *maximum* power will go up. It is still
an issue of whether the *total* power curve will go up or down over the range
that must actually be used.) This is the reason our teeny sports cars usually
have high-reving, low torque engines. The same would apply for acceleration
and for a street car too, except for the dynamic effects of point 2 above.
For a given limited displacement, more power is obtainable only by pushing the
revs up. And more power is *always* what you really want until the effects of
point 2 above start to intercede in low-speed acceleration. It is up to the
manufacturer to decide the tradeoffs and engineer the entire car, engine,
tranny, weight, mpg, etc. accordingly. But this begs the question of whether
the marketing-analysts in the company want to sell it as a dashabout (Elan) or
a high-speed GT (300-ZX) or an economy commuter (Fiero), etc., and for what
manufacturiong costs (new gearbox?? new manifold castings?? ha!). Unless you
are building a dedicated machine, you can't divorce the engineering from the
marketing. And even if you could, you have to consider the application.
Which is more important, lower-speed acceleration and driveability or higher
top end? If you (the reader) choose the former, go back and re-read point 2
above. Your choice of torque over power (within limits, of course) is quite
valid, and that paragraph explains why.
An example:
Triumph made a GT-bodied Spitfire go 140 mph on the Mulsanne, but to do so,
they needed more power. They could have gotten that power from either more
torque or higher rpms. Given the constraint of the existing 1147cc engine,
they had a torque limit, so raising the rpms at which that torque was produced
was the only option. But do you suppose they could have sold a 140 mph version
of a Spitfire GT to the American public? No way, even if they could have met
muffler laws! So they dropped the 6-cyl into it instead. As a streetable car,
it made less power, but it was much more driveable and therefore sellable.
Jim Muller
|