I've always been confused about the concepts of torque &
horsepower, so did some research on google.
Torque and hp are not really intuitive to grasp, and here are some
of the easier-to-understand links I've found:
http://www.nissanperformancemag.com/november98/torque.shtml
http://www.allpar.com/eek/hp-vs-torque.html
http://vettenet.org/torquehp.html
Essentially, the articles come to the same conclusion as many of
you have mentioned - torque is responsible for acceleration (how
hard your head is thrown against the seat when you floor it), while
hp is a representation of how much torque your engine can produce
at high rpms - and would be more important for top speed.
They say that torque and hp are mathematically related by the
formula:
Hp (bhp) = Torque (lbs.ft) x rpm / 5252 (if torque is given in US
units) OR
Hp (bhp) = Torque (Nm) x rpm / 7023.5 (if torque is given in
Nm)
I think this means that the max torque & hp specs we read so
often are basically 2 snapshot views of an engine's torque curve,
at different rpms. The max torque shows the engine's peak torque,
which happens at a certain rpm level, while the max hp indicates
how much torque it can pump out at a higher rpm.
Eg. Car A has max torque of 145Nm at 4000rpms & max hp of
100bhp at 6500rpms. From the formula, when the car is generating
max hp at 6500rpms, it is pushing out 108Nm of
torque.
Car B has same max torque of 145Nm at 4000rpms but
higher max hp of 120bhp at 6500rpms. From the formula, Car B is
pushing out 130Nm at 6500rpms. So Car B can
accelerate harder at 6500rpms than Car A, although they accelerate
just as hard at 4000rpms.
The formula also means that all hp-torque charts will always
intersect at 7023.5rpm, IF the units are given in bhp and Nm, AND
the scales are the same, regardless of car, engine, tuning, TC or
NA.
So for the WRX Sti graph posted above, which shows 206kW = 280bhp
at 6000 rpm, the formula predicts that at 6000rpm, the torque
should be 328Nm. See where the 6000rpm line intersects the blue
torque curve and yup, its around 328Nm.
The articles go on to recommend that its better to upshift
at redline, rather than anywhere else based on the shape
of the curves.
What do you guys think of the analysis given in the links?