Measuring Real-Time Horsepower In Vehicle
 

For years I’ve been contemplating how I might measure the torque of the engine in my truck. This information would enable me to calculate horsepower and fuel efficiency real-time. I’ve considered load cells and other deflection measuring schemes and they all have major drawbacks. Now I think I’ve come up with a simple scheme that’ seems like it should work. The idea is to use the rotational inertia of the flywheel as an indication of torque by measuring the acceleration of the flywheel during the power stroke and the deceleration of the flywheel between power strokes. A hall effect gear tooth sensor on the flywheel teeth would give me an RPM update at every tooth -- plenty of resolution to pick out the accel/decel of the flywheel. One concern I have is how the vehicle mass will affect the measurement as it is reflected back through the manual transmission. Any thoughts? Better ideas?

You only need to know your fuel consumption to give you a good idea.
I have seen live horsepower readings on a cummins engine analyzer.

That's true, if you know your efficiency. I want to physically measure the power so that I can verify engine mods and also to drive at the engine's peak.

ScanGauge II has an "X gauge" that displays horsepower. It may not work with all OBD II vehicles.

horsepower
 

The power could vary infinitely with no change at the flywheel.The information you'll be most interested in will occur at a constant velocity,rpm.
Most inertia and momentum is in the vehicle itself.Loads will vary with aero,rolling resistance,distance,grade,curves,wind,atmospheric conditions,road surface,etc..
In times past,a 'Tapley Meter' (torsional strain gauge) was inserted in the driveline,which would measure torque as a function of load.From this,knowing rpms,the horsepower could be derived.
Fuel mass consumed indicated mpg.

Measuring horsepower can also be done by accurately measure the acceleration of the vehicule with acceleration sensors.
With the weight Cr and the Cd.A value into the program a horsepower curve can be established.
A friend of mine made such a device which installed in the car can measure the acceleration every some many milliseconds.
Measurement should take place on a flat road .
I used to accelerate in first gear from zero to max rpm.
Probably those measurement boxes sell also in the States.

Thanks for the replies. Aerohead, if my power source were theoretically perfect and provided smooth torque, I'd agree that the power at the flywheel could vary infinitely and not be seen by this scheme. But in that case I'd have no need for a flywheel. In reality, my power source is a 4 cylinder engine with a power impulse every 180 degrees. As you know, the flywheel stores energy between power impulses and since the flywheel is a fixed mass I think I should be able to measure power by measuring the relatively imperceptible acceleration and deceleration of the flywheel itself. This is the same basic idea as the scheme that janvos39 mentioned, but applied to the flywheel. Varying loads will not be a problem and I'm pretty sure it would work perfectly on something like a boat with a fixed prop where the load looks purely resistive. But as you mention, almost all of the mass is in the vehicle itself and I'm unsure how this will affect it. My hunch is that it won't be a problem...