From: Alistair Bell <albell@uvic.ca>

   >
   >You would actually need all terms in the polynomial.
   >Contributiong to the force is:
   >0th order: static friction.
   >1st: surface friction in laminar flow, tire and bearing friction etc.
   >2nd: Turbulent flow.
   >The "fitted curve" on my graph is done this way.
   >

   Ahh, sorry, once again I jumped the gun! I have to admit that i didn't look
   at your "methods" on your webpage (writing this at home using a 10year old
   SE/30, good for mail but not so good at web browsing, that'll have to wait
   'tll i get to work). Anyway, I assumed that the drag was calculated from
   actual "road" measurements, not a calculated value.

The road test data is used to find the coefficients in the model.
The raw data was also plotted, but of course only gives a few
point values.

Here are the links if you forgot them:

A power curve:
http://www.cs.rochester.edu/u/jag/vw/engine/diesel1.9/Diesel1.9l_12_98.gif

Torque curve:
http://www.cs.yale.edu/~jag/vw/engine/diesel1.9/Diesel1.9lTorque_12_98.gif

Drag curve:
http://www.cs.yale.edu/~jag/vw/engine/diesel1.9/WestyDrag_12_98.gif

Tire friction is (somewhat) significant.
After some (a lot of) search you can find coefficients for this from
the major tire manufacturers. I posted some years ago, but forgot
the details. Think that the linear term is about 10kW or maybe 10 hp at
60 mph. Not dominating, but not insignificant either.

/Martin