instantaneous hp

[49078]

I'm looking to get some opinions on a home project idea. We are wondering if it is possibile to measure instantaneous hp on a car.

Our current thoughts include using the drag force, mass, and friction to determine the total resistive force on the car. Then using the speed and engine revs calculating hp.

I have done a bit of research online and have not found much so I'm wondering if anyone has done this before. Any thoughts on our current idea?

We could get a dyno reading, have a computer find the engine speed and look up the hp from a programmed curve (found using a dyno) probably. Of course, we don't have a dyno so we're looking for our ways.

Cheers

 

[ivymike]

instantaneous hp on a driving car?

You could probably get within 10% percent by getting instantaneous fuel rate from the ECM and taking the ratio over peak HP rate (from a quick run down the highway), then multiplying by advertised peak HP.

 

[GregLocock]

Or by strain gauging the propshaft, or measuring the load in the engine mounts.

However, neither of these is especially easy, your approach of modelling the instantaneous performance of the car is pretty good, but you'll have to factor in gradients and wind, which will ruin the accuracy.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[JStephen]

You can surely come up with some sort of number, but I don't know how meaningful it would be.

Normally when people want engine power, they want just the engine power, not the net output at the wheels. So your not necessarily measuring the same thing other people are, so can't directly compare numbers.

I don't know how you'd accurately measure drag force or friction, either. You can measure the rate of deceleration with the car coasting, and get the sum of aerodynamic drag + friction, but you won't get an accurate number necessarily.

We did this for a lab in college (measuring drag by deceleration, that is). One problem is wind. You can do the test in two different directions to cancel out wind exactly parallel to path of travel, but can't account for effect of crosswind that way.

 

[GregLocock]

Oh, you can measure it properly, using wheel force transducers. At a million dollars a set I think they are not the right answer.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[MikeHalloran]

A dynamic on-car HP measurement system was offered commercially, in the late sixties I think. The price was pretty steep, like two grand when that was a lot of money.

The operating principle was simple enough; it put displacement transducers across the engine mounts. Because engine mounts are not linear, you'd need to lock the crank and apply a range of known torques, including large ones, to calibrate such a system. Today's microprocessors could take care of display and any calculations needed, but the transducers would still be a little pricey.

Accelerometer- based systems sold today are potentially as accurate, easier to install, and much cheaper.



Mike Halloran
Pembroke Pines, FL, USA

 

[MWPC]

49078,

Maybe I am missing something but several companies make accelerometers for cars. The ones made by G-tech have built in programs that calculate torque and hp based on the weight of the car and the engine rpm. They sense rpm through the electrical feed even on CI engines. You do have to add hp due to wind and drag but their web page has some approx. formulae. Hope this helps.

 

[GregLocock]

Now work out the error due to small gradients.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[ivymike]

Of course, instantaneous hp and peak hp are rarely the same thing... I thought the OP was looking for a way to measure output HP (engine, tire, or elsewhere) at any given time, including when the vehicle is traveling at a constant velocity (when accn-based techniques won't work).

 

[49078]

Hey guys,

I was looking for a way to measure the HP output. The friend of mine who i'm 'working' with on this actually has done up a pretty basic accelerometer. I'm not sure how good it -- or how bad it is -- but as Ivymike pointed out, we were worried it wouldn't work at constant speed.

Actually, what we want to use this the most for is for a long distance commute, on highways...with constant speed.

 

[MintJulep]

Dick Johnson developed a drag rake for use in flight testing aircraft. Google will learn you about that. Interesting challange to adapt to automotive.

Wind could be reasonably estimated by looking at the difference between the vehicle's speedometer and a pitot tube airspeed indicator.

Since I'm pitching aircraft stuff, grade could be estimated using the vehicle's odometer against a variometer.

Bearing drag is a function of speed and load. You should be able to get an approximation formula from one of the bearing companies.

Is there a real need for this, or is it just to see if you can?

 

[49078]

Just to see if we can.

 

[MintJulep]

Sounds like an interesting project. Keep us posted on your progress.

 

[carnage1]

could you measure the spring constant of the engine mounts as instaled and then datalog the position of the motor relative to the frame using some basic digital output gagues. (two cheap pair of calipers with holes drilled in the jaws for mounting, a tach, and a datalogger) with one caliper per side you should be able to write some basic code to get good data even with vibration.

 

[GregLocock]

That's sensible. Even better is to install the callipers as suggested, but measuring the roll of the engine, and then calibrate them for torque by appling a torque to the crank pulley. This will be more sensitive and does not rely on knowing the the possible non linear rates of the engine mounts. Bear in mind that you are measuring driveshaft torque, not engine.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[49078]

Carnage and Greg,

What sort of calipers are you referring to? Some sort of vernier caliper or am I overlooking something basic to the autoworld?

 

[49078]

One more method we have considered is essentially all electronic...

using engine speed, fuel consumption, maybe boost, and an efficiency table to figure out how much power the engine is producing

My problem with that is the efficiency table. Either constructing one or finding one. Besides that, i don't see how it wouldn't work.

Which reminds me, i should mention the car has a turbo which may make things even more difficult

 

[GregLocock]

Yes, digital callipers or linear pots. You'd actually be better off with pots in my opinion, some are designed for long lives. I doubt that digital callipers are.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[mattsooty]

What EMS system are you using? Any torque based system, such as Bosch ME7, should be able to tell you modelled instantaneous engine brake torque (obviously) and then its a hop-skip-and-jump away from instantaneous engine power.

MS

 

[awhittle]

I have heard that it is possible to put a vr pickup at each end of a driveshaft and install a 36-1 tooth ring gear at both ends. Then use a megasquirt (

http://www.msefi.com

) like blackbox to determine the degrees of twist in a driveshaft of a known strength. I have been out of engineering way to long to remember the math. This way at least you can get the torque that is making it thru the transmission and is being applied to the differential.

AW