#6 Torque VS. HP, and how its measured:
In the first article i wrote i went over what HP is, sort of...As I already mentioned, its very difficult to explain...particularly in the area of 'which is more important: torque or HP?'...i'll touch on that before i talk about ways to measure these 2 outputs...
If you read #1, you've seen the basic formula, and have the idea that all an engine makes is torque firmly in place...When it comes to an engine making 'twist', you basically want 2 things. You want to make that twist as 'strong' as possible, but you also want to make it at a high frequency...THAT is Horsepower...HP is simply how frequently you can get the engine to make torque. You can make a huge pile of torque at low rpm with a diesel truck engine, but you won't be making that torque very frequently...as the revs are low...where as a modern F1 engine will make literally motorcycle engine like torque down low, but can spin up to 18,000 rpm...where the 2.4L V-8 will be making nearly 800hp...The F1 engine can apply its torque many times more than the diesel engine in the same given time...
Luckily though, there is a slightly easeir way to think about what HP actually does for a car...HP is a measurement of torque application, as in how much torque is applied over a certain amount of time....where as torque is simply a measurement of force at a given instant...acceleration of the car is mostly influenced by the HP, NOT the torque...despite numerous myths...
Since HP is measure over time...its not effected by gear ratios or anything that create 'instant' benefits...acceleration of a car is effected by 2 things...the HP output of the engine (not just the peak HP, but the 'area under the curve')...and the mass of the car...you always hear about a power to weight ratio, but how often does someone say a torque to weight ratio?...almost never...as torque is multiplied by gears, and since those are very different car to car...its not a comparable measurement...
to make a car accelerate faster...you have 2 options...increase hp, or decrease mass...or in the best case...do both...
To measure a car's acutal HP, its actually VERY simple math wise...you only really need 2 things. You need to know the mass of the car as accurately as possible...and you need a accurate way to measure the acceleration over a specific amount of time....and to measure that acceleration, you only need to know the change in vehicle speed over a specific amount of time in seconds...if you have a 3000lb car, you can find its HP output, at the wheels (which will be the number after all parasitic loss is factored from the drivetrain), by measuring the amount of time it takes to accelerate from say 20 mph to 100 mph...the specific speed change doesn't matter...it can be 10 to 20mph, it can be 1 to 200 mph...the time in seconds will be different, but the derived HP measurement will stay the same...the only thing that is important is to have no wheel slip during the test, and no gear changes, as those will foul up the measurement..that is why most testing equipment measures output from a 'rolling start', and commence the test in only one gear...
RPM actually isn't important for this test...you can still find a peak HP number form the above without engine rpm, but rpm is needed to know exactly where peak hp occured within the engine's rev band...also, by knowing the engine rpm change through the test, you also can then reverse engineer the engines torque output as well.
As far as equipment to measure this stuff...its done by some form of dynameter. There is something called a brake dyno, which is used on engines outside of the vehicle, and does it completely differently than what i just explained...but i'll leave that alone for now, as its measured output isn't really relevant as many 'parts' of the car are not factored properly...an inertial dyno is what will give you the best illustration of your engine's output, as it is done with the vehicle in your actual car.
a industrial automotive dyno simplifies things a little, despite it being a very expensive machine...Instead of needing the exact vehicle weight; most dyno's will use a balanced 'roller'...of which its mass, diameter, and circumference are measured and known...it doesn't matter what those numbers are, they just need to be known...For practical reasons, most rollers are extremely heavy, in some cases more than the car being tested...for the best measurement of output, the longer the test takes...the better...if you had a 50 lb roller, even a weak engine could accelerate it so fast the test would be over in less than a second...making it too hard to take acceleration measurements of the roller at hundreds of different points.
so with the mass known, the diameter and circumference are needed to know its change in speed relative to the car's drive wheels turning it....if you know the start and stop speeds, you know the total acceleration, and if you measure the time it took to change that roller speed...you know the horsepower output of the engine...but in order to know peak horsepower, you need to measure that change in speed many many times...the point at which the speed change is happening at its maximum is where the engine makes its peak power.
that is all a dyno does...it measures how long it takes your car to accelerate a known mass over a specific amount of time...the faster it accelerates that known mass, the more power its making...The dyno techs will also use an adapter of some sort to get a engine speed reading for the test...by knowing engine speeds and correlating that with 'where' the engine made the most acceleration, they can plot a graph in engine rpm of what the acceleration curve looks like...they can also then find the torque curve with a calculator reversing the formula i showed in article #1. Its important to remember that this torque value given is misleading, because its torque from the crank AFTER parasitic loss is factored...its NOT actual torque at the wheels, because of gear ratios multiplying that force...your tires are actually twisting the roller with MORE force than that value no matter what gear you're in (the final drive will always make torque at the wheels higher than at the crank)...but it is a good way to see how efficient your drivetrain is, as long as the factory output at the crank is accurate...
But since we now know the measurements needed to accurately find hp at the wheels...do you really need a $750,000+ machine to do it? Actually, not really...
You can use Earth as a pretty good roller...and instead of trying to measure how much Earth rotates from the force of your engine, you can instead measure how much the car rolls on its surface...You need to know get an accurate measurement of your cars mass, though. You also need a few devices to measure acceleration accurately. This can be done by an accelerometer, a gps system, or in some cases both (the torque app i have on my phone does exactly this)...if those 2 variables are measured accurately, you'll get a hp number every bit as accurate as a real dyno...since you're measuring total acceleration, you won't be able to figure out torque or where peak hp occurs without incorporating engine speed into the test (torque does this, too)...and lastly, you need to do the test on something as flat as possible, because gravity will increase or decrease the acceleration time depending on going up or down hill...
overall...its pretty simple stuff...but its surprising that there are people that have had their car on a dyno many times, but still aren't sure of whats actually being measured.